Practical DIY or modest cost Vapor Phase - does it exist yet?

[rx8pilot]

I have perused the bowels of the internet and found a small handful of vapor phase DIY experiments. None of them seem practical, effective, or well thought out so far.

After looking at some of the entry-level and above commercial models, I have started to appreciate the challenge to some extent. The cost of the commercial units, even the entry-level, is rather steep. I would not, however, mind spending $1k on parts and building a small system to replace my current convection batch oven. Most of my work is mixed power and signals which is really challenging to reflow.

Any good DIY projects or functional entry-level options? My boards are all small - roughly the size of my hand or so.

Perhaps this is wishful thinking.

[tautech]

Some threads I remember:

https://www.eevblog.com/forum/projects/vapour-phase-soldering/
https://www.eevblog.com/forum/reviews/mini-blog-imdes-vapour-phase-purchase/

[Kjelt]

Most expensive part is the active cooler to keep the upper area cool.
You could use watercooling and shop for a chinese watercooler for lasers to get there.
Still stainless steel holder, elevator plateau, closing lid, watercooler, steppers and drivers, PSU, DIY electronic control, $1k DIY would be feasible but will cost you much more in time.

[spongle]

I've been thinking about getting into it myself the last couple weeks. I plan to start with a very basic system, no platform motion or PID heating, fancy recovery/condensation systems etc. I hope it can still be an improvement over my current butchered toaster oven. The main concerns I see are not exceeding the decomposition temperature (~280C) at the heating surface, and of course not creating an explosion with a sealed vessel.

I have found a source that supplies small quantities of Galden HT-230 at a reasonable price, although I've not confirmed yet they will sell to individuals: http://www.synquestlabs.com/product/id/52503.html

If it goes well I would be interested in collaborating on a more sophisticated system that could be produced as a kit/open hardware product at some point.

[SMTech]

Nothing fancy going on with this tiny one https://eleshop.eu/vapour-phase-mini-condens-it.html, looks a bit like a fryer .

[MagicSmoker]

Meh... How Hard Can It Be? At risk of angering the EE gods with that dreaded phrase, vapor phase reflow really doesn't seem like brain surgery to me. The only real tricky aspect I can see to constructing one is to dish the top of the lid upwards so that condensed vapor flows off the sides rather than straight onto the board. Another caveat is to vent the container through a reflux condenser to maximize the recycling of what is undoubtedly a very expensive solvent. Given the quality of work the OP has achieved in the past I don't see this being too much of a challenge for him.

Holy Frijoles - this stuff is expensive: https://www.appliedthermalfluids.com/home/shop/galden-hs-240-vapour-phase-fluid/

$1580 for 7kg?!?! You gotta do a lot of boards yourself to justify that kind of cost!

[spongle]

Did you check out my link? $75 for 250g.

Other threads have mentioned unavoidable losses around 1g per cycle, obviously more for a sloppy system.

[Kjelt]

 

Nothing fancy going on with this tiny one https://eleshop.eu/vapour-phase-mini-condens-it.html, looks a bit like a fryer .

Nothing fancy and your components can worst case be exposed to 230 degrees for over 4 minutes 
You should follow the reflow profile.

[MagicSmoker]

Did you check out my link? $75 for 250g.

Other threads have mentioned unavoidable losses around 1g per cycle, obviously more for a sloppy system.

I didn't before, but now I have and hey, they are located about 2-2.5 hours north of me here in Florida. Pretty good pricing, relatively speaking, and starting off at much smaller quantities, too. I don't think 250g will be quite enough for what the OP is after, but 1kg should be plenty (7kg would be downright excessive).

It's easy enough to reduce losses using a reflux condenser as I mentioned before and, besides, this stuff is apparently 10k worse than CO2 as a greenhouse gas (even high boiling point liquids like this give off some vapor at room temperature).

That said, Kjelt makes a pertinent observation about exposing the boards to reflow temperatures for too long. I'm not liking any of the ways I've come up with off the top of my head to get around this (basically involving putting the board in *after* the solvent has come up to temperature then removing it after all the paste has reflowed).

So, yeah, I guess it might be a fairly hard problem to solve and probably not worth the trouble.

[rx8pilot]

Nothing fancy going on with this tiny one https://eleshop.eu/vapour-phase-mini-condens-it.html, looks a bit like a fryer .

The size seems interesting, but I feel that having control over the PCB position is critical. I don't mind (at least initially) having a manual movement so that I can control the heat profile the PCB is exposed to without having to have some complex active cooling system.

Meh... How Hard Can It Be? At risk of angering the EE gods with that dreaded phrase, vapor phase reflow really doesn't seem like brain surgery to me.

In concept it is simple. From what I can see so far, the challenge is primarily mechanical since the system needs to be fairly well sealed under operating conditions and needs an elevator arrangement to control the thermal profile. The electrical system and PID is trivial in comparison. Getting PCB's in/out of the system while the galden is still hot needs a relatively complex/clever arrangement.

I like the idea of starting with some sort of existing commercial kitchen appliance that I could modify and construct an elevator. Something fairly tall would be nice. Since I have a CNC machine shop and a lot of motion control experience, it is doable. i would say that half of my motivation is necessity and the other half is doing something fun and unique. The goal is not specifically to be ultra-cheap or to make a business out of it.

[spongle]

Getting PCB's in/out of the system while the galden is still hot needs a relatively complex/clever arrangement.

I am thinking now about having auxiliary reservoirs for preheating/chilling the galden.

If coupled by steel flex hose, they could be raised/lowered by an external mechanism to transfer the liquid, or have fixed positions and use ball valves to dump the hot galden into the main chamber and drain it for cooling.

However, this might be unnecessary if enough cooling power can simply be applied to the fluid with a copper water coil or similar.

[OwO]

Is vapor phase really necessary at all or do you just simply need a fan in your oven 

[Kjelt]

Is vapor phase really necessary at all or do you just simply need a fan in your oven 

My opinion:

- an IR radiating oven ( that is the one I have) has a major disadvantage that it is direct heat to the parts, some parts mostly connectors don't like that at all and can melt.
- an conventional oven with (non IR) heating elements is too slow, it can not follow the reflow profile IMO.
  it takes too long to heat up to the desired temp and when there it takes way too long too cool down again.

So the pro's solve this to place multiple ovens on multiple temperatures behind eachother in a line so the reflow profile is matched.
Hobbieists can either go for IR oven and in case of connectors they have to shield them, or vapor phase.

[OwO]

Yes IR ovens are 💩.

I don't see why it isn't possible to simply use hot air in an oven to get fast heating and cooling, the air intake can be from inside the oven during heat up and outside during cooldown. Or simply an oven with a fan and vent port for cooldown. Either way it should be easier than building an airtight chamber and using dangerous and expensive fluids.

[spongle]

Actually, Galden is quite safe, you can even ingest it without harm. It's only a danger if you get it above decomposition temperature, certainly quite a lot better for you than breathing the exhaust fumes from a convection oven...

In addition to the intrinsic temperature limiting to prevent component damage, VP provides very even heating, excellent heat transfer for component with high thermal mass, and shields the board from oxygen (vapor displaces the air) for better solder joints.

[Kjelt]

I don't see why it isn't possible to simply use hot air in an oven to get fast heating and cooling, the air intake can be from inside the oven during heat up and outside during cooldown.

If you have a tiny oven that might succeed, if it is a normal size pizza oven you will not get an even temperature distribution unless you put a 8kW heater with large blower on it and then the airflow will blow off the components.

[helius]

Without heat control, vapor phase will not achieve good results. The vapor cloud does not have enough heat capacity to reflow the board; it is only a small heat reservoir between the liquid heater and the components. The reflow profile is programmed, just like with a convection heater.

[Koen]

I use a cheap asparagus cooker, a cheap hot plate and a small bottle of Galden. 40 boards a week. Easy, cheap, reliable, perfect not to melt connectors. I switch between asparagus cookers. One cools down, one heats up.

If I wanted to do more, I'd buy a big deep rectangular stainless container. The stuff in which massive kitchen cook vegetables and put it on the same cheap hot plate.

[spongle]

Do you do anything to calibrate the temperature of your hot plate? Or is it just "set and watch", and pull off the heat when you see reflow?

How are you suspending the boards above the galden, and how far above?

[Koen]

I stack five boards in the cooker. And use a small TM-902C thermometer to monitor it. It's all very easy and convenient.

There's a small hole in the asparagus cooker lid that fits the thermometer probe perfectly.

[rx8pilot]

I use a cheap asparagus cooker, a cheap hot plate and a small bottle of Galden. 40 boards a week. Easy, cheap, reliable, perfect not to melt connectors. I switch between asparagus cookers. One cools down, one heats up.

If I wanted to do more, I'd buy a big deep rectangular stainless container. The stuff in which massive kitchen cook vegetables and put it on the same cheap hot plate.

This the first level of experimenting I am expecting to do. Hoping to gain some knowledge of the process nuances by trial and error before I get any expensive parts.
Looking around for common items that can be repurposed for the heating chamber, lift, etc.

[mrpackethead]

Moving the boards is possible but you'll have to do it very gently.. Any bumps etc will result in displaced parts.

My experince, is that you can get good temp control with just heating up the galden with enough heat.   The IEMS i had did'tn have enough capacity to heat up quick enough.
Cooling is the more tricky thing.  However i've foudn running cold water through a some stainless steel pipe coiled in the bottom of my tank works pretty well,  trying to air cool the tank like the IEMS does' results in a long cool down period.. ( way too long )..      I just send the water to the drain,  you only need a few literes per cycle..

At the simplest however a pot on a hotplate is all you need.

[Kjelt]

Reading this you wonder why the pro machines have:

- Height controlled platform with multiple temperature sensors feedback
- Active cooling
- Galden recovery (vacuum/suction)
- Double lids, one on top of the entry/exit/cooling chamber and one closing of the heating/Galden chamber.

such a waiste of materials 

[SMTech]

I think little units from the likes of Roland Hecht (who are no longer?) Exmore & Asscon (some of which may be the same unit) use a built in ballast for the cooling. Incidentally the 1g/cycle number I have used previously came form a Roland Hecht daatasheet (or their sales representative).

Big machines are plumbed into the water supply.

[mrpackethead]

I think little units from the likes of Roland Hecht (who are no longer?) Exmore & Asscon (some of which may be the same unit) use a built in ballast for the cooling. Incidentally the 1g/cycle number I have used previously came form a Roland Hecht daatasheet (or their sales representative).

Big machines are plumbed into the water supply.

iv'e processed about 1200 panels ( approx A4 ) size  and used aprox 900ml of galden.   The galden does'nt pour out the top of the tank. It does'tn dissapear when its cold.  It is quite hydroscopic and the first cycle you do, quite often will produce quite a bit of steam which will try to escape.

[Kjelt]

iv'e processed about 1200 panels ( approx A4 ) size  and used aprox 900ml of galden.   

That is pretty good:

900ml = 1650g = €338.- so about 1,4g per board = €0,28 per board.

[nisma]

I'm passed from the tefal (similar to Clatronic FR 3587)  to Steba DF 300 fryer, the first allows 14x22cm boards, the latter A4 boards ( 20x30cm). Dimension are from memory.
Overal cost is small 150 to 200 Euro.

[mrpackethead]

While converting a deep fryer might intially seem a good idea, the immersion heater is a non starter.   If you are making frys, you have 3-4L of oil which submerges teh heater.   Using galden, you'll only have 2-3mm of galden in teh bottom of the tank.    and the heater wont' be in it. 

[nisma]

The heater and electric block is unused. It can be simple slided out. I use a heater similar to 3D printers on that i drop the galden using a 3d printed perisaltic pump.
On the outside of the inner container there is a soft copper tube used for the cooling water liquid (200/400ml) that i use for cooling down after process end.
 

[mrpackethead]

I did start with soft copper tube, because its really easy to work with.. However,   water + heat + pressure ends up resulting in quite signifincat errorision over time.. ( once temp gets over 80c )..   I see however your tube was on the outside of the tank.. I'd put mine on the inside of the tank.  I' have now replaed it with thin wall stainless steel.     Its harder to work/ costs more and is not thermally as conductive, but does'tn erode. so its pretty good from that respect.

Sorry i did'tn release you had actually built a system.. a few people have suggested using a deep fryer in the past ( complete with heater )..

This thread plus a couple of other discussions, and some practical experience has given me an idea for a very simple DIY grade system that could deliver good results.


INduction cooktop  like this ( https://www.breville.com/au/en/products/induction-hotplates/lic400.html )
Suitable saucepan to work with it, probalby something like a stock pot ), but needs somethign that has a decent sealing lid. ( ideally glass )

make a coil of stainless steel pipe to go into the bottom, and then bring the inlet / outlet out through a hole and then weld/braise it. That can just sit on teh bottom.    you'll need a couple of thermocouples..  and a way to control the waterflow.

I've seen teh induction cookers for as little as $50..       The heating side is easy.. The cooling side is a bit more tricky. 

[nisma]

I have tested the induction plate. It have 10 levels of power, I have used level 5.
Heat up time is fantastic, but as induction does packet modulation, the galden rises and sink 5cm depending if the plate transmit power or not. Level 1 is the same,  just the off interval is a lot longer. I think with 4 fans the cooling is enough if the vapour could climp higher up and water is not needed. Basically water is needed only if there is no lift and if the container z height is limited. Clearly if z height is limited, less ml of galden is needed to operate the reflow.

[mrpackethead]

My practical experience has been with a 'sealed'  ( has a lid ) static stainless steel chamber ( ie one with no active lift in/out for the pcb ) that air cooling was insufficent to get enough of a cool down gradient.         It was onoly getting me to 1.5C/sec which is just not enough ( really wanted at least 4 ).   With the water, it was really easy to cool it down nice and quick.

Some of the induction cooktops have very slow PWM rates, but are effectively continous control.   Idf you were getting a rise'sink of 5cm then somethigns clearly not running fast enough.     

[3roomlab]

anybody like to give your pencils a workout?

https://chemistry.stackexchange.com/questions/87904/thermophysical-properties-of-liquid-galden-ls-230-between-25-c-and-230-c

    boiling point: 230 °C
    density: 1820 kg/m³
    kinematic viscosity: 4.4 cSt = 4.4 × 10-6 m²/s
    vapor pressure: 453.3 Pa
    specific heat: 973 J/kg K
    heat of evaporation at boiling point: 63 kJ/kg
    thermal conductivity: 0.07 W/mK

1.8kg for 1L. assume the hobby pot is 1L. 25C to say 200C = 306kJ energy. so based on the curve, this 306kJ needs to be removed too in whatever the profile? in 5 minutes? some clever heat sunk hobby pot? wrapped when in use? uncovered when cooling down?

[nisma]

why you think of using 1L instead of 50ml or 100ml.

[mrpackethead]

as a starting point considering putting about about 3mm depth of galden in the bottom of your pot. Its all you need.

[3roomlab]

why you think of using 1L instead of 50ml or 100ml.

well hopefully it is easier to math out instead of 337ml or something

[nisma]

Theory for 100ml, the practice is different and involve setting time, heat transfer and loss from oven construction.
Ideally for setup, amplified pressure sensor with 12bit, better 14bit is needed in order to detect phase3 of vapour phase.
As you can see on the numbers below, if you multiply it with 10, the power requirement is a lot and this is the reason big VPH that can load 1 Lt of galden have 3.2Kwh heaters.
As Galden is a bit hygroscopic, the first heating to 150 degree is with reduced power in order to give the alcoholic and water contens from flux and from absorbed air the possibility to
escape. This assumes a Heating power of 1200W wile monitoring the main voltage using packet modulation as usual on heaters.

Step 1: Supply energy for heater with power 514 W in 43.3 second to increasing liquid temperature from 20°C to 150°C.
Step 2: Supply energy for heater with power 1028 W in 13 second to increasing liquid temperature from 150°C to 230°C.
Step 3: Supply energy for heater with power 112 W in 15 second to keep liquid temperature at 230°C
Step 4: Cooling liquid with speed 6.1 ml/second in 18 second to cooling hot iquid from 230°C to 120°C

This is the theory and pratice is that cooling must be doubled (thermal mass of oven heat it up again, repeated later two times for 8.2 sec) 
resulting in 270 sec for reflow and 530 sec for reaching 120 degree.
In theory, it is 71,3 sec for the reflow and 89,3 for the whole cycle.  If you make a simulation with thermal mass, thermal loss of oven construction, .... maybe you get the same
values as the reality. This cooling is fine because it is cheap and simple to do, but not for production, only for setup.

