Reflow soldering in kitchen with ceramic hub

[tatus1969]

I do have a reflow oven (the old PCB-Pool mini oven "convertible"), but I realized that it has one big disadvantage when soldering prototype boards: I cannot access them while reflowing. As I find this possibility very handy to fix a tombstoned part or to check correct floating of a QFN chip, I came up with another solution.

Here is the complete process in a series of pictures, plus a video. The board shown is the battery management system as part of my current project https://www.eevblog.com/forum/projects/introducing-my-project-'kicksurfer'/msg1170253/#msg1170253.

Preparing a fixture for solder paste application


Aligning the stencil, fixing it with tape


Inspection of correct alignment, pushing down the stencil


Using leaded solder paste for my prototypes, that just makes the process easier; store in refrigerator, let it warm up and thoroughly stir before use


Applying a blob of paste - note the board added to the right, which is to support the stencil and keep it from flexing in the next step


Spreading the paste, the right angle is important, move slowly, distribute pressure evenly, go all the way in one pass without stopping


Finished spreading


Stencil removed - be careful here as well, move slowly but steadily


Close inspection of critical structures - this is a 0.4mm pitch QFN


Preparing component placement


Placement - my hands are not too steady (you'll see that in the video), so I use my other hand for stabilization, creating sort of a bipod


Fully populated board, not all components are spot on but that can be fixed during soldering


Video showing the reflow process in our kitchen. I did some tests with an infrared thermometer before, in order to find the right power setting. It's a pity that it is limited to single-sided boards
https://youtu.be/hptbgbDvask

Finished board!

[KL27x]

Mmmm. Nothing like the smell of burnt FR4 and solder flux in the kitchen. You're not married, are you?

[tatus1969]

Mmmm. Nothing like the smell of burnt FR4 and solder flux in the kitchen.

Not if you do it right And clean up with flux remover afterwards...

You're not married, are you?

Yes, I am! Since almost 17 years now. I have done worse things in our house, like having a 250W power resistor explode in our living room (that one in my profile picture). While she was watching TV. We couln'd use that room for the rest of the evening.

[ar__systems]

Your wife deserves a medal for patience

Not only it is in-hygienic, it stinks the kitchen, contaminates surface, but more importantly you can't control heating rate. Look, your board was done at 2:20. Compare that time to the recommended re-flow profile.

[Nusa]

So the iron in the pins and contacts is being used to solder the board. That sounds like an extremely high thermal cycle for the chips themselves. Obviously getting away with it, at least so far.

[ar__systems]

So the iron in the pins and contacts is being used to solder the board.

No. Solder paste.

[tatus1969]

Your wife deserves a medal for patience

Yes she does. But she sees and acknowledges that electronics is my passion.

Not only it is in-hygienic, it stinks the kitchen, contaminates surface

Basically yes, although we do not cook our food directly on the hub I am using no-clean solder paste though. The flux component does not smell, and can easily be cleaned.

but more importantly you can't control heating rate. Look, your board was done at 2:20. Compare that time to the recommended re-flow profile.

I also saw that the reflow time was too long, when I watched the video. But that is not caused by the soldering "process", but by me tinkering around with the components. I should have done the boards sequentially.

The peak temperature duration can easily be controlled by removing the boards from the hot area. It is in fact shorter than with the PCB pool oven. And the thermal stress could be lower than using infrared (nicely heating all the black packages first and much later the reflective metals), because board and components are more evenly heated.

Surely, is is far from any JEDEC profile, but I have been using it for a while without problems so far.

[tatus1969]

So the iron in the pins and contacts is being used to solder the board. That sounds like an extremely high thermal cycle for the chips themselves. Obviously getting away with it, at least so far.

As mentioned, it is using solder paste. Peak temperature was set to roughly 200 ~ 210°C.

[Nusa]

So the iron in the pins and contacts is being used to solder the board. That sounds like an extremely high thermal cycle for the chips themselves. Obviously getting away with it, at least so far.

