Let's Play 'Dump the Cap' - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State

Solid State Talk all about solid state amplification.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 16th February 2003, 10:14 AM   #1
diyAudio Member
 
EchoWars's Avatar
 
Join Date: Dec 2002
Location: Left of the Dial
Default Let's Play 'Dump the Cap'

Have a preamp I'm messing with (attached partial schematic). I want to just flat get rid of the output capacitor. I'm hoping ya'all can help me here...

Root problem is...if the amp has been off for a while, on powerup I get a strange 'scratchy' sound at the output of one channel. It's almost like a dirty volume control would sound, but kinda fluttery and mainly in the frequency range of about 80-120Hz. Lasts perhaps 1 minute or so.

I suspect that the large 100uf polarized cap at the output is the problem. It only does this in one channel. Afterwords, the amp sounds and behaves just fine. When the amp was set up and running in my system, I swear it was the right channel. After disassembly, and changing a few resistors (needed to increase gain a bit), and checking the opamp socket (a nice gold plated Amp socket, not a machined socket, but a decent one nevertheless), now the right is fine and wierd startup 'flutter' has moved to the left.

Yes, it is possible that I had channels reversed somewhere down the line and it was left channel all along, soooo....

With the circuit as it is:

On startup, the opamp output on each channel first swings to about +7 Volts, then drops to about -2 Volts, finally settling at about +1mV on left channel, and about +2mV on the right. I suspect that, if the output was DC coupled, it would settle much faster. I also suspect that the startup 'flutter' is related to the reverse-biasing of this polarized electrolytic, as well as its large size (original size was 3.3uf, but I added pre-out/main-in jacks, and with 3.3uf I couldn't drive a very low impedence at output).

Wierd thing is that the 'flutter' is intermittent. I suspected the tone control switch (S6 in drawing), and have cleaned the hell out of it with DeOxit and Pro Gold. Made no difference. Eight times out of ten I can turn it on and no flutter is there. The other two it will flutter for about 1 min in the left channel (used to be the right...??) and finally calm down.

Since the DC offset of the opamp is so low (1mV and 2mV), why can't the output cap just come out? Is there a reason to have it there? The next amp stage has a 4.7uf cap at it's input (it's a emitter-follower in a Szilaki configuration, so input impedence is high). And most devices that the preout would connect to would have a cap at the input (notable exception: I use an active crossover with this preamp, and the input stage is a DC coupled buffer, feeding an AC coupled filter...seems the initial DC swings shouldn't bother it too much...) Tone controls (which don't get used much) are basic NFB stuff.

As a test, I'd like to simply short across the 100uf cap and see what happens with the wierd startup flutter. Can I do that? Or should I take the cap out before shorting??

Is removing the cap going to be a problem since then both the 120K feedback resistor and the 820K resistor /5pf cap will then both be DC coupled to the opamp output?

I'd really like to toss the 100uf cap...can I?
Attached Images
File Type: png opamp_mod_2.png (7.6 KB, 807 views)
  Reply With Quote
Old 16th February 2003, 11:18 AM   #2
diyAudio Member
 
Join Date: Mar 2001
Location: Brazil
Default Re: Let's Play 'Dump the Cap'

Quote:
Originally posted by EchoWars

Root problem is...if the amp has been off for a while, on powerup I get a strange 'scratchy' sound at the output of one channel. It's almost like a dirty volume control would sound, but kinda fluttery and mainly in the frequency range of about 80-120Hz. Lasts perhaps 1 minute or so.

I suspect that the large 100uf polarized cap at the output is the problem. It only does this in one channel. Afterwords, the amp sounds and behaves just fine. When the amp was set up and running in my system, I swear it was the right channel. After disassembly, and changing a few resistors (needed to increase gain a bit), and checking the opamp socket (a nice gold plated Amp socket, not a machined socket, but a decent one nevertheless), now the right is fine and wierd startup 'flutter' has moved to the left.

Yes, it is possible that I had channels reversed somewhere down the line and it was left channel all along, soooo....

With the circuit as it is:

On startup, the opamp output on each channel first swings to about +7 Volts, then drops to about -2 Volts, finally settling at about +1mV on left channel, and about +2mV on the right. I suspect that, if the output was DC coupled, it would settle much faster. I also suspect that the startup 'flutter' is related to the reverse-biasing of this polarized electrolytic, as well as its large size (original size was 3.3uf, but I added pre-out/main-in jacks, and with 3.3uf I couldn't drive a very low impedence at output).

