El-cheapo reflow oven

[toomas]

Sharing my excitement and some useful data. I just bought one el cheapo 650W 5L oven for 29 EUR for some home reflow soldering. I love it being small, it is so easy to move around. It also has a metal plate, good for more even heat. Has two quarz heaters, up and down. Glass door. On eBay, it's ridiculously cheap, just 8.50 EUR

Looks like this:


I did some temperature profiling just by turning it on, letting it heat up and then opening the door to finish. Temperature is measured with thermocouple on board. Blue line is when I had one 50x50mm unpopulated board directly on metal plate and I didn't open the door. Red and yellow had eight 50x50mm boards raised by 1.6mm above the metal plate so that there's air between the plate and board. The difference between yellow and red is just that in case of yellow, I started with boards in oven at 70C.



More pics:

All the measured temperature data and how well it fits the NC254 SAC305 lead-free solder profile (the same paste that Dave used from Beta Layout) is in this google spreadsheet:
https://docs.google.com/spreadsheet/ccc?key=0Am57pQmYi317dDI0a3VfSUYxcmdoVHhOeXJteW8tN2c&usp=drive_web#gid=1

To me it seems that even without a controller that's good enough for home use. The only real worry I have is that the temperature ramp up until 150C takes about 2x longer than that lead free profile recommends (do you think this is an issue to worry about? Flux vaporizing too early?). Everything else fits in the solder reflow profile. You just have to know when to exactly open the door. I really don't see a need for the controller, because I can always do one test run and figure out when and how slowly (or fast) to open the door and that's basically it

The actual solder paste should arrive in a week, then I can make real runs.

[Frank]

I was wondering if a 650W oven would do the job. I have one sitting around with no use and after watching the videos from Dave I think that's time to get it to work. So I will follow this topic and see if I can learn something

About the longer temperature ramp, these ovens lose a lot of heat to the ambient, as they do not have a good thermal insulation. Maybe adding some insulation would help with this matter.

[toomas]

About the longer temperature ramp, these ovens lose a lot of heat to the ambient, as they do not have a good thermal insulation. Maybe adding some insulation would help with this matter.

That could help, because this thing gets quite hot from outside. A simpler solution I'm going to try first is to put the boards in a preheated oven, at 80C or so.

[mikeselectricstuff]

About the longer temperature ramp, these ovens lose a lot of heat to the ambient, as they do not have a good thermal insulation. Maybe adding some insulation would help with this matter.

That could help, because this thing gets quite hot from outside. A simpler solution I'm going to try first is to put the boards in a preheated oven, at 80C or so.

Not sure how effecting preheat would be - it's using radiation and convection, and you lose the hot air when you open the door.
May lose a little less if the walls are already hot but doubt it would make a huge difference.

[mjkuwp]

This is a very common DIY project, almost comically so!

anyway, I have done my own oven and googled upon one really good idea - boost the heaters..

The resistance heaters are very simple devices.  My oven has 4 heaters with two circuits and each has two elements in series.   It is a 1200W oven so it could be converted to 2400W by re-wiring the heaters.  At this level it is close to tripping the circuit breaker (USA 20A circuit) so one would want to have a test to find the safety margin.

It is not the method that I did.  I took apart the elements and trimmed the heater wires shorter (original idea found elsewhere).  With the shorter wires, the resistance is lower and the power higher.  I changed from 1200W to 1700W and I program a 1.5C per second ramp. The oven can almost do this.

I insulated my oven but that is a lot of work.  If I did another one I think I would boost it more and not do the insulation.

[moemoe]

On eBay, it's ridiculously cheap, just 8.50 EUR[/url]

+49€ shipping, or 19€ inside Italy.

[mark03]

anyway, I have done my own oven and googled upon one really good idea - boost the heaters..

Yep, I'm not sure how much power is really needed if you are using IR, but if you go the convection route, I can testify that 5 kW is about right!  It does require a 220V circuit.

http://www.keteu.org/posts/reflow_oven.html

I do think PID control is a bit overkill for the IR toaster-oven approach.  The temperature on-board must vary wildly from part to part depending on IR absorbance, so I am skeptical that fine-grained control of the heaters will make much of a difference.

[hlavac]

Did try this with a 700W toaster oven, and it was too slow to be usable.
I calculated I'd need something like 2kW to ramp up fast enough to the peak temp and not exceed the maximum time over temperature constraint...

