bias/idle current adjusting

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello, i am trying to set bias current first time and amplifier is Denon PMA 980R.
It has 4 small yellow potensiometers for adjusting bias.
So far i have managed to set 2 potensiometers to 5mV that only require multimeter, but there are 2 more yellow potensiometers it can be seen below on some images and those require 10mV 1khz signal to set.

I also noticed something strange while seting bias , there are + and - on circutboard but one channel has reverse polarity (i put multimeter leads on right) and also i measured power outputs one had small positive voltage and the one with negative bias had also small negative voltage on output, What can cause this?
Manual says i need to wait about 10 min and set bias on 2 potensiometers then feed in 1kHz 10mV signal and set other 2 to 30mV, nothing abuit negative voltage was mentioned.

Also did few more measurments and transformer gives out 2x 39,5V,both diode bridges take in 80v and give out 106v, on all transistors 54V.
now i feeded about 10mV sinus and channel that had positive voltage on bias checking terminals showed 30mV but other is stuck on 5 and wont go up , i moved potensiometer both ways.
If i put music on and measure then it switches to 30 mv on medium volume and on high or low vol its 5.
On other channel with negative bias 30mv wont work at all .
Last edited:
I downloaded german manual with schematics on hifi engine, i cannot upload it on forum but here is link
Login | HiFi Engine
What is reason for this amp to switch bias up on 30mV on signal ,if i undrerstand right many amps in same series do this? But does it help with sound quality?

For me left channel bias is positive and it is 5mV then switches to ~30mv on medium signal , on full signsl bias goes very low again.

Right channel bias i adjusted to -5mV and speaker output also has small negative voltage compared to left channel with positive voltage. Also it wont go to 30mV i tryed different signals strength but nothing.

Could it be that some voltage is leaking from something else?
Login | HiFi Engine
This german manual has schematics.

Id like to ask what is reason for 30mV bias that comes on with signal?
It probably is used on many amps at least same series but why is extra bias voltage needed?

Left channel works 5mV then with signal turns ~30mV and om maximal signal goes lower than 3mV. This bias seems to work fine
Right channel had before -4mV bias i adjusted it to -5mV but on board it shows +/- for checking points and bias is voltage is reversed. It also wont turn 30mV bias on i tryed different signal levels.

Would it be damaging if one channel has negative bias -5mV and other positive 5mV that also turns to 30mV with signal?
One with 30mV heats up on full volume 1khz signal/ no load otherone wont heat much.

Please help me understand what to check on schematics.
Maybe there is some sort of power regulator for bias potensiometers VR501-VR504 that could be checked.
Sorry about my english.
Joined 2007
Paid Member
A quick look at the circuit shows a conventional Vbe mutiplier that also has input from an opto isolator, presumably to alter the bias dynamically, hence the need to adjust with signal. This signal dependent bias will be some marketing ploy (Class A that isn't :D)

You won't have "negative bias" anywhere but you will have to follow the manual to set it up correctly.
A quick look at the circuit shows a conventional Vbe mutiplier that also has input from an opto isolator, presumably to alter the bias dynamically, hence the need to adjust with signal. This signal dependent bias will be some marketing ploy (Class A that isn't :D)

You won't have "negative bias" anywhere but you will have to follow the manual to set it up correctly.
Could easily be some sort of marketing, it has golden text front New optical class A.

One channel works as it shoud and adjusts itself to 30mV with signal and has positive output.
One channel on has -5mV bias and -6mV from output. Also wont adjust bias up.

Could it be voltage leak?
Joined 2007
Paid Member
You can't have a negative bias current. Make sure you are consistent with your meter leads and that you measure between identical points on each channel. 5 millivolts or -5 millivolts across a resistor gives the same current.

If the bias doesn't increase on one channel with signal then you need to trace the opto control signal back and see if there are any differences between the channels.
Used multimeter leads right way , on board it has + and - also set multimeter on both power outputs ground and swaped + lead.
Somehow it gets negative and maybe thats why it wont adjust itself.
Opto is sensor near radiator?
What happens to sound if it wont increase bias voltage or has negative?
My problem is that i am more used working on car electronics and it rarely involves details, usualy wiring.
Test point connected to R584 is suposed to be negative then and other one positive.
+ and - on circutboard for testpoints confused cause of info in manual but schematics shows one should be +5 mV and other -5mV. First bias adjustment currents are fine then.

