Vapor Phase reflow oven

[hobbyelectronics]

On a side note but still about vapor phase, what would be considered the best solution to control the heater, assuming its powered by 230VAC?
To control the heater with a percentage power in the profile control itself, for each step, PID control would not be required.

Some of my thoughts are:

1, SSR with Zero crossing. (Perhaps, changing the on/off ratio over 1 second, giving maybe 0 to 100%, e.g. on for 200 mS, off for 800 mS to give approx. 20%)
But with a zero crossing SSR the switch on time could be up to 10 mS, depending where it is in the AC cycle. (Do they also switch off at the zero cross over?)

2, Same as above but maybe use a zero crossing detector circuit so that a normal non zero crossing SSR is turned on and off close to zero volts.

3, Use zero crossing detection, then fire a TRIAC to give phase control, so each profile cycle could give the correct percentage of power for each heating step. (e.g., 80% heat. For either a timed step or temperature endpoint is reached)

4, Use a TRIAC instead of a SSR but trigger it with a zero crossing opto (Such as MOC3083M with built in ZCD), but still use on/off times over one or 2 seconds. So the TRIAC is only fired at zero, but the ratio is controlled by the on/off times.

5, Simple mechanical relay with varying on/off times over a few seconds.

6, Any other method to control proportional power, maybe FETS, or another way.


I am very interested in this and may cobble something together as an experiment, maybe use a small custom ARM or PIC based controller with a small display. Perhaps driving either Halogen heater tubes (2x 500W) or maybe a 750W Silicon heater mat.
Any thoughts on this or ideas about mains proportional power control would be interesting.

Thanks

[wraper]

Just t show what sort of garbage IMDES is.

[wraper]

There are multiple old threads on this forum about vapor phase, but what worked for me when modifying mini condens is described here, you may want to read more of that thread.
https://www.eevblog.com/forum/reviews/mini-blog-imdes-vapour-phase-purchase/msg1104140/#msg1104140

[andrea.longobardi85]

by being larger and having moving parts. The vapour zone sits at the boom of the tank and doesn't extend to a surface the galden could condense on and then drip back on the board. The baords are lowered into the vapour zonestay there until they reach temp and then are raised out again, There will typically be an assisted cooling and drying phase ->IBL have this presentation or you could lookup Asscon.

I wander what is the price gap between a professional Vapor phase as an IBL and the vaporflow…
Another question I have…what is the drawback of a Vapor phase oven compared to a convection reflow oven beside maybe throughput…as far as I understand Vapor phase consume less energy, require less space and the thermal stress on components is lower…

[SMTech]

by being larger and having moving parts. The vapour zone sits at the boom of the tank and doesn't extend to a surface the galden could condense on and then drip back on the board. The baords are lowered into the vapour zonestay there until they reach temp and then are raised out again, There will typically be an assisted cooling and drying phase ->IBL have this presentation or you could lookup Asscon.

I wander what is the price gap between a professional Vapor phase as an IBL and the vaporflow…
Another question I have…what is the drawback of a Vapor phase oven compared to a convection reflow oven beside maybe throughput…as far as I understand Vapor phase consume less energy, require less space and the thermal stress on components is lower…

I don't have up to date numbers but based on old quotes, an top loading machine like a VP 510 should be in the €20-30K bracket, the IBL perhaps at the higher end of that range. That is still just a batch reflow oven, for the same money you can get something forced convection that can handle continuous throughput. Inline VP does exist but it is quite expensive and might need things like water supply for cooling. You will find a number of Electrolytic capacitor datasheets tell you not to use vapour phase reflow as they suffer damage and capacity loss.

[andrea.longobardi85]

Thanks, indeed I have seen also SMD sensors that could not been soldered with Vapor Phase…

[glenenglish]

Andrea : The sensors that could not be soldered- did they not have tape over the sensor port ?

The IMDES yes, you do need to do some modifications.  . I cut a slot in the PCB between those two pins- note :  my 1st IMDES was like that, but my 2nd IMDES had 3mm spacing. I pointed this out to them and they modified it.

But the basic idea works well.

