Manually controlled toaster reflow oven

[Peabody]

I manually solder SMD boards now, but would like to be able to reflow maybe two or three times a year, all just prototype stuff.  The cheap Chinese ovens apparently require major modifications to work right, and the DIY Arduino-controlled toaster reflow ovens I've seen on Youtube are pretty elaborate and actually not all that inexpensive.

But I found this video by Andreas Spiess where (at about 2:20) he describes just manually controlling a cheap toaster oven with the aid of a $4.50 thermocouple temperature gauge:



I wonder whether this might be a practical solution for my very limited needs.  I've seen posts about using a hotplate for reflow, and I believe I remember a Mike's Electric post from a while back where he says he just turns on the oven and lets it go until the solder flows, then turns it off and opens the dore.  If those methods work, then it seems the Speiss method certainly ought to work.  I would be using only leaded paste.

I can't imagine that Spiess' Aldi oven is still being sold, so I would like to get something from Walmart, probably Black & Decker.  I assume it should have quartz heating elements, but I don't know whether there should be two or four elements, or whether it should have a convection fan.  Is anyone else using the Spiess method?  If so, can you tell me what oven model you use?

[tggzzz]

I use:

• gas hob, for easy/fast heat control
• saucepan with glass lid, to keep the air above the board hot and still be able to see when the solder flows
• handheld IR thermometer
• a few mm of sand in the bottom of the saucepan, to spread heat and avoid hotspots, and possibly to buffer the heat source and make the temperature more controllable

It works well enough for my purposes. The IR thermometer is optional.

Given the low cost (one dedicated saucepan), easy availablity, and small space, it is probably worth trying before spending a lot of time/money on a reflow oven. OTOH, if you already have an oven and are prepared to thoroughly clean it afterwards, then consider just "sucking it and seeing".

[Peabody]

Well your method is an indication that Spiess is right that the temperature protocol isn't that exacting.  But of course the maximum temp would matter.

I don't see any consensus online about the number of heating elements or the convection fan.  But I guess four elements and a fan would let you reflow larger boards, assuming it can heat up fast enough.

[t1d]

My brother and I built a DIY unit. I bought two of the exact same Quartz element model toaster oven, at the thrift store. I doubled the elements, in one of the ovens; 2 above and 2 below. We used a PIC18f4550 Microcontroller, with a custom .jar GUI program, running on a laptop. It has PID control. The chip is able to drive, directly, the control input, of a Crydom solid-state relay. A Max6675 converts the K-type thermocouple.

The build went well and all facets operated as planned, except two... The oven does not heat fast enough, nor does it cool fast enough. This is due to needing more elements and forced air cooling. Actually, convection heating would be a big help. Insulating the box improves heating, but slows cooling. So, the result is that the cycle, if not controlled manually, suffers significant overshoot and time lag.

However, it does a wonderful job! I really like it and use it all the time. But, I do not have to certify the heat cycle, to anyone.

The initial heat rise period lags and, when it recovers, the temperature rise is too fast. This could lead to shock. The rise to the reflow stage is usually acceptable. The fast rise to reflow lags, but is usually on-slope. But, there is extreme overshoot, here, due to the short 10 second duration. I manually operate the door, to prevent overheating; that works well enough. I have never had a component failure.

The heat cycle is stated by each component manufacturer, in the components' data sheets. The purpose of heating at certain temperatures and speeds is to eliminate thermal shock and over-heating stress, during the high temperature solder reflow period.

I have wondered, if a convection heat gun, in an old toaster oven body, sans heating elements, might not do the trick.

I say keep it simple, but certainly do it. SMDs are cheaper and take up less PCB real estate. The process is actually easier than hand soldering, IMO. I turn out some beautiful boards.

[Peabody]

Thanks very much for the post, t1d.  I guess if there was some prospect of using an oven "all the time" I would be more open to undertaking a major project like yours.  But that's just not in the cards.  However, I belong to a local OSH group, and just confirmed that none of the other members have any reflow facility either.  So maybe a group project will make sense.

[Clear as mud]

My girlfriend and I used a manually-controlled toaster oven to reflow probably about 200 boards.  Most of the boards were 50x80mm or smaller.  We were going to build a controller for the oven, and actually bought most of the parts for the controller.  But once we started using the toaster oven, we realized it would be better to start with a different oven, something that wouldn't require so many modifications to be a suitable reflow oven.  I still am a bit upset about one nice laboratory convection oven that I didn't manage to purchase, because it was auctioned off at the same time my brother was graduating from college, and I went to his graduation instead of the auction.

