As the title says, I've got bad hum, my LFO led is not lighting up, and my LM13700s are all really hot. I've done the usual component and solder bridge check and don't see anything wrong.
I'm afraid I'll damage IC 3, 4, and 5 so I haven't left power connected for long enough to measure their voltages. So, I pulled them out, and here's my voltages on IC 1 and 2:
IC1:
1: 2.85 2: 2.85 3: 2.62 4: 6.3 5: 3.14 6: 2.85 7: 2.86
8: 3.4 9: 2.85 10: 2.81 11: 9.1 12: 2.86 13: 2.86 14: 2.85
IC2:
1: 5.7 2: 2.83 3: 5.7 4: 10.2 (seems high?) 5: 6.4 6: 0.9 (should be 0?) 7: 5.6 8: 2.86
Pull out all the ICs. Measure the voltage on the bare socket for pins 2, 4, 5 and 8 of IC2. What do you have?
Quote from: madbean on May 11, 2017, 05:15:20 PM
Pull out all the ICs. Measure the voltage on the bare socket for pins 2, 4, 5 and 8 of IC2. What do you have?
Thanks, Brian!
2: 0
4: 0
5: 0
8: 9.6
Quote from: AntKnee on May 11, 2017, 06:04:01 PM
2: 0
4: 0
5: 0
8: 9.6
Your voltages are not making sense. Are you sure you listed them in the correct order when you had IC2 installed?
Quote
IC2:
1: 5.7 2: 2.83 3: 5.7 4: 10.2 (seems high?) 5: 6.4 6: 0.9 (should be 0?) 7: 5.6 8: 2.86
Without IC2, you should be reading about 4.5v on pin2 and pin5, 0v on pin4, about 9v or so on pin8. The voltages listed with the IC installed are all wrong (for instance you said you have 10.2v on pin4 when it should be ground).
So, one of three things is happening here (or a combination thereof). Your original voltages with the IC installed were listed in the incorrect order, your actual IC is bad, or there are more extensive power problems on your board. What you need to do first is verify you have about 4.5v on pin7 of IC1B. This is the VB supply that should be feeding pin2 and 5 of IC2. If not, you need to look for problems with IC1, and that you have installed 100k resistors for both R53 and R54. Next I would try a different part in IC2 if you have one. A plain TL072 is fine. Test the voltages again and see what you get.
I rechecked the voltages. These look better. I'm not sure what happened earlier.
IC1:
1. 4.7
2. 4.7
3. 4.3
4. 9.4
5. 4.7
6. 4.7
7. 4.7
8. 8.7
9. 4.3
10. 4.6
11. 0.3
12. 4.7
13. 4.7
14. 4.7
IC2:
1. 8.8
2. 4.7
3. 8.8
4. 0.3
5. 4.7
6. 8.8
7. 1.1
8. 9.5
With IC1 left in (is that what you meant for me to do?) I get the voltages you said I should get on IC2.
With IC1 in, and IC2 out, my LFO led comes on. When I insert IC2, the led does not come on. Not sure if that is good or bad.
I tried inserting a TL072 in IC2, and I get the same result.
R53 and R54 are definitely 100k.
Lemme think about it a bit and get back to you tomorrow. Have some things I need to finish up today.
Ok. Thanks for the help!
I got my 13700s, 2n3904 and 2n3906 from Tayda. I ordered another set from Mouser, just to swap them out and rule out possible fakes or junk. I will update after that happens.
Ok, popped in all new ICs, and still have the same issue.
I replaced all the electro caps, too.
I noticed that sometimes the led will come on when fiddling with IC2, but I cant find a short or solder bridge on it. I have reflowed the joints twice. Its the same with multiple chips. Maybe the socket itself is bad? Regardless of when it is on or off, if it doesn't come on, IC3, 4, and especially 5 get hot. If the led comes on, I get no effect, but I can get signal through the circuit.
Well I'm not really sure what's happening here. Sorry to ask you to do this, but try this -
Take out all the ICs. Put in only IC3. Does it get hot?
If not, put in IC1. Does IC3 get hot now?
If not, put in IC2...IC3 gets hot now, doesn't it? Just determining if IC2 is the current sink here (which I suspect it is).
If you get to step 3 and it turns out how I think, then let's see a pic of your board and wiring. Actually, we should probably see that anyway at this point.
I thought you sounded correct, but surprisingly, IC3 got hot really fast when installing IC1.
(https://dl.dropboxusercontent.com/u/13215910/Web%20Images/stagefright/20170516_083140.jpg)
(https://dl.dropboxusercontent.com/u/13215910/Web%20Images/stagefright/20170516_083205.jpg)
(https://dl.dropboxusercontent.com/u/13215910/Web%20Images/stagefright/20170516_083315.jpg)
R5 is wrong. Replacing it now.
Quote from: AntKnee on May 16, 2017, 02:15:47 PM
R5 is wrong. Replacing it now.
Wait, no, its right. It measures 47k, but it the striping is odd.
hard to say for sure from the pix, but check your solder on R25 near connection to IC3. Looks like it might be touching the ground plane? Could be a similar thing at R24 near the pcb edge.
Quote from: AntKnee on May 16, 2017, 02:29:20 PM
Quote from: AntKnee on May 16, 2017, 02:15:47 PM
R5 is wrong. Replacing it now.
