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

Troubleshoot a complex build
MUFF WIGGLER Forum Index -> Music Tech DIY Goto page 1, 2  Next [all]
Author Troubleshoot a complex build
mongrol
Hi All,
I've just finished my NLC Null-A2 and as expected, it doesn't work. I've proven power to the board but have a dead short between +12 and -12 and my cheapo multimeter (on continuity) is showing about 600ohms between GND and +12 and also GND and -12. This is measured straight off the power connector.

Is there a methodical way to track down the short? Or is it a case of tedious inspection and reflowing of suspect parts?

Does the measurements above point to any particular piece? If we have a dead short between + and - 12v's perhaps there's a component where these tracks meet?
batchas
I don't know where exactly you measure this dead short, but when you measure with your DVM straight off the power connector, you get +12v when you measure between +12v and GND (0v) ist that correct?
-12v when you measure between -12v and GND (0v) ist that correct?
+24v when you measure between +12v and -12v ist that correct?
I guess "proven power to the board" means all this is correct.

Then to start I'd first check the orientation of the caps right after the power connector (see if polarity is ok). Are these ok?

Maybe you can post a photo? Might be easier to help.
mongrol
Thanks for replying. The dead short is between the +12 and -12 pins in the power socket.

Power is proven by ensuring the correct voltages are coming out the cable. When the power is connected up the +12 and -12v's are both dragged to 0.1v

All caps on the Null-A2 are MLCC and not polarity conscious. There are a couple of diodes on both rails next to the power connector, before the first set of caps, but they are round the right way according to the docs.

Had a good stare at the board and also continuity tested every components. Didn't find any shorts across a component but then it's pretty tricky to test with SMT. Reflowed about half the board so far.

Edit: Just to be clear. The dead short also occurs when the power socket is disconnected from the board. So the socket is proven.
NANO-Jorge
When I start troubleshooting a module, I check first the power voltages, then the ICs voltages.
search64
Post pictures!
Revok
Is there a schematic?
mongrol
Thanks folks but I'm afraid posting pictures isn't going to help. It's a pretty big board with 13 modules on a single PCB and 13 pages of schematics to go with it. I've poured over it several times now cleaning up various blobs and reseating a few bits. Tried a can of freeze spray with it plugged in with no luck either. I can trace volt drop in a general direction but when it hits components my knowledge fails me. I have no idea how to troubleshoot around capacitors.

My next plan of attack is to take each module separately and remove the first components on the rails to try and isolate the problem.
andrewF
I think posting a bunch of clear close up pics of your build would be the best thing to do.
Freeze spray won't make any diff. Removing components will only work if you remove the actual component causing the short.
Hopefully you just have a chip in backwards, which someone should spot in the photos.

Of course there is a schematic, this is NLC w00t
mongrol
Righty ho. The maker says post pictures, so pictures it is. I've since removed the 100k trimmer and power socket but will cease removing anything else. IC's appear to be in the right way. Will check IC's are in the right placements next.

https://photos.app.goo.gl/UXskf8kbGwdQGB4n6

[/img]
Revok
I'd give the board a solid cleaning. You have little solder specs on it.
mongrol
hmm, not really. Those are little holes that have filled in (are they called via's?), or white screenprint. There's some flux still on it where I've been testing but no solder specs. Wil give it a good clean again anyway.
Boogie
mongrol

as far as the pictures show, I propose to check the following:

C11 is only soldered on 1 side
R96 seems to be soldered on 1 side only
R100 also ?
Check R55 looks the same
pin14 of the chip at R240 can use resoldering
you may have a short at Q6 C to left via

That's what I can find in a quick scan of the pictures.

Good Luck !
andrewF
Thanks for the pics
Looks a good build, ICs in correctly.

Perhaps R131 needs a touch of solder, not that this will solve your problem.

There are not many places +12 can short with -12.
Unlikely would be a dodgy trimpot, these would be the two VCO range trimpots, the one you removed and TR6 (freq offset state VCF).
also check your soldering of the topmost 2 pins of the 10 pin connector directly below TR6, this is about the only place where + and - are in adjacent pins.

