• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Amplifier Distortion, DC-Offset, and You!

AFAIK you can still post pics as long as you have them stored someplace else (like on your own server). If you drop the link into the posting, you should be fine.
 
EW, does this measurement work on tube units? I just bought a Pioneer SX-2000 and want to test before I hit speakers with it...

The details are HERE

TIA!
 
Acually, Voltage is best thought of as the pressure of the water. The speed and amount of water flowing is the current and the size of the opening is the resistance.
Click to expand...

I suggested this analogy to a couple of classmates while getting my Electronics Technology degree many years ago. They very snidely suggested that I go back and re-take a general physics class. I could have mentioned that I had just aced a very difficult college physics course the previous quarter, but I didn't think these guys were worth wasting more of my time at that point. Just the same, thanks for re-confirming that thought in my mind!

Of course, it's not a perfect analogy, which is true of ANY analogy, but it works well for the most part.

btw: I got those service manuals on order for correcting my Sansuis, but at this point, it might have to wait until winter. I really want to see what effect it has on the sound of my au-717 (100 mv off-kilter!)

Echo -- do you think the DC correction could help with the overall dynamic range? The 717 seems to have less "punch" than my older Sansui. It also seems as though excess DC voltage would in turn create excess DC idle current when connected to speakers, which might push the speaker diaphrams out to a position where they might not have quite the room or flexibility to respond to extreme signal changes as quickly. Or worse yet, could speakers even start clipping?

I don't know -- maybe I've just proven that I really need that physics refresher after all?
 
Originally posted by outlawmws
EW, does this measurement work on tube units?
Click to expand...
No...nearly all tube amps are transformer coupled at the output, so unless there is something horribly wrong with the iron, you'll measure a millivolt or so...and that's where it'll stay.

john_w...

Any DC at the output does indeed displace the woofer cone in proportion to the amount of voltage present. The exact effects on sonics probably depend a bit on the speakers and the amp themselves, but my own observations include a blurring of transients at all levels, and excessive distortion which is most noticeable at low levels. And it makes the speaker unhappy to be unable to center itself where its mechanics want it to, but the voltage present won't let it.

Less 'punch' might be a way to describe it, but we're assuming that there are no other problems with the internal electronics. However, what you describe could also be the result of aging electrolytics or transistors with C-E leaks (tough to track down until they finally give up completely).
 
My recently acquired and cleaned '78-'79 (not sure) Sansui 9090DB reads:

14-15 on the left
11-12 on the right

I also refoamed yesterday two AR90L and are now driven by the Sansui, what a great sound but a tough load for the receiver. I even tripped the Sansui protection... :dunno: :rolleyes:

Anyway a much better sounding combo than the AR 48S I was used to listen to.
 
Is this Thread ever gonna end? Took me hours to read through it. I need a :beer: :finger: Great thread though! Guess my 9090's are all :finger: up. The DB has less than 15dcmv and the 9090 has over 70 on one channel. The crazy thing is the 9090 sounds great! :rant: :hdphones:
 
Bobby4244 said:
Guess my 9090's are all :finger: up. The DB has less than 15dcmv and the 9090 has over 70 on one channel. The crazy thing is the 9090 sounds great! :rant: :hdphones:
Click to expand...

The spec on both 9090 series are

0mv ±10mv

It is possible to get them down to 0 with the pots that are on the driver board but it is better to replace them with bourne 25 turn pots for better control.
 
Wow, what an immensely invaluable thread! :thmbsp:

Just tested a couple of my units with my Rat Shack DMM (cat.#22-813). The results...
Rotel RA-913 integrated:
Left: -3.0 mV (w/brief occasional variances between -5.7 to +1.2)
Right: -2.9 mV (w/brief occasional variances between -4.8 and +2.8)

More troubling results came from my newest acquisition, a Telefunken TR-600 DC-coupled receiver I bought for $20:
Left: 71mV
Right: 83 mV
It also got noticably warm in the general amp section, in the rear, whilst I was conducting this measurement. A related symptom? :dunno:
(BTW, it has yet to be hooked up to speakers.)

:beerchug:
-AlphaLima
 
I finally got around to checking my old stuff on this rainy day.
There are no dashes-those are negative numbers.

MARANTZ 2330B
R -26.5
L -44.5

MARANTZ 4100 int amp
FRONT MAIN R -11.8
FRONT MAIN L -10.7

KENWOOD KA-7100 (2)
R 11.3
L -4.5

MARANTZ 4430
FRONT MAIN R -21.1
FRONT MAIN L -16.7

KENWOOD KA-7100 (1)
R -5.3
L 27.1

What's with the negatives on the Marantz units? Or the others, for that matter. The test leads were hooked up + to +.
What's the worst unit-the 2330B? I don't like the variation in the #1 Kenwood, either.
The best?
 
What's with the negatives on the Marantz units? Or the others, for that matter. The test leads were hooked up + to +.
Click to expand...
'+' to '+'?? I assume you mean '+' to '-', the same way you'd connect a speaker.

RE: The first question regarding reading a negative value. See post #33, #38 and #39 of this thread. :twak:

Summary: Maintaining '0V' at the output is a balancing act with DC coupled amplifiers using split supplies (such as a Kenwood 700M, for instance, in which the output stage is powered by '+65V' and '-65V' supplies). Audio is an AC signal, with half of the signal being greater than 0V, and the other half less than 0V. Obviously we use the positive supply to generate positive-going output voltages, and the negative supply to generate negative-going output voltages. Part of the job of feedback to the input differential pair is to maintain the average DC output at 0V. Think of the differential pair as kids at the end of a see-saw, and DC offset represented by an imbalance of that see-saw. The see-saw is certainly capable of swinging either way ('+' or '-' AC signal), but the differential pair is supposed to keep the output balanced at at an average DC value of 0V at all times. The transistors cannot be perfectly matched, but the closer the better.

