30.00 Amp Degraded

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Worked one of the SI boards fairly heavily yesterday. Changed the input coupling caps, on board filter cap, and switched to 10 uH chokes from Miller.
Powered up and no smoke. Checked for ac and dc on the output and found 34 mv on on channel as before. But the other channel now reads 9 volts dc offset.
Can anyone direct me to where to start. I yanked it apart again and did not locate any problems with the items subbed. Must have hit one of the surface mount parts with the soldering iron.
I have tried checking it cold with a properly working board. So far have not located the problem. Any ideas?

George
 
I also managed to "degrade" one of my boards two nights ago. Lifted the solder pad on R01 on top (the pad that goes over to L1), and also lifted the trace going to R2. I can probably get it working again, but it's going to be a PITA.

Did you bridge a trace with solder? Can you post some closeups of the board? Also, did you remove the flux on the board with alcohol and a toothbrush after doing your soldering work?

I have a slight problem on the second board that I'm working on. After replacing R4 and R5, two of my resistors don't measure properly. I only read 16.6k across R2 and R4, but get the proper 19.8k on R1 and R5. I know the chips measure properly when desoldered. Maybe someone here will have an idea on that as well?

Good luck!
 
Ok,

To start figuring out what is wrong with the chanel with offset we need to determine if the problem is on the output side of the chip or the input side. Not much else to go wrong if the other chanel is working.

A couple of things come to mind.

1. There is an open circuit between the output of the amplifier and the speaker output pads. This could be a bad connection or cracked run caused during the replacement of the output chokes. Verify that you have continuity between the snubber diodes and the speaker output pads. The end of the diode closer to the center of the board is the end to measure.

D1 to R+
D2 to R-
D3 to L+
D4 to L-


2. The new input capacitor C2/C4 is leaking DC. What kind of caps did you install? If they are electrolitic capacitors verify that the polairity is correct. A backwards cap will leak DC and the amp will see this as input signal.

The input signal is feeding into an inverting opamp circuit. We have access to all the nessasary nodes to figure out if it is working properly.

Schematic of input circuitry:
An externally hosted image should be here but it was not working when we last tested it.


Pin 16 is the non inverting inputs shared by both channels and biased to 2.4 volts DC internally.

If the input cap C2/C4 is leaking current R1/R2 will be pulled down by the leakage current. This will cause the output of the opamp pins 10/14 to go positive in voltage. The safest thing to check in this case is to measure the DC voltage across R1/R2. In a happy world there should be no DC voltage across either resistor.

If DC voltage is measured on R1/R2 of the bad chanel then there is either a short to ground at the R1/R2 side of C2/C4 or C2/C4 is leaky/backwards.

Check these things out and let us know if things improve.

Gary
 
motherone said:
I'm intentionally shorting C2/4 and replacing it offboard with some 4.7uF Solens =)

From my checking, the input caps are C3/C4. c1,c2 appear to be part of the input filter. If not, this is my problem, I removed C1 and C2. My preamp is transformer coupled, the input sees the 38 ohm dcr and it effectively loads down the input to reduce rfi pickup.
The reason C1 and C2 were removed was that I knocked one off while soldering in a replacement cap for C3/C4. I am having a tough time working with parts this size.
Used a 4.7 ufd ceramic cap from Digikey for the input cap. It checks fine in to the 10K shunt resistor on one end and the 20K inline resistor on the other. It does not appear to be leaking. May be open though.

George
 
If you want to short C3/C4 and run a larger input cap off board you have to remove R01/R02. There can be no parts capable of passing DC current to ground connected to R1/R2.

When my amps get here I'm planning on removing R01/R02, C1/C2 and L1/L2. C3/C4 will get replaced with jumpers then off board caps will be used for input duties with direct connection to the input RCA's via shielded cable.

Still thinking about the power supply details. Just found Vinnie's page yesterday. Lots of info there to digest.

Gary

Gary P's DIY page
 
Voltage pull down

theAnonymous1 said:


That is exactly why I shorted C3/C4 and it was a no go. I put external caps in but even with no input signal connected having them shorted still causes DC on the output. I'm not sure why, but one of you smart fellers will probably tell me:D .


The input of the 2024 is biased up to allow the output to be dc coupled with a single rail supply. Somewhere around 2.5 volts is applied to the input and this is why the c3/c4 is needed where it is at. The 10k shunt resistor is pulling down the 2.5 volt bias and this causes the dc on the output.
I need to dig back in mine, something must be pulling it down to zero to get 9 volts dc. I am back to using a stocker for now.
I have been hesitant to probe around with the chip powered up. All the problems with letting the smoke out of the 2024 chips proves to me that it is easy to kill one if not extra carefull.

George
 
Re: Voltage pull down

Panelhead said:



The input of the 2024 is biased up to allow the output to be dc coupled with a single rail supply. Somewhere around 2.5 volts is applied to the input and this is why the c3/c4 is needed where it is at. The 10k shunt resistor is pulling down the 2.5 volt bias and this causes the dc on the output.
I need to dig back in mine, something must be pulling it down to zero to get 9 volts dc. I am back to using a stocker for now.
I have been hesitant to probe around with the chip powered up. All the problems with letting the smoke out of the 2024 chips proves to me that it is easy to kill one if not extra carefull.

