News:

Forum may be experiencing issues.

Main Menu

Sunking problems...

Started by pauloman80, March 20, 2011, 07:46:22 PM

Previous topic - Next topic

pauloman80

Yes, I did check it.  My friend brought his DMM over and we used its continuity checker to inspect all the specific points discussed thus far in this thread (thanks to you fine smart gents) and also several additional ones he thought to check.  According to the continuity meter, everything appears to be as it should be, except for the voltage not being what it should be.  No clue.  He's at a loss, and I've been at a loss for weeks now.  I've got another board, I suppose I could start all over again... argh.  At least the wiring is pretty much done, I would just have to reattach it to the new board.  Gonna take extra time getting all the parts and making sure the values are as correct as possible.

But I'd still love if someone on here is willing to take a look at this thing.
--------------------------------------------------
Regards,
pauloman80

jkokura

If the voltages aren't right, there are only a limited number of problems there could be. Either a part is wrong, backwards, or faulty. It's in there. I'm sorry you might be feeling so frustrated, I certainly understand.

Only one project has beat me as of yet. I have solved every problem I've come across, except one where I had to use a new board. I have one project that is currently not working but I hope to get it working. It's perhaps the biggest single project I've every pursued, and I'm not going to be getting another board.

Jacob
JMK Pedals - Custom Pedal Creations
JMK PCBs *New Website*
pedal company - youtube - facebook - Used Pedals

gtr2

What type sockets are you using?  It seems like a silly question but I'll explain...

Machine-Pin  http://www.smallbearelec.com/Detail.bok?no=100

or

Dual Leaf  http://www.smallbearelec.com/Detail.bok?no=144

I've had the occasional problem with the dual leaf version.  Once you install, pull out, and reinstall an IC in this type a few times or even once the leafs bend back.  I've tracked down some problems with the IC not making contact with a "broke in" socket.  Test for continuity between the pins on the top of each IC and the solder joints on the underside of the board.  Do this for each pin on each IC.

It's worth a shot if you have this type of socket.

Josh
1776 EFFECTS STORE     
Contract PCB designer

pauloman80

#48
Quote from: jkokura on April 08, 2011, 03:20:46 AM
If the voltages aren't right, there are only a limited number of problems there could be. Either a part is wrong, backwards, or faulty. It's in there. I'm sorry you might be feeling so frustrated, I certainly understand.

Only one project has beat me as of yet. I have solved every problem I've come across, except one where I had to use a new board. I have one project that is currently not working but I hope to get it working. It's perhaps the biggest single project I've every pursued, and I'm not going to be getting another board.

Jacob

Thanks, Jacob.  I know for a fact that all my parts are correct; I've quintuple-checked every single one over the course of this troubleshooting foray.  I also know that none of my polarity-critical parts are hooked up backwards, and the pots are also wired up properly.  I had a brief moment yesterday where I thought I did it wrong, but I didn't, much to my relief and frustration (relief because it was right, frustration because I still haven't tracked down the issue).  

It is certainly possible that a part is faulty, but we haven't been able to track that sucker down.  I gotta say, seeing Josh's post right after yours regarding IC sockets and knowing that I ended up with IC sockets that look kinda "bargain bin" in comparison to the machine-pin version, it's very possible there's an issue with the sockets I used.  Pin 6 on IC3 and Pin 3 on IC1 are still not what they should be, and we followed the entire VB trail to figure out if Pin 6 on IC3 was grounding, and it's not.  We checked the continuity of the circuit from the input all the way through the negative feedback loops of both halves of IC1, and we checked the entire charge pump circuit.  Everything seems normal except for those two particular values.  

That leaves me with four options as to why it's not working:

1.) The board is bad/has a bad trace/is being a big fat stupid doo-doo head;
2.) One or more of the IC sockets I used are bad as Josh explained in his latest reply, I'll still need to do a continuity check between the IC pins and their solder joints to determine if that's it (thanks for the suggestion, Josh!);
3.) There's a ridiculously obvious/easy/simple reason other than the ones submitted thus far that we simply keep glossing over and missing and will feel incredibly embarrassed/stupid for missing, but we'll do so while playing through the pedal with the pots all fully dimed;
4.) There's a infinitesimally obscure and bizarre reason other than any other explanation submitted thus far that we haven't uncovered for whatever reason and would be perfectly understandable if we didn't figure it out until wayyyy after the fact.  

Conclusions: I'm going to do a refresh of my parts and start building another one with the other board I have, checking and double-checking the values of every single part before populating it onto the board.  Speaking of which, I noticed that nearly every build I've seen thus far has these thick red rectangular caps for C4, C5 and C6.  Where can I find those?  Took me forever to find the ones I used.  Please share!!

If ANYONE wants to take a stab at troubleshooting this existing board for me, I'll detach it from the wiring and ship it to you.  I'm 99.999997% positive the issue is with something on the board and has nothing to do with the wiring.

