Amplifier Distortion, DC-Offset, and You!
- Thread starter EchoWars
- Start date
Okay EW.....
Read the whole thread last night, ll I can say is :thmbsp:
Thought I'd join in anyhow. Only thing I have running on the bench was my recently acquired 8080db which didn't look to have ever had the cowver off recently anyhow. Amp hasnt been worked on - made in May 76
Right DC started out -85mv and gradually floated to a +16mv then floated between that and an occasional bounce to the neg. side. The only consistancy was not going more than +16 The left channel floated slowly between about +1mv and +8mv I can't allow for the right channels float to the degree it did, the meter's a good OTC, I would have thought if there was any loading effect it would have stayed more consistent? Whatchya think? :scratch2:
Ron
Read the whole thread last night, ll I can say is :thmbsp:
Thought I'd join in anyhow. Only thing I have running on the bench was my recently acquired 8080db which didn't look to have ever had the cowver off recently anyhow. Amp hasnt been worked on - made in May 76
Right DC started out -85mv and gradually floated to a +16mv then floated between that and an occasional bounce to the neg. side. The only consistancy was not going more than +16 The left channel floated slowly between about +1mv and +8mv I can't allow for the right channels float to the degree it did, the meter's a good OTC, I would have thought if there was any loading effect it would have stayed more consistent? Whatchya think? :scratch2:
Ron
EW,
I checked the board after seeing the recommendations you had. Appreciate the partial print, have the 9090 only. I went over the other boards I had sitting around this afternoon for grins as well.... The other two boards (7070,9090) ran about 50% bad or marginal caps. All I found on the 8080db board in question was only a 470uf that's marginal on esr. Everyone else checks out okay, (I use a VuData to quick check the trans) course the pot could easily be a little flakey but I have to admit of the three boards I have here (8080db, 7070, and a 9090) the 8080db board in question is so new looking it scared me. The other two look their 30 years but this one either has been replaced with a new part or it's been used so seldom it loks new. The caps are different than I'm used to seeing as well, they used Elnas for the 330's rather than UCC's. What do you usually replace the trim pots with? Any specific recommendation or jsut any multi-turn cermet pot? Thanks for your time!
I checked the board after seeing the recommendations you had. Appreciate the partial print, have the 9090 only. I went over the other boards I had sitting around this afternoon for grins as well.... The other two boards (7070,9090) ran about 50% bad or marginal caps. All I found on the 8080db board in question was only a 470uf that's marginal on esr. Everyone else checks out okay, (I use a VuData to quick check the trans) course the pot could easily be a little flakey but I have to admit of the three boards I have here (8080db, 7070, and a 9090) the 8080db board in question is so new looking it scared me. The other two look their 30 years but this one either has been replaced with a new part or it's been used so seldom it loks new. The caps are different than I'm used to seeing as well, they used Elnas for the 330's rather than UCC's. What do you usually replace the trim pots with? Any specific recommendation or jsut any multi-turn cermet pot? Thanks for your time!
A 5K Bourns trimpot would be fine. Digikey PN 3296P-502-ND.
It still is quite possible that one transistor of the diff pair is suffering from a bad case of the thermal jitters. If you have something to match transistors, put a closely matched pair of Zetex ZTX694B's or ZTX1056A's in there. They work great, but they are a EBC layout, wheras the originals are ECB layout, so care must be taken when installing 'em.
It still is quite possible that one transistor of the diff pair is suffering from a bad case of the thermal jitters. If you have something to match transistors, put a closely matched pair of Zetex ZTX694B's or ZTX1056A's in there. They work great, but they are a EBC layout, wheras the originals are ECB layout, so care must be taken when installing 'em.
Last edited:
EW,
You're a gentleman and a scholar! Thanks for the info. You guys here seem to like Digi-Key a lot I see. Perhaps I'll give it another try, I pretty much quit dealing there after getting fake trannys a couple years ago. anyhow back on the subject at hand, I'll start using the pots in all my updates, I recap em anyhow. If it was one of the trans then it should be consistant. It had been running for several hours when I checked it.
You're a gentleman and a scholar! Thanks for the info. You guys here seem to like Digi-Key a lot I see. Perhaps I'll give it another try, I pretty much quit dealing there after getting fake trannys a couple years ago. anyhow back on the subject at hand, I'll start using the pots in all my updates, I recap em anyhow. If it was one of the trans then it should be consistant. It had been running for several hours when I checked it.