Is that what you had wanted or not ?
Basically it start at 20 deg. and need to go up to 230 deg. this are 210K temp. diff.
Now 100ml = 182 gr and , the specific heat is  177 J/K and specific heat for producing vapour is 114.66 J/ml, heating element is 1000W as assumption.
That means 230 deg. heated 100ml fluid can be evaporated with 1kwh heater in less then 11.5 seconds or 1kg in 63 seconds.
 210° * 177J = 37170J / 1000w = 37,17 sec for reaching 230 deg and then if you want evaporate 20.93ml you need 2.4 sec heating power of 1kw.

[rx8pilot]

As I was doodling some design ideas I considered the concept of moving the galden/heating element and leaving the PCB position static. The galden does not have to move as delicately as the PCB. The best part of that idea for me is the possibility of having a three-layer Pyrex window to observe the re-flow status at fairly close range. Only a single level of thermocouples would be needed at the PCB level plus one on the moving galden level. On paper, it seems like a simplified approach that reduces the complexity/precision of the mechanics.

The next consideration is a cooling system that is reasonably simple to execute. I have some ideas that don't seem too bad. Trying to get a drawing together but the basics are having a DI water reservoir (no fancy chemicals), a radiator, a pump, and a winding of copper coils thermally bonded to the main chamber. The main chamber would be wrapped in insulation and placed in a larger enclosure. Hoping this would be enough to overachieve on positive and negative temp gradients. So far - this in the territory of a single weekend to build depending on how many off the shelf parts I can identify and how much I have to custom machine/fabricate.

Still unsure how to maximize the galden recovery.

[mrpackethead]

Still unsure how to maximize the galden recovery.

where are you expecting to loose the galden?

[rx8pilot]

Still unsure how to maximize the galden recovery.

where are you expecting to loose the galden?

leaks, cracks, opening the top while still hot, etc....not really sure honestly. It just seems that the commercial designs bend over backward to get the galden back to the reservoir. Perhaps it is a non-issue.

[mrpackethead]

Still unsure how to maximize the galden recovery.

where are you expecting to loose the galden?

leaks, cracks, opening the top while still hot, etc....not really sure honestly. It just seems that the commercial designs bend over backward to get the galden back to the reservoir. Perhaps it is a non-issue.

its not a non-issue *if* you are not trying to keep the galden vapour hot and raising / lowering the pcb into it, so you can increase your throughput..   Hot galden vapour will move easily and you would loose it.    By leaving your pcb fixed in a closed container, and letting it cool down before you open it,  that issue really just dissapears and becomes a non issue.

[Koen]

Damn, try the thing with a pot and a hot plate first. I never bought more Galden. I don't feel like I'm losing any. Most condenses back in the pot.

I do filter it with a coffee filter from time to time though.

[mrpackethead]

Damn, try the thing with a pot and a hot plate first. I never bought more Galden. I don't feel like I'm losing any. Most condenses back in the pot.

yes, its a whole lot simpler than lots of people appear to be wanting to make the problem.    I'm curious though Koen,  how do you find your cool down goes?  I find that just letting it passively cool just takes too long

[rx8pilot]

  I'm curious though Koen,  how do you find your cool down goes?  I find that just letting it passively cool just takes too long

My expectation is that it needs forced/active cooling.
Also - my goal is not to do one PCB and call it a day. Hoping that it will be at least as fast as my current JEM-310 batch convection oven but offering a better final result.

[mrpackethead]

My expectation is that it needs forced/active cooling.
Also - my goal is not to do one PCB and call it a day. Hoping that it will be at least as fast as my current JEM-310 batch convection oven but offering a better final result.

My experience has been that it takes too long, in the stainless steel baine marie container i use, forced air, was just not enough too cool it down.      THats why i have experimented with thin pipe in the bottom carrying water for the cool down.   

My currnet project is to build three or four units side by side,  which will give me the capacity to keep up with the PNP line.

[3roomlab]

well here is quite a fun looking simulation
100ml of water mrpackethead?

(edit corrected error, to ml instead of grams of material)

I am guessing the pro machine must have flow controls
additional simulation, cool water rate of change @ 100ml per sec gives about 110C drop in about 9 sec.
correction, looks like only 25ml/s needed for 70C drop in 9s, much closer to the profile needs
but this is assuming the interface of water to galden is really really good
the scenario looks like the reverse of a capacitor welding machine but discharging heat !

[mrpackethead]

no, quite a lot of liters of water, running through it too cool it down.

[mrpackethead]

Been thinking about how to possibly DIY this today easiy and while i was wondering around briscoes today, i saw this.. Its an electric fry pan, its got a nice curved lid..

The galden cloud can be controlled well, and even though this has reletively low sides, it shoudl be quite possible to hold the galden in their when its hot.  Some of the soft vapour will hit the lid, but it will condense, roll down teh lid and back into the pan.          The frying pans heat control should not be too hard to hack.       Theres a vent hole in the lid, that would need to be closed, but its small.    Even if the main cloud reached the glass, it would just roll down.

A a few loops of stainless steel tube in the bottom for cooling water.   

A suitable frying pan should be obtainable for less than $100,  ( this is 2400W one, and was $89NZ ).. i could get an A4 page sitting on the bottom so it is big enough for me.   I think we probably need a coupel of thermocouples on this, possibly one attached to the base of the pan, and one a bit higher than PCB height.

Control should be easily done with something like an Adunio,     inputs would be a few push buttons,  the thermocouples..  output would be driving a Zero crossing Triac Driver, and a solinod for the water.    You might want to pump water,  so you get some variable control of flow..    Ii've just been running my water from the tap, throught he tube and into the drain.. Not very green

[3roomlab]

I was just making soup at the induction cooker
Id say an induction cooker + sealable pressure pot.
because the induction cooker has small thermal mass
so all the cooling is now on the pot
the heating profile of induction cookers are not bad in simulation.
after cooking the pcb, take the pot and dip the whole thing in a sink of water
but this is so uncontrolled temp drop
but I do not know how to calculate the pressure of hot galden soup
lets hope the pot leave no after taste for the chicken stew.
ok no, this cannot be happening haha

[3roomlab]

A suitable frying pan should be obtainable for less than $100,  ( this is 2400W one, and was $89NZ )

I had a peek at some pictures of the temp profile

and I think you hit a spot with that 2400w
the slow ramp according to simulation needs around 250w, then the hard ramp needs around a 2000w rate. 
this is looking very interesting
but the graph rate look like 100C in 3s, while the simulation is 100C in 10s. I do not suppose there is a 6kW fryer? 
edit : after compensating for 100g of PCB, the wattage to ramp to peak 225C is now not 2000W but 2700W. the PCB was not properly accounted for in previous simulations.
a 100g PCB is estimated at around 40-50 J/K extra
with the 2400w fryer, and same conditions, the peak temp is around 210.

[mrpackethead]

that is a very aggressive curve,  a ramp of of around 6C/sec  from 160 would suffice.  30C/sec would be way too fast i thik

[trevwhite]

This video might be of use. Does not seem to be a very complicated setup.


Trev

[nisma]

If you need short process times, don't cool down to 100 degree (120 deg. + open lid = near immediatly 100 degr. )
Instead use a gate and lift. 
As reference, the process chamber is 30x30 cm of this oven, usable pcb size is 25x25cm .
 

[Kjelt]

If you need short process times, don't cool down to 100 degree (120 deg. + open lid = near immediatly 100 degr. )
Instead use a gate and lift. 
As reference, the process chamber is 30x30 cm of this oven, usable pcb size is 25x25cm .

Yes that is how the pro companies do it, the three temp sensors in the picture are solely for the heating control of the Galden, the vapor should stay above level 1 and around level2, it should not reach level 3 (overtemp).
On the carrier, best on the pcbs there are thermocouples for the reflow temperature control which is a seperate control loop.

[nisma]

No, the vapour need to reach level3, as otherwise the pcb is not at temperature, ie it heat up. 

[Kjelt]

That is a different machine I saw then.
The one I saw regualtes the temperature of the Galden so the vapor at level2 is about 230.
The temperature at level 3 is much lower so the pcb can be lowered conform the reflow profile (60 seconds or more soaking at around 160).
So where will the pcb be placed in the soaking stage in your example?
It has to be below the gate but there the te perature is not allowed to be 230C.
So yes there will be some vapor gases there but not the thick 230C vapor for the final reflow stage.

[nisma]

You statement can be true. In reality one can select sensor 1 (for glue hardening or pcb baking), sensor 2, sensor 3 or sensor 4 (external) as heating termination and start of cooling.
The lift lower the pcb to the reflow height. The reflow height is between sensor 1 and sensor 2, more near sensor 1 .
If the pcb have a certain height, sensor 3 or eventually sensor 4 is needed, otherwise sensor 2 can be used.


 

[mrpackethead]

A suitable frying pan should be obtainable for less than $100,  ( this is 2400W one, and was $89NZ ).. i could get an A4 page sitting on the bottom so it is big enough for me.   I think we probably need a coupel of thermocouples on this, possibly one attached to the base of the pan, and one a bit higher than PCB height.

Control should be easily done with something like an Adunio,     inputs would be a few push buttons,  the thermocouples..  output would be driving a Zero crossing Triac Driver, and a solinod for the water.    You might want to pump water,  so you get some variable control of flow..    Ii've just been running my water from the tap, throught he tube and into the drain.. Not very green

So, i coud'nt help myself and i found a 'refurb' one of these for sale.    Its a breville BF560.   Stuck about 100ml of galden in teh bottom and heated it up.. ( with the lid on ).. I stuck a thermocouple probe through the lid hole and put some fire putty around it to seal it..     

Its got plenty of capacity to heat the galden. It boilded and created a vapour cloud..  I suspect that it willa ctually need more like 300mL    Some of the soft vapour did rise as high as the lid, but it condensates on teh glass, rolls down the lid, and back into the pan.  Its got a lip around the edge so it seals up well and any liquid drops back..    Without being absolutely certain it appears to run without looseing any galden.   I did this all with the frying pans own controller, but clearly thats not quite going to be whats needed to do the real job.

With it being a bit more shallow than your typical setup, leaving the lid on is going to be critical.      Its cool down is too slow and thats going to require some cooling.  I'm going to put about 3m of stainless steel 4mm tube in there and set it up for water cooling.   Possibly you might get away without this, but i'm not so happy with leaving my boards at elevated temp for exstended periods.

On the face of it, this certainly seems to have potential.   There are some saftey considerations,   Galden Vapour is very dangerous from a burns potential. You absolutley do not want to get in contact with it..  so, a bit of a lock out on the lid is going to be needed as well.

[Kean]

Its cool down is too slow and thats going to require some cooling.  I'm going to put about 3m of stainless steel 4mm tube in there and set it up for water cooling.   Possibly you might get away without this, but i'm not so happy with leaving my boards at elevated temp for exstended periods.

I would think you'd probably need to thermally bond a water cooling "jacket" to the outside of the pan.  Having something that sits inside would take up precious space, and if not bonded it wouldn't be very effective overcoming the thermal mass of the metal pan.

[3roomlab]

how do the factories get rid of the galden stuck on the pcb removed from the pot?

[Kjelt]

how do the factories get rid of the galden stuck on the pcb removed from the pot?

Probably the cooling fans blow it off 

[mrpackethead]

Its cool down is too slow and thats going to require some cooling.  I'm going to put about 3m of stainless steel 4mm tube in there and set it up for water cooling.   Possibly you might get away without this, but i'm not so happy with leaving my boards at elevated temp for exstended periods.

I would think you'd probably need to thermally bond a water cooling "jacket" to the outside of the pan.  Having something that sits inside would take up precious space, and if not bonded it wouldn't be very effective overcoming the thermal mass of the metal pan.

the tube is only 4mm high, and its not really an issue space wise.  Your board does sit on the bottom of the pan, you sit them on a frame that lifts it approx 15mm above the bottom.     My experience with using water cooling tube like this previously is that its very effective in getting the heat out of the galden quickly.   

[mrpackethead]

how do the factories get rid of the galden stuck on the pcb removed from the pot?

my practical experienec is that by the time the board has cooled down sufficently so you can pick it up, theres very little ( if any ) galden left on it.  I'm assuming that it probably evaporates off the board

[mrpackethead]

A quick test..   Electric frying pan at 100% power,  100mL of galden put in the pan.      It took about 115 seconds to get enough heat to create a vapour cloud about 60mm deep.     Too fast, but shows theres more than enough power to do the job.   By changing the power level from 'sear' to 'simmer' I was able to control the height of the cloud quite easily..    Obviously doing this by hand is not really a good plan, but a thermocouple at the right place, plus a small microcontroller can do this without hassle.

I did have a small amount of visisble vapour, get out of the pan, but this is very similar to the IMDES system and I am pretty confident that it is water that is being boiled off.  It only occurs at the start of the the heating.. The galden itself never gets anywhere near the edge.

I'll paste up a board and make a video later. I'm very confident that this will work and work well, with a bit of smart control.   As i have continously mentioned, MUCH care must be taken with Galden Vapour, it will burn you very badly.. If you are going to do this,  be very thoughtful about this.

I've now seen enough to know that this will work really well.,

[mrpackethead]

So, i thought i'd just try this out, and see how it goes. Found an old spare pcb and stuck some paste on it very quicky and very messily...    (* yes, i know, now i want to do it for real with a proper stencil printed board ).  (Its Henkel CG10 paste which is a very poor choice for hand application, it does not flow well out of a syringe, and it was cold

It reflowed nicely, with just 100mL of galden. 

these were only fesiblity checks to see if it is workable.   The next thing to do, is to put a control loop around the heater, with a thermocouple and control the heater.

 

[mrpackethead]

This is what happened with the fryer power at 100%.   I stopped heating at 390Seconds after i started.   There was 100ml of Galden in the tank.

The tip of the thermocouple was positioned so it was where the pcb would be.

[mrpackethead]

so,

(1) Its pretty clear that the ramp up is too fast, and it needs to be slowed down. This is done by controlling the amount of energy that goes in.

(2) There is a lot of energy left in the system once it reaches ~230C..  This is way too long for a sensible reflow and will need to be actively cooled.

[ChristopherN]

Has anyone tested using Teflon / Anti Stick coated pans? I have a very large board that I need to reflow and nothing to do it.

I did find this: https://shop.metro.de/Hendi-Partypfanne-239605-2810.html?c=185

It would be large enough and there is nothing too sensitive on the board so a slow ramp will not really hurt.

[mrpackethead]

I suspect that frying pan vapour phase is quite a new thing.  :-)   

I can't see any issue with using it. It shoud'tn get hot enough to come off. THe only thing i'd worry about is if the flux gets stuck on it.

[mrpackethead]

Clearly being able to control the power settings in this thing will be pretty crtical.   For some testing i'm going to use this DMX Controlled DImmer pack, its got some pretty big triacs, which will do the job nicely.

[ChristopherN]

Ok, I'll get one tonight and report how it goes. I have some temperature controllers laying around that can be controlled using RS485, so it should be easy enough to rig something together.

[mrpackethead]

I see that pan is only 6cm deep.  The one i've got is nearly 10cm deep and its only just deep enough.. THat might be a problme, as your galden might get up to the top very quickly.

[Kjelt]

In the old vapor phase topic I read that Galden also vaporates on room temperature.
They advise to store it in an airtight container after use.

[spongle]

Where did you end up sourcing your Galden?

[3roomlab]

I see that pan is only 6cm deep.  The one i've got is nearly 10cm deep and its only just deep enough.. THat might be a problme, as your galden might get up to the top very quickly.

you should totally try a soup pot, a deep 1

[mrpackethead]

some more experimenting, and some interesitng observatrions;

(a) the height of galden vapor above its liquid is can be controlled, ( once its vapourizing ), by controlling the amount of power you supply to the system..    With a PID control loop it appears that it is quite possible to maintain quite a tightly controlled vapor cloud height.   (+/-5mm is easy )

(b) putting an object above the galden liquid ( such a a pcb or a wire mesh ) largely 'stops' the vapour from rising further utill the object is nearly at the vapourisation temp.   