Nevertheless, the toaster oven worked pretty well, we just had to watch it carefully.  It has 4 heating elements, two on top and two on bottom, and a built-in switch that turns them off when opening the door.  I would preheat the boards slowly by turning it off and on for a while, then once it got pretty warm, I'd leave it on full time until I saw the solder paste start to melt and bubble.  Watching carefully, I would barely open and close the door to cycle the heating elements, trying to keep it at a constant temperature and prevent overshoot, letting the solder paste bubble for 30 seconds or so.  Then I would turn the heat off, and every 10 or 15 seconds I would open the door for 5 seconds or so, allowing it to cool down, but not too fast.  It was all pretty "seat of the pants,"  I didn't have even a temperature gauge, although I think I had a clock or a stopwatch to judge time.

We used lead-based solder paste, and that probably helped to keep the components from overheating, too.  The other thing that helped was that on most of the boards, we had a couple of tactile buttons, but we had gotten them on closeout and they had a white finish instead of the more standard black, and they would turn brown instead of white if we overheated them.  I think the browning of the buttons probably happened at a lower temperature any other component damage (probably why they were cheap), and so we got a good feel for exactly how to operate the oven door to get it hot enough to melt the solder paste, but not hot enough to discolor the buttons.

Hopefully this was not too .

[Peabody]

@Clear as mud, thanks for the information.  It appears that manual control is a reasonable possibility.  But I'm curious whether you ever considered following what was going on with a thermocouple attached to your board.  My thought was to use something like a cheap TM-902C thermocouple with digital readout to try to follow the correct profile for the solder paste being used, but do that manually in real time.   I think the critical part is placement of the thermocouple tip.  One suggestion I've seen is to insert it into a plated-through hole so you would read the temperature of the board itself.

Do you have the model number of your oven?  Are the heating elements quartz?  Is there a fan?  And I assume you place the board so that it is pretty much equi-distant between the elements, and not directly above/below one pair.

[coppercone2]

Should a proper reflow even expose the pcb to direct ir heating? Toaster ovens kinda have guards but would the best result not be from a thermal wind tunnel that takes in clean air and blows it over the pcb so you dont have it festering in its own flux vapors and stuff?

I imagined something like a heater with thermally pressure flow matched channel that bring the hot air over the pcb in a duct system?

I figured i was gonna build one a few times but i always wondered how it should be done. Maybe even with a ir sensor matrix to regulate flow over the pcb with slide vents. You could put thermal measurement feducials on yhe pcb design with black mask or whatever so it matches to the sensor grid

I dont care at all about excessive power consumption for a non high production solder job.

I get a feeling that a bigger oven will work better so long you stay away from the edges though. Maybe 1/4 oven diameter around the pcb?

[Peabody]

Should a proper reflow even expose the pcb to direct ir heating? Toaster ovens kinda have guards but would the best result not be from a thermal wind tunnel that takes in clean air and blows it over the pcb so you dont have it festering in its own flux vapors and stuff?

My understanding is that commercial reflow "lines" operate that way.  They have successive heat zones with circulating air at the specified temperature.  However, the cheap Ebay/Aliexpress commercial reflow boxes appear to to be IR-only, with two elements overhead, and none underneath.

But I think the big problem with convection ovens for home use is that they take far too long to ramp up and properly follow the heating protocol.  At least that's what I've read various places.  So the general recommendation you see is to use ovens with quartz elements, which do produce rapid changes in temperature in both directions.

[coppercone2]

you could just butcher it put heavier wiring and double the number of elements from two ovens to make a frankenoven, for a toaster oven it would work no problem I dismantled it before.. just current in their crappy circuits rises

[Peabody]

I decided to build a manual oven or hot plate controller.  It controls the AC power to the device over periods of 5 seconds, and adjusts the duty cycle during that period according to a manual setting using a rotary encoder and 2-digit LED display.  So if the setting is 40, it would turn the oven on for 2 seconds, then turn it off for 3 seconds, then repeat that until the setting changes.  Think of it as very slow PWM.

I'm waiting for the SSR to arrive, and will initially be working with a hot plate instead of a toaster oven, but the controller could be used for either.

At this point there's no temperature feedback to the controller.  I will be reading the thermocouple temperature in real time and adjusting the duty cycle as needed.  If this "tuning" works, eventually I would hope to just load the controller with a sequence of duty cycle settings that are automatically executed to reproduce the desired temperature profile for leaded solder paste.  In theory there would still be no need for temperature feedback assuming the hot plate always begins at room temperature.