Then it comes down to a dead IC, most likely an op amp as the CMOS are connected to +V only.
No real procedure to find this without cutting traces to isolate sections of the PCB. When I get pissed off I hook it up and wait to see which one gets hot. Wear safety goggles and lick your finger before touching the back of any IC, when it sizzles you have found it.....remember to lick your finger or get a friend to do it for you.

Also what is your power supply? The null-A draws around 100mA per rail, if your PSU is already stretched it may be shutting down on you.
andrewF
Boogie wrote:
mongrol

you may have a short at Q6 C to left via

these are connected with a trace anyway
Revok
mongrol wrote:
hmm, not really. Those are little holes that have filled in (are they called via's?), or white screenprint. There's some flux still on it where I've been testing but no solder specs. Wil give it a good clean again anyway.


I mean stuff like this. Especially that larger bead circled between the two legs.
mongrol
Thanks a lot guys. Very thorough. Will check over all these items later.

Edit: The PSU is a frogleg beetle which runs 420mA per rail. https://modularaddict.com/frogleg-synthesis-beetle-power-supply-pcb . It get's pretty hot when powering this board.
mcbinc
mongrol wrote:

Edit: The PSU is a frogleg beetle which runs 420mA per rail. https://modularaddict.com/frogleg-synthesis-beetle-power-supply-pcb . It get's pretty hot when powering this board.


Wouldn't recommend powering this up with a known problem *but* if the supply is getting warm, the offending parts are likely getting warm.

* One trimmer leg is unsoldered (top of board)
* I could imagine a short underneath C69
mongrol
In todays installment I've removed all the TL072's one at a time, cleaned up when I found some unmelted solder paste under them and retested. The short remains so I put them back on. In tomorrows episode we'll do the TL074's.

I'm sure the fact that I'm getting a partial short between both rails and GND (some 600ohms) means something. It probably means somethings broken. smile
plushterry
Or you've put the wrong resistor in somewhere eek!
fuzzbass
mongrol wrote:
Thanks a lot guys. Very thorough. Will check over all these items later.

Edit: The PSU is a frogleg beetle which runs 420mA per rail. https://modularaddict.com/frogleg-synthesis-beetle-power-supply-pcb . It get's pretty hot when powering this board.


I'm not saying your psu is the issue here, but I did find this amusing:

"The “Beetle 1.3” provides a simple way to generate 12VDC bipolar power without the need for an AC wall wart [...] The recommended input is 12VDC Wall Wart.."

Otherwise this PSU should be appealing to me. My mom dropped me and my buddy off at the cinema to see "The Love Bug" the week it premiered. I'm that old.
BugBrand
mongrol wrote:
In todays installment I've removed all the TL072's one at a time, cleaned up when I found some unmelted solder paste under them and retested. The short remains so I put them back on. In tomorrows episode we'll do the TL074's.


I think that's your biggest hint - I always had to check like crazy when doing solder-paste at home. Both unmelted paste / or issues if there are vias under the IC body (shorts can occur there).
fuzzbass
Since AndrewF said you might have to work to short V+ and V-, its possible you have multiple problems. Unflowed paste trapped under any of the components might give you a partial short. If I was doing this, I would look first at the power decoupling caps flanking the ICs. The circuit should work (perhaps not well) with all of them removed. Push half of them off and check. If not, push the other half off. If you have hot air, this won't take long. You can run your ohmmeter while you do this and see the results in real time.
mongrol
Raah! I've removed all the TL-074's and found multiple cases of solder paste that could've been potentially shorting. As I went the resistance between +-V's slowly went up, 800ohm, 1500ohm's. Once all were removed and cleaned up I'm getting no shorts. Applying power gives me a good +-12v's on both rails.

They'll be put back down with a tack/drag method instead of paste this time.

I expect other IC's will be in the same state so may do them as well to be safe.

Thanks to all. It's been a good learning experience so far.
andrewF
Well done, glad you had the patience to go back thru it all. Guinness ftw!
mskala
fuzzbass wrote:
"The “Beetle 1.3” provides a simple way to generate 12VDC bipolar power without the need for an AC wall wart [...] The recommended input is 12VDC Wall Wart.."


I think the key word there is "AC." You need a wall wart, but not the harder to find AC-output type.
MUFF WIGGLER Forum Index -> Music Tech DIY Goto page 1, 2  Next [all]
Page 1 of 2
Powered by phpBB © phpBB Group