As mentioned, it is using solder paste. Peak temperature was set to roughly 200 ~ 210°C.

I thought inductive heating only worked on ferrous metals? Are you saying the tin or lead in solder is being heated directly as well?

[tatus1969]

So the iron in the pins and contacts is being used to solder the board. That sounds like an extremely high thermal cycle for the chips themselves. Obviously getting away with it, at least so far.

As mentioned, it is using solder paste. Peak temperature was set to roughly 200 ~ 210°C.

I thought inductive heating only worked on ferrous metals? Are you saying the tin or lead in solder is being heated directly as well?

it is a conventional hub, not inductive. the hub heats the pcb, the pcb heats the solder paste.

[ar__systems]

I also saw that the reflow time was too long, when I watched the video. But that is not caused by the soldering "process", but by me tinkering around with the components. I should have done the boards sequentially.

Actually I meant that it the boards were done too fast

Every type of flux stinks Those fumes that you can see in the video, are you saying they are odorless? Well even so you don't want to inhale those.

[rs20]

I'm surprised tombstoning is such a big issue for you that it warranted such a radical change in process for you. Having reflowed hundreds of SMD 0805's and 0603's, I've only ever had one tombstone, and even then it was trivial to fix with a hot air gun (two soldering irons would have sufficed too). And I do all my soldering with a hot air gun, which ought to be unusually susceptible to tombstoning since the spatial temperature gradients across my board will be much higher than in a half-decent reflow oven. Is it possible that you're using bad paste? 

Interesting from here:

Solder paste solution

Eliminating tombstoning or at least greatly minimising it can be achieved in part through the solder paste recipe. First, use a solder paste with a thermally stable flux system that will maintain tackiness. Then, use a solder paste with metal particles with two different eutectic points: 50% melting at 179°C and the balance melting at 183°C - alloy after reflow: Sn62.5/Pb36.5/Ag1.0.

The angling effect caused by the wetting force in the joint that melts more quickly is mechanically hindered by solid particles of the alloy that melts at 183°C. This provides the alloy that melts at 179°C on the other pad with a fraction of a second more time to wet as well, thus restoring equilibrium.

As for QFNs, I just use extra-long pads and drag solder any bridges away along the lengthened pads.

[Nusa]

So the iron in the pins and contacts is being used to solder the board. That sounds like an extremely high thermal cycle for the chips themselves. Obviously getting away with it, at least so far.

As mentioned, it is using solder paste. Peak temperature was set to roughly 200 ~ 210°C.

I thought inductive heating only worked on ferrous metals? Are you saying the tin or lead in solder is being heated directly as well?

it is a conventional hub, not inductive. the hub heats the pcb, the pcb heats the solder paste.

Aha, that explains my confusion. My brain heard inductiive when you called it a hob on the video title. Partly because common cooktop terms are a bit different in American English.

So it's the frying pan method without the frying pan. That I understand. I cook with gas, so I'd still need a frying pan to do that.

[tatus1969]

So the iron in the pins and contacts is being used to solder the board. That sounds like an extremely high thermal cycle for the chips themselves. Obviously getting away with it, at least so far.

As mentioned, it is using solder paste. Peak temperature was set to roughly 200 ~ 210°C.

I thought inductive heating only worked on ferrous metals? Are you saying the tin or lead in solder is being heated directly as well?

it is a conventional hub, not inductive. the hub heats the pcb, the pcb heats the solder paste.

Aha, that explains my confusion. My brain heard inductiive when you called it a hob on the video title. Partly because common cooktop terms are a bit different in American English.

So it's the frying pan method without the frying pan. That I understand. I cook with gas, so I'd still need a frying pan to do that.

That makes sense. I googled that word because I didn't know it. But shouldn't Google be American biased?

[tatus1969]

Actually I meant that it the boards were done too fast

Solder starts melting at 1:09, and I remove the boards at 2:31, is that too short?