But the first step is to see if I can reproduce the target temperature profile at all - whether the hot plate can be made to rise from soak to reflow temperature fast enough without overshooting.  We will see.

[tom66]

I did reflow in my household oven.  Used a thermocouple and ramped it up to 200C then held it around 220-230C for about 3 minutes, that was enough to flow solder, then turned oven off and left it to cool naturally.

It may not produce superb joints but worked well enough to solder an "FMC" connector used for FPGA dev boards with no solder bridges, as well as a 0.4mm pitch connector with only a few bridges which were easily removed with solder wick.

[blacksheeplogic]

My understanding is that commercial reflow "lines" operate that way.  They have successive heat zones with circulating air at the specified temperature.  However, the cheap Ebay/Aliexpress commercial reflow boxes appear to to be IR-only, with two elements overhead, and none underneath.

Might give you some idea of the differences, mine is not a high end table top reflow oven (table size of 350 x 240 mm). It uses Quartz IR & Forced Hot Air Convection and required a dedicated 30A circuit. 

[Peabody]

I've made some progress with the reflow controller.  The original mode just let me adjust the duty cycle manually, but I've now added another mode that executes a specific time-based duty cycle sequence that's embedded in the firmware.  That sequence was derived from my experimenting in manual mode.  There is still no thermocouple feedback to the controller.  I'm using the Walmart $10 hot plate with exposed heating elements, topped by a circular saw blade about 3/8 inch above the elements.  The duty cycle is varied over a 2-second period.

The pictures below show the temperature profie I'm now generating with my controller, along with Kester's profile for 63/37 leaded solder paste.  The biggest difference I see is that my controller takes twice as long to ramp up into reflow from soak as Kester specifies.  Because of the thermal inertia of the hot plate and blade, if I try to ramp up faster, it overshoots.

The other major difference, for the same reason, is that cool-down is way too slow.  However, I believe the answer to that is to put on oven mitts, and simply remove the saw blade (with PC board) from the hot plate as soon as reflow is completed.  Maybe have a square of ceramic tile to put it on for cooling.

What do you think about the slow reflow ramp?  Is that likely to damage the board, or the parts?  It seemed to me less likely to cause problems than a big overshoot.  But I have no experience with reflow.

For those who may be curious, here's the duty cycle sequence that produced the profile:

68 seconds at 100% 
20 seconds at 75%   
20 seconds at 50%   
28 seconds at 25%   
76 seconds at 0%   
10 seconds at 16%   
44 seconds at 100%   
off

My idea was to have a controller automatically switch the power on and off per that sequence, with no need for feedback, PID controllers, displays, etc.  Maybe a couple LEDs to indicate what's going on, and a buzzer to indicate that you maybe need to move the saw blade off the hot plate right now.  So basically, you would need an Arduino or something similar (I used an MSP430), a voltage regulator powered by a 9V battery, a solid state relay, the hot plate, the saw blade, three screws on which the blade rests, and whatever hardware is need for connectors, a case, etc.  And you would maybe want to have a $4.50 TM-902C thermocouple with display so you could see exactly what's going on.  My controller is more complicated than that because I have a rotary encoder and two 7-segment displays,  but neither of those would really be necessary.

Well I don't know if this is all going to work or not.  The world doesn't need another PID controller with thermocouple feedback.  I was just hoping a simpler, less expensive reflow alternative would be feasible.

[blacksheeplogic]

My re-flow oven only the center portion is used.

[kellogs]

Hey everyone, I am looking to build myself such a DIY reflow oven too.

About the heating elements - to quartz, or not to quartz. This is the (ethernal) question....

Aside from that, I have got myself a quartz-infrared unit. 9 liter @ 1 kW of power. When both top and bottom heaters are in use, the damn thing shuts them down with a 18s on / 8s off duty cycle. Other times it is 8 on / 18 off and other times some other duty cycles. Its user manual has a line about the on-off duty cycle above, and it is there to avoid overheating.

Now this could be good as I might be able to reuse that temperature sensors into my controller.
It could also be bad because the entire thing seems not to be built to whitstand much heat :-|
User manual says nothing on its maximum temperature capabilities. It does say to leave the sides and back some distance from the walls for it to vent through the side/rear slots.

Do I stand a chance into successfully converting this thing into a steady and repeatable-results reflow oven or will it probably melt down on my first modified run ?