2

Background

I tested some Bourns PEC11L encoders before soldering - they were working fine. Once soldered to my PCB at 300C using Sn99.3,Cu0.7 solder, they stopped working. Looking at the datasheet, I see that I did it all wrong...

PEC11L datasheet excerpt

(In case you're interested, the encoders now almost always skip some of the states in the cycle 00-01-11-10-00.)

So now I'm considering Cui's ACZ11, TT Electronics' EN11. Neither of the datasheets for these mention soldering considerations.

Wurth's mentions wave soldering in the datasheet, but not hand soldering. I don't have a wave soldering machine.

CTS's 11CE gives manual soldering advice, but has a plastic shaft, so I can't use it.

Question(s)

Should I try the Bourns again at 250C, use no-clean flux, and make sure the iron's only applied for a short time? (How short?) Should I get some SnAgCu solder? In the absence of information on the Cui and TT datasheets, can I assume anything?

OutstandingBill
  • 355
  • 1
  • 10
  • 4
    When soldering by hand, what matters way more than your temp is your dwell time. If these are pins that go into a plastic housing, man it's super easy to melt the plastic. But it can definitely be done. The trick is to not even touch the pin. You get a flat chisel tip, clean it, put a small drop of solder on it (like the amount a fly would expell if he sneezed), and touch that little blob to the PCB's copper ring but NOT THE PIN. Let it dwell for one second, heats up your PCB. Then touch the solder directly between the soldering tip and the pin. Be quick to feed it in, and pull away. – Kyle B Dec 29 '21 at 07:50
  • 1
    Practice on scrap parts, and magnification is your friend --- Even some cheap reading glasses from the drug store if you can't afford a good loupe. – Kyle B Dec 29 '21 at 07:56
  • Ah, after reading a few times, I think I get the idea. 1. Place all the encoder pins into the PCB. 2. Put enough solder on the tip to allow it to make a good head-transferring connection to the PCB. 3. Heat the PCB ring. 4. Once the PCB ring is hot, feed in a bit of solder so that it fills the gap between pin and ring. Thus the encoder only gets conducted heat during 3 from touching the ring, and from 4 as the solder flows around the pin. That would make a great answer. – OutstandingBill Dec 29 '21 at 08:18
  • 1
    @OutstandingBill: Try it, verify it, write an answer, profit. – JRE Dec 29 '21 at 08:33
  • :) I only care it works for you brother. I felt it appropriate as a comment. You got the idea exactly right. A flat chisel tip is almost mandatory, sharp pointy tips can't conduct enough heat fast enough. – Kyle B Dec 29 '21 at 08:48
  • @KyleB that sounds like a recipe for a cold solder joint. Depending on the failure modes, that might be preferable to destroying the part, but I wouldn't recommend it. – Scott Seidman Oct 09 '22 at 15:20
  • @ScottSeidman Please do share what you find "wrong" about that and how you would do it differently. – Kyle B Oct 10 '22 at 03:58
  • @KyleB you're recommending hot solder be pulled to a cold pin. I would 1-use lead, 2-use the right iron and tip, 3-dab a tiny bit of solder on the tip for thermal conductivity, 4- heat all parts that need solder until solder melts into the joint. – Scott Seidman Oct 10 '22 at 10:52
  • @ScottSeidman Steps 1-3 reiterated what I said. Step 4 you're using solder that's now had a chance to become contaminated with oxygen. Probably not a whole lot better.... – Kyle B Oct 10 '22 at 13:31
  • @Kyle -- no, the big difference is step 3. You clearly say "The trick is to not even touch the pin", and "..and touch that little blob to the PCB's copper ring but NOT THE PIN. Let it dwell for one second, heats up your PCB. Then touch the solder directly between the soldering tip and the pin. " YOU NEED TO HEAT THE PIN. Without heating the pin to at least the eutectic point of the solder, you have a cold solder joint. – Scott Seidman Oct 10 '22 at 15:07
  • @ScottSeidman I meant your step 3... put a dab of solder on it. I didn't number the steps. The thermal mass of that pin is very low, it'll heat very fast. If the fresh solder wicks, it's hot enough. Anyhow his issue was melting the plastic.. that was happening because of his dwell time. I prescribed something a beginner can do. If it's me, I'm hitting both pieces at the same time and being very fast. What you're describing could be considered "painting with solder". You burn off the flux before you even start soldering. So maybe I get a cold pin, you might have a dirty one ;) – Kyle B Oct 10 '22 at 18:43
  • 1
    BTW - we're neighbors! I see you're in Rochester. I'm just west of you on the 90. Go Bills! (Rite?) – Kyle B Oct 10 '22 at 18:44
  • I may not have been real descriptive in comments, but my step three uses fresh solder once the joint is hot. The dab of solder I described just is to increase thermal conductivity so the pin and pad heat faster -- it's not the only solder used in the joint. – Scott Seidman Oct 10 '22 at 20:39
  • 1
    ... and this may be the Bill's big year! – Scott Seidman Oct 10 '22 at 20:39
  • 1
    I really appreciate the detailed discussion on this subject. It's been a while since I tried this approach, and I didn't have a chisel tip back then, but I thought I'd managed to keep the contact between the iron and the leg pretty brief. Have you tried soldering one of these encoders? I mean, maybe they're the most difficult thing in the universe to solder : ) – OutstandingBill Oct 10 '22 at 22:03

2 Answers2

3

I found soldering was destroying too many of my encoders, and messing up the whole PCB as a result. Instead I designed a little daughter board (pictured below) which connects to the main board via four blue wires. The sockets are SSW-103-01-T-S from DigiKey.

rotary daughter board

daughter board with encoder

OutstandingBill
  • 355
  • 1
  • 10
1

Problems like this should alleviate with experience, but I really recommend prototyping delicate parts with lead solder.

Scott Seidman
  • 29,274
  • 4
  • 44
  • 109