Wait, no, its right. It measures 47k, but it the striping is odd.
Can't measure it while it's still in the circuit.
Sent from my iPhone using Tapatalk
Quote from: wgc on May 16, 2017, 04:45:35 PM
hard to say for sure from the pix, but check your solder on R25 near connection to IC3. Looks like it might be touching the ground plane? Could be a similar thing at R24 near the pcb edge.
Thanks for taking a look!
I rechecked those and don't see anything suspect. I cleaned those up and reflowed them anyway.
With IC1 in place, here is the voltages at the sockets, (without any other ICs installed):
IC2:
1. 3.8
2. 4.7
3. 3.8
4. 0
5. 4.7
6. 3.8
7. 3.8
8. 9.4
IC3:
1. 4.75
2. 0
3. 4.75
4. 4.6
5. 0
6. 0
7. 0
8. 0
9. 0
10. 0
11. 9.4
12. 0
13. 4.4
14. 4.75
15. 0
16. 4.75
Quote from: jimilee on May 16, 2017, 05:43:08 PM
Quote from: AntKnee on May 16, 2017, 02:29:20 PM
Quote from: AntKnee on May 16, 2017, 02:15:47 PM
R5 is wrong. Replacing it now.
Wait, no, its right. It measures 47k, but it the striping is odd.
Can't measure it while it's still in the circuit.
Sent from my iPhone using Tapatalk
I didn't, I pulled the package of matching resistors and measured them. However, I am assuming the one installed is not defective. I did not remove it.
Those new voltages seem spot on, for the most part. This is a stumper. If this were me, the next step I would take is looking at the traces around the LM13700s with a magnifying glass. What you are looking for is a manufacture error like a micro trace under the solder mask connecting it from a pad to the ground plane...or just something that doesn't look right. Not the pads that are grounded, though - those each have 4 little bridges that connect from the pad to the ground place. I'm talking something smaller that looks out of place. Can't be sure it is a board error but with different ICs from different sources and some good voltages the possibilities start narrowing.
I've looked at your trace side pic and I don't see anything so far, but it will easier for you to get a better look.
Looked it all over with a magnifying glass, didn't see anything that looked out of place. My eye is not trained to spot this sort of thing, though. I checked for continuity to ground on every solder pad and marked them wherever I got a tone with my tester. If anything looks odd, or if something is not marked that should be, let me know and I will double check it.
(The two x marks on IC4 are a mistake by me, they should not be marked.)
(https://dl.dropboxusercontent.com/u/13215910/Web%20Images/stagefright/20170516_172718.jpg)
It looks like you feed quite a bit of solder on to pads, as a lot of the joins are a bit blobby on the component side of the board.
Any chance there is contact between things on that side of the board? Its not easy to tell in the picture as the round shiny surfaces throw some odd reflections.
Edit/addition:
For example, I think this is just a bit of reflection, but maybe worth a look?
(http://i.imgur.com/r1OcicR.png)
Quote from: somnif on May 16, 2017, 11:38:51 PM
It looks like you feed quite a bit of solder on to pads, as a lot of the joins are a bit blobby on the component side of the board.
Any chance there is contact between things on that side of the board? Its not easy to tell in the picture as the round shiny surfaces throw some odd reflections.
Edit/addition:
For example, I think this is just a bit of reflection, but maybe worth a look?
Thanks for taking a look for me. Its a reflection, though. I have seached both sides of the board multiple times with a magnifier trying to find possible bridges and haven't found a culprit yet.
I also went ahead and removed and replaced my IC1 socket, just in case some debris or something was shorting underneath it.
It appears you may have used 2Ms (or 2 Ohm) in place of 20ks. The fourth band should be red. I can't quite make out what color is on that band in your pic (is it yellow or maybe silver?)
edit - Your 62k also appears incorrect. Do you have more from those batches you can check with a DMM?
Shiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiite!!!!!!!!!!!!
I know what happened.
I placed an order for 300k, 200k, 62k, 47k, 39k, 30k, and 20k.
BUT, I ordered ohms, not Kohms. So all of those resistors are wrong. When checking them, I just overlooked that they weren't kohms and just looked at the number value.
Dammit.
Sorry everyone. I'll swap all those out and update. Let this be a lesson to everyone!
Well, I'm sorry I didn't check the resistors sooner. It would have saved us some time. But, I believe it will solve your problem. If you need some resistors, I can send you some to save you yet another order.
Quote from: madbean on May 17, 2017, 12:49:12 AM
Well, I'm sorry I didn't check the resistors sooner. It would have saved us some time. But, I believe it will solve your problem. If you need some resistors, I can send you some to save you yet another order.
Brian, I really appreciate that offer, but I'll just order the right stuff. Thank you so much for your personal time and help on this.
Now, I have to pull 20 resistors without causing damage! No worries, I have cool tools. Cheers!
I almost made that same mistake last night. Wanted a few resistors with thicker leads than Tayda's so chucked a bunch of singles and doubles on my Mouser order last night. Thankfully 2.2 ohm 1/4 watt resistors were on 2 week back order (and I noticed that fact on the checkout page) or I would have been just as confused.
Its up and running. 8)
I had to replace a total of 23 resistors, which was a pain, but well worth it.
Sounds great!
Thanks again, B!
Cool. Man what a pain that was I can imagine!