Second one when i feet 1kHz signal from laptop around 10mV then volume up left channel turns bias around 30mV and radiator heats quite hot but right channel wont turn it 30 it stays on 5mV and radiator wont heat much.

This amp seems to be cold when no signal on both radiators but when listening to music low level one radiator heats up quite hot and another wont heat at all.
One not heating wont get 30 mV bias as well.

Does it reduce sound quality? , surely reduces power consumption.
Joined 2007
Paid Member
Bias for a class ab output stage (which this is) has only one optimum point. The varying bias is just marketing. Just because it makes it run as an "overbiased" class ab stage doesn't make it better.

You need to compare the control voltages to the opto to see if/where any problem is. They are derived from IC402 and TR402 and (apart from the opto and a couple of resistors) are common to both channels.
I connected 1 speaker and put 100hz test tone 1000hz 5000hz 10000hz etc.
I hear no difference on low freq when i turn bias up or down.
Seemed some difference on 5000Hz 10 000Hz.
I am too beginner to make those bias voltages work as they should but il just turn secondary ones down and set first ones to 5mV, then it wont heat radiators up and more energy goes for power.
Joined 2007
Paid Member
In practice the smallest amount of bias will stop audible distortion. Set the main bias correctly to give the 5 millivolts.

If the dynamic bias doesn't operate correctly on one channel then its up to you if you want to look for a fault. I can tell you what to do and what to check.
Thanks for taking time and all help, i learned bit more how to read schematics today and understood little more about this amp.
Right now i have turned 47K variable bias resistors both down but one that worked earlier still gives 10mV on its channel so on certain volume levels one bias is 5mV other 10mV on low levels both 5mV.

Il try to understand schematic and how IC402 and TR402 are delivering bias voltage.
Is one of those delivering power for base bias and other for dynamic one?
Are those heat sensing elements near radiators adjusting its voltage?
If i manage to understand it il try to check components and how far voltage goes on schematics.

This dynamic bias system is quite strange and i think il leave it on minimal settings.
Could dynamic bias reduce distortion or increase power output or something?
If not il give my best to fix it and leave it on minimal settings if it has no real point other than to write some complex text on amplifier that most customers wont understand anyway.
Last edited:
Joined 2007
Paid Member
Looking at one channel TR517 is a conventional "vbe" multiplier. All it does is generate a volt drop between collector and emitter, the actual voltage depends how hard the transistor is turned on, and that depends on the setting of VR501.

So far so good...

The opto isolator works by "turning off" TR517 (by pulling the base and emitter "closer together" voltage wise). The opto is turned on (which increases the bias) by TR402 conducting and allowing current to flow in the LED part of the opto. That turns on the transistor (light sensitive in the opto) and that increases the bias current in the amp.

So the first step (that I would do) would be to isolate the opto (by removing R543, the 15 K in series with the preset) and setting the normal bias correctly to give the 5 millivolts.

I'll be back... food awaits :D
Joined 2007
Paid Member
Does that make sense ? You need to first set the normal bias with no input from the opto to confuse things.

Having done that you then confirm that when the recommended signal is applied that the collector of TR402 falls.

As the collector voltage falls current flows from the positive rail and through the LED in the opto via R414/5/6 and 7 to ground. So you need to confirm that happens.

Then, assuming the normal bias adjusted OK, the dynamic bias should too. If it doesn't adjust on one channel then the only parts involved are the opto and the preset and series resistor. The optos could be swapped to prove if there is a faulty one.
Looking at schematic i see that TR402 turns both optos on when reaches needed input signal.
Now i measured optos and both get ~11kOhm from TR402 middle leg and there are 2x 5,6K resistors.
Also both opto leds gets +53V power.
Then i measured opto led on diode settings and one shows ~1500 and other one that didnot work showed infinity.
So i belive opto led is out, il replace it and see if it burns again or works.
Thanks for all instructions , without i would not have known even how to spot opto.
There are so many transistors on schematics its confusing to look if i dont know what they do exactly, now i know what some do.

Id like to ask is it fine if to turn dynamic bias down to 10mV as its minimal setting or best to use original 30mV?
I have read few topics about bias voltage and optical class A amps but i still dont understand what increased bias on load gives?
As far as i understand higher bias makes crossover more smooth and heats radiators more.
If it would heat less would it be more powerful amplifier?
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.