Every type of flux stinks Those fumes that you can see in the video, are you saying they are odorless? Well even so you don't want to inhale those.

Not odorless, but nothing that fills the room noticeably. You have to stick your nose into it to smell it. I also use hot air gun soldering and that produces the same amount of fume. Really not a big deal to me.

[tatus1969]

I'm surprised tombstoning is such a big issue for you that it warranted such a radical change in process for you. Having reflowed hundreds of SMD 0805's and 0603's, I've only ever had one tombstone, and even then it was trivial to fix with a hot air gun (two soldering irons would have sufficed too). And I do all my soldering with a hot air gun, which ought to be unusually susceptible to tombstoning since the spatial temperature gradients across my board will be much higher than in a half-decent reflow oven. Is it possible that you're using bad paste? 

I do not have excessive tomb stoning, there is none in the video if you look closely. I just adjusted some parts that I had not placed straight enough before. Please don't treat this thread too seriously, I had this idea out of curiosity and just wanted to share it with you. I thought it could be interesting for someone not owning a reflow oven or a hot air gun. Soldering SMD with an iron is a PITA. Stencils have become so cheap nowadays and I encourage everybody to try this or another type of DIY reflow technique, instead of thinking that SMD stuff at home is out of reach and stick to vintage all through hole component boards. This was also the reason for the picture series, I hope it encourages the one or other to try this out. All you need is solder paste and the stencil, maybe some spare FR4 material, The rest should be available in a typical household, including the kitchen 

[tatus1969]

As for QFNs, I just use extra-long pads and drag solder any bridges away along the lengthened pads.

Thanks, I also noticed that, I had quite a few solder bridges and will change the design in this direction. I again did the mistake to follow the datasheet recommendations. It was a 100u stencil, maybe a bit too thick for that pitch either. It works perfect for .5 QFN though, so I was optimistic.

[janoc]

Nice hack, but I think the last thing I would want in the kitchen is leaded solder paste. Leaded solder alone is not very toxic unless you literally eat it but the tiny particles of lead in the paste are a different matter. They are both more reactive than normal solder wire due to their much larger surface area and the small size allows them to get in the body much easier. It is very easy to smudge it somewhere by accident and never clean it completely, especially if the surface is not perfectly smooth. Then you (or your kid/wife/etc) puts food there ... Or it dries out after a while and you inhale the dust when it gets disturbed. It won't kill you outright but a slow lead poisoning is not fun.

Do yourself a favor and get a dedicated hotplate for this kind of work - and keep it out of anywhere where food is routinely handled.

[tatus1969]

Nice hack, but I think the last thing I would want in the kitchen is leaded solder paste. Leaded solder alone is not very toxic unless you literally eat it but the tiny particles of lead in the paste are a different matter. They are both more reactive than normal solder wire due to their much larger surface area and the small size allows them to get in the body much easier. It is very easy to smudge it somewhere by accident and never clean it completely, especially if the surface is not perfectly smooth. Then you (or your kid/wife/etc) puts food there ... Or it dries out after a while and you inhale the dust when it gets disturbed. It won't kill you outright but a slow lead poisoning is not fun.

Do yourself a favor and get a dedicated hotplate for this kind of work - and keep it out of anywhere where food is routinely handled.

Don't worry, I am aware of the toxicity of leaded solder paste, and I will not allow that anyone gets in touch with it and work cleanly. And I don't smoke, so I am also not afraid that the fumes will harm me. If I would be doing this day by day, it would be a completely different story. I don't feel too good when cleaning the boards with flux remover and no gloves, but that is nothing compared to what women do with their nail polish remover.

[janoc]

Don't worry, I am aware of the toxicity of leaded solder paste, and I will not allow that anyone gets in touch with it and work cleanly.

I know you are not an idiot but you obviously don't understand - accidents and mistakes happen even to the best of us. Imagine that you trip over your cat (=dog/kid/wife/own shoe ...) on the way to the hob and drop a pasted up board on your kitchen counter. You clean the mess up but it is never perfect and traces of that lead will stick there. And then you or your wife will be handling food there. I somehow doubt that you are covering everything in the kitchen up before working there.