What the IMDES experience shows that even if you do a sh1t job at implementation, results are still quite good.

[Doctorandus_P]

What the IMDES experience shows that even if you do a sh1t job at implementation, results are still quite good.

Yes, of course it is.
The soldering quality is mostly guaranteed by the boiling point of the fluid (which is a pretty stable temperature), and the energy release when the vapor condenses on your PCB. Oxygen displacement is also inherent in the process, and even pre-heating is done by holding the PCB horizontal when the hot vapour rizes. (This fluid has a lot of gravity in it, and I guess the vapor is also quite heavy)

Because (nearly) the whole process is controlled by the fluid and it's vapor, the machine itself is a very simple thing, and that is why it's just a horeca pan with a heating element, with a few extra considerations:

1. You do not want to over cook the vapor. When it decomposes because of over heating you get nasty gasses.
2. You want to reclaim as much of the vapor / fluid as possible, because the stuff is expensive. (And those EUR1500 ovens do a pretty mediocre job at that).

[hobbyelectronics]

I am so tempted to build my own. I have pretty much all the parts, including a stainless tank, but its a lot of effort and time.
I am also considering buying the VaporFlow, but feel put off by the cost for what you actually get. Yes is an out of the box solution, but its still a lot of money for what it is and for small volume or hobby use, its not cheap at €1,595.00 Plus carriage at around €70.
it would be good to try one as their does not seem to many first hand reviews about it.

I used to have a UniFab 6 zone top/bottom IR oven, proper production unit, but it was so problematic. It worked, but for different boards it needed a lot of tweaking and tomb stoning was always a problem, no matter what I did.
Would be nice to have a VaporFlow, small, low hassle and hopefully as good as or far better than the IMDES (From a soldering point of view).

As these small ovens loose vapor/steam, what spec does a fume extractor need to be?

[trevwhite]

I didn’t have time to build my own. I learnt a lot from buying an Imdes oven. I learnt I was over thinking how to build my own. That oven has 400w or 800w heater that it switches on and just measures the temperature. When it gets to a certain temperature it might half heat on. It holds temp for a while and then switches fans on to cool everything down. That’s about it. You won’t be getting much more than that for your money.

[hobbyelectronics]

Yes, that is so true, its a lot of money for what you are actually getting. However, it depends on if you need a vapor phase soldering solution. Then the lower cost options are limited to maybe two.

My thoughts, were to control the power with phase control (say 0 to 100%), and PWM the fans (5 to 100%), either have two halogen heater tubes (500W each) or a silicon heater mat and have a touch screen for the control.
A simple method to edit / create the profile in steps. Each step would endpoint with temperature or time and be either cooling or heating, with a percentage value for power (Heat or cool), when the step endpoint is reached (time or temp) it moves to the next step.
That way the control could be pretty good, with multiple steps I think you could achieve a good temperature profile.

And perhaps with a keep warm function, maybe to just keep the fluid above ambient, for faster throughput.

Maybe some sort of better lid / cooling design to try to minimise the loss of fluid.

Just thoughts really.
Maybe convert an ultrasonic tank/bath unit. They can be about the right size and have lids and picked up fairly cheap.
I am surprised that their is not a low cost, maybe smaller, version available aimed purely at hobbyist. I think it would really work. Could be fairly small really.

Interesting. I did look around to see if stainless tanks are available off the shelf that may suit, but not found much.


Interesting, what's available:
https://www.ebay.co.uk/sch/i.html?_from=R40&_trksid=p2332490.m570.l1313&_nkw=untrasonic+tank&_sacat=0

[ace1903]

Interesting approach using deep fryer  shown on video on YT
https://youtu.be/jiEU2uqPdsE?t=1531
I am also overthinking solution with special heaters and cooling system.
Seems that deep fryer with PT100 sensor and on/off control will do the job perfectly.

[woody]

I didn’t have time to build my own.

Same here. I started to realize that, while I might be able to make anything (within limits) it sometimes makes more sense to just get an acceptable solution and start using it. Most aspects of a VP oven are relatively simple as the smart is in the fluid, and it would not be too complicated to build one, but buying one enabled me to solder right away. I never looked back.

