MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

Ryo Penta troubleshooting
MUFF WIGGLER Forum Index -> Music Tech DIY  
Author Ryo Penta troubleshooting
hmurdoch
Hi all

I got Penta kit recently from Thonk and proceeded to build it over the afternoon. The build seemed to go fine although my soldering iron was playing up a bit (have since sorted this out). At times the fine tip wasn't quite hot enough to melt the solder, so it's possible that I kept the iron on the PCB for too long at times, potentially damaging some parts. Anyway, when I finished putting everything together I plugged it in to test it out. Almost as soon as I did I saw smoke coming from underneath and unplugged it. I've had a look and can clearly see burn marks on R1, R2, C1 and C2.

Does have any idea what the possible cause could be? One potential problem (I think it probably will be this) is that I had only soldered the pin and socket strips to PCB2 when I plugged it in (completely slipped my mind that they would need to be soldered to the other PCB). Does anyone reckon this could have caused the problem? If so, could I simply replace the damaged components and give it another test? I've since soldered the pin and socket connectors to PCB1 but haven't tested it out again yet. The damaged components don't look completely destroyed, so maybe will still function, but there are clear black marks on them. I can post a picture if necessary.

Thanks!
Monobass
hey, missing stuff out is unlikely to make the magic smoke come out, it's more like in this case that you either attached the power incorrectly or got a polarised component backwards somewhere. posting photos of both sides of all PCBs would be best place to start.
hmurdoch
Thanks for the response, here are pics of the two PCBs (phone quality I'm afraid). As far as I can tell everything has correct polarity, and when I plugged the module in I made sure the polarity was correct. For context, I'm using Befaco power.

I also noticed when taking the photos that it doesn't look like the capacitors mentioned in the OP post are actually damaged. I think they just have burn marks from the damaged resistors.


hmurdoch
bump for tips
Luap
Its hard to tell from the picts, but is the small 8 pin IC in the correct way? Have a good look for bridged solder joints too.. I can see a couple of suspect looking joints there.
You'll almost certainly need to replace the fried resistors too.
SmartBits
IC3 (LM358) looks to be inserted the wrong way round indeed.
PhineasFreak
i agree with these other fellas - it does look like the lm358 was inserted wrong way round.

if so you will at minimum need to replace r1 and r2 and c1 and c2 that burned out and also possibly the lm358.
hmurdoch
Thanks for the advice wigglers, I'll get hold of those parts and report back once I've tried something out.
hmurdoch
Just replaced the 2 resistors, 2 capacitors and the chip and it all seems to be working well, except pot 3, which doesn't seem to be doing anything at all. the LED works but when the sequence gets to that step all that plays in the base note of the oscillator. Anyone got any ideas? Should I try reflowing each connection to that pot? Or could the pot be faulty?

Really happy with this module regardless, just had a lot of fun making a track with it.
stabilt
Apart from POT3 the bottom pin of the upper left connector between the PCBs (when looking from the front of the module) and the mixing resistor R12 is relevant for step 3.

The pictures are a bit low in resolution but I have tried to point out the relevant solder points apart from POT3 in the attachments.

Pots can be sensitive to prolonged heat exposure, especially smaller pots.
antf4rm
Just finished this build. It all works now. At first it did nothing at all. i traced the issue back to the LM2931 realizing that it was not getting 12v on the input. Beeped out the board until i found that there was no connection from C3 to C5 due to a bad trace on the pcb. After installing a jumper from the non-ground leg of C3 to the non-ground leg of C5 everything works like a charm. applause
stabilt
antf4rm wrote:
Just finished this build. It all works now. At first it did nothing at all. i traced the issue back to the LM2931 realizing that it was not getting 12v on the input. Beeped out the board until i found that there was no connection from C3 to C5 due to a bad trace on the pcb. After installing a jumper from the non-ground leg of C3 to the non-ground leg of C5 everything works like a charm. applause

Wow that is surprising, especially since the trace thickness is well over the limits of the PCB manufacturer and there is no sharp (90 degrees) angles. That should make it very low risk for etching errors. But we have noticed a decline in quality from Project PCB on the panels but this is the first time an etching error have occurred from them on a couple of thousand PCBs.

We have already decided to switch manufacturer before this. I'm glad you figured out the error. thumbs up
antf4rm
stabilt wrote:
antf4rm wrote:
Just finished this build. It all works now. At first it did nothing at all. i traced the issue back to the LM2931 realizing that it was not getting 12v on the input. Beeped out the board until i found that there was no connection from C3 to C5 due to a bad trace on the pcb. After installing a jumper from the non-ground leg of C3 to the non-ground leg of C5 everything works like a charm. applause

Wow that is surprising, especially since the trace thickness is well over the limits of the PCB manufacturer and there is no sharp (90 degrees) angles. That should make it very low risk for etching errors. But we have noticed a decline in quality from Project PCB on the panels but this is the first time an etching error have occurred from them on a couple of thousand PCBs.