Wierd thing is that the 'flutter' is intermittent. I suspected the tone control switch (S6 in drawing), and have cleaned the hell out of it with DeOxit and Pro Gold. Made no difference. Eight times out of ten I can turn it on and no flutter is there. The other two it will flutter for about 1 min in the left channel (used to be the right...??) and finally calm down.

Since the DC offset of the opamp is so low (1mV and 2mV), why can't the output cap just come out? Is there a reason to have it there? The next amp stage has a 4.7uf cap at it's input (it's a emitter-follower in a Szilaki configuration, so input impedence is high). And most devices that the preout would connect to would have a cap at the input (notable exception: I use an active crossover with this preamp, and the input stage is a DC coupled buffer, feeding an AC coupled filter...seems the initial DC swings shouldn't bother it too much...) Tone controls (which don't get used much) are basic NFB stuff.

As a test, I'd like to simply short across the 100uf cap and see what happens with the wierd startup flutter. Can I do that? Or should I take the cap out before shorting??

Is removing the cap going to be a problem since then both the 120K feedback resistor and the 820K resistor /5pf cap will then both be DC coupled to the opamp output?

I'd really like to toss the 100uf cap...can I?
You certainly can and should toss the output cap away. Particularly if you have a cap in the next stage.

In fact most times it's better to pull out all output caps, particularly the last cap before the power amp, and put a very good quality cap there (some prefer film types, some BG).

In this case just use a wire to short the output cap. Check if you have DC at the chip input and also short the input cap.

But I don't think what you are telling is related with output cap failure. The IC sockets might be a reason, and you should change them to machined types. As the chip is a dual type, have you tried another dual to see what happens?

The intermittent problem might be related to a cold solder too, so if you didn't yet go and use your solder pen on all pads.

The next thing you should do is bypass all the pots you have (volume, balance, treble, bass, etc.) and see how the signal gets through. Perhaps even bypass this whole stage, feeding a CD signal direct to the power amp, and see if the problem vanishes.

In fact this is where I would start trying: bypassing this stage.

Another thing: when you say the amp was off for a while, how long is that? Where was it kept?


Carlos
  Reply With Quote
Old 16th February 2003, 01:09 PM   #3
diyAudio Member
 
EchoWars's Avatar
 
Join Date: Dec 2002
Location: Left of the Dial
I really doubt the socket is the problem. Again, it is a very good Amp part. I did not have any machined parts at the time.

I have also changed the 2604 with another, and receive the same results with DC offset (close anyway) and also have the same problem with the wierd flutter.

I have resoldered many connections looking for a bad solder joint, and have even closely examined the board with a 10x maginifying glass to see if I might see a poor joint. They all look good, and I reflowed quite a few anyway.

The problem originates here in this stage. The volume control is before this stage, and changing the volume setting (or balance for that matter) makes no difference in the level of the flutter. This opamp stage is the first active stage in the amp. and the flutter goes away (in the speakers anyway) if I pull the pre/main jumper (as it should if it was coming from the preamp). Not only that, but when I can get the flutter to appear, I can see it on the scope at the opamp output.

The 4.7uf cap in the schematic and the 47uf cap in the feedback loop are both BG non-polar. The cap at the next stage is also a 4.7uf BG. The 100uf cap was a simple Panasonic FC polarized, since there is not much room to work with here. Peter Daniel sent me a couple of Panasonic HFQ's (thanks Peter) that I think sound a bit better and certainly fit a lot better, but the problem with the flutter persists and I believe if I use no cap at all then the problem will go away.

By 'off for a while', I simply mean an hour or so. This amp has been in near continuous use for 25 years.

So, dumping the output cap is a reasonable thing to do...and the 820K resistor and 5pf cap won't affect anything?
  Reply With Quote
Old 16th February 2003, 01:55 PM   #4
diyAudio Member
 
Join Date: Mar 2001
Location: Brazil
Quote:
Originally posted by EchoWars

So, dumping the output cap is a reasonable thing to do...and the 820K resistor and 5pf cap won't affect anything?
I have done that many times on CD players, shorting the output cap. You can put a series resistor in the vacant place, something like 220 ohms or so.

It will work great.


Carlos
  Reply With Quote
Old 17th February 2003, 05:24 AM   #5
diyAudio Member
 
EchoWars's Avatar
 
Join Date: Dec 2002
Location: Left of the Dial
So...what might be the pros or cons of either simply putting in a jumper wire, or adding a 220 ohm resistor?

Anyone?
  Reply With Quote
Old 17th February 2003, 11:22 AM   #6
diyAudio Member
 
Join Date: Mar 2001
Location: Brazil
Quote:
Originally posted by EchoWars
So...what might be the pros or cons of either simply putting in a jumper wire, or adding a 220 ohm resistor?