Then again, everybody's miles, dollars and entrepreneurial instincts vary.
 

[wraper]

Interesting approach using deep fryer  shown on video on YT
https://youtu.be/jiEU2uqPdsE?t=1531
I am also overthinking solution with special heaters and cooling system.
Seems that deep fryer with PT100 sensor and on/off control will do the job perfectly.

Modifying deep fryer should be fine IMO. However you must not use something with immersible tube heater as it will require a ridiculous amount of extremely expensive Galden and thermal inertia will go through the roof too. You want cooling BTW, otherwise it will cool down forever and more Galden will be lost.

[trevwhite]

Whilst the Imdes build quality is questionable I think the oven works and for that reason I would copy its construction as a starting point. I would build one for immediate use and another for experimentation.

It really is a simple oven and it works okay, especially for hobby use.

[hobbyelectronics]

Whilst the Imdes build quality is questionable I think the oven works and for that reason I would copy its construction as a starting point. I would build one for immediate use and another for experimentation.

It really is a simple oven and it works okay, especially for hobby use.

Yeah, all very true, the Vaporflow is not cheap (Or the IMDES), but its an out of the box solution - That works.
And, I still have to say the IMDES Mini we purchased in work just to do a few boards, amazed me. The results were incredible really. Nice shiny joints, that looked spot on. Had no issues at all with the actual finishes boards.

LS230 is magic!.

[trevwhite]

Vapor phase really is magical stuff. It's a very clever solution ( pun intended ).


I was drawn to it because one profile works for any size/density of board.

[sam512bb]

Good day All,

I do not know if anyone found this past , but dated, thread on VP:

https://www.eevblog.com/forum/projects/vapour-phase-soldering/

I thought it might be of interest.

Cheers,

Sam

[glenenglish]

Interesting thing happened recently
I had identical boards in two IMDES machines . In one VP  oven,  the hermatically sealed relays (10mm long, 5mm tall, 8 mm wide, quite light weight )  seemed to float off the PCB onto other areas of the PCB - they moved about 20mm....
The Galden level was higher  (and 50%  higher than the depth probe they give you)  than the one that had no problems, I'm putting that down to some bouyancy occuring in the process.
Have not investigated much yet on that apart from noticing the levels was higher. fix the galden level, and my relays stopped floating.

[hobbyelectronics]

That's interesting info, thanks.
I assume the the top of the liquid was still below the PCB?

One thing I did notice is that with the imdes we have at work, we never measure the liquid level, just pour it in and hope for the best.
Its never changed the out come of the boards. But then we never placed sealed components that may have air inside and could maybe move around more easily in the dense vapor???

Looking at the measuring stick that is supplied with the VaporFlow, you really don't need much liquid in the bottom of the chamber. Looks like its only a few mm. They also give a figure in grams and ml I think as well as the measuring stick.

Out of interest, we use very fine micron, metal disposable paint strainer to filter the fluid (like a coned coffee filter and much cheaper). We also clean the tank when we empty the fluid to filter it.

[jpanhalt]

Voted based on my very limited experience with a T962(modified).  A VP reflow would have to get ready to run fast and be stable for long periods of non-use.  It would have to be clearly better than my T962 that I got new for $100.

[woody]

Out of interest, we use very fine micron, metal disposable paint strainer to filter the fluid (like a coned coffee filter and much cheaper). We also clean the tank when we empty the fluid to filter it.

May I ask, how often (in relation to how often you use the oven) do you clean the Galden fluid?

[IconicPCB]

Do any oven manufacturers specify volume of liquid versus resulting volume of vapor?

[wraper]

Do any oven manufacturers specify volume of liquid versus resulting volume of vapor?

There is no such thing, garden must never evaporate entirely. If it happens, Galden will overheat and decompose into highly toxic hydrogen fluoride https://www.dir.ca.gov/dosh/dosh_publications/Hydrogen-Flouride-fs.pdf and Fluorophosgene .

[IconicPCB]

It is well known that Galden fluid is thermally liable, but that was not in question.
The problem is to determine the absolute minimum volume of fluid in an oven.