Oops...please note that the first PN I posted was for a 500 ohm pot (501 part number). The correct number would be a 502 part. The number in the post has been corrected.
Digikey was not exactly alone in the fake transistor scandals...however, you can pretty much count on anything you get there nowadays is the genuine article.
Digikey was not exactly alone in the fake transistor scandals...however, you can pretty much count on anything you get there nowadays is the genuine article.
Kegger
R.I.P. 1/12/1966 - 6/1/2017
So I suppose 20 volts DC on both channels is to much aye? LOL
Yah working on an sx-990 that the main wired in fuse underneath was blown.
Jumpered the fuse for now and get an actual 20v DC not millivolts but 20 volts!
"only leaving the unit on for about 25 seconds at a time" The voltage keeps droping
from the 20 volts but for fear of something else going I don't leave it on long enough
to find out how far down it goes. That's either channel by the way measuring at the
speaker outputs on the back of the unit. A or B speakers do the same.
Well I got some tracing to do.
-----------------------------------------------
"edit after doing some digging around"
Allright did some tracing and I got some nice crispie fried resistors by the outputs!
----------------------------------------------------------------------------------
Yah working on an sx-990 that the main wired in fuse underneath was blown.
Jumpered the fuse for now and get an actual 20v DC not millivolts but 20 volts!
"only leaving the unit on for about 25 seconds at a time" The voltage keeps droping
from the 20 volts but for fear of something else going I don't leave it on long enough
to find out how far down it goes. That's either channel by the way measuring at the
speaker outputs on the back of the unit. A or B speakers do the same.
Well I got some tracing to do.
-----------------------------------------------
"edit after doing some digging around"
Allright did some tracing and I got some nice crispie fried resistors by the outputs!
----------------------------------------------------------------------------------
Last edited:
Okay, I'm a little confused here, please help me out here EW as I am trying to learn this bias and DC adjustments, what's the difference between the too?
how would I go about adjusting these two settings in my Pioneer sx880 unit. yes, i do have the service manual on hand and do know how to read the schematic, thanks.
how would I go about adjusting these two settings in my Pioneer sx880 unit. yes, i do have the service manual on hand and do know how to read the schematic, thanks.
abpeep said:
Both what? 30mV where?WingNut said:
I accidently quoted the Fig. number at the bottom of the page (Fig 14) instead of the page number, which is pg 17.
I see no reason to perform the actions in the first paragraph...where VR7 and VR8 are turned fully counterclockwise. Simply start off with the 'DC Balance' adjustment, and move on to setting the bias.
The procedure is simple in the extreme...follow the book.
Their intent on having you turn VR7 and VR8 fully counter-clockwise is to set bias to a minimum. There is no real need to do this, and I have...from time to time...seen where turning the pot in the direction indicated actually was cranking bias up to the maximum instead of down to the minumum, due to some error in writing the manual or some last minute change in the amp itself. Power-on after this, and all the magic smoke comes out.
Any thoughts on this article?
http://www.tnt-audio.com/clinica/bias_e.html
Playing with fire perhaps?
http://www.tnt-audio.com/clinica/bias_e.html
Playing with fire perhaps?
Morden2004
Relaxin'
Sansui manuals are, well, suspect
Absolutely! The Sansui AU-999 manual, for example, has as step #1 in the bais adjustment as "Set V R-803 and VR-804 to minimum -clockwise position".
Wrong. But a careful inspection of the text shows that there is a word missing!
As EW says, smoke will quickly follow.
I was OK as I had the o'scope connected and when the sudden transistion occured I still had my hand on the switch. Lucky me.
Paul
EchoWars said:
Absolutely! The Sansui AU-999 manual, for example, has as step #1 in the bais adjustment as "Set V R-803 and VR-804 to minimum -clockwise position".
Wrong. But a careful inspection of the text shows that there is a word missing! As EW says, smoke will quickly follow.
I was OK as I had the o'scope connected and when the sudden transistion occured I still had my hand on the switch. Lucky me.
Paul
Just picked up a Sansui G-5700 last weekend and got a SM right away. Tried to set bias and DC offset tonight but looks like there are probably other problems. Had no problem setting the bias (besides touchy pots). The DC offset readings on both channels - no speakers, no source - varies up and down by roughly 40-50 mV in an erratic fashion. Sounds like I can trip the protection circuit with just a slight turn of the pots.
Oh well, I probably needed a project.
Alan
Oh well, I probably needed a project.
Alan