(c) the 'air' thats in teh 20mm above the vapour ranges from 120 to 160C.. Nearly perfect for the pre soak.

[rx8pilot]

some more experimenting, and some interesitng observatrions;

(a) the height of galden vapor above its liquid is can be controlled, ( once its vapourizing ), by controlling the amount of power you supply to the system..    With a PID control loop it appears that it is quite possible to maintain quite a tightly controlled vapor cloud height.   (+/-5mm is easy )

(b) putting an object above the galden liquid ( such a a pcb or a wire mesh ) largely 'stops' the vapour from rising further utill the object is nearly at the vapourisation temp.   

(c) the 'air' thats in teh 20mm above the vapour ranges from 120 to 160C.. Nearly perfect for the pre soak.

This is good information.
It seems that moving the Galden tray as a partial control for the temp profile is a better option than moving the PCB itself. A modestly tight PID of the galden temp and a simple motion system for the cloud could yield just about any profile.
For my larger scale experiment - I am looking at commercial oven windows so I can easily see what is happening from the side. So far, that does not seem all that difficult and should help get things dialed in quickly by having both thermocouples and visual cues.

[mrpackethead]

It seems that moving the Galden tray as a partial control for the temp profile is a better option than moving the PCB itself. A modestly tight PID of the galden temp and a simple motion system for the cloud could yield just about any profile.

I suspect that might be more complex than you expect.   If you 'dropped' the galden tray, you will still have 'energy' in the surrounding container, which will mean that the cloud probably will get wider for a little time, and then sink down and find a new point of equilibrium.   Like wise,  if you move it up, you'll intially compress the cloud and then the surrounding container reaches temp, it will rise..    It does not take much mass to cause the cloud to 'stall' while that mass comes to the vapor temp.     

All that said, its an interesting idea and one that might have some merit.   I'm pretty convinced that i wont' need to implmenet any moving parts,  and the system will be able to be kept exceptionally simple, and low cost.   The sacrifice might be in throughput.  Having said that though,  This is system will be so cheap, that scaling out will be a very viable option.

[/quote]
For my larger scale experiment - I am looking at commercial oven windows so I can easily see what is happening from the side. So far, that does not seem all that difficult and should help get things dialed in quickly by having both thermocouples and visual cues.
[/quote]

The galden vapour itself does not show itself up visibly that much..   Most of any visible cloud tends to be water vapour.

[ChristopherN]

The shallow pan I bought is too shallow. It heats up fine but the lid leaks a lot of vapor. It seems like big stainless vessels are kind of hard to find, I need something that can fit a 450*350mm board.

Any ideas?

[SMTech]

Probably not as easy to find online but cheesemaking & brewing equipment would be an interesting place to look. Lots of big stainless containers and the engineers/service people in those industries often have the fabrication skills to make it too. Hanging out at a few bankruptcy sales could also work...

[Kjelt]

trying to imagine a large pot, but what came to mind is the caterers thingy. made a little search, saw some chafing dish 22x14inch @ amazon. could be cheaper on aliexpress. 

The issue is to heat this thing to 230C without putting a heater in.
These things usually go to 90C to keep food warm, this can be easily done by heating the bottom. But bypassing the thermostat and temperature fuses and to heat the bottom to 230C could give problems, perhaps the whole outside will warm up over 150C for instance.

[mrpackethead]

The shallow pan I bought is too shallow. It heats up fine but the lid leaks a lot of vapor. It seems like big stainless vessels are kind of hard to find, I need something that can fit a 450*350mm board.

Any ideas?

If you can visibly see much  vapour its more than likely to be water vapor,..  I've run my 10cm deep pan now 12 time or so, and i've not lost any noticeable galden from it. 

[mrpackethead]

trying to imagine a large pot, but what came to mind is the caterers thingy. made a little search, saw some chafing dish 22x14inch @ amazon. could be cheaper on aliexpress. 

The issue is to heat this thing to 230C without putting a heater in.
These things usually go to 90C to keep food warm, this can be easily done by heating the bottom. But bypassing the thermostat and temperature fuses and to heat the bottom to 230C could give problems, perhaps the whole outside will warm up over 150C for instance.

This is why the frying pan is so good.

[mrpackethead]

I was able to achieve a much better curve with some PID based ( in python ) control of the heater, with a thermocouple.. I was able to slow down ( and stop the heat up ) to create a short soak..   This is certainly not an ideal profile curve yet, but i'm now confident that i can get there with some more iterations and tuning. 

In the mean time, i've decided the the quickest way to move this off a PC platform is to use an Ardunio.. Theres some good PID librarys for that as well.      Because i'm using a DMX512 lighting control pack to control the pan, i'll use a DMX sheild to control it.   For a more practical solution this just needs to be a phase controlled triac circuit.   But i'm using what i have, and I eventually think i'll run several of these side by side, and it has 12 channels, and its spread across 3 phases.

[mrpackethead]

Pile of bits and peices for hte frying pan!

Will start building this up now,  planning to make it run on a ardunio.

[rx8pilot]

Oooooh! This is awesome. Can't wait to see the follow up.

Typet purly on my fone.

[mrpackethead]

Aliexpress the home of cool useful low cost, sometimes dubious gear.   All came with valid descriptions though.

[pisoiu]

While reading various information about vapor phase soldering, I noticed that in almost all discussions/videos/examples, it is assumed that one board is lowered into vapor for soldering. But how about stacked boards/panels put one above each other in a supporting frame, for increased throughput? Is any reason not to do this, as long as vapors encompass the whole stack from top to bottom ? Thank you.

[mrpackethead]

I have stacked four boards high, and had good results. Its a good way of upping your through put significnatly.

The challenge is stacking them and not disturbing the components while they are still on unfryed paste.

[rx8pilot]

I have stacked four boards high, and had good results. Its a good way of upping your through put significnatly.

The challenge is stacking them and not disturbing the components while they are still on unfryed paste.

Did you have to fabricate some sort of fixture to accomplish this? I like the idea but only if I can easily and reliably stack them without bumping parts around.

[pisoiu]

I purchased some aluminium profiles, that kind used in frames of 3d printers. They can be easily assembled together and have some channels in which a panel could slide easily. I don't see why this would not work.
https://www.tme.eu/ro/en/details/k2020-i5_2m/aluminium-profiles/kraftberg/k2020-i5-2m/

I also got some galden from eleshop, I am thinking to make some tests with vacuum vapor phase soldering, to see if I can build something usable for small batches.
My starting point would be to use a stainless steel pressure cooker. It has a solid frame, sealed container, usually sloped lid to drain top condensed galden, the lid has ports where I can connect vacuum pump, etc. Also easy to heat with induction.
Problems so far is to find a big enough vessel for my panels. Biggest I was able to find so far has 36cm in diameter. I also have some concerns about the sealing ring, I'm not sure it will hold 230 degrees.

[mrpackethead]

I have stacked four boards high, and had good results. Its a good way of upping your through put significnatly.

The challenge is stacking them and not disturbing the components while they are still on unfryed paste.

Did you have to fabricate some sort of fixture to accomplish this? I like the idea but only if I can easily and reliably stack them without bumping parts around.

I just used some nuts as spacers. :-)    not exactly high tech, and  very prone to bumping, but it did work.

[Kjelt]

Again I look at the pro machines and see that none AFAICT stack the pcb's, why ?
My guess is that this is because the temperature differs with the height/distance in the mist.
For DIY projects this is perhaps not a problem, you probably keep heating till the heighest boards also reach the target 230C and reflow but by then the lower boards probably have been too long in the 230C region.
The pro machines have traceability of the reflow curve, they need this for their customers to proof they kept to the standards in case some boards prematurely fail due to heat stress or other causes.

[pisoiu]

That's a good observation. I imagine that could happen in the heating phase where I assume that vapors displaces the above atmosphere gradually, therefore different heights have different temperature profiles. Perhaps things are different when using vacuum. I have some galden on the road, I still have to find a closed container and a vacuum pump, I'll try to make some tests.

[KaneTW]

Be aware that the boiling point of liquids changes with pressure. Generally, the vacuum is pulled after the solder paste has melted (this will require a cold trap or similar to prevent Galden loss, and your pump will need to be able to hot process gasses).

[pisoiu]

My plan was to create vacuum at the beginning. I actually think to use boiling point variation vs pressure to my advantage, lower pressure means lower boiling point.
I make some assumptions here, maybe I'm wrong, but the timing of events would be:
-put boards, close container
-create vacuum, not very hard, what allows a cheap 1 stage pump
-if I start to heat galden, boards won't actually increase temperature until vapors reach their surface. Lower boiling point means galden will evaporate sooner than 230C, (boiling point at 1 bar). This should help with temperature gradient. Avoids situation when vapors touch boards only when they are very hot and cause sudden increase in temperature. During this process, pressure inside container should gradually increase due to presence of vapors. By controlling carefully quantity of liquid galden versus container volume, should be easy to calculate needed galden qty to obtain 1 bar when vapors reach 230C, in this situation, inside pressure is equal with outside pressure, therefore we have same boiling point. Am I wrong?
-stop heater, cooling by whatever means necessary to control cooling slope, creating vacuum again inside
-when temperature is low enough, slowly allow air inside until pressure equalizes, then open container.


le: thinking of it, I may need to create a system to insert galden into container only after vacuum is created, otherwise I may extract galden vapors during vacuum creation process...

[KaneTW]

Unless you keep pumping off vapor, the pressure in the tank will eventually increase due to evaporation of Galden. A 10 liter volume will be at atmospheric pressure after 250g of Galden have evaporated (via ideal gas law, with Galden having a molar mass of 1020g/mol). You'd need to run some simulations or calculations to determine the temperature gradient.

Galden's boiling point at 10mbar is 112C (http://www.trimen.pl/witek/calculators/wrzenie.html + datasheet values for Galden). However, it rises pretty quickly, with a BP of 207C at just half atmospheric pressure. That seems like too steep a gradient to me, so you'd have to continue to pump off and control the vacuum meticulously to ensure a good profile.

In most commercial machines, a profile is implemented by hovering the boards above the vapor front. I'm not sure if you could mix several different Galden types and get a layered system that way, though.

Some further reading re: profile:
http://www.torenko.com/pdf/Todays-Vapor-Phase-Soldering-Tech-Paper.pdf
https://ieeexplore.ieee.org/document/6102707

[pisoiu]

In most commercial machines, a profile is implemented by hovering the boards above the vapor front.

And probably this is why they do not stack boards.

[mrpackethead]

Again I look at the pro machines and see that none AFAICT stack the pcb's, why ?
My guess is that this is because the temperature differs with the height/distance in the mist.
For DIY projects this is perhaps not a problem, you probably keep heating till the heighest boards also reach the target 230C and reflow but by then the lower boards probably have been too long in the 230C region.
The pro machines have traceability of the reflow curve, they need this for their customers to proof they kept to the standards in case some boards prematurely fail due to heat stress or other causes.

My observation is that the delta T in the dense vapour is very small.   The height of the dense vapour is largely proportional to the amount of energy that you continue to supply once the liquid galden has reached boiling point.,   The vapour cloud will rise very quickly to its 'level',  definatly a short enough time that you can create a useful cloud of say 50mm within 3-4 seconds. 

[Kjelt]

Again I look at the pro machines and see that none AFAICT stack the pcb's, why ?
My guess is that this is because the temperature differs with the height/distance in the mist.
For DIY projects this is perhaps not a problem, you probably keep heating till the heighest boards also reach the target 230C and reflow but by then the lower boards probably have been too long in the 230C region.
The pro machines have traceability of the reflow curve, they need this for their customers to proof they kept to the standards in case some boards prematurely fail due to heat stress or other causes.

My observation is that the delta T in the dense vapour is very small.   The height of the dense vapour is largely proportional to the amount of energy that you continue to supply once the liquid galden has reached boiling point.,   The vapour cloud will rise very quickly to its 'level',  definatly a short enough time that you can create a useful cloud of say 50mm within 3-4 seconds.

So next experiment is to stack lets say three levels of boards attach a thermocouple to each board and log the temperature reflow curve from start to finish. Show the results. I can predict they are way off the ideal reflow curve esp the soaking time at 160C and the final 230C time will be exceeded for the lower board and not met with the upper board or any shift depending in the actual movement of the boards and rise of the mist.
It is a nice idea but not one that companies building VP machines since the 80s would not have thought off.
Instead they create the extra space in the existing 2D tray space.
Just my two cents, but its a science forum, do the experiments.

[pisoiu]

I can predict they are way off the ideal reflow curve esp the soaking time at 160C and the final 230C time will be exceeded for the lower board and not met with the upper board or any shift depending in the actual movement of the boards and rise of the mist.

I am quite sure this will happen as long as there is air above the vapors.

[mrpackethead]

The answer is that you get a step function in the height of the vapour, if you ahve stacked boards,  if you have a 'static' system ( no moving parts ). 

A quick test, The vapour rises to the first board, then 'stops' as the board absorbs energy and then the vapour will start riseing again to the next board and so on and so on.  So yes, the 'curve' would be different for each board.   The time differnece depends on how big ( and how much energy it takes to heat ) the boards are.   

If you had a dynamic system ( boards being dropped into vapor )  and you had a big enough cloud.

At the price of a frying pan and $30 of control stuff, if you want to double up production, buy another frying pan. The currnet problem to solve is cooling.    Its easy enough to heat up the boards to a nice controlled curve, and do it repeatably, and not loose lots of galden.     The cooling side is much more difficult.


 

[mairo]

...     The cooling side is much more difficult.

How do they do it on a commercial machines? I think I have seen that the board tray is moving up and down on commercial ovens, I guess they lift the board of the galden once cooled and introduce cooling from the top? Can't you introduce CO2 into the system for cooling. Many ovens have CO2/Nitrogen inlets for rapid cooling.

[Kjelt]

How do they do it on a commercial machines? I think I have seen that the board tray is moving up and down on commercial ovens, I guess they lift the board of the galden once cooled and introduce cooling from the top? Can't you introduce CO2 into the system for cooling. Many ovens have CO2/Nitrogen inlets for rapid cooling.

Depends on the design. I have seen machines with seperate chamber where the board is transported to cool down, so it is lifted and then moved sidewards to a seperate room,
or it is lifted and a mechanical "lid" is moved to seal of the galden chamber and on top there are pipes around the walls with actively (compressor) cooled water flowing trough them.
I have not seen active air (fans) or that sort, probably to prevent that components get blown out of position before the solderpaste has cooled down enough.

[mrpackethead]

The moment you start adding vaccum and or board transport everything is much more complicated by a factor of ten. While not impossible I do think it’s out of scope for a simple low cost machine.

I have been testing some cooling methods and after quite a bit of trial and error the best solution so far with my electric frying pan is simply a decent fan that blows a lot of air from under the pan. It’s very simple.  With this I hit 3.8c/sec peak temp decrease.   This doesn’t hit the target of 6.   

[KaneTW]

Buy an (a couple) aluminum watercooler heatsink(s) from Aliexpress or your local machine shop and watercool it? That's what my plan was for my vapor phase oven.

[suj]

The shallow pan I bought is too shallow. It heats up fine but the lid leaks a lot of vapor. It seems like big stainless vessels are kind of hard to find, I need something that can fit a 450*350mm board.

Any ideas?

If you have not yet found one, try to use the catering container according to the EN631 Gastronorm standard. They are available in the normalized dimension 2/1 (length x width x depth: 650x530x200 mm), made of thin stainless steel and also lid are available for them. Sometimes, the lid have a silicone gasket. You should fit in <100 Euro. These containers are used by manufacturers of ultrasonic cleaners and imdes probably used it also for his "invention".

[KaneTW]

Oh, that's a good idea. Nice.

[mrpackethead]

The moment you start adding vaccum and or board transport everything is much more complicated by a factor of ten. While not impossible I do think it’s out of scope for a simple low cost machine.

I have been testing some cooling methods and after quite a bit of trial and error the best solution so far with my electric frying pan is simply a decent fan that blows a lot of air from under the pan. It’s very simple.  With this I hit 3.8c/sec peak temp decrease.   This doesn’t hit the target of 6.   

Removing the cover off the base, gets me to 5.1    Some extra heatsinks stuck to the side of the pan and a bigger fan might be enough.  THat will make build extremely easy.  Converting kitchen appliances to soldering ovens is a thing.