So I'll answer your question with a question: Seeing as how the amp runs off of split supplies (as is required to reproduce music), why would measuring a negative value raise an eyebrow when a positive would not? :)

it's the absolute value (remember your math??) that matters here, not the sign
 
Hey EW, SUCCESS!

Man.......I'm glad you posted this thread, and I'm even more glad I took your advice and got a service manual for my SX-1980. As you may remember, my receiver was over 400mV on each channel. I made the DC offset adjustments as well as the idle current adjustments. I got the idle current right on the money at 30mV on both channels and the DC offset at 2.2mV/right and 6.1mV/left. Man, those DC pots sure were touchy......I might have got it down to zero otherwise. The verdict? Wow......this thing sounds so much better, especially at low volume, and it runs much cooler. Thanks so much for the advice and the great thread! :banana: :banana:
 
EchoWars said:
'+' to '+'?? I assume you mean '+' to '-', the same way you'd connect a speaker.
Click to expand...

I meant + on the multimeter to + on the speaker outputs.
I went back and read the posts on page 3, I understand how the negative numbers are irrevelant. You don't have to beat me on the head! :whip: :D
Have you got time for any more winter jobs?
 
You don't have to beat me on the head!
Click to expand...
Lol!! I just like that one, being a Three Stooges fan... :D

Understand about the meter now...

I dunno if the email from profile works or not, but the PM does. If you have something that needs attention, then you can send me a PM and we'll get emails exchanged.
 
Denon DRA 95A Receiver no workie

Two decent receivers in two months. no good.

So, what does it mean when you don't hear the "click" sound shortly after power up? Just prior to total meltdown, this unit would intermittantly emit a very loud, short buzz even without volume up. Worked fine otherwise.

Thanks,
Ed
 
Onkyo TX-65 Left 3.0 Right 2.6 millivolts.... :banana:

Sansui AU-G99X Left 26.0 Right 34.0 :scratch2:

I found two dc offset knobs on EACH channel, and what appears to be two bias knobs on each, for a total of 8 knobs.....anyone have any idea how I should go about adjusting and cleaning, perhaps?? Or am I just being a fussy idiot?? For example, how would you adjust TWO knobs?? And if you just start turning those bias and dc offset knobs, is something gonna blow if you aren't careful?? I don't think I'm so impaired that I cannot do this, but I was scared to even turn my amp on and put meter leads in the speaker outs, much less start turning things. Maybe I should leave this to Echowars or some other resident brain. :yes:
 
fropiler said:
So, what does it mean when you don't hear the "click" sound shortly after power up? Just prior to total meltdown, this unit would intermittantly emit a very loud, short buzz even without volume up. Worked fine otherwise.

Thanks,
Ed
Click to expand...
The click is generally the output relay. If it's not engaging, then there's a problem...either in the relay circuitry, or in the amp (relay will not engage if there is excessive DC at the amp output). The buzz is a wierd one...perhaps an internal ground has been compromised...?
I found two dc offset knobs on EACH channel, and what appears to be two bias knobs on each, for a total of 8 knobs.....anyone have any idea how I should go about adjusting and cleaning, perhaps??
Click to expand...
Not unusual to find two offset pots on units with a FET at the amp input...one being a coarse adjust, and the other a fine adjust. The two others?? Only one would be a bias adjust, the other is possibly part of the V/I limiter and adjusted to start limiting output at some level.
For example, how would you adjust TWO knobs??
Click to expand...
For offset, the coarse is simply adjusted to get the offset in the ballpark, and tweaked to perfection with the fine adjust
And if you just start turning those bias and dc offset knobs, is something gonna blow if you aren't careful??
Click to expand...
Yes. Until you know the exact function of the trimpots, leave them alone.
 
Great thread,thanks EW.Finally, read all 24 pages before asking. Can't remember, hope I'm not being redundant. I adjusted the Offset and Bias (using the service manual) on my Pioneer SX 838 today. Offset values were 170mV L/20mV R. I got the offset to around zero (touchy trimpot) moved on to the bias, brought it up to 30mV from around 25mV. Double checked the offset once again. Phone rang :dammit: and I left the meter hooked up checking the offset. When I looked at the meter is was all over the place + to - 300mV both channels :wtf: . The bias was steady and stayed where I had adjusted it to. What might be the suspect component here? I was hoping that these adjustment would help with the protection circuit kicking in and out. Been on now for two hours and havn't heard the relay once. Still got the meter jumping around though on the offset. Thanks in advance for your time in response.

mac
 
Gave it another shot after my last post. I was able to get the offset set and it remained stable. I hooked it up to some speakers and it played for a bit. The protection relay started kicking in and out again in no particular pattern. I checked the offset again and it was all over again :cry: . Thought I might have had it there for a while. :grumpy:
 
Is the offset stable in either channel? Does the protection trip in a regular and repeatable manner? Will it idle (i.e no music) for long periods and not trip? See if you can find the amp output before it gets to the protection ckt and see what is amiss. Can you disconnect the preamp from the amp?

EW - Could this be a protection circuit issue? Or is it more likely an amp problem?

Theoretically it could be something as trivial as dirty pots. Swap 'em out for multi-turns and if that isn't the issue, you'll be glad you did anyways.
 
You must log in or register to reply here.
Back
Top Bottom