And I don't smoke, so I am also not afraid that the fumes will harm me.

I didn't mean fumes, those are not likely to be too toxic (or at least any different from normal leaded solder). I meant dried leftovers of the paste, e.g. after an accidental spill, aka dust particles. Those you will inhale or eat if you disturb them.

Toxic chemicals and food just don't mix - a small electric hotplate for your bench will cost you perhaps 20 euro and save you all this aggravation.

If I would be doing this day by day, it would be a completely different story. I don't feel too good when cleaning the boards with flux remover and no gloves, but that is nothing compared to what women do with their nail polish remover.

That's just acetone - not good for your skin but that's about all.

[KL27x]

If it's done in 2:00 that is pretty fast. I hardly ever use my reflow oven. I forget how long it takes, but it's more like 5 to 10 minutes before the "flash" and then another 5 minutes cooldown? And boy does it stink to high heaven. Smells like burnt capacitors.

I would be worried about the solder paste splattering when it reaches sizzling temp, though.

I suppose a "hob" reflow would be mostly IR + direct contact heating? I'm sure it works fine. But I wouldn't be able to de-tombstone the parts while the pcb is on the hob. Without somewhere to anchor my hand and at least a mag lamp, I wouldn't stand a chance.

Honestly, I would probably hand solder those boards faster than stencil and pasting.... tinning and hot air for the QFN's. This why my oven doesn't see much daylight.

Yes, I am! Since almost 17 years now.

Heh, Congrats! My SO is the kitchen gestapo. I rule the garage. I win.

[tatus1969]

I forget how long it takes, but it's more like 5 to 10 minutes before the "flash" and then another 5 minutes cooldown? And boy does it stink to high heaven. Smells like burnt capacitors.

Same problem with the PCB pool oven. And when you want to accelerate cooldown by opening the door, the vibration ruins the soldering.

I would be worried about the solder paste splattering when it reaches sizzling temp, though.

I have never observed anything like this.

Honestly, I would probably hand solder those boards faster than stencil and pasting.... tinning and hot air for the QFN's. This why my oven doesn't see much daylight.

Possibly. But I consider the reflow process being more reproducable, reliable and less heat stressing both components and board. Total time for three boards was roughly 30 mins for paste application, 1 hour to place all components, 10 minutes for soldering. And it scales better when doing more identical boards.

Yes, I am! Since almost 17 years now.

Heh, Congrats! My SO is the kitchen gestapo. I rule the garage. I win.

If I had a garage with electricity, I would immediately emigrate to it. Needs to have a postal address then for the pizza delivery

Toxic chemicals and food just don't mix - a small electric hotplate for your bench will cost you perhaps 20 euro and save you all this aggravation.

I totally agree to that, and that is also a good idea. But again, I am not too concerned about poisoning from leaded solder paste, there is probably more heavy metal in food like fish or pork liver and I still like to eat them. Lets summarize for others, reading this thread, that working with solder paste in the kitchen is not advisable.

[janoc]

I totally agree to that, and that is also a good idea. But again, I am not too concerned about poisoning from leaded solder paste, there is probably more heavy metal in food like fish or pork liver and I still like to eat them.

Considering that leaded solder paste is 60% lead, umm, I think your math is wee bit off. If a fish contained 60% of heavy metals, I don't know ...

[tatus1969]

I totally agree to that, and that is also a good idea. But again, I am not too concerned about poisoning from leaded solder paste, there is probably more heavy metal in food like fish or pork liver and I still like to eat them.

Considering that leaded solder paste is 60% lead, umm, I think your math is wee bit off. If a fish contained 60% of heavy metals, I don't know ...

does your math take into account that I don't eat solder paste like fish?

[james_s]

does your math take into account that I don't eat solder paste like fish?

You're missing out, I like to spread it on crackers