Amplifier Distortion, DC-Offset, and You!

[abpeep]

Hooked up speakers to the G-5700, let it sit for about an hour and I get the same type of results - erratic fluctuation of about 10-12 mV on the left channel, about 3-4 mV on the right.

I can dial it in to have these fluctuations around 0mV - left channel goes about +9 to -3, right goes -1 to -5.

This thing has a number of problems - tuner doesn't seem to work and definitely needs calibration, while trying to hook up speakers some of the tabs came loose (had to hook up speakers inside), a couple of the caps seem to have some discoloration on the sides (leakage?), the volume control seems to have corroded severely in one position (two treatments of deoxit has helped but there is still a stiff spot in there).

Probably continue to play around with this one as I get up the nerve to take on another challenge. Thanks for this thread - I have learned something while adjusting each of my amps/receivers (and haven't blown anything up yet!!!).

Alan

 

[Balthazarr]

The other night when I chased down the source of crackling in my NAD 3140 Int I noticed what looked like charring (soot) on the legs of a couple of OP transistors. Almost like exhaust. lol
Transistors are located at opposite ends of the heatsink. One pair each side.
I assume that those on the right are for the rt channel and those on the left for the left channel.

Don't know if it was there when I adjusted offset and idle a few weeks ago, but when doing so I had a problem with the rt channel when I fired it up causing the prot circuit to activate.
Took me a while, a long while, to realize the leads on the back of spkr had contacted each other bc I had been pulling the thing around to access rear panel of amp.
During this time I ran the amp trying to determine the cause of shutdown.
Thus the rt channel received a lot of abuse.

I would think that if any damage done the two on the rt would show something?
Not opposite sides.
The amp works fine now with no crackling and equal strength in both channels when balance at TDC.
Any ideas as to the origin of the carbon black?

 

[EchoWars]

Just small-signal transistors with black leads? Pretty common on older gear as time and corrosion attack the tin on the legs.

 

[Balthazarr]

Only those two, but they are 2 of the large ones.
There is 1 smaller on the inside of each pr of large trans's.
So you have 2 large on small on the left and 1 small 2 large on the rt.
The inner 2 on each side were not affected.

 

[EchoWars]

Take a pic...start another thread. :thmbsp:

 

[tschanrm]

Finished up replacing all caps in my sx-1010 a couple months ago including power caps. After a long time of adjusting the bias pots (which were also replaced), the best I could do was a left/right value floating between 1-2mv.

Does anybody know how long it should take for the dcoffset to stabilize? On my 1010 it would take around 10 minutes or so after I adjusted the dc offset pot, and sometimes longer for the dc bias pot.

I have some cheap digital multimeter, could this be part of my problem? (it's this exact model: DT-838 Link

 

[Balthazarr]

It doesn't have to be "0".
I have a Craftsman Multi-scope (osc function in add to Multimeter) and depending on scale, mine will float as well.
If I set it on V the mv reading is stable. On mv it bounces around.

 

[EchoWars]

Finished up replacing all caps in my sx-1010 a couple months ago including power caps. After a long time of adjusting the bias pots (which were also replaced), the best I could do was a left/right value floating between 1-2mv.

Don't confuse bias and offset. Bias has little or nothing to do with DC offset (just tossing that in since too many have read this thread and blown up their amps because the difference was not clear to them).

Does anybody know how long it should take for the dcoffset to stabilize? On my 1010 it would take around 10 minutes or so after I adjusted the dc offset pot, and sometimes longer for the dc bias pot.

Takes a while...offset and bias are at the mercy of thermal parameters. They drift as the amp heats and cools. I let an amp warm up for an hour, check and set, and look at it again a while later. It's all you can do and not worth losing sleep over.

I have some cheap digital multimeter, could this be part of my problem? (it's this exact model: DT-838 Link

They don't get much cheaper than that...but offset does fluctuate...nothin' you can do about it except replace the adjustment pot with a DC Servo (way beyond the scope of this thread).

 

[john_w]

Based on Echo's original post, only 1-2 mV is pretty darn good! If it's anywhere near that after maybe a few hours, I'd quit while I was ahead.

 

[david_a]

Marantz 1070:
L: 34mV
R: 28mV

I suppose I'll ask around in case anybody has the manual. It's got four nice pots internally that are just asking to be tweaked, but I'm not about to go frying the poor thing over values like that.

What happens if you tweak the offset without bothering with the bias? Nothing bad will come of that, right? Given the readings I have, I think the unit is in good health, so I'm wagering the bias is probably reasonable...

Oh, and would the offset have any effect on the headphone output? I know most of these old amps don't have dedicated headphone amps in them.

 

[EchoWars]

Unless you want to spring for the manual so that you know exactly what you're doing, I'd leave it alone.