[gasmeter]

Hi Gastronorm's do work.

You need the 200mm deep ones and don't fit a sealed lid its not needed or it will blow off.
I simply fitted a swagelok bulkhead fitting in the lid. Then a tube up into the air to condense .
You get a lot of steam at first.

Make sure the tube has a decent bore though.

You can cool it down just with some decent power fans and if you can water cool the lid.

I have gone through several iterations from the flask (which worked amazingly well). To induction heating   huge catering  size pots (trying to work out what to do with them now ).
For home use the Gastronorm is by far the best choice in my opinion.
I had a lot of success with these.

I'm moving in a different direction now but every prototype I made would solder anything I put in it.
I used a metal screened can as a pcb stand on (mins lid) early tests and it soldered the entire can too using the solder in the corner overlap.

Its fun to play with but after every prototype test I weighed my Galden back into the bottle .

I've lost about 13g so far in the last 8 months transferring it between different test vessels from a 900g bottle.

I have also experimented with various types of heater , induction, external, internal.
For an internal heater you can pad out the volume with ceramic beads.

Peter

[gasmeter]

I should add

It also put a metal plate on top of the gastro norm (Al) with a cutout for a bora silicate glass window.
Then I bonded silicone to this so when it sat on the tank, the weight of the metal panel compresses the silicon enough to seal.

Then its easy to run thermocouples etc out of the tank via the plate .
I made a PID controller and really it worked just fine .
I monitored the temp of the heater, the temp of the Galden, the temp over the pcb and the temp 50mm or so above the pcb.



Peter

[mrpackethead]

Have you got a profile?   

[jeremy]

Hi Gastronorm's do work.

You need the 200mm deep ones and don't fit a sealed lid its not needed or it will blow off.
I simply fitted a swagelok bulkhead fitting in the lid. Then a tube up into the air to condense .
You get a lot of steam at first.

Make sure the tube has a decent bore though.

You can cool it down just with some decent power fans and if you can water cool the lid.

I have gone through several iterations from the flask (which worked amazingly well). To induction heating   huge catering  size pots (trying to work out what to do with them now ).
For home use the Gastronorm is by far the best choice in my opinion.
I had a lot of success with these.

...

I have also experimented with various types of heater , induction, external, internal.
For an internal heater you can pad out the volume with ceramic beads.

Peter

What heater do you use with the gastronorm pans? What about heating the flask?

[mrpackethead]

The IEDMS uses a bay marie ( you call them gastronoms )..  for heating its got ceramic elements under the pan. It does lead to quite uneven heating, which has over time distorted the pan.  Its just too thin for extended heat/cool cycles i think..   

[mrpackethead]

A Few more tweaks and i've now got what i think is a very satisifactory curve with the frying pan.

I reflowed a couple of pcbs today and got very good results.   I'm not loosing any noticable galden.

The next part of this, will be to make a controller board. I've decided in order to make it as flexible as useful, i'm going to make a board that can be controlled via USB.   So, your options would be to control it with a PC, or a Raspberry PI or any other number of thingss...

[gasmeter]

Hi

I put cartridge heaters into an aluminium block, then stood it off about 1mm from the base with washers.
I tried heating from below .It wasn't easy the base wasn't flat enough to get good heat transfer but it did work.

The advantage of the AL block is you can also mount thermocouples on it.

Peter

[gasmeter]

Hi

I designed a PID controller and I was able to achieve any profile I wanted.
The heater was over 2KW so no shortage of power.

When I first started testing in a flask I bought a 3l one from chine that was as big as I could get.
The advantage of the flask is that you can see the vapour forming and rising.

I simply used one of the Chinese hot plates initially, before designing my own multi sensor pid heater controller.

Peter

[mrpackethead]

I've started a new thread to document the design and build for a "practical DIY modest Cost Vapor Phase" Frying pan project.  That post will be for the specific project


https://www.eevblog.com/forum/projects/diy-vapour-phase-frying-pan-project/

[mark03]

Came here from your post in Projects.

I wonder how well this would work in 120V countries?  Generally our heating appliances don't reach 2400W power levels.

[spongle]

Yeah I tried to find a US source for that fryer, no luck.

[mrpackethead]

Came here from your post in Projects.

I wonder how well this would work in 120V countries?  Generally our heating appliances don't reach 2400W power levels.

I had a quick look on Amazon, and i see what you mean.  I suspect that 1800W or so will more than suffice.  With 2400W,  its overpowered, and my gut instinct is that an 1800W frying pan would suffice.   I've been running mine at a maxium of 70% duty cycle.   ( https://www.breville.com/us/en/products/woks-skillets-deep-fryers/bef450.html ).   THat one has a stick free surface.. I can't see why that woudl be a problem, I've just not done it, so dont' know, but it appears to be the same shape/size as the one i have.
 

[KaneTW]

Galden might be incompatible with PTFE coatings (see their FAQ), so test before use. PTFE also begins to pyrolyse at 200C, so be careful. If you can only source nonstick pans, burn it off (in a well ventilated area!) by heating it to around 500C.

[mrpackethead]

Galden might be incompatible with PTFE coatings (see their FAQ), so test before use. PTFE also begins to pyrolyse at 200C, so be careful. If you can only source nonstick pans, burn it off (in a well ventilated area!) by heating it to around 500C.

There you go,  Probably not a good idea to use a non stick pan, and try to find a stainless steel one.

[mrpackethead]

Yeah I tried to find a US source for that fryer, no luck.

take a walk to walmart and see what you can find. :-) Let us know what you find.

[Towger]

take a walk to walmart and see what you can find. :-) Let us know what you find.

Walk to Walmart!  Yanks don't walk to Walmart, they go by car. [emoji43]  If seen walking with a couple of Wallmart bags in your hands you get strange looks from drivers going by.  A regular occurrence when staying with the mother inlaw in Texas.  A big Wallmart is just 6~7 minutes walk down the road.  You actually cannot walk into it without going the final 100m on the road, as the footpath ends!!

[gasmeter]

Hi

Well I have just ordered some custom heaters.
They are 2400W because in the UK we can use 13A or 16A feed , even 20A via the correct connector combination.

For 120V is either 1/2 the power and in fact if the heating system / insulation is designed well 1200W with no wasted energy is plenty.

My plan with the 2400W is just to speed things up and its over engineered.

My test systems using a Gastronorm used less than a Kw and it worked .
I didn't insulate anything and heat was being lost everywhere.. It just took a little longer.

Peter

[mrpackethead]

Hi

Well I have just ordered some custom heaters.
They are 2400W because in the UK we can use 13A or 16A feed , even 20A via the correct connector combination.

For 120V is either 1/2 the power and in fact if the heating system / insulation is designed well 1200W with no wasted energy is plenty.

My plan with the 2400W is just to speed things up and its over engineered.

My test systems using a Gastronorm used less than a Kw and it worked .
I didn't insulate anything and heat was being lost everywhere.. It just took a little longer.

Peter

Insulating it, could make your cool down really problematic,  You dont' want to be sitting at temp for too long.   I agree that having 1200W is probably enough

[gasmeter]

Hi

Cool-down is by water cooling .. So very fast.
Your right you can't use fan cooling when its insulated . Insulation though helps massively with the efficiency.

What I learned was that any system works be a tall pot or gastronorm, the issue comes in applying the heat efficiently and also efficient cool-down in order not to keep the board at elevated temperatures beyond the recommended profile time period.

Peter

[mrpackethead]

Hi

Cool-down is by water cooling .. So very fast.

You dont' want to cool down too fast either.  The cooling phase determines the grain structure of the solder joint. A fine grain structure provides the most reliable mechanical bond. To achieve this structure, a rapid cooling rate as the solder transitions from liquid (liquidus) to solid (solidus) is needed (the first ~50°C of cooling). HOWEVER The limiting factor for the maximum cooling rate is the stress that is exerted on the solder joint if the rate is too fast. This thermal stress, depending on the differences in CTE (coefficient of thermal expansion) of the joining surfaces, can fracture ortear the solder joint. The greater the difference in CTE of he joining materials and the cooling rate, the greater the thermal stress generated. A cooling rate of ~4°C/second is pretty typical.

Your right you can't use fan cooling when its insulated . Insulation though helps massively with the efficiency.

I'm really not to stressed about the efficiency of the system.   A very simple build is what i want to acehive.   Are you taking the container off the heater and then dunking in a water bath or similar?    My concern woudl be moving a board while the the solder is still molten, as it would be very easy to bump and then its all over!

[Smallsmt]

a rapid cooling rate as the solder transitions from liquid (liquidus) to solid (solidus) is needed

As I know, the first 20 degrees cooling should take place more slowly so that there is an ordered crystallization.
For this reason our reflow oven has a hold time in which the heating switches off but the fan for cooling does not start until a preset cooling temperature is reached.

[gasmeter]

Hi
My Mk1,2,3 sub release a - d (so many huge pots and gastronorm's left over from experiments ) anybody like a 100 Litre pot , or maybe a sweet little 60 Litre ?

On all my test builds I put the pcb on a static carrier and used forced air cooling on the outside of the tank / heater.
This did work but its very easy to spend too much time at a elevated temperature .

My latest build is meant to be for small business's like my own to use so it has pcb handling and it can lower and raise the pcb, so the pcb handling is more complicated and automated with safety interlocks etc.

The best system I saw for a home machine for raising and lowering is this excellent video / idea

https://youtu.be/8JRXbwYZmrU

Plus you get to practice your south African accent.

Amazingly only 5 likes on his genius video

Replace his can with a Gastronorm and Bob's your Uncle , Nellies your aunt !

Peter

[mrpackethead]

This did work but its very easy to spend too much time at a elevated temperature .

Once i removed the cowling off the bottom of the frying pan to expose it to more air flow, and put a decent fan on, i'm getting cooling of 6degC/sec which is fast enough..    I did try some experirments with water, which worked, but massively increased the complexity of the build.     

My latest build is meant to be for small business's like my own to use so it has pcb handling and it can lower and raise the pcb, so the pcb handling is more complicated and automated with safety interlocks etc.

my project really was about answering this question..  and i think i've found out that now a Practical and low cost DIY solution is avaialble.   However in the mean time, i'm pretty confident that im going to be able to scale it sideways so that i have three pans side by side, so i can use it for my low volume production.   

The best system I saw for a home machine for raising and lowering is this excellent video / idea

https://youtu.be/8JRXbwYZmrU

This video does demonstrate how simple it is to get resonable results.  He does mention that you can bump the board while its cooling and as your panels get bigger ( mine are A4 page size ) that is more and more of a issue.   And at the end of the video you see lots of tombstones.
I've been finding that its quite possible to get very good curves with a bit more effort with the controls, and that really sorts out tombstoning, Generally this is becuase you've got thermal inbalance between the ends..  ( one end goes liquid first ).     

[gasmeter]

a rapid cooling rate as the solder transitions from liquid (liquidus) to solid (solidus) is needed

As I know, the first 20 degrees cooling should take place more slowly so that there is an ordered crystallization.
For this reason our reflow oven has a hold time in which the heating switches off but the fan for cooling does not start until a preset cooling temperature is reached.

Thanks for this sage advice.
When my next system is built it will have a cooled inner lid section running continuously and a more severe directly coupled water cooler to the heating assembly for the Galden.
My plan untried as yet is

at the end of the cycle

1/ Shut off the Galden heater
2/ Raise the board at a controlled rate into the cooler region above the Galden and below the cooled plate suspended from the lid.
3/ Use water cooling to reduce the Galden temperature to the point where the lid can be opened and the board removed .

My plan is to reduce the Galden to say something around 100deg C rather than cooling it right down between cycles to speed up cycle time over multiple panels. Safety aspects of this have been thought through.

Peter

[mrpackethead]

For this reason our reflow oven has a hold time in which the heating switches off but the fan for cooling does not start until a preset cooling temperature is reached.

Thanks for this sage advice.

I have never seen anything that suggests that a 'slower' rate at the start of cooling is appropriate. In fact the opposite. The faster you can cool the better, and this is the first time i've ever heard this.    None of the datasheets for the solder i've seen suggest this.   I'd go and check this out before i made any more design decisions about it.     THe limiting factor for your solder cool down rate is the mechanical stress's, and most pastes will suggest cooling rates of between 4 and 10 C/sec

[KaneTW]

Cool down should be done quickly, but not too fast. I see ramp rates of rec. -4C/s to not faster than -10C/s

The cool down rate of the profile should be controlled within 4°C per second. In general, a faster cool down rate will result in a finer grain structure and a stronger and shinier solder joint. However, exceeding 4°C per second could result in thermal shock to the assembly.

https://aimsolder.com/technical-articles/reflow-profiling-time-above-liquidus

Would have to do some paper search later on for something more detailed.

[Kjelt]

Usually the max. ramp down speeds is aprox. twice the max. ramp up speed.
Example from WE says:
-  max ramp up     speed is 3^oC
-  max ramp down speed is 6^oC

[Smallsmt]

For this reason our reflow oven has a hold time in which the heating switches off but the fan for cooling does not start until a preset cooling temperature is reached.

Thanks for this sage advice.

I have never seen anything that suggests that a 'slower' rate at the start of cooling is appropriate. In fact the opposite. The faster you can cool the better, and this is the first time i've ever heard this.    None of the datasheets for the solder i've seen suggest this.   I'd go and check this out before i made any more design decisions about it.     THe limiting factor for your solder cool down rate is the mechanical stress's, and most pastes will suggest cooling rates of between 4 and 10 C/sec

It influences the crystallization of the tin, the solder joints look better and have a flat surface.
When the tin is solid, the cooling ramp starts, which triggers heating phases in our reflow oven if it cools too quickly.

[Smallsmt]

How about a galden chamber that can get closed when elevator moved the PCB up to speed up processing?

[KaneTW]

For this reason our reflow oven has a hold time in which the heating switches off but the fan for cooling does not start until a preset cooling temperature is reached.

Thanks for this sage advice.

I have never seen anything that suggests that a 'slower' rate at the start of cooling is appropriate. In fact the opposite. The faster you can cool the better, and this is the first time i've ever heard this.    None of the datasheets for the solder i've seen suggest this.   I'd go and check this out before i made any more design decisions about it.     THe limiting factor for your solder cool down rate is the mechanical stress's, and most pastes will suggest cooling rates of between 4 and 10 C/sec

It influences the crystallization of the tin, the solder joints look better and have a flat surface.
When the tin is solid, the cooling ramp starts, which triggers heating phases in our reflow oven if it cools too quickly.

I've never seen this recommended in a profile. While this sounds plausible, I'd like some references stating that the resulting structure is better.

[Smallsmt]

You can check the examples on our oven page https://www.smallsmt.biz/reflow-oven-hr-300/

I can do some photos using inspection system in the next days.

All of the photos show used this kind of cooling scheme. The pastes was GC10 a lead free paste.

[mairo]

You can check the examples on our oven page https://www.smallsmt.biz/reflow-oven-hr-300/

I think you need to add the max temperature that the oven can achieve on the main specs shown on the provided link. How is the cooling done? Is this your design / manufacturing, or is a re-branded Chinese oven?

[mairo]

Reading the metal plate at the back of the oven 'wenzhou small technologies" I guess, answers my question in regards to the manufacturer.

[Smallsmt]

I think you need to add the max temperature that the oven can achieve on the main specs shown on the provided link.
How is the cooling done? Is this your design / manufacturing, or is a re-branded Chinese oven?

MAX Temperature is 300C°. For constant temperature mode you should keep below 220C° max.
Cooling is done by exhaust vent and temperature rate controlled using heater.
Yes the oven is a modified re-branded oven.

Whenzhou Small Technologies is our chinese registered name for SMALLSMT in china you can't use a simple brand name for registration for chinese government.

[mrpackethead]

the project is moving ahead ( albeit slowly as its an 'extra' afterhours thing ). .

I have posted some python code and the json that describes the reflow curve.   While we are doing this specifically for VP, the controller coudl infact be used for lots of things.    I'll add the pid to the code, and improve the timing loop very shortly.

https://www.eevblog.com/forum/projects/diy-vapour-phase-frying-pan-project/25/

[gasmeter]

Hi All

I'm ordering some Galden LS230 for myself from europe (5Kg ) for myself in 2-3 weeks.
If enough people want smaller amounts I am happy to pad up my order and combine postage etc.