We have already decided to switch manufacturer before this. I'm glad you figured out the error. thumbs up



I'm pretty sure its an error in manufacturing and not my fault (of course i could be wrong). I have noticed that the ryo PCBs are fairly easy to damage if you have to do any desoldering. They are all working now so this is by no means a complaint. I'm so glad you are aware of the pcb quality and that you are taking steps to make improvements. Right on! applause

Here is what i ended up with. I totally burned R1 when my probe slipped during troubleshooting. So i had to replace R1 during which i destroyed one of its pads. So then i had to make a second "blue heat shrink" jumper. very frustrating

Works now. Can't wait to start making some 5 step things happen.

antf4rm
There might be something funky happening. Look where those 2 traces come together at the pad on the non-ground leg of C3.
One trace coming from a via the other from C5.
That junction might have something goofy going on that you'd have to zoom in to see in the cad software. It's the only place on the board that i can see where 2 traces come together in that manner. Just guessing? hmmm.....
stabilt
We have only had one etching error before and that was some early Trig Xpander PCBs on a 90 degree angled trace only 0.22 mm thick made in one of the very cheap (maybe too cheap) chineese PCB houses. It was a misstake that there even were a 90 degree angle there in the first place as those are prone to etching errors if they are too thin.

We haven't noticed any problems with desoldering the boards. We use a weller wd1000 iron set to 365C and lead free solder with 3% silver. If you use the regular 99.3% tin / 0.7% copper lead free solder you would need 380C or if you have an iron that drop in temp alot while soldering. With leaded solder lower temperatures can be used.

We also use braid and a soldapullit solder sucker. We recently got a tiny japaneese solder sucker that was praised by many but haven't tried it yet. The huge soldapullit have an outstanding sucking force but yeah, its huge and thats a disadvantage.

Another favourite of mine is heating up the solder and quickly smash the board on the table so the solder and component lead comes swooshing out. I always destroy the component before desoldering also, it makes things so much easier, so there is only small leads left in the pads and nothing else of the component.

Damaging the boards can come from too hot iron, too long (very long, the component should be damaged before the pad is) heat exposure or pressure put on the pads. But your solder joints in that zoomed in picture looks quite good so I'm sure you have plenty of experience.


I cant see any strangeness in Eagle. The trace comes from R1 going past C1 for power noise filtering decoupling, then to C3 for current storage decoupling and then 2 traces go from there to C5 that is decoupling the input of the power regulator and the other trace via the via to one of the PCB to PCB connectors. The goof would be a trace going straight from R1 to the LM358 decoupling instead of from C3 to the LM358 decoupling but it's not a huge issue. Trying to make a star configuration but missed one power trace. As it's not an audio module I didn't put too much effort in to make it a star.
stabilt
I have gone through that trace on all 74 remaining PCB2 from the same batch with both visual inspection under a magnifier light and electrically tested them with a dmm and at least that trace is intact on the remaining ones.

Thanks once again for the heads up! SlayerBadger!
antf4rm
Wow! right on. Thanks for all the info and for looking into the issue. I'm sorry if i've sent you on a wild goose chase and wasted your time. Guinness ftw!
stabilt
antf4rm wrote:
Wow! right on. Thanks for all the info and for looking into the issue. I'm sorry if i've sent you on a wild goose chase and wasted your time. Guinness ftw!

No waste of time. If there is an etching error chances are it's only going to be on 1% or so of the boards, just wanted to make sure. It's too common that PCB manufacturers cheat with the e-testing.

Cheers! Guinness ftw!
erstlaub
I just grabbed a built one of these off the ebay but it seems to have an issue with step 3 either skipping entirely or sometimes with a bit of wiggling, it will light but send out fluctuating voltages



I'm starting a return procedure but this is a brilliant little module otherwise and I'd prefer to suggest a possilbe repair to the seller rather than having to buy another kit myself and most probably turn that into a completely non functioning pile of melted plastic with my rubbish soldering abilities.

I've given it a quick visual check and there don't appear to be any obvious bridged or bad solder joints but I'm no expert and haven't taken it apart fully to inspect.

Any thoughts I can suggest to him? You can sometimes get the wonky step to sound if you wiggle the pot a little but not always (it also *sometimes* works stably for a bit then stops, sometimes works regularly but the voltage seems to rise and fall, sometimes works but causes trills etc).

Gutted, it's exactly what I've been needing lately. Thanks in advance for any suggestions.
analoc
Soldered my Penta yesterday, what a lovely kit!

Everything works with one exception: Step 4 starts at a much higher voltage than the others, resulting in a higher pitch. Range seems pretty much the same to my ears.

Does anyone have an idea what goes wrong here? Did I fry something while soldering?

Seems also step 4 can’t hold the exact pitch - it’s floating. Any ideas?
digix
Built my Penta kit today - my first true Eurorack DIY project. With the exception of soldering in an incorrect resistor which had to be carefully desoldered and replaced (go slow and check before soldering!) everything went just fine. I powered it up outside my case to ensure no magic smoke incidents.

On the first try it was stuck at step 1 and wouldn't cycle through when clocked. I checked over all my soldering and yep - missed one of the pins of one the IC sockets. Quick fix and now working like a champ!

Looking forward to my next challenge - the Optodist build. Have all the parts - that one's going to take some time.

Cheers!
OB1
digix wrote:

On the first try it was stuck at step 1 and wouldn't cycle through when clocked. I checked over all my soldering and yep - missed one of the pins of one the IC sockets. Quick fix and now working like a champ!


I had numerous issues with mine - LEDs not behaving properly, some LEDs lit dimly while at rest, steps 1 and 4 giving a much smaller range than 2, 3 and 5, 3 step mode not working... I reflowed everything with lots of flux last night and now seem to be in the same state as you were - step 1 seems to be working fine, but it won't cycle - which pin of which IC fixed this for you?

Have really struggled with these PCBs - the quality doesn't seem as good as a lot of others I have built and have found it hard to get the solder to flow well. Hopefully I'll get this working because it seems like a nice little sequencer!
MUFF WIGGLER Forum Index -> Music Tech DIY  
Page 1 of 1
Powered by phpBB © phpBB Group