Very few amps have a separate headphone amp. No need...just pop in a couple of resistors to drop the voltage level to the headphones a bit and you have a perfectly fine headphone source for $0.10. Yes, offset affects the phone jack too, but would read considerably less at the phones due to the dropping resistor.

 

[Kerb]

AB International 600LX 270 watts per channel 8 ohms

left channel= 8.7 mv
right channel= 4.8mv

I got this amp on ebay a few weeks ago, for 80 bucks.

It sounds awesome BTW :yes:

 

[david_a]

Many thanks to Dominic (oldmarantz) who sent me the relevant scanned page from the manual. Adjusting the offset and idle on this thing was very simple and actually pretty fun. After a thorough DeOxit cleaning and some dial twiddling the readings are >2mV on each channel and the idle is right where it's supposed to be... :boink:

The sound seems a bit clearer, but I haven't had much time to listen to it since adjusting it. Anyway, thank you EchoWars for this wonderful thread!

Very few amps have a separate headphone amp.

I'd believe that about vintage amps, but I was under the impression that nearly all modern amps just use a cheap opamp thrown in as an afterthought. But I guess those amps don't really count around here :thmbsp:

 

[von.ah]

Noob in awe

Hi everyone. This is my first post here in AK, and would just like to say AMEN -- I am officially diseased with the vintage bug. What an amazing resource this place is! And EchoWars, :thmbsp: to you , my friend, for this excellent and needed info thread. So, soon I hope to pick up a Marantz 1152DC, and of course, this is the first thing I hope to test with it. Anyone know what to expect with the 1152's internals for adjusting?
Cheers,
~Marc

 

[Morden2004]

Sansui Service Manuals are great

to use as fire starters :yes: because they are usually WRONG :rofl: .

Yet again I am confused trying to set the bias on my 7070 (following a surgical procedure to install two new{old} power meters). Following the manual is NOT an option because the text doesn't even match the diagrams; for example pulling the leads off F06 and F07 (the documented test points )for measuring the BIAS current (manual says current!!) do not interrupt the circuits as one would expect. Inserting your meter between, say, F05 and F06 to measure the requisite 30 mA on the RIGHT channel (but documented as LEFT channel in the text) results in nothing - no current - ZERO, nada.

However the diagram shows those jumpers removed to do the test (as I expected) but guess what? It the channel runs happily along with the jumper removed? This can't be right. Damn all Sansui manuals right to hell.

I'm taking a break. I'll see if I can figure out this mess.

Anyone else run into this with a Sansui 7070, 8080 or 9090?

Paul

 

[akahn]

My Yamaha CR-820 reads at about 20mV for each channel. My CR-620 does better, looks to be about 5-10mV.

 

[Morden2004]

Re: Sansui Service Manuals & settings

Boy, is my face red!

I have successfuly adjusted my Sansui 7070 and achieved the following readings:

LEFT: -0.16
RIGHT: +0.30

BIAS: both channels set to 30 mA +/- 1 mA

Now, as to why I was in a rant about the manual and what I discovered as to why I couldn't get a BIAS reading, you'll have to jump on over to the [Exclusively Sansui] forum where I'll shortly post about my work replacing the power meters (there WILL be pictures) and why I couldn't get a reading for the BIAS settings on my 7070.

I'll post there shortly - I don't want to hijack this thread with chatter that's not on topic.

Paul

 

[hpsenicka]

KA-2002A in trouble...

I just received a KA-2002A integrated, and as a matter of course decided to check offset before connecting anything to it.

Well at power on the readings were off the 200mV scale.... as high as 18V actually!! :jawdrop:

As the unit warmed up the voltage dropped steadily, levelling off at just under 1.5V... still way too high!

Interestingly, if I leave the voltmeter leads attached when I turn off the power, I can observe the voltage falling to zero, then the polarity reversing and the voltage increasing.... as if the capacitors are discharging??

Cosmetically the unit is nice and relatively compact, and I was intending to use it to drive my computer speakers.... So I would be interested in attempting to fix it myself if the cost of parts is not prohibitive, else I may be looking for a Sonic Impact T-amp to use instead.

Any suggestions where to start? And no, I don't have a service manual for this unit yet...

 

[_Attila_]

Marantz 2235B:
Right: -20.3 mV
Left: -15.0 mV

Could be bettrer, but, ok, I can live with for now....

 

[EchoWars]

Any suggestions where to start? And no, I don't have a service manual for this unit yet...

Kenwood wasn't big on cap-coupled designs, but I don't know squat about the 2002A. Got a dummy load? Hell, even a 1K resistor connected from '+' to '-' will drain the cap if that is the cause of the voltages you're seeing.

If you can connect a resistor (anywhere from 100 ohms to 1K) at the terminals and the voltage drops to something normal, then it should be OK to connect speakers. If the resistor burns up, then that would be a no-go instead.