I don't know the landed price in the Uk as yet but it's better than the prices I was quoted from anywhere else.

If enough people are interested then I can work out a firm-ish price (subject to euro rate etc).

Let me know if interested in Kg you would like, then I will buy bottles and decant / weigh from the 5l container.

I just want to see if there  is any interest at the moment.


Peter

[mrpackethead]

For the frying pan, you only need about 80ml.    Thats about 130g.      250ml would last most hobby level folks for a long time, if you are careful with it.

[kony]

Virgin from manufacturer, or used and recycled?
Preliminarily I am interested in 1kg of fresh - but that depends on exact asking price.

[mrpackethead]

Virgin from manufacturer, or used and recycled?
Preliminarily I am interested in 1kg of fresh - but that depends on exact asking price.

I'd probalby not buy recycled, just becuase i dont' know whats been in it..  However, it does clean up very well.   I've cycled mine 100's of time.  I perodically run it through a coffee filter folded in quaters, through a funnel.    You get flux and other mis bits of crap in it, and the paper seems to clean it up pretty well..

[kony]

The process of recycling is not a problem - that you don't get assay of what the recycled material is in what boiling point fractions is. Hence why am I interested only in fresh one from manufacturer.

[mrpackethead]

It certainly is a mix.   The difference between the HS and LS grades is simply how tightly it was distilled.    I have not tryed to use the HS grade, but its quite a bit cheaper and it might actually be just fine.

[kony]

Any update on the possible group buy? I might be open to taking more than 1kg if the price is right.

[spongle]

A tip - I was able to get some "reprocessed" galden in 1kg quantity relatively inexpensively. Look for companies that do reclamation of vacuum oils. Was told the 5kg minimum size is a manfacturer mandate.

[spongle]

Anyone know if I should expect problems using vapor phase for boards with modules?

Will the high heat capacity and vapor defeat metal shields that would protect them during normal reflow?

[rx8pilot]

I finally escaped the grasps for my mechanical engineering responsibilities for a solid run back in EE. So, I can dig back into this topic. Very happy to see the conversation has provided some great information and interest. Now, more than ever, I need the benefits of Vapor Phase for the boards I am making and in some cases repairing.

First step from here is to procure some galden.

[spongle]

I got mine from https://inlandvacuum.com/

Just done a few trials so far with the vegetable steamer method, but it's worked nicely on some connectors that I kept frying with my toaster oven

[Koen]

I've been very happy with vapour and as the last message said, it's brilliant for plastic connectors.

Now, I've recently had issues where parts would jump out of alignment before soldering. Even a massive 7x7 QFN. Would anyone have experienced this ? Could my boards be too close to the liquid Galden ? Could my Galden be too old ? It's been two years. Could it be caused by water mixed in the Galden ? It's weird and I can't really confirm it but it seems to me that there is more liquid in the pot than before.

In the meantime, I'll filter it again tomorrow and see where that gets me.

Thank you ! Koen

[mrpackethead]

I've been very happy with vapour and as the last message said, it's brilliant for plastic connectors.

Its well worth testing though, before you load up an entire board of them.   Some plastics are awesome...   Some.. still melt. 

Now, I've recently had issues where parts would jump out of alignment before soldering. Even a massive 7x7 QFN. Would anyone have experienced this ? Could my boards be too close to the liquid Galden ? Could my Galden be too old ? It's been two years. Could it be caused by water mixed in the Galden ? It's weird and I can't really confirm it but it seems to me that there is more liquid in the pot than before.

That is not an issue i've expereinced.   THe galden i'm using is ~3 years old since i bought the big bottle of it.  I' just top mine up as needed. ( add 50ml or so not that often, i figure i 'loose' about 1 - 1.5mL per cycle.  ).   

Galden is hydroscopic, and so def you will have water in it.  I give my system a cycle before i use it... let the steam escape... it floats out the top, and the galden vapour happioly stays in the bottom.



In the meantime, I'll filter it again tomorrow and see where that gets me.

Thank you ! Koen
[/quote]

[rx8pilot]

Virgin from manufacturer, or used and recycled?
Preliminarily I am interested in 1kg of fresh - but that depends on exact asking price.

I am still looking for experimental amounts, but no luck yet in the USA. 7kg is about $1,000USD.

[spongle]

Did you see my message about https://inlandvacuum.com/ ?

You can also try http://www.synquestlabs.com/product/id/52503.html

[mrpackethead]

Am doing soem work on this today.. :-) Finally..

[rx8pilot]

Did you see my message about https://inlandvacuum.com/ ?

You can also try http://www.synquestlabs.com/product/id/52503.html

I totally missed that, thanks for the repost. Excellent options!

Typet purly on my fone.

[mrpackethead]

Just a bit of useful information for you.  Galden Gas is relatively heavy..  LS230, is about 24kg/m3.   COmpare this to air which is about .7kg/m3.    Thats why it stays in the frying pan if you dont' stir it up.

[Kjelt]

I heard that Galden also evaporates with roomtemperature, very little but enough to keep it in a closed environment.
Can anyone comment or share experiences ?

[mrpackethead]

I heard that Galden also evaporates with roomtemperature, very little but enough to keep it in a closed environment.
Can anyone comment or share experiences ?

I leave my frying pan with the lid on it, between use,  and dont' notice any appreciable loss.   If there is, its minimal.

[Kjelt]

That is good to hear, can you as an experiment when the liquid is roomtemperature, place cellophane (thin plastick) over the pans opening and then the lid on it, and check a week later if there is condensation on the inside ?

[mrpackethead]

That is good to hear, can you as an experiment when the liquid is roomtemperature, place cellophane (thin plastick) over the pans opening and then the lid on it, and check a week later if there is condensation on the inside ?

I could' but i'm not going to.  I have a glass lid on my pan, ( it is curved, so if there was any significiant moisture, it would roll back to the pan ).. and i've not seen any significant condenstation on it.   

The biggest issue is that the Galden is quite hydroscopic, and the first time you heat cycle it, after its been sitting for a while you will often end up driving a bit of steam out.. ( which is easy to confuse for it being 'galden vapour'.   If i've not used my pan for a few days, i'll give it a cycle first, the water is driven of.

[Lukas05]

Hi everyone,

I´ve been following this thread with great interest. Very useful information hidden in here 
It seems that some of you had a similar issue as myself. I´ve soldered my PCBs with my hot air station for many years now. The molten plastic base of SMD electrolytic capacitors where the ugly result 
For my current project this was no longer an option. Unfortunately there was no cheap solution at an acceptable price or quality (China chip heater haven't worked too well  )

This situation motivated me to work on my own vapor phase oven design. Since I´ve learnt a lot in this thread I thought I should share the progress of the project.
The construction is actually quiet simple:
An old stainless steel pot from a gastronomy auction is used as the process containment. I´ve used GPU watercooling blocks on each side of the container, to cool the inner walls at a height of about 15cm above the liquid. Two 120mm radiators keep the water cooling loop at about 5K above ambient. A perforated aluminum sheet is suspendet on four steel wires, wich are fed through a stiff tube to the actual "lift".
By varying the hight of the aluminum sheet it is possible to realise a solder profile.
The cover of the container ist liftet by rolling up some steel wire wich is connected to the suspension of the cover. This way I´am able to lift out the finised PCBs whithout running the risk of burning myself.

The first results don´t look too bad in my option. The solder stencil needs some tweaking considering the aperature size. Also the solder profile needs to be implemented properly for shiny solder joints.
This is just the current prototype... The wiring will be cleaned up when the controller board is finished. I´m also planning on some nice stainless steel sheets tho cover up the whole apparatus. Next step would be to implement the software and a fancy color lcd 

https://www.flickr.com/photos/136915492@N02/49030571321/in/dateposted/

https://www.flickr.com/photos/136915492@N02/49030571011/in/photostream/

https://www.flickr.com/photos/136915492@N02/49030789407/in/photostream/

https://www.flickr.com/photos/136915492@N02/49030789177/in/photostream/

https://youtu.be/Du0YXhiLl0w

[rx8pilot]

This is just the current prototype... The wiring will be cleaned up when the controller board is finished. I´m also planning on some nice stainless steel sheets tho cover up the whole apparatus. Next step would be to implement the software and a fancy color lcd 

That is awesome! 

[mrpackethead]

Ok, that is seriously cool.  Sorry lots of questions.

What is providing the heating?
How deep is that container, and is that the lid that came with it, or something else.
Whats pulling the steel wires?

Are you dropping the temp of the galden before you open the lid?

[Kjelt]

Nice build Lukas  Very impressive.

If you have to do serious production you could opt for an active watercooler that keeps the temperature at a set point. They are not that expensive ($300) mostly used for CO2 laser tube cooling and you can use carcoolant instead of water to keep it a closed system and prevent corrosion.

[Lukas05]

Hi all, I´m glad you like my approach to the problem so far 

What is providing the heating?

I´m using a 1100W heating element from a small tabletop oven. This form factor fits quiet nicely in the bottom of the container.
The big aluminum chunk in the middle is there to displace some of the galden. This way I only need 200ml of liquid wich benefits the BOM cost of the system 
The specific heat of aluminium is also lower than that of galden, wich means the big block doesn´t adversely effect the thermal impedance of the system. (excluding the limited surface and lack of convection)

How deep is that container, and is that the lid that came with it, or something else.

The container is about 35cm deep. The lid is the one that came with it. I just cut out the opening for the window, an added a gasket around the lid.

Whats pulling the steel wires?

The steel wires for the inner lift are attached to a spindle driven actuator.
Behind the spindle you can see the roll up mechanism for the lid lift.

Are you dropping the temp of the galden before you open the lid?

Yes, I drop the galden temperature to about 80°C. One design goal is to loose as little galden as possibile. I use four fans under the container that ramp up once the solder process is finished, and the inner lift is in it´s final position under the lid. Due to the relatively small combined mass of the galden an the aluminium block (and the high temperature difference) this is enough to cool the system down to 80°C in about 3-4 minutes.
Since the PCB inside just looses heat through convection (and a minor part due to radiation) it stays slightly above the gas temperature directly under the lid. This way the galden won´t condense on the PCB, and I get a dry board after soldering and cooling.

If you have to do serious production you could opt for an active watercooler that keeps the temperature at a set point. They are not that expensive ($300) mostly used for CO2 laser tube cooling and you can use carcoolant instead of water to keep it a closed system and prevent corrosion.

I´ve thought about active cooling. At this point though since I´m only using this for prototypes I don´t see the need to. But I will definitely switch the water out for something more corrosion inhibiting.
But even before that I have to get the brass pipe fitting out of the water tank. Copper alloy, aluminum water blocks, and plenty of electrolyte won´t go well for very long

[mrpackethead]

You've done somethign very clever there.

The container is really interesting, iv'e looked all over for a container that deep, and just can't find one.

[JPlocher]

Would something like a 20q/l roasting oven be a useful starting place?    At $US 50, they won't break the bank...

example: https://www.walmart.com/ip/Hamilton-Beach-28-lb-Turkey-Roaster-Oven-Model-32231/54056097

  -John

[ignilux]

Damnit you guys, the last thing I needed was another cool thing to spend money on. And yet... I must have one!

[Kjelt]

Would something like a 20q/l roasting oven be a useful starting place? 

too shallow unless you are willing to loose Galden, which is much more expensive than that oven.

[mrpackethead]

Would something like a 20q/l roasting oven be a useful starting place? 

too shallow unless you are willing to loose Galden, which is much more expensive than that oven.

Actually, maybe not.  My work with using the frying pan says that you can have a shallow pan.  The galden is actually reletively contronable.   that is not the problem that you solve with using a deep container.    Using the deep container lets you build MUCH better profiles, becuase you can build the elevator platform. That is the game changer.   I had avoided it, because I could not come up with a mechanically simple way to build it.  What Lukas05 has done is come up with a mechanically simple way to make it work. That is the game changer.

I scoured Aliexpress looking for a deep Dish, and have failed. :-(

[pisoiu]

I scoured Aliexpress looking for a deep Dish, and have failed. :-(

Same here with aliexpress but there may be another way. I made some tests with that kind of rectangular containers which are used to serve icecream and food in malls. Don't know what they are called. But that format has maximum 200mm depth. I wanted deeper and I was able to find at a local dealer of stainless steel supplies. Among other things (tubes, bars, various containers), they sell containers for dish washing in restaurants. I find those suitable, they are quite big, and they have another feature, the bottom is slightly sloped, which in my opinion is a good thing to use the deepest part for heater.
Check this pdf: http://www.italinox.ro/images/pdf/gastronorm.pdf , at page 4 and 5 you will find them. Texts are in Romanian but images and drawings are self explanatory. Walls are 1mm thick and 330mm depth seem ok for this application.

[tautech]

I scoured Aliexpress looking for a deep Dish, and have failed. :-(

Really ? 

Plenty here:
https://www.aliexpress.com/wholesale?spm=a2g0o.productlist.0.0.44464cbeo8t7Ji&SearchText=stainless%20steel%20pot%20with%20lid

275L one big enough ?
https://www.aliexpress.com/item/4000070766006.html?spm=a2g0o.productlist.0.0.44464cbeo8t7Ji&algo_pvid=b151105f-39f8-4a2d-8913-e8c848d8aecb&algo_expid=b151105f-39f8-4a2d-8913-e8c848d8aecb-4&btsid=95242914-9696-4d11-9c60-e6e31836fb72&ws_ab_test=searchweb0_0,searchweb201602_9,searchweb201603_52

[Kjelt]

rectangular containers which are used to serve icecream and food in malls. Don't know what they are called. But that format has maximum 200mm depth.

those are standard for the horeca and called Gastronorm.

You can use it as the base that contains the fluid and get a stainless steel plate chassis welded upon it as high as you want.

[mrpackethead]

I scoured Aliexpress looking for a deep Dish, and have failed. :-(

Really ? 

Dish as opposed to Pot.  Being round, it makes it difficult to attached flat plate water coolers ( as used for GPU's etc ) which are readily avaialble and cost effective.     It seems that Gastronorms max out at 200mm deep.

[mrpackethead]

I scoured Aliexpress looking for a deep Dish, and have failed. :-(

Same here with aliexpress but there may be another way. I made some tests with that kind of rectangular containers which are used to serve icecream and food in malls. Don't know what they are called. But that format has maximum 200mm depth. I wanted deeper and I was able to find at a local dealer of stainless steel supplies. Among other things (tubes, bars, various containers), they sell containers for dish washing in restaurants. I find those suitable, they are quite big, and they have another feature, the bottom is slightly sloped, which in my opinion is a good thing to use the deepest part for heater.
Check this pdf: http://www.italinox.ro/images/pdf/gastronorm.pdf , at page 4 and 5 you will find them. Texts are in Romanian but images and drawings are self explanatory. Walls are 1mm thick and 330mm depth seem ok for this application.

Yes, it seems that the 'Gastronorm' maxes out at 200mm.  They all come in standard sizes.

[Lukas05]

Hi,

since some of you liked the project, I thought I would continue documenting the progress

A problem occured in the last run of the vapor phase oven. The container is sealed pretty good at the moment. There is only a small hole in the side wall for the heating element cables.
This poses a problem when it comes to pressure equalisation. The relatively cold PCB is lowered into the vapor phase. A lot of the gas condenses onto the cold surface in the first few seconds.
This causes the pressure in the system to drop, and air from the outside rushes into the chamber. This by itself isn´t a problem. But it gets problematic when lifting the hot PCB out of the vapor phase.
By lifting up the hot board into cooler parts of the chamber, the pressure rises. This causes galden vapor to get pushed out of the chamber. This of course means unwanted galden loss over time.
Currently I´m designing sorf of a reflux condenser. This will probably just be a 6mm copper pipe bent into an air cooled coil. The galden condenses in the pipe, and flows back into the container, while also providing a controlled pressure equalization.

I´m goint to verify this design in a FEM simulation. I´ve already startet so simulate parts of the project, to get some ideas of the material stresses and also transient behaviour of the system.
This will also be the foundation along with the measurements for the temperature profile regulator. I will post the results once I have calculated them through.