George


George,

Why not pull R01/02 and C3/4. Then lay a 10k resistor across each channel of your volume control to ground, followed by your 4.7uF cap? If you're not using a volume control, you can always mount the resistor across your RCAs.
 
motherone, your a genious! Ok, so this is what I'm gonna do.....

R1/R2 Remove and place external 10k resistors across RCA terminals
L1/L2 Remove and jumper
C3/C4 Remove, jumper, and place external elecltrolytics in line with input from RCA terminals.

Does this sound ok? Am I missing anything?

TIA
 
theAnonymous1 said:
motherone, your a genious! Ok, so this is what I'm gonna do.....

R1/R2 Remove and place external 10k resistors across RCA terminals
L1/L2 Remove and jumper
C3/C4 Remove, jumper, and place external elecltrolytics in line with input from RCA terminals.

Does this sound ok? Am I missing anything?

TIA

A cleaner signal path is to just remove C3/C4. Then connect the coupling cap to the solder pad that is connected to the 20k resistor. Think is will try this myself. Should be the same and leave room for a better cap than the miniscule surface mounts.

George
 
theAnonymous1 said:
motherone, your a genious! Ok, so this is what I'm gonna do.....

R1/R2 Remove and place external 10k resistors across RCA terminals
L1/L2 Remove and jumper
C3/C4 Remove, jumper, and place external elecltrolytics in line with input from RCA terminals.

Does this sound ok? Am I missing anything?

TIA

I don't know about the genius stuff, but hopefully the suggestion'll help.

I'm going to remove R01/R02 and C3/C4. My signal path will then be:

RCA Jack -> 10k noble pot -> 10k 1% dale between pot out and ground -> 4.7uF Solen poly cap -> input on board

L1/L2 along with C1/C2 are probably not neccessary. They seem to prevent RF from coming into the amp, if I'm not mistaken? This isn't specified in the datasheet. Maybe it was something added by SI in order to prevent RF from coming through the little crappy plastic box they put it in.

Panelhead said:


A cleaner signal path is to just remove C3/C4. Then connect the coupling cap to the solder pad that is connected to the 20k resistor. Think is will try this myself. Should be the same and leave room for a better cap than the miniscule surface mounts.

George

Based on the posts above, isn't a DC bias injected by R01/R02?


I agree, but the solder pads are very very weak. I've already lifted 1 pad and 1 trace -- I would bring them to the input and jumper the unneccessary parts on the board. At least the input pads are fairly beefy considering the size of the board.
 
Oh, almost forgot, what type of electrolytics are you using, Anonymous? Most folks are of the opinion that electrolytics are very bad in the signal path. You should take a look for at least some 2uF poly caps. These can be purchased even at radio shack (though they bend you over for it). These would probably sound alot better than electrolytics.
 
Well, success at last. I removed or relocated all but C1/C2 and and the little amp is workin just fine. So does it sound better with these input mods? Who knows..... was that really the whole point of this experiment?

Oh, almost forgot, what type of electrolytics are you using, Anonymous? Most folks are of the opinion that electrolytics are very bad in the signal path. You should take a look for at least some 2uF poly caps. These can be purchased even at radio shack (though they bend you over for it). These would probably sound alot better than electrolytics

The only reason I'm using electrolytics is because its all I had laying around. Guess I need to take a trip to ratshack and get some polys.

BTW, should I just go all the way and remove C1/C2 also? Is high frequencies on the input really that big of a problem with a good source?
 
theAnonymous1 said:
Well, success at last. I removed or relocated all but C1/C2 and and the little amp is workin just fine. So does it sound better with these input mods? Who knows..... was that really the whole point of this experiment?



The only reason I'm using electrolytics is because its all I had laying around. Guess I need to take a trip to ratshack and get some polys.

BTW, should I just go all the way and remove C1/C2 also? Is high frequencies on the input really that big of a problem with a good source?

If you're putting the amp in a metal enclosure, I would guess that the impact of RF should be less. I'm not an expert at this, but like I said, my theory is that SI did this because of the plastic enclosure.

Do you have an unmodified amp to compare it to?
 
Yes and no. I didn't mod any of the electronics on board, but its in a new enclosure with all new wiring soldered in. I guess I could do a side by side comparison, but unless there is a profound difference in the sound I'm afraid my lead ears won't be able to tell. A bit more low end is pretty much all I was hoping for.
 
theAnonymous1 said:
Yes and no. I didn't mod any of the electronics on board, but its in a new enclosure with all new wiring soldered in. I guess I could do a side by side comparison, but unless there is a profound difference in the sound I'm afraid my lead ears won't be able to tell. A bit more low end is pretty much all I was hoping for.

Heh. I'm probably with you right there. I'm sort of modding these things just to mod 'em.. Take them to their "max", even if I can't necessarily hear the difference :D

I wonder how many folks I made cringe with that statement!
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.