Anyone?
Not many much cons for using a 220 ohm resistor, and many pros. The most important pro is that the resistor will protect the preamp output stage by limiting the current.

Most CD players already have those resistors.

Though I must say this is more useful with headphone amps.


Carlos
  Reply With Quote
Old 17th February 2003, 11:50 AM   #7
diyAudio Member
 
jan.didden's Avatar
 
Join Date: May 2002
Location: Great City of Turnhout, Belgium
Blog Entries: 7
Default lets dump the flutter

Sounds to me that at startup you have an oscillation burst. Did you change the opamps from the original ones?
Try with C133 & C134 47pF or so.

Jan Didden
  Reply With Quote
Old 17th February 2003, 12:01 PM   #8
ALW is offline ALW  United Kingdom
diyAudio Member
 
Join Date: Jun 2001
Location: UK
Quote:
if the amp has been off for a while, on powerup I get a strange 'scratchy' sound at the output of one channel. It's almost like a dirty volume control would sound, but kinda fluttery and mainly in the frequency range of about 80-120Hz. Lasts perhaps 1 minute or so.
Capacitor leakage current can cause exactly this problem, although it may not be the problem in this case. Larger capacitors exhibit the problem more (leakage is proportional to value), higher voltages are best.

It's a common problem in old AC-coupled preamps, in the inter-stage DC blocks (e.g. Naim).

Andy.
  Reply With Quote
Old 17th February 2003, 12:12 PM   #9
diyAudio Member
 
EchoWars's Avatar
 
Join Date: Dec 2002
Location: Left of the Dial
Quote:
Not many much cons for using a 220 ohm resistor, and many pros. The most important pro is that the resistor will protect the preamp output stage by limiting the current.
If you look, there is a 665 Ohm resistor at the pre-out jack for current limiting. Adding the 220 Ohm will simply raise output impedence slightly. I guess what I don't understand is the relationship of the 120K feedback resistor, and the 820K/5pf capacitor (there to limit gain at HF, I assume)...and whether it is better to have them separated by a cap, a small resistance, jumper wire, or perhaps it makes no difference at all.

janneman,
The 2604 replaces a very old 7-pin inline opamp from the mid 70's, a specialty part made for Kenwood, so yes, it is an upgrade (on a small custom PC board with the eight leads running to the original Kenwood board, leads perhaps 2.5cm long, power bypassed with .18f film cap) ). Would seem strange then an oscillation burst would show itself as mainly low-to-mid frequency noise. Regardless, I have a few 39pf caps around, and may replace the 5pf cap with one.

Right now I have temporarly jumpered over the 100f cap, and I cannot get the turn on flutter to return. I will have to experiment some more and ensure that this is a true fix.
  Reply With Quote
Old 17th February 2003, 12:25 PM   #10
diyAudio Member
 
EchoWars's Avatar
 
Join Date: Dec 2002
Location: Left of the Dial
Quote:
Originally posted by ALW


Capacitor leakage current can cause exactly this problem, although it may not be the problem in this case. Larger capacitors exhibit the problem more (leakage is proportional to value), higher voltages are best.

It's a common problem in old AC-coupled preamps, in the inter-stage DC blocks (e.g. Naim).

Andy.
You see, just by listening I am almost convinced this is a problem with leakage, or a problem with using a polarized cap in an environment where it is getting no bias. The 100f/50V cap at the output has had a Panasonic FC, a Panasonic HFQ, and I even stole a Black Gate 100f/50V K series cap from another piece of equipment as a test, and all show the same problem with the wierd turn on flutter. All these caps are polarized, and here I've got a whole 1 or 2mV on them for bias. They do not seem to like this, and I can't get a decent bi-polar small enough to fit where the polar comes out, which is why I want to toss the cap out completely.

Again, I have the cap jumpered across for now, and I'll have to continue experimenting to see if the flutter is gone. But still want to know if there should be a small resistance separating the 820K/5pf loop, and the 120K feedback resistor...
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
a big toy to play with.... angeloitacare Multi-Way 68 19th March 2013 08:23 PM
My dump is better than your tip jackinnj Everything Else 6 9th April 2009 08:49 PM
can't play stereo yngwie Solid State 2 17th August 2007 05:05 AM
How does a PC play WAV files? amo Digital Source 2 22nd October 2004 12:44 PM
If there was an instrument you would play what would it be? gigatron Everything Else 45 4th February 2004 12:48 AM


New To Site? Need Help?

All times are GMT. The time now is 09:05 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2