The post process of the coupled cfd simulation looks beautiful already I think
Here you can see the convection of the galden until 30 seconds after turning on the heating element.

https://www.flickr.com/photos/136915492@N02/49070612068/in/dateposted/

[Kjelt]

Wow you are really taking it a step further 
Just trying to think along could it be that your base carrier plate is to massive, taking too much Galden up?
If I look at this pro machine the boards carrier is only a wire mesh
https://youtu.be/jVQTcViEaUk

[mrpackethead]

Wow you are really taking it a step further 
Just trying to think along could it be that your base carrier plate is to massive, taking too much Galden up?
If I look at this pro machine the boards carrier is only a wire mesh
https://youtu.be/jVQTcViEaUk

You'd get the same issue, if you have a pcb that is large ( its also solid ).     

[mrpackethead]

A problem occured in the last run of the vapor phase oven. The container is sealed pretty good at the moment. There is only a small hole in the side wall for the heating element cables.
This poses a problem when it comes to pressure equalisation. The relatively cold PCB is lowered into the vapor phase. A lot of the gas condenses onto the cold surface in the first few seconds.

This would absolutely be correct.  This raises another problem. If you lower the PCB into the Galden too fast, your re flow profile is going to be way too steep on the way up.    With the 'static' frying pan method, i was effectivel 'preheating' the pcb. ( giving it a soak period ) by regulating the amount of energy that was going into the system.    If you were to monitor the temp of the PCB,  ( you'd need a thermocouple attached it ), you' would be able to control the rate at which it heated up quite well.  Above the 'thick' galden cloud, is a 'soft' cloud, and that vapour provides a pretty useful way of getting teh board up to a temp 10 lower than the melt point of the solder paste.

Galden vapour is easy to disturb. you only need a tiny draft and you'll blow it out of the tank.  However, If its 'still', it really does sit nicely, and unless you put LOTs of energy into it, its easy to keep it in the tank.     Moving PCB's in and out of the vapor does create a lot of movement.


This causes the pressure in the system to drop, and air from the outside rushes into the chamber. This by itself isn´t a problem. But it gets problematic when lifting the hot PCB out of the vapor phase.
By lifting up the hot board into cooler parts of the chamber, the pressure rises. This causes galden vapor to get pushed out of the chamber. This of course means unwanted galden loss over time.

Currently I´m designing sorf of a reflux condenser. This will probably just be a 6mm copper pipe bent into an air cooled coil. The galden condenses in the pipe, and flows back into the container, while also providing a controlled pressure equalization.

That has proved to be an effective way of keeping galden in teh tank as well.  I tryed using it to cool my frying pan project, and it worked.     The big problem with teh frying pan project is that while i can get really good results,  its a very long cycle time..  as the galden goes from cold to nearly cold, and thats the best part of 15 minutes.

Your design is really very clever using the cable lifts.    Some of the commerical VP machines bring the board out of the vapour into the top of tank, and then a door closes between teh board and the bottom of the tank.   I think thats doable with another cable lift.   I'll try and draw a picture tommorrow of what i mean. 
If that could work, you'd be able to get great cycle times, becuase you'd probably nto even need to cool the galden down.

[Lukas05]

Hi,

If you were to monitor the temp of the PCB,  ( you'd need a thermocouple attached it ), you' would be able to control the rate at which it heated up quite well.

I think this is the route I will go with this. The mass of the PCB varys a lot potentially from small PCBs with 0.8mm thickness, to 2.4mm multilayer boards with huge inductors.
This of course influences the time constant of the thermal low-pass wich ist formed with the heat capacity of the PCB. I´m thinking of an easy mechanism that presses a thermocouple on the PCB.... (not so sure how I would tackle this right now)
Another approch I had in mind would be to fixate a thermocouple on a small PCB with a ground plane that is installed permanently somewhere where it doesn´t take up much space.

Some of the commerical VP machines bring the board out of the vapour into the top of tank, and then a door closes between teh board and the bottom of the tank.

Yeah the two chamber design is very common it seems. I´m not sure if I want to install something similar, since it adds quiet a bit complexity an potentially BOM cost to the project.
Since I don´t need very low cycle times I think I will first evaluate the quick cool concept in detail. If I´m not satisfied with the result I will investigate the two chamber hybrid.

Just trying to think along could it be that your base carrier plate is to massive, taking too much Galden up?

Yes I think I will also swap that base out for something that has a higher "hole-to-metal" ratio.

[Koen]

I tried Gastronorm and the walls were too thin. I needed a bigger area for bigger panels and pots + glass lids + 5cm high spacers won again. 30 EUR for each set.

[Lukas05]

Hi everyone,

the project progresses fast which means it is time for an update 
(I think I´ll start a thread in the "projects, designs" sub forum next time)

The prototype got a temporary casing, which will be used for thermal performance testing of the system. The Airflow of the axial fans get´s hindered quite a bit by the casing, which means the steady state temperatures will also move a bit. I wanted do make a thermal simulation at fist, but the meshing process of this intricate geometry with all the cables and tubes would have taken a lot of time.
Thermal performance measurements are coming up soon.

https://www.flickr.com/photos/136915492@N02/49148444541/in/dateposted/

The display doesn´t function at the moment. I still have to throw the controller PCB together.
For quick testing I´ve decided to use the classic perfboard approach. The Board is located directly under the water pump  Which is kind of okay since it is only a 12V system...
The final PCB of course will be placed somewhere else in the case.

https://www.flickr.com/photos/136915492@N02/49148444301/in/dateposted/

As mentioned earlier I also wanted to swap out the carrier plate. The new plate has a much lower thermal mass, and also allows much more vapor to come through.
I used this opportunity to max out the available space, and increased the overall area of the carrier plate.

https://www.flickr.com/photos/136915492@N02/49148653877/in/photostream/

On the run I´ve installed a bigger window, and equipped the lid with some LEDs for a more pleasant PCB stalking experience during the solder process.

https://www.flickr.com/photos/136915492@N02/49147953273/in/photostream/

https://www.flickr.com/photos/136915492@N02/49147953628/in/dateposted/

[mrpackethead]

The prototype got a temporary casing, which will be used for thermal performance testing of the system. The Airflow of the axial fans get´s hindered quite a bit by the casing, which means the steady state temperatures will also move a bit. I wanted do make a thermal simulation at fist, but the meshing process of this intricate geometry with all the cables and tubes would have taken a lot of time.
Thermal performance measurements are coming up soon.

Do you even need to do encase this?  I found that it really impacted the cooling phase a Lot.   I dramatically increased my airflow, to get a resonable cool down.       The other thign was to not place the fans just above a bench. ( like another brand i wont' mention ) did.. It almost rednered them useless, and only slightly better than passively cooling.

[Lukas05]

Do you even need to do encase this?

Hmm if I were the only one to use this oven, I wouldn´t encase ist. My colleagues will probabely use the oven for prototyping as well... It feels much saver this way

The other thign was to not place the fans just above a bench

This is something I noticed as well. The inlet of axial fans in general is really sensitive to obstructions of the air flow. This effect seems to be highly dependent on the fan blade design, the hydraulic diameter of the fan itself, the rotational speed of the fan, and the pressure differential across the fan. The fans I use need at least 4cm - 5cm space to deliver acceptable performance.

[cgroen]

Any updates of your nice project Lukas ?

I have been looking at the IMDES VP "ovens", I soon stopped again when I saw the pictures of the internals.....
So, the more I think about it plus the amount of advice here on the forum by all those that have tried the vapor phase, I'm beginning to realize that this might be a possible solution for me (although I'm pressed on time). I have been lurking around first for a decent reflow oven, didn't find "the perfect one".
If one were to go with vapor phase, what is the "minimum" height of the container for the Galden ? I see that 300mm is mentioned, is it so that this is the minimum, or should f.ex 200mm also work fine without too much loss of galden ? (Gastronorm containers are usually max 200 mm). Also regarding cooling, as I understand it, you will need active cooling ?
I'm looking at something in the size of 320 x 260 mm PCB area (at least)

[Koen]

Regarding height, just close the lid.

Gastronorms I tried were too thin and bowed or popped during the process.

[cgroen]

Regarding height, just close the lid.

Gastronorms I tried were too thin and bowed or popped during the process.

Thanks Koen! I found some 0.7mm stainless steel ones, but guess they are no good then, pity as they are easily available (and claim they are good to 300 degC) !
I wonder what type the container Lukas uses, didn't find any hints so far

[Koen]

His lid looks like this : https://www.nisbets.be/fr/couvercle-hermtique-inox-et-silicone-vogue-gn-12/cp269

I found I lost a lot of space in Gastronorms due to the rounded corners (versus my expectations) and I hated how unsafe it felt. I didn't want to deal with a 230C issue nor nanny-watch it. Regular pots don't cause me concerns and my panels fit so I thought "k, good enough".

[KaneTW]

Yeah, the IMDES ovens are horrifying.

When I was engineering a vapor oven design, I arrived at the conclusion that you pretty much need active cooling to get the dT/dt required.

Gastronorms will warp under the temperature--whether that's acceptable for you or not is another matter. Since the requirements for the container aren't super strict, I'd just weld together a container of acceptable thickness stainless if a Gastronorm doesn't work out.

Deep friers just use Gastronorm afaik.

[cgroen]

Thanks guys!
I was looking at these: https://www.expondo.dk/royal-catering-gastrobakke-gn-1-2-200-mm-10011039
They claim +300 degC working. Guess its time for experiment....

[KaneTW]

I've done a quick simulation for thermal stresses when going from 300K to 600K. A (slightly simplified) GN 2/3 0.7mm container with fixed constraints on the rim will buckle about 2mm outwards on the sides and 1mm inwards on the bottom. The stresses on the body are acceptable (<80MPa). The constrained rim gets hit with a lot of stress but that's a modelling artifact. Allow for about +- 1mm of warping around the rim.

[cgroen]

I've done a quick simulation for thermal stresses when going from 300K to 600K. A (slightly simplified) GN 2/3 0.7mm container with fixed constraints on the rim will buckle about 2mm outwards on the sides and 1mm inwards on the bottom. The stresses on the body are acceptable (<80MPa). The constrained rim gets hit with a lot of stress but that's a modelling artifact. Allow for about +- 1mm of warping around the rim.

Thanks Kane,
thats not too bad I think ?
Just trying to figure out if I go with the GN 1/1 or the GN 1/2.
I found a 2000W heating element https://horecatiger.eu/en-eu/shop/heating-element-2000w-230v-heating-circuits-1-415669
that would fit the GN 1/1 quite nicely, but I'm afraid it wont heat the Galden too well as the heating element is right at the edge of the tank (the tank is 530 x 325 mm externally, the heater is 452 x 256 mm) ?

In the pciture below I have placed a 8 mm thick alu plate (the heater is 6.3 mm and will sit around 1.5mm above the bottom because of the size and the corners in the tank) in the center (gray area) to consume some volume (and lower the amount of Galden, like on Lukas design), if I put Galden in it so that the alu plate has 5 mm Galden on top of it I need a little short of 1 liter Galden.
But again, I'm not sure this will work with the large size of the tank (and the heater at the perimeter)....
Maybe the GN 1/2 is easier to get to work in this way, although I could really use the size the GN 1/1 gives me.....

[KaneTW]

Quote from Solvay's Galden FAQ:

For Galden® PFPE grades, a general recommendation is not to install a heater watt density higher than 4 ÷ 5 W/cm2, in order to avoid instability in heater surface temperature for low flow units (i.e. fluid reservoir in the TCU). Having the heater section where fluid flow is present near the outlet of the pump, it will allow higher watt density heaters to be used. The critical breakdown is ~ 14 W/cm2. We recommend 4 W/cm2. The decision to go beyond this level should be made by the customer.

I would not use a insertion heating element due to the low surface area. That heater will give you 9 W/cm2, which is fairly high.
Like you mentioned, it's also at the edge of the tank, so heating will be fairly localized.

My recommendation is to heat the entire bottom of the tank. Get a bunch of mica heating strips so that the bottom of the tank has a watt density of 4-5W/cm2 (remember steel is fairly poor thermal conductor). Epoxy them to the bottom of the tank with a heat-resistant epoxy. Same with aluminum water cooling heatsinks. Mount the container with a flexible high-temperature adhesive (e.g. Loctite 596).

Unless you have the money to spend, I strongly recommend sourcing the elements and heatsinks from China. The prices I saw for those at Western sites are absurd (>200EUR for a heatsink etc)

[cgroen]

Quote from Solvay's Galden FAQ:

For Galden® PFPE grades, a general recommendation is not to install a heater watt density higher than 4 ÷ 5 W/cm2, in order to avoid instability in heater surface temperature for low flow units (i.e. fluid reservoir in the TCU). Having the heater section where fluid flow is present near the outlet of the pump, it will allow higher watt density heaters to be used. The critical breakdown is ~ 14 W/cm2. We recommend 4 W/cm2. The decision to go beyond this level should be made by the customer.

I would not use a insertion heating element due to the low surface area. That heater will give you 9 W/cm2, which is fairly high.
Like you mentioned, it's also at the edge of the tank, so heating will be fairly localized.

My recommendation is to heat the entire bottom of the tank. Get a bunch of mica heating strips so that the bottom of the tank has a watt density of 4-5W/cm2 (remember steel is fairly poor thermal conductor). Epoxy them to the bottom of the tank with a heat-resistant epoxy. Same with aluminum water cooling heatsinks. Mount the container with a flexible high-temperature adhesive (e.g. Loctite 596).

Unless you have the money to spend, I strongly recommend sourcing the elements and heatsinks from China. The prices I saw for those at Western sites are absurd (>200EUR for a heatsink etc)

Thanks for the hint with the heating density! Just to be clear, the mica heating strips you are tlking about, are they "solid" ie encapsulated in some metal like these: https://au.rs-online.com/web/p/general-heating-elements/9210035/
or is it something else ? If not, I would be worried that there will be problems if these are bonded to the tank with epoxy and the tank bulges under heat (they might slip/crack) ?
I wonder what heating method IMDES is using on their/his tanks ?

[KaneTW]

Yeah, something like that. It needs a fairly decent epoxy to hold up to the stresses, but it should be doable. Would need to verify with simulations. The big issue there is temperature. Duralco 133 goes up to 350C, but I'm not sure if the strength is sufficient.

 From what I remember commercial ovens bolt them on using a backplate, but that would require drilling through the Gastronorm container.

It's still possible to make a liquid tight seal under those constraints, but there's more failure points and you need a machined part anyhow which kills all the advantage a Gastronorm has.

If a bonded strip heater turns out to be unviable, you could immerse them. That should give you a high surface area. Bonding in this case can be done with a high temperature silicone if it's chemically compatible with galden. Otherwise fix it from outside.

[cgroen]

I'm trying to dive a little more into this....
I see the Weller WAM 3000 has a smaller container below the "floor" of the large chamber, where it has the Galden/heater.
Is this a possible (or am I misunderstanding their drawing..) solution ? Doing this, one could much easier find a suitable heater, and the amount of Galden would be lower (compared to having the whole bottom of the large chamber covered with Galden). Will this still enable the vapor to move/fill the bottom of the large chamber ?
(sorry for asking probably stupid questions, I'm not a "gas wizard" )

[KaneTW]

That's a good idea, mostly because it cuts down on heater surface area. From the drawing it seems like they're bolting it on.

The vapor is denser than air, so there will be a vapor blanket up to some level. Level control is done via temperature sensors.

[cgroen]

That's a good idea, mostly because it cuts down on heater surface area. From the drawing it seems like they're bolting it on.

The vapor is denser than air, so there will be a vapor blanket up to some level. Level control is done via temperature sensors.

Thanks Kane!
Ï'm looking at the various models ASSCON VP450 and the Weller, a plan starts to materialize

The process chamber, I will get that done in stainless steel (I'm not a TIG welder, so I will leave that to the professionals  ). Its very easy then to find a suitable heater and have that incorporated into the design. I'm aiming at something like 400 x 400 mm PCB area, I wonder how much area around the PCB is needed for the vapor to get "above" the PCB ? I'm guessing 50 mm around larger than the PCB on all 4 sides, but this is just a guess !!

The "Galden" container that will be placed in the middle/below of the large chamber will be designed around whatever suitable heater I can find, I will get something with a large area (many "windings") per the recommendations you gave about approx. 4W/cm2. I will be looking for something around 2KW or so (should be plenty).