I gave up and kept using my cheap pancake making hotplate (nice big flat surface, max temp just about right for lead free solder paste)

Anyone tried these glass pot hot air circulation ovens for reflow? Looks they have the right power, and one can sort of see whats going on inside...

[mjkuwp]

...

I do think PID control is a bit overkill for the IR toaster-oven approach.  The temperature on-board must vary wildly from part to part depending on IR absorbance, so I am skeptical that fine-grained control of the heaters will make much of a difference....

The toaster oven that I have has a convection fan.  PID is a little overkill for sure but I did not know that until I watched the PID loop run many many times.  I still think it is better to have it because it frees my brain to be concerned about other things.  In my early uses of the machine I always mounted a thermocouple directly to the board i was soldering.  The solder would melt extremely close to the time at which the sensor showed the board was passing the melt temperature.  However, it takes a lot of effort to attach a wire to each board.

Recently I did three boards (one at a time, three runs) in which I left a sacrificial blank PCB in the oven next to the board I was manufacturing.  This did not go as well.  There was a big offset (maybe 10C?) between when I saw the solder melt and when the thermocouple said it would.  As you said, the absorption of the radiated energy is different between the boards.

I need to think of a better way to do this.  I may consider using one large copper plate underneath the board to be produced and also under the board with the sensor on it.  One problem with that is that any thru-hole components sticking below the board will have to be accounted for by drilling holes in the copper plate.

a work in progress for sure..

my oven is documented here, only about the 100th person to do this 


http://mjkuwp94.tumblr.com/ReflowOven

[hamster_nz]

I've done much the same, but using a home-built controller and an SSR to control the heater. It is quite a fun project

I've done a little bit of thinking and have come to a few conclusions.

* Convection has only minimal effect in the process. It is almost all  IR from the top element, and a bit of conduction through the tray. This is why Dave ended up with uneven heading when the PCB was just on the wire rack - some sort of tray is a must.

* It takes a lot of thinking about how to get an accurate temperature reading off of the PCB. In the end I opted to keep the thermocouple in the center of a 50mmx50mm PCB, using a coin about 10mm from the tip to keep it in place. I would like to use Kapton tape but a coin far is cheeper than a roll of tape!

* Even with the power off and the door open it cools quite slowly unless there is some some active air movement. Unaided convection isn't all that efficient at moving heat around.

* An in-spec cooldown is hard to achieve without a fan to move the heat out of the oven. Cooling slower than spec should not be a big problem as long as you don't speend to long around the peak. The slower the solder solidifies the finer grain structure, and should give a more robust joint.

[IanJ]

With my own home made reflow oven I changed out the heaters for ones from a bigger oven as the originals were far too slow.
Soak is still a bit slow as is the final ramp up to reflow, but it works. I don't use PID but instead a few simple way-points that the control electronics does y=mx+c between (Arduino). You can kinda see that in the profile below.

Seconds versus degC.



Ian.

[sleemanj]

I would like to use Kapton tape but a coin far is cheeper than a roll of tape!

Hamster_NZ, if you are in NZ, PM/email me your address and I'll send you some kapton offcuts from my roll :-)

[mark03]

IR is the only way a toaster oven can possibly be used for reflow.  If you take away the radiative component (e.g. by shading the heating elements and circulating the air), 1-2 kW is just not enough power for pure convection reflow.  IR does seem to work well enough for lots of people, though.

Your hair-blower idea is conceptually very similar to my reflow oven (see reply #6, above).  If anyone else tries this it would be fun to compare notes.  I'm sure mine could use further optimization of the airflow, but for now it is "good enough."

[nali]

About the longer temperature ramp, these ovens lose a lot of heat to the ambient, as they do not have a good thermal insulation. Maybe adding some insulation would help with this matter.

That could help, because this thing gets quite hot from outside. A simpler solution I'm going to try first is to put the boards in a preheated oven, at 80C or so.

I took the heaters out from my toaster oven and replaced them with the heater elements from a scrap laser printer fuser - linear halogen lights about 9 inches long and 750W each IIRC, but very low thermal inertia. Using a PIC to control the temperature gave an almost perfect profile apart from the last (reflow) stage where the last 60 degrees or so were a bit sluggish 

[JimmyM]

It just so happens that my convection toaster oven is 2.6kW and ramps like crazy. It also overshoots like crazy too since it uses 8 standard heating elements. I'm working on that. I think I need a larger Kd value in my PD controller.

Convection heats very nicely. The whole boards reflows within seconds of the first component to reflow.