Stove Top Reflow Experiment

I have been meaning to post something about this for a while …

When I made the prototype for the Servo Boosterpack, I soldered the parts using a stove top reflow method

Preparing to add some solder paste

I ordered solder paste from Proto Advantage. I ordered a syringe of Chip Quik 63/37 Sn/Pb paste and a needle kit. It was delivered right away.

I originally placed an order from Zephrytronics due to some positive mentions I saw on some electronics sites. However, after hearing nothing for several days after placing my order I contacted them and they said they had to make up some more and it would probably be another week. I didnt realize they didnt stock it when I placed my order. So, I don’t have any opinion about the quality of their products but I recommend an immediate followup email after you place an order to find out when you can expect it.

To apply the solder paste, I did not have a stencil. Since I am not planning on going into any kind of production, I didn’t think that was worth the trouble and expense to prepare a stencil. However, having been through it with the syringe now, I can understand the appeal of a stencil. Anyway, I used the syringe to apply a small blob of solder paste on each pad. I had to judge how much to add to each pad and based on the size of the pad. The solder paste was fairly thick and it was a little hard to control the application at first. Once I got the flow going though it worked pretty well as long as I kept up a rhythm and didn’t try to stop and restart.

To place the parts, first I laid out the parts on a sheet of paper with reference designators. This way I could place the parts quickly without needing to look up the kind of part for each place on the board.

Parts laid out and sorted

I used tweezers to place the parts. I started off using a magnifying glass, but ended up just taking off my glasses and using my eyes. I am extremely near sighted and without my glasses I can see very small things very close up really well (it is my superpower). That is why if you watch the video (below) you will see my face very close to the board. As with the solder paste, once I started going I developed a kind of rhythm for placing parts and got better at it as I went along.

Once my parts were placed I was ready to try to heat the board for solder reflow. I have read a number of articles about people making their own reflow ovens or just using off the shelf toaster ovens. I also read about a seemingly common technique using an electric skillet. You can read about these at Sparkfun or Adafruit or just google it. I don’t have an electric skillet so I decided to use my gas stove top.

I placed an aluminum plate down to spread the heat as uniformly as possible. Then I started the burner on as low a setting as I could and let it slowly start to warm up. I used an IR thermometer to measure the temperature of the circuit board. Note that the IR thermometer is not useful for measuring the temperature of the aluminum plate because it is shiny. The reflow temperature is 183 C so I waited until the board temperature approached that temperature. I could see the solder start to flow. Once that started I turned off the heat on the reasoning that my measurement was probably lagging and the board would continue heating for a bit before it started cooling. I let it start to cool for several minutes. Once I felt sure that the solder was not longer liquid I transferred the board to another sheet of metal that was thinner that would cool faster. The aluminum plate was taking a long time to cool.

Aluminum plate on stovetop

WARNING TO READERS: Before you try this yourself, you need to educate yourself about the risks of using leaded solder. If you are not completely certain that what you are doing is okay and are willing take whatever risks it entails, then DONT TRY THIS AT HOME! You have been warned.

A note about my use of leaded solder this way: I am aware of the hazards of leaded solder. So please save you and me the time and don’t post any OMG! you used leaded solder on your stove! comments. To reduce any risk of contamination, I covered the stove top and area around it with aluminum foil. This prevents any possible contamination of surfaces with lead. However, I think the chances of there being any real risk here are vanishingly small. First of all the solder would have to be physically transferred to the stove top or other surface. And then the same solder would have to be transferred to some food in order to be ingested by me. I don’t think this is very likely (as in extremely unlikely) but just to be sure, that is why I used the foil.

Once the board cooled off I inspected it. The solder melted for all of the components. However it appeared that the solder joints closer to the edge of the board did not look as nice as in the center so I assume there was uneven heating. I used a soldering iron to touch up where needed and also hand soldered the big inductor. Finally, I had to add a couple of jumper wires because I made a mistake in the power supply design (the board design on github has been corrected).

Finished prototype with jumper wires

One of several lessons that I learned is that “no-clean” flux does not really mean you do not need to clean the board. When I first tested the power supply it was behaving very strangely and I was getting weird voltage output that was varying and changed depending on how I was measuring it. I now realize that the flux was leaving enough conductivity on the board that it was messing up the feedback circuit. Once I cleaned the board well it was all working correctly.

Here is a video I made of the process.

Thanks for reading.