Cooling, I see some have a bunch of fans for cooling (IMDES etc), some use a real compressor/chiller (ASSCON) and some use simple water circulating system (Lukas). I also saw one of the members here (I think it was Iconic-PCB) that had 2 coils inside the process chamber with what looked to be simple water cooling. If that system is running or not I don't know.

I'm also temped to make the tray moveable by a winch system, ASSCON seems pretty neat. I know this is not absolutely needed, but why not overdo it once you have the possibility
I know this is probably going to take me too long, I really need this to be finished rather sooner than later, but I figure I might as well do it as good as I can...

Thanks a lot for the inputs so far, really appreciated !! 

[cgroen]

So, something along these lines....
The bottom chamber has an extra bottom so there is a small compartment where the heating element will be placed. This will give me a good even surface to mount isolation on. This also enables me to select whatever heating element is suitable.
Only "issue" is that the heating element is just touching the top of the small chamber, its only stuffed in between the top and bottom. I hope this will transfer the heat.
Otherwise I can use the mica strip heaters, might even be better!!

[KaneTW]

Heat transfer with an air gap is really bad. For that type of application I strongly suggest a strip heater with some (suitably rated) thermal grease and clamps or a backplate fixing it to the bottom.

[Lukas05]

Hi,

the project on my side is also coming along quite nice so far.
I´ve opened up a new thread a while ago, if you want to have a look at the current stage of development.

https://www.eevblog.com/forum/projects/automatic-vapor-phase-oven/

The results are pretty good. The profile also looks a lot cleaner now. The modelling of the lift mechanism combined with the behaviour of the vapor phase itself where a bit challenging.
I ended up modelling the system in LTSpice to play around with the regulation loop parameters.

This is the most recent board I soldered in the oven:

https://pbs.twimg.com/media/EQSbkagWsAAwvSB?format=jpg&name=4096x4096

I´am using gastronorm at the moment since it really is the cheapest option. Until now I can´t confirm any buckling of the container. My simulations also don´t show any excessive amount of stress. I will post-process the simulation results and post them in here. I don´t know If I will stick with the container in the future. The reason why I didn´t publish any source of the container is that it was an ebay fund.

I wouldn´t recomment drilling any holes in the bottom of the container. Or at least I don´t see the need to. I could imagine this action would come along with a lot more manufacturing an qualification effort. The immersion heater I´m using works good so far. The phase change at the heating element can transfer a lot of energy. By nature of convection the resulting vapor phase is very homogeneously. A PCB that covers most of the tray are creates a lot more disturbances in the heat distribution profile than the location of the heating element. (these disturbances aren´t very large still)

[cgroen]

Heat transfer with an air gap is really bad. For that type of application I strongly suggest a strip heater with some (suitably rated) thermal grease and clamps or a backplate fixing it to the bottom.

Kane,
I agree, it needs to make contact. The idea was that _perhaps_ the heater could be pressed up against the upper plate, but its still not optimum. I'm currently looking into mica strip heaters, by combining a bunch of those in series/parallel I can reach f.ex 1700 Watt on a 350 x 300 mm "tray" where the galden then would be located. That will give me 1.6W/cm2 which is a lot lower than the max recommended of 4 to 5 W/cm2.

Hi,

the project on my side is also coming along quite nice so far.
I´ve opened up a new thread a while ago, if you want to have a look at the current stage of development.

https://www.eevblog.com/forum/projects/automatic-vapor-phase-oven/

The results are pretty good. The profile also looks a lot cleaner now. The modelling of the lift mechanism combined with the behaviour of the vapor phase itself where a bit challenging.
I ended up modelling the system in LTSpice to play around with the regulation loop parameters.

This is the most recent board I soldered in the oven:

https://pbs.twimg.com/media/EQSbkagWsAAwvSB?format=jpg&name=4096x4096

I´am using gastronorm at the moment since it really is the cheapest option. Until now I can´t confirm any buckling of the container. My simulations also don´t show any excessive amount of stress. I will post-process the simulation results and post them in here. I don´t know If I will stick with the container in the future. The reason why I didn´t publish any source of the container is that it was an ebay fund.

I wouldn´t recomment drilling any holes in the bottom of the container. Or at least I don´t see the need to. I could imagine this action would come along with a lot more manufacturing an qualification effort. The immersion heater I´m using works good so far. The phase change at the heating element can transfer a lot of energy. By nature of convection the resulting vapor phase is very homogeneously. A PCB that covers most of the tray are creates a lot more disturbances in the heat distribution profile than the location of the heating element. (these disturbances aren´t very large still)

Thanks Lukas,
I'm looking in your thread 10 times a day hoping for updates
Can you tell me the size of the gastronorm you use ?

[cgroen]

Just got a couple of mica strip heaters (I have never seen these in real life, so had to get my hands on a couple to get a feeling for it)
My good friend is a brilliant mechanical engineer/CAD designer, I'm hoping for his help on this
I wonder why the Weller WAM 3000 is "dead", it seemed like a reasonable idea, either the market was not there, or it didn't work 

[nisma]

It works, the reason it is dead is because of several design decisions and its high price.
Later a cheaper model with better specs came out, VP50 relabeled to some names , I don't remember the exact name from non OEM version. Having double eurosize panels on WAM the tact time is 17 minutes and on VP50 it is just 3min /panel. If PnP have tact time of 6-12min do you really consider the more priced WAM for more then 3 seconds
after you have realized this?

[cgroen]

It works, the reason it is dead is because of several design decisions and its high price.
Later a cheaper model with better specs came out, VP50 relabeled to some names , I don't remember the exact name from non OEM version. Having double eurosize panels on WAM the tact time is 17 minutes and on VP50 it is just 3min /panel. If PnP have tact time of 6-12min do you really consider the more priced WAM for more then 3 seconds
after you have realized this?

Thanks Nisma,
good to know that it at least works (do you happen to know what the selling price was at the time?) !
Do you know about which "design decisions" that was the problem ?
Incredible that the cycle time was 17 minutes on it, I mean with a gate and all that.

[nisma]

No, i don't remember the selling price.
The so called bad design decision was not bad itself, it was just bad in relation to the concurrency that emerged later.
The 30x30cm instead of 50x50cm for batch processing.
Probably this was the limit for forced cooling.
The other is the use of forced air cooling instead of liquid cooling with laser chiller.
liquid cooling is more efficent and a lot more quite as forced air cooling.
Annother problem with the massive air movement is that dust is moved around giving additionl headcaches.
The Fan starts during galden heating process.
Third is not to use fan cooler on top, this lead to increased galden usage.

[cgroen]

No, i don't remember the selling price.
The so called bad design decision was not bad itself, it was just bad in relation to the concurrency that emerged later.
The 30x30cm instead of 50x50cm for batch processing.
Probably this was the limit for forced cooling.
The other is the use of forced air cooling instead of liquid cooling with laser chiller.
liquid cooling is more efficent and a lot more quite as forced air cooling.
Annother problem with the massive air movement is that dust is moved around giving additionl headcaches.
The Fan starts during galden heating process.
Third is not to use fan cooler on top, this lead to increased galden usage.

Thanks Nisma!
sounds like good valid points you make there, i can imagine the large fans and dust, didn't think of that before, but makes perfect sense.
I see on most commercial units that the coolers (water cooler) is mounted at various points compared to the vapor cloud, some shows them just above the cloud, and some show them higher up in the chamber, I wonder the effect of that?

[cgroen]

I'm trying to get  a feeling for the heater needed for my system. Does any of you have an idea of how much "heating area" is needed to get this going, lets say I have a chamber of 500 x 500 mm, and that the area where the liquid galden is located is maybe 300 x 300 mm (as an example). Is that way too much area to heat on, or is it way too small (I kknow that Galden is heavy, and I hope the vapor will "level out" so it will fill the larger compartment from the bottom and up when forming, so the area where the galden is located does not need to be huge) ?
I'm not a "vapor/fluid" expert (software is my main trade ), and I'm having a hard time figuring all this out. Also, lets say that the "container" where the liquid galden is located is 300 x 300 mm, how much power is needed for a system like this to work ? are we talking 1000W, 1.5Kw, 3 KW or ?? (assuming 230 degC, and vapor height of maybe 150 mm or so )
My good friend is arriving in a couple of days, and then the design starts, I need some ballpark figures to aim at for a start, I know that this will be a lot of experimenting, but would be nice to have a starting point.
Originally I needed a finished system, luckily I have taken care of the immediate need in another way, so I have the time now to make something that hopefully will work.....
All inputs are welcome 

[KaneTW]

When I was designing a system, my main considerations were:

* Heater/fluid well area determined by minimum fluid level
=> You want at least some fluid (how much depends on your controller skills) to provide a safe operating margin so that the heater can keep running during the process

* Maximum fluid level determined by process volume (and well area)
=> You can estimate the amount of galden needed using the ideal gas law and your process volume, then add the minimum fluid amount determined above

* Vapor level determined by target heat capacity
=> You want enough vapor so that there's always vapor around the PCB during the actual soldering process. If there's not enough vapor, you'll have difficulties with maintaining a profile.

* Heater power determined by cycle requirements, maximum heating density, and process volume
=> You need to fill the vapor level back up to nominal after a cycle finishes, which is determined by how much heat you can dump into the liquid. If you can wait for half an hour for the vapor level to go back up, you don't need a strong heater. If you need to do it in 2 minutes, things get more complicated. You can calculate the minimum energy required from the Galden datasheet (specific heat + heat of vaporisation), but that doesn't include losses.

[cgroen]

Thanks for all the input Kane, much appreciated 
I know what direction to go with a lot of the stuff now, need to go study!
But, one thing I have a hard time figuring is how large the part that holds the Galden needs to be compared to the area of the large chamber. My idea is to make a smaller area with the liquid Galden, put the heaters below that, and let the vapor rise up and around in the large chamber (exactly like the Weller WAM3000 in the drawing below). From that drawing, the "hot area" looks much smaller than the actual process chamber. The good thing about that is the lower amount of Galden needed (smaller area). I just need some kind of idea if that is possible and "allowed" (a different approach is Lukas and the IMDES devices, they have liquid Galden over the complete area of the chamber.
I have a couple of mica strip heaters here, each is 6 x 4 inches, and 950 Watt. An ideat situation would be to make a small chamber of perhaps 8 x 8 inches or so, the heaters could then cover the bottom of that, and heat the Galden in there. I'm just not sure if a 8 x 8 inch (200 x 200 mm) chamber would be enough for "cooking" the Galden for a process chamber of 500 x 500 mm, or if the vapor just will fill it fine anyway

Again, thanks a lot Kane for your input (and your patience )

[cgroen]

Doing (at least trying to) some math here....
I'm NOT sure I got all this under control, I might be waaaay off and wrong in a lot of the following, but hey, at least I'm trying


If I set the process chamber to 500 x 500 mm, the height of the vapor to 200 mm (volume is 0.05 m3)
The temperature is set to 230 deg C, and pressure of 1 atm, I get 1.211 mol required using: https://www.chemicool.com/idealgas.html
According to https://www.solvay.com/en/brands/galden-pfpe , the molecular weight of LS/HS/HT-230 is 1020 amu (LS/HS/HT-200 is 870 amu)
So 1 gram of HT-230 is 1/1020 mol (1/870 mol for HT-200).

So to fill the 500 x 500 mm chamber with a 200 mm high "cloud of vapor", I need 1.211 mol * 1020 => 1235 gram of HT-230
If using HT-230 instead (870 amu), I will need 1.211 mol * 870 => 1054 gram of HT-200

Density at 25 deg C of HT-230 is 1.82 g/cm3, so 1235 gram is 679 mL.
Density at 25 deg C of HT-200 is 1.79 g/cm3, so 1054 gram is 589 mL.

Just for fun, compared to water: molecular weight of water is 18.02 amu. 1.211 mol of water weights 21.6 gram
So 21.6 gram of water can fill the chamber with vapor, Galden HT-230 requires 1235 gram/679 mL for the same !

Getting fluid up to boiling temperature:

Specific heat @ 25 deg C for Galden is 0.23 Cal/g deg C => 0.96 J/g*deg C.
To raise 1253 gram Galden from 20 deg C to 230 deg C requires: 1253 * 210 * 0.96 => 254 kJ.
So if 253 kJ is required, I can bring the Galden to boiling point (from 20 to 230 deg C) in 1 minute using 253 kJ/60 seconds => 4.2 kW

Specific heat for water is 4.186 J/g*deg C.
To raise 1000 gram of water from 20 deg C to 100 deg C requires: 1000 * 80 * 4.186 => 335 kJ
So, the water will boil in 60 seconds if I inject: 335 kJ/60 seconds => 5.58 kW.

Vaporizing the fluid completely:

Heat of vaporization for Galden when already is at boiling point, for both HT-200, HT-230, LS-200 and LS-230 is 63 J/g (the different types are almost all the same value).
So to vaporize 1235 gram of (already boiling) HT-230 I need 1235 * 63 => 78 kJ.
If I would need to vaporize all the Galden in 1 minute I would be required to supply 78 kJ/60 sec => 1.3 KW of heat during that 1 minute (sounds like an awful small amount ??).
(for HT-200, this would be somewhat smaller, the amount of Galden is smaller, and it would need to be raised 30 deg less than the HT-230)

Heat of vaporization of water already at boling point:2260 J/g
To vaporize 1000 gram of water (already boiling) I need 1000 * 2260 =  2260 kJ
Vaporizing in 60 seconds requires: 2260 kJ/60 sec => 37 kW for 60 seconds

All of this assumes ideal conditions, no loss to the surroundings etc.

Can anyone perhaps confirm the calculations, at first I thought it is a large amount of Galden needed to fill the chamber, I know it is quite large (500 x 500 x 200 mm), but still?
Also, I don't quite understand that the heat of vaporization is 63 J/g for Galden and 2260 J/g for water, is that really true ?

On top of that amount I will need to add the minimum level as Kane pointed out (also a 200 mm high vapor cloud is very much, could probably be much smaller).

EDIT:

If chamber is 500 x 500 with only 130 mm cloud height:
Volume = 0.5*0.5*0.13 = 0.0325 m3
At 230 deg C, 1 atm, 0.0325 m3, 0.787 mol is required.
0.787 mol * 1020 => 803 gram => 441 mL

Raising 803 gram from 20 to 230 deg C in 60 seconds: 803 * 210 * 0.96 => 162 kJ = 162 kJ/60 seconds => 2.7 kW

Vaporizing it in 60 seconds when at boiling point: 803 gram * 63 J/g => 50.6 kJ => 50.6 kJ/60 seconds => 840 Watt

[cgroen]

Been busy here
My vapor phase oven is taking shape. I have a very good friend that is a mechanical engineer in his day job, so I'm lucky 

Its a ongoing work, so far most of the chamber is designed, the lift system and the heaters. Still missing the details for the lid (the sealing has been taken care of with rubber gasket that goes down in the chamber, the sides at the top is at a slight angle to make it tight.
Below is some screenshots of whats been done so far. The cooling devices are right now being produced (in china), they are made with channels inside and Festo couplings for the water. My plan is to use a "laser tube chiller" (CW5200 with compressor, https://www.teyuchiller.com/co2-laser-tube-water-chillers_p31.html), they can go down to 5 degC (I see ASSCON's device runs at 6 degC, with a low alarm at 5 degC, if I'm not mistaken).
All of the temperature sensors are not going to be mounted, only 3 of them. But as I don't know the exact height, there are 10 different positions to chose from.
In the Galden container, there is a stainless steel tube that will allow water cooling of the Galden (can be activated with a solonid during cooling sequence).

The controller for it are also (almost) finished, still need to figure the best algorithm for soldering, couple of graphs for the temperature sensors etc, right now there is simple support for solder profiles, the lift moves to the correct position etc. Will need to be tweaked once I have a physical oven

Controller uses a NXP processor (LPC54628), has 32 MByte SDRAM, 128 MB filesystem (I know, overspec'd ), H bridge for the lift motor, encoder input for lift, 6 x 3 wire PT100 inputs, 4 x 2 A outputs, 4 x analog/digital 24V inputs, Ethernet and a 4.3" capacitive touch screen (GUI is using littlevGL).

I have uploaded a small video of a run, all with simulated operation. And as I wrote, the actual soldering sequence/profile is no where near how its supposed to be, this was only done to help developing the GUI etc.