***EDIT: I'm remembering something odd we encountered yesterday that leads me to believe I might have installed crappy IC sockets.  We removed all three ICs when we started troubleshooting, my friend wanted to take them totally out of the equation on our first go.  Upon putting them back in, we discovered Caps 19 & 20 got really hot.  IC3 wasn't quite as hot, but was getting there.  Methinks the socket might be crap like Josh explained.  I've been successful so far at carefully removing parts from the board without damaging or lifting the solder pads, so I think I need to get some better IC sockets from my friend here (he snatched up all the good machine-pin sockets from our local electronics surplus store, the hoser  ;D ) and replace the one that's holding IC3 to the board.  Might not be until next week, but we'll see.
--------------------------------------------------
Regards,
pauloman80

gtr2

#49
Hopefully you'll get it working.  Here are my opinions in a different order than asked in your post.

2)It's not that the sockets are bad per say, it's just that the leaf type isn't good for a lot of swapping.  You can however get some sort of small tool in there and bent the leafs back if you find this is your problem.  Desoldering sockets is a pain.  You have to destroy the socket on the top so you can isolate each pin to remove.  I'd opt for bending back the leafs if this is your problem and not replacing it unless you have no other options.

1) The board being bad isn't even an option to consider...

Conclusion) You may need a different IC in IC3.  This build seems very picky with the charge pump (IC3)  This could be the problem.  I would try the ICL7660SCPA. I believe you have a comparable substitute right now?  Go with what is verified at this point.

Josh
1776 EFFECTS STORE     
Contract PCB designer

pauloman80

#50
I hope I can get it working as well.  I started work on this thing months ago.  Meanwhile, my building buddy has built his own (which fired right up) as well as about 15 other pedals, including one he redesigned himself based on a Landgraf Tubescreamer clone.  Sheesh.  

Responses to your opinions:

2.) I don't even know if the ones I have are the leaf type, but then again, I don't know how many different kinds of sockets there are.  The ICs go into each socket pretty smoothly but very securely, and they appear to fit nice and snug.  If removing the socket is that difficult I don't think I want to go that route, especially considering how tightly Madbean was able to condense the layout.  I'd wreck the components surrounding the socket.  Not worth it.

1.) I didn't rule out the board being pad for two reasons: first, there's been a total lack of any other explanation for the problem up to this point; and second, my building buddy has encountered bad fabbed/manufactured boards before, and he's been building a lot longer than me.  In spite of the fact that Madbean (as well as you and a good portion of the other members of this site) is a veritable genius compared to me and I love what he's done for the building community and that he's the man, mistakes happen in the manufacturing process.  

The LTC1044 is the closest comparable substitute I currently have for the charge pump.  I originally couldn't find any 7660SCPAs, so I got 7660CPAs instead and wasn't aware they wouldn't work properly until I put them in and got the high-pitched squealing, which was the first issue I discovered when powering up the build.  It's very possible that having the wrong charge pump may have fried something else, but I've replaced the TL072 in IC1 twice since switching to the LTC1044 (which according to the build doc is an equivalent substitute for the MAX1044 as long as I sub a 9V Zener in place of the 12V Zener at D3, even though I tried both and it didn't seem to make a difference either way).  

I think the main conclusion I can reach is this: the signal just doesn't seem to be getting the clipping it should through IC1, and it appears to be bypassing D1 and D2 entirely.  It's almost as if it's skipping that entire part of the circuit and is going straight from C6 to the junction at Pin 2 of IC2, because Tone and Volume are both performing as they should, which would be great if this were just a boost pedal with a tone circuit.  :P  The signal path from Input to Pin 3 of IC1 all has the right continuity and appears to have the right voltage until it actually reaches Pin 3, where the voltage is barely half what it should be.  Looking at the schematic, I don't quite understand how Pin 6 of IC3 would cause Pin 3 of IC1 to act strangely, but then again, we haven't been able to determine why Pin 6 of IC3 isn't showing the right voltage either.  If I can get my hands on some SCPAs and put them at IC3, I'll be back to the verified parts list.  Then, who knows?  Gotta try it.  But I need to find those friggin' elusive 7660SCPAs!  No idea where to find them, and I have searched exhaustively both online and locally.

The only other thing I can think of is perhaps the 68n caps I have are bogus or mislabeled.  Maybe their value is much too high and there isn't enough signal getting to the rest of IC1.  But, I don't have a meter to check their values; it seems to me I'd see incorrect voltages in more places than just two pins on two separate ICs; and I don't see how it could explain the wrong voltage on Pin 3 of IC1 as the cap comes after it in the circuit.  Thoughts?  Where do you get your 68n and 390n caps from?  Those were the caps I had the biggest amount of trouble getting my hands on.

If switching to the verified IC3 and replacing those two 68n caps fixes the problem, SWEET.  If not, well, I have no idea.  
--------------------------------------------------
Regards,
pauloman80

gtr2

How about this:

http://www.mouser.com/ProductDetail/Microchip/TC7660SCPA/?qs=iPotdFlfUx98hlix9UcWJA%3d%3d

I got lucky at futurlec and I got a 7660scpa but I wouldn't go that route.  It's not labeled as a scpa on the website, just lucky...  Unfortunately I only ordered one for some reason or I'd share.  :)

68nF take your pick, you just need to convert to uF to find them

http://www.mouser.com/Passive-Components/Capacitors/Film-Capacitors/Polyester-Film-Capacitors/_/N-5g7w?P=1z0wq35&Keyword=capacitors&FS=True

Conversion chart http://www.justradios.com/uFnFpF.html

Josh

1776 EFFECTS STORE     
Contract PCB designer