I need this signal to inject in an amp to set the bias. I have a B&K 3022 generator, but I can't quite figure out how to get such a low signal. Is there easy way to get this done?
It is easily done with 2 resistors, say a 10K in series with a 10R. The generator is set to 10V and this divider will have an output of 10mV (1000:1).
The output impedance will be 10R, which is usually no problem with any reasonable load. 1% resistors are usually quite good enough.
Adjust values / amplitude to get other results.
If the output of the generator will not go down that low you could build a simple attenuator to get what you need. Here is where a way to measure AC RMS volts would be most handy, like a scope or a good RMS voltmeter. I poked around a bit and it looks like the manual is available. If you have it check for the minimum output level and report back.
I measure from .065VAC to 7.260VAC on the output of the 3022 via a Fluke 77 (an older one) which is not rms. So I guess what I need to do is attenuate this range down and feed into an oscilloscope? Now to figure out how to do that, I have a Tek 2245.
Can I also just set it to read 28.29mv (the Vpp of the rms value)? Is the Fluke reading peak to peak or must I halve that value?
Can I also just set it to read 28.29mv (the Vpp of the rms value)? Is the Fluke reading peak to peak or must I halve that value?
In looking at that old thread, there appears to be a discrepancy between the adjustment procedure and some notes near the optical biasing device, procedure says 100mvrms input and at device 10mvrms input, perhaps that is why getting this adjustment to work has not been smooth for me, nor the individual in this thread:
http://www.diyaudio.com/forums/solid-state/224708-trouble-denon-pma-520-bias-adjustment.html
I wonder if the actual value is 10mvrms. Does 10mvrms seem like a more reasonable input from a source (ie cd player) than 100mvrms?
http://www.diyaudio.com/forums/solid-state/224708-trouble-denon-pma-520-bias-adjustment.html
I wonder if the actual value is 10mvrms. Does 10mvrms seem like a more reasonable input from a source (ie cd player) than 100mvrms?
OK, the manual has line input sensitivity at 150mv, so an input of 100mv seems reasonable to set the bias.
By the way, it appears Nelson Pass has a patent on this optical biasing approach used in the Denon products of that time.
By the way, it appears Nelson Pass has a patent on this optical biasing approach used in the Denon products of that time.
Hi,
No amplifier uses 1KHz 10mV sine waves to set bias.
The wrong end of the stick is being firmly grasped.
The voltages are most likely DC across the output
emitter resistors, either one or both of the pairs.
The inanity in that thread is that bias varies with
an applied input signal, it doesn't, but can vary
with a high signal level using a dummy load.
Only really bad bias circuits need testing with an
input signal and dummy load, most you just
adjust at idle to the correct DC reading.
Which will vary as it warms up.
rgds, sreten.
I can't see the real circuit but if bias has two trim pots its :
a) It doesn't, one is bias the other is output offset
b) Its some form of reducing thermal bias modulation,
if so it cannot really work with no applied dummy load
c) Note the instructions are inane, as no initial setting
for the latter two trim pots are given, for adjusting
the initial two trim pots, it makes no sense
d) For certain written by an idiot
Best guess is two pots control bias and two pots
control a Vbe multiplier and thus they interact.
Best guess is you start with all pots in the middle,
but sadly perhaps they should be all minimum.
I'd try both and see if either works.
No amplifier uses 1KHz 10mV sine waves to set bias.
The wrong end of the stick is being firmly grasped.
The voltages are most likely DC across the output
emitter resistors, either one or both of the pairs.
The inanity in that thread is that bias varies with
an applied input signal, it doesn't, but can vary
with a high signal level using a dummy load.
Only really bad bias circuits need testing with an
input signal and dummy load, most you just
adjust at idle to the correct DC reading.
Which will vary as it warms up.
rgds, sreten.
I can't see the real circuit but if bias has two trim pots its :
a) It doesn't, one is bias the other is output offset
b) Its some form of reducing thermal bias modulation,
if so it cannot really work with no applied dummy load
c) Note the instructions are inane, as no initial setting
for the latter two trim pots are given, for adjusting
the initial two trim pots, it makes no sense
d) For certain written by an idiot
Best guess is two pots control bias and two pots
control a Vbe multiplier and thus they interact.
Best guess is you start with all pots in the middle,
but sadly perhaps they should be all minimum.
I'd try both and see if either works.
Last edited:
For the sake of completeness, definitely yes.
Normal amp input impedance is 10K to 100K. ish.
Input loading error with a 10R source will be 1/1000 to 1/10000 ish.
Small beer error.
This amp, and the "Optical Class A" Denon amps of the 1980s (ie some PMA-, POA-, DRA-) DO require a signal injection to set the bias. It is a two step procedure outlined in the thread I referenced in Post #9. The first step (adjustment of VR403 and VR404 to get 5mV at a test point) is rather conventional and goes smoothly. The second step is to inject a 1KHz, 100mVrms into a line input, turn the volume to maximum, and adjust a second set of trim pots (VR405 and VR406) to a different, higher value (20mv) at the SAME test point. The problem comes in that as one monitors the test point with the signal being injected and sweeps towards the full volume position (labelled "40" on this amp), the measured value increases and then suddenly drops off (does this at the "34" setting) to the previous 5mV value. The poster referred in my Post #9 experienced exactly the same thing. Now, what I have found is that by reducing the input amplitude from 100mVrms downward (to 60mVrms, the lowest I can get out of the 3022), this drop-off shifts to further down the volume pot position, the dropoff now happens at position "36." What I have done is made sure the maximum value is 20mV measured at the test point before the dropoff, but the procedure says to do this at the 40 position, not a lower one, as I have done.
In the schematic, there is a note at the optical device (IC503) referencing two voltages that can be checked at pin 4, one for when there is no signal injection (step 1) and when when there is. That note says 10mVrms, instead of the 100mVrms that is in the adjustment procedure. Since it appears that the strange step 2 behavior (the drop off) gets better when the input signal is lower, perhaps it is possible that there is an error in the procedure and the input value should be 10mVrms. I did check the SM of another Denon amp and it also has 100mVrms in the procedure, but that schematic failed to provide the note for 10mVrms that is on mine.
In the schematic, there is a note at the optical device (IC503) referencing two voltages that can be checked at pin 4, one for when there is no signal injection (step 1) and when when there is. That note says 10mVrms, instead of the 100mVrms that is in the adjustment procedure. Since it appears that the strange step 2 behavior (the drop off) gets better when the input signal is lower, perhaps it is possible that there is an error in the procedure and the input value should be 10mVrms. I did check the SM of another Denon amp and it also has 100mVrms in the procedure, but that schematic failed to provide the note for 10mVrms that is on mine.
Last edited:
Is this an attempt to adjust the bias on some optical inputs?
I am totally confused. Bias adjustment is normally done to set the idle current on the amplifier's output transistors (or tubes). I keep reading "optical" ...
Please be more specific about what you are attempting to do.
I am totally confused. Bias adjustment is normally done to set the idle current on the amplifier's output transistors (or tubes). I keep reading "optical" ...
Please be more specific about what you are attempting to do.
It is an "optically controlled bias system" per the unit's face plate.
I am reading the proper voltages at the optical device (IC503), both with no input and the 1KHz input by setting the trim pots at the stated voltages at a volume position just before the drop off, just not at the full volume position, as described in the procedure. The amp works fine and I will leave as is.
I am reading the proper voltages at the optical device (IC503), both with no input and the 1KHz input by setting the trim pots at the stated voltages at a volume position just before the drop off, just not at the full volume position, as described in the procedure. The amp works fine and I will leave as is.
- Status
- Not open for further replies.