I still have a few issues/worries, one of them if the galden container is "large" enough, if its close enough to the walls of the chamber, I guess the vapor will find its way regardless.

Video of controller here:

https://youtu.be/1f__h6oHO9U

[cgroen]

Couple pictures of controller

(still need to implement the graph of the temperature sensors etc)

[rx8pilot]

Couple pictures of controller

(still need to implement the graph of the temperature sensors etc)

Holy hell.....I have some catching up to do 

Looks awesome.

[KaneTW]

Very nice--glad it's coming along well.

[cgroen]

Very nice--glad it's coming along well.

KaneTW,
a lot of it because of your input 

My major concern at the moment is if the heaters will do the job, if they fail, the idea is to make a solid alu block and embed "heating cartridges" in it, clamp it to the bottom of the Galden container, and use that instead.

Once I get something constructed, I will do a proper description, I will also probably publish the files for the mechanics and BOM file

[suj]

Looking at previous professional projects, I was wondering whether to show my realization (still under construction), but maybe someone will be interested. I decided to use the GN container, heating from the outside with an IR heater and cooling with water in an open system. The lid has a gasket and a mounted quartz glass window. A coil made of a capillary tube serves as venting, and a cooling coil is bent from an aluminum tube. I used high-temperature silicone as sealants and assembly adhesive. As a control, I used a PLC from my shelf, I only bought a touch panel for it. As the PLC had analog inputs for temperature conversion I used a K to voltage converter which I bought on ebay. One of the thermocouples is glued with Kapton tape to the bottom of the container (red in the graph), the second thermocouple on the thin wire arm touches the PCB (green) and the third is placed a few centimeters above the PCB (yellow). I used temperature gradients as the setpoint for the PID controller. The slightly strange color of the container on the outside is due to the fact that in order to reduce the reflectivity of IR radiation I used chemical oxidation on stainless steel. I set everything up in the kitchen to have close access to the drain, and it has stayed that way for the last few months  . I need to mobilize myself to put everything in a small control cabinet and do the final cabling. The effects on my amateur level satisfy me. Finally, a few photos of the effects.

[rx8pilot]

 

[julian1]

These consumer "Deep fat fryers", that combine enclosure, top window, heating element, and thermostat look a bit interesting.

https://www.trustedreviews.com/wp-content/uploads/sites/54/2019/10/3-VonShef-1.5L-Stainless-Steel-Deep-Fat-Fryer-768x512.jpg

https://www.trustedreviews.com/reviews/vonshef-1-5l-deep-fat-fryer

There are a bunch of model varations/sizes, but most regulate to 190C as top temperature. Run it open loop and maybe it would hit 230C without futher mods.

Attach a vertical length of copper tubing out the top, like a still/ air-cooled fractionating column to condense the Galden back to liquid. Or glue flexible tubing to the lid, and then run tap water through it.

[pisoiu]

Hi all,
I need a bit of advice regarding my tests with vapour phase oven. My tests are made in a 200mm deep gastronorm pan.
Problem #1, heating. I was deeply concerned about heaters not exceeding 300 degrees celsius (and breaking galden into toxic compounds) and I decided to use PTC heaters from aliexpress, those sold for LED desoldering, with 260 degrees limit. I mounted under the pan 4 pcs x 300W heaters, which seem to do the job at the beginning. What I missed from calculations is the fact that the PTC heaters cut power drastically once temperature goes up. It heats up everything up to 100-150 degrees but at 200 degrees it stops, cannot go further. At that point they take around 250W (all 4 pcs) and that is not enough to raise temperature further. It is a balance point between input power and what is lost. I need to change the heater and to dump the PTC idea.  I was thinking to put a 260 degrees resettable thermal fuse on a normal heating element to prevent it to overheat. But I am unable to find something usable at this temperature. I made some experiments with something like this: https://www.tme.eu/ro/details/hts16-230-500/rezistoare-de-incalzire/telpod/hts-16-230-500-3-6-3/ but they were not ok, I had problems with the contacts desoldering, even if the manufacturer claims the element is ok up to 400 degrees on surface.
Can anyone recommend me a suitable heating element, preferably not round? I need a flat surface to install the thermal fuse.
Problem #2. During tests, without any lid, from the heated galden vapors were raising constantly, but I assumed they are only water vapors from atmosphere, since galden was under 200 degrees all the time.  Trying to build up temperature inside chamber, I decided to put an aluminium lid above. After some time, since aluminium plate was cooler than the vapors coming from the chamber, they sticked to its surface. But when I put the finger on the lid surface, there was not water, I think it was galden. The substance on the lid surface has the same oily consistency as galden. Am I losing galden? Why, since its temperature is under boiling temperature?
Thanks.

[IconicPCB]

cgroen.

i have continued with the build lately and am getting good results with galden230.

I am working with four solid hot plates from an electric stove. The heater assembly is pulling about 11 Amps at 240V and the production cycle takes about 15 minutes.

I have designed a controller for the oven but have not implemented it yet. I am operating the oven manually on the understanding that i am NOT EXCEEDING 4W/cm2 applied power density.

The oven is in early development, I wish to introduce a cooling fan into the chamber to scavenge vapors prior to opening the lid.

I have introduced a separator between top and bottom half of the chamber and the product basket and the lid allow for the basket to be lowered into reflow one and be lifted into cooling zone while the separator gate is closed between the two zones.

I still get  vapor loss however i expected the vapor to condense on cooling coils much more so than it does.I think a recirculating fan inside the chamber will scavenge the vapor.
I use only one thermocouple in the basket to monitor rise of vapor.
Once the temperature reaches 230 C i time the reflow cycle and then turn of the power, turn on water and raise the basket into cooling zone followed by closure of barrier between the two zones.

I wait for the temperature in cooling zone to reach a manageable level and then open the lid and remove soldered boards.
The basket is 350 x 450 mm and the overall throughput is acceptable due to the large area.

[cgroen]

cgroen.

i have continued with the build lately and am getting good results with galden230.

I am working with four solid hot plates from an electric stove. The heater assembly is pulling about 11 Amps at 240V and the production cycle takes about 15 minutes.

I have designed a controller for the oven but have not implemented it yet. I am operating the oven manually on the understanding that i am NOT EXCEEDING 4W/cm2 applied power density.

The oven is in early development, I wish to introduce a cooling fan into the chamber to scavenge vapors prior to opening the lid.

I have introduced a separator between top and bottom half of the chamber and the product basket and the lid allow for the basket to be lowered into reflow one and be lifted into cooling zone while the separator gate is closed between the two zones.

I still get  vapor loss however i expected the vapor to condense on cooling coils much more so than it does.I think a recirculating fan inside the chamber will scavenge the vapor.
I use only one thermocouple in the basket to monitor rise of vapor.
Once the temperature reaches 230 C i time the reflow cycle and then turn of the power, turn on water and raise the basket into cooling zone followed by closure of barrier between the two zones.

I wait for the temperature in cooling zone to reach a manageable level and then open the lid and remove soldered boards.
The basket is 350 x 450 mm and the overall throughput is acceptable due to the large area.

Sounds good!
My project is on hold at the moment, everything is ready (mechanically) to be produced, but until now, I simply have not had the time to finish it I hope things will clear somewhat in the future so I can get on with the project.
If you have any more info/pictures etc of your setup I would love to see them!

[kylehunter]

Sorry for the bump, figured it's better here than a whole new topic..

I'm considering building an oven for our shop. It'll mainly be for prototypes, small batches, and for complex boards when I don't want to profile our big ovens (we have two heller convection ovens). End game, I'd like to have one that has a fast enough batch time to where it becomes feasible to use the vapor phase vs powering up the main oven for all but production runs.

Cost isn't going to be super important, as the alternative for us is to spend $20K  for an Asscon oven.

I'm planning on modeling a lot of this off of vapor degreasing systems. They use incredible toxic solvents, as a result, limiting vapor loss is a safety concern. https://www.baronblakeslee.net/vapor-degreasers/baronette/index.html is an example of a nice batch style vapor degreaser. They don't rely on a sealing lid, instead they go full force with the condensing coils to make sure vaport can't leave. Not having a sealing lid helps massively with pressurization, and cycle time.

Here's a quick mockup of what I'm think of thusfar: (Ignore the exact dimensions, this was before I found the mcmaster container!)




Basically, when doing multiple panels, the Galden will be constantly boiling, so the vapor cloud will always be present. The first set of condensing coils will constantly be condensing vapor. This effectively will stop most vapor from being above this point. Then there will be a 75mm gap, which is called "freeboard" in the vapor degreasing world. There will still be a bit of vapor here, but it will be cooling as it goes. Above this point, there will be another set of coils. This should condense the rest of the vapor that is still around.

This "layered" system allows for cycle times to be pretty fast, since you don't have to wait for the Galden to boil. It also allows there to be a nice temperature gradient to help with reflow profiling.

My current idea for the heater is to have a custom machined heater block with cartridge heaters inside it. (We use this for a reflow hotplate currently) This will be affixed to the bottom of the container, to heat from the bottom up. The coils will be liquid cooled using some sort of laser cooler or similar. The carriage system for the PCB I really haven't put much thought into yet, as that is a pretty easy part and will worry about it later. https://www.mcmaster.com/3763K211/ is what I'm planning on using for the container.

Please let me know your thoughts! I literally just started brainstorming this yesterday, so I'm sure I'm overlooking a lot.

[helius]

There's a very important difference between a vapor degreaser and a VPS system, which is that the former runs open-loop and the latter can't. The Galden cannot be "constantly boiling" because there would be no control over heat transfer rates, and your boards would be terrible with tombstoning and solder voids everywhere. This has been gone over repeatedly in this thread and others, take the time to read them.

[kylehunter]

There's a very important difference between a vapor degreaser and a VPS system, which is that the former runs open-loop and the latter can't. The Galden cannot be "constantly boiling" because there would be no control over heat transfer rates, and your boards would be terrible with tombstoning and solder voids everywhere. This has been gone over repeatedly in this thread and others, take the time to read them.

So either I'm wildly missunderstanding how commercial solutions operate, or I miss explained my idea to you..

So, commercial, batch or inline systems (I don't mean the cheap few thousand $ hobby systems) the Galden is constantly boiling, just like I described. You control the ramp rate of the PCB by controlling how it is lowered into the cloud. This is the only way to have somewhat fast cycle times. If not, you have to wait for it to boil and cool down each time.

Here's an article showing how it works (sorry for the crap image!):

https://www.electronicproducts.com/vapor-phase-reflow/# (See section "2.  Preheat zone")


Since the area above the 230c vapor will of course still be hot, by lowering the gantry at a controlled rate, you effectively control the preheating of the system.

Hope that explains it!

Also, this system will absolutely be closed loop. There will be multiple thermocouples at each layer making sure no temperatures exceed safe operating temps.

[helius]

So either I'm wildly missunderstanding how commercial solutions operate, or I miss explained my idea to you..

So, commercial, batch or inline systems (I don't mean the cheap few thousand $ hobby systems) the Galden is constantly boiling, just like I described. You control the ramp rate of the PCB by controlling how it is lowered into the cloud. This is the only way to have somewhat fast cycle times. If not, you have to wait for it to boil and cool down each time.

Yes, I think you do misunderstand how they operate. They don't "control how it is lowered into the cloud", instead the elevator moves down when the cycle begins and raises when the cycle ends. The speed or depth of the elevator is not servo controlled. The heat input, on the other hand, is: yes, the Galden is hot all the time, but the thing you seem to be missing is that once a vapor cloud exists, every additional watt added to the fluid is a watt that is released on the board by condensing fluid. If this heat flux is not controlled according to a profile, the results will be inconsistent.

[KaneTW]

The last time I spoke with an ASSCON rep the position was controlled in the higher end batch units. There's docs somewhere.

[kylehunter]

So either I'm wildly missunderstanding how commercial solutions operate, or I miss explained my idea to you..

So, commercial, batch or inline systems (I don't mean the cheap few thousand $ hobby systems) the Galden is constantly boiling, just like I described. You control the ramp rate of the PCB by controlling how it is lowered into the cloud. This is the only way to have somewhat fast cycle times. If not, you have to wait for it to boil and cool down each time.

Yes, I think you do misunderstand how they operate. They don't "control how it is lowered into the cloud", instead the elevator moves down when the cycle begins and raises when the cycle ends. The speed or depth of the elevator is not servo controlled. The heat input, on the other hand, is: yes, the Galden is hot all the time, but the thing you seem to be missing is that once a vapor cloud exists, every additional watt added to the fluid is a watt that is released on the board by condensing fluid. If this heat flux is not controlled according to a profile, the results will be inconsistent.

I'm sorry, I just don't think you're correct... Like KaneTW, I also have spoken with Asscon some time ago, and they do control the soldering like I said. You may be right on the cheaper models, again, that's not what I'm trying to compare here. Just look up online "vapor phase reflow preheat" or "soft vapor phase reflow". There's a lot of articles discussing this method. Here's a really good one, which even show's a diagram of the board being moved into and out of the cloud to control temperatures: https://smtnet.com/library/files/upload/Vapor-Phase-Soldering.pdf

> every additional watt added to the fluid is a watt that is released on the board by condensing fluid.
No it's not... Since there's a cold condensing coil in the chamber, that coil removes excess heat, causing the fluid to drip back down in the tank to be reheated. I could constantly boil the liquid, without a board inside it, and as long as my cooling coil is powerful enough, it would sit indefinitely at a steady state. I think you're thinking that this is just a closed chamber, with a heater on the bottom. That's not at all what I'm going for here...

[KaneTW]

The whole point of vapor phase reflow is that the vapor is not heated, only the liquid is. Primary method of heat transfer is condensation of the Galden onto the assembly. The amount of heat going into the liquid phase doesn't matter as long as there's sufficient vapor and the liquid phase doesn't locally overheat.

That being said, I don't think batch units are pressurized. There's a condensate coil and a lid but it's more like a pot lid, not a pressure cooker lid.

I'd run a fluid sim with your approach.

[kylehunter]

The whole point of vapor phase reflow is that the vapor is not heated, only the liquid is. Primary method of heat transfer is condensation of the Galden onto the assembly. The amount of heat going into the liquid phase doesn't matter as long as there's sufficient vapor and the liquid phase doesn't locally overheat.

That being said, I don't think batch units are pressurized. There's a condensate coil and a lid but it's more like a pot lid, not a pressure cooker lid.

I'd run a fluid sim with your approach.

Yeah absolutely, that's what I'm going for here. It won't be pressurized at all.

[KaneTW]

Fundamentally the main issue I see is the condensate coils not providing enough of a temperature gradient. Rest seems close enough to existing batch units.

That being said, this is getting into the territory where it'd be cheaper to buy an inline vapor phase system than develop and iterate on this system.

[kylehunter]

Fundamentally the main issue I see is the condensate coils not providing enough of a temperature gradient. Rest seems close enough to existing batch units.

That being said, this is getting into the territory where it'd be cheaper to buy an inline vapor phase system than develop and iterate on this system.

Gotcha. Yeah my plan for the start is just to run straight tap water and back down the sewer. At least then I know that the cooling capacity won't be an issue.

As for the pricing, IDK man, those inline machines start at $100k - $200k. This won't even come close to that. Especially the first prototype setup. We'll see though.

[Kean]

Kyle, have you looked at the Vapor Phase One?  https://pcb-arts.com/en/vapor_phase_one

I'm not sure if it is big enough for your needs, or if it can run in a batch mode similar to what you describe - but it is open source hardware and might be worth looking at.

Their video is kind of cool...

[kylehunter]

Kyle, have you looked at the Vapor Phase One?  https://pcb-arts.com/en/vapor_phase_one

I'm not sure if it is big enough for your needs, or if it can run in a batch mode similar to what you describe - but it is open source hardware and might be worth looking at.

Their video is kind of cool...

Yeah definitely have! He actually posted a lot of his progress in this thread and another on EEVBlog.

It seems like a solid idea, with some odd design choices. I'd do a custom one to be able to have the continuous boil approach which they don't do, and also content for a new YT series!

[Kean]

Yeah definitely have! He actually posted a lot of his progress in this thread and another on EEVBlog.

Ah, yeah... 

I did check this thread for a couple of search terms to see if it was already mentioned, but it didn't come up as Lukas doesn't actually mention it or his company by name in this thread.