Problem faced with Valhalla2 as pre with Audiolab 8200A as power amp - Help needed

  • Thread starter Deleted member 15865
  • Start date
"DC leak" and DC Offset are two different things.
Both can be tested easily with a decent multimeter.

To measure DC offset, remove any input from your headphone amp, short the +/- of the input, and measure the DC voltage across the +/- leads of the headphone amp output (check separately for left and right channels). You'll need a multimeter that has a range of 200 mV DC.

For DC leak, put negative probe of your multimeter to metallic conducting part of your chassis and probe the positive probe around various parts of the circuit and chassis, input and output connectors.

An you please guide on how to measure DC leak and DC offset in a DAC? Will the procedure be the same as above for the preamp? The DAC has a coaxial input... so how do I short the +/- for it?

Also, for the preamp itself, when you say short the +/-, you mean I should hold the probe (which one?) so that it touches both the + and _ of the preamp input? Is that safe to do?

And should I have the load connected or removed while I perform these tests?

Also, what level of DC voltage leak/offsets are normal? In other words if the DC voltage is above what value then we can conclude it might be causing the said problem?
 
Last edited by a moderator:
DC offset is something different. That is measured across the speaker terminals, and should be under 20mv. As close to 0mv the better.
 
For shorting you will need bare RCA plugs, you can buy some cheap cables and cut it close to the connector and join the two connector cables for shorting.
Just put the black probe of the DMM to any ground point and the red probe to the inner metal of the input on your tube pre to measure.
Set the tolerance to 20 on the DC voltage section of the DMM. Ideally you should not get any reading.
Test it under all conditions, cold as well as warmed up.
 
Your experience suggests that I need to check both the preamp (Valhalla 2) and the DAC (Bifrost) checked for DC.

I was able to isolate the problem unit (not the cause at that point) by using an alternate DAC - which of course did not cause the protection mode fault. The probable cause was identified by the mfr, through the distributor (they did not respond to me directly, though I had written to them - very strange because they are usually very communicative).

Later I found similar reports on the 'net. I have added a few links which I had saved

https://www.head-fi.org/threads/schiit-gungnir-dac.603219/page-308#post-13930698
on measuring for DC in output

https://www.head-fi.org/threads/schiit-gungnir-dac.603219/page-309#post-13949801

https://www.head-fi.org/threads/schiit-gungnir-dac.603219/page-309#post-13951114

https://www.head-fi.org/threads/schiit-gungnir-dac.603219/page-310#post-13951316
0-20mV DC bias is considered very good. 20-50mV is typical, over 100mV can be a problem.

ciao
gr
 
To a person doing the measurement ?

I did not imply or even consider harm to operator, I was talking about what fried my equipment.

Since you asked, however I attempted to refresh my memory on how many volts or amps are lethal , https://www.quora.com/How-many-volts-will-kill-you. (I'm sure that there are more authoritative and informative sources but this is the best I could come up with at the start of the workday)

ciao
gr
 
For shorting you will need bare RCA plugs, you can buy some cheap cables and cut it close to the connector and join the two connector cables for shorting.
Just put the black probe of the DMM to any ground point and the red probe to the inner metal of the input on your tube pre to measure.
Set the tolerance to 20 on the DC voltage section of the DMM. Ideally you should not get any reading.
Test it under all conditions, cold as well as warmed up.

I conclude that’s too technical for me to do. And I don’t find myself competent to do the same. Will it suffice if I simply take the preamp (and also the DAC to check it) to an electronics expert/repairer for the testing? Or would I need to carry along the CD transport too?
 
@yogibear : I think a multimeter is not the right tool to measure the output of an amp or DAC since they usually output AC which is ever changing with the music. A workaround is to play a single frequency sine wave as it will output a fixed AC output. An oscilloscope would be a more appropriate tool, though. However, a multimeter is perfectly adequate to measure DC offset as it is a fixed voltage.

@SachinChavan : don't be scared off. Measuring DC offset is much easier than you think. The idea behind it is like this: a device like your headphone amp must amplify and output only audio signal which is essentially an AC signal. So it must not generate any DC component on its own and pass that on to the next component in the audio chain. However, no device is perfect so the headphone amp (or any other device, for that matter) can generate DC internally and amplify that just like it amplifies the audio signal. By stopping any input signal into the headphone amp either by physically shorting input or keeping it open, we are trying to determine if the gear under test is responsible for internally generating DC. If readings under shorted input is within safe limits, you need to check further to the source.

@GR, thanks for chipping in with your experience. Your input brought a direction for the OP to focus on.
 
@jls001 yes sir, Oscilloscope is the most scientific approach, also ensures accuracy.
But since I don't process one, I follow the multi meter way as a guide to help me set me bias, DC offset and amp gains.
Though bias and DC offset are straight forward, the power amp gain setting is a tough job with multi meter.

Since multi meters are usually reasonably accurate for 60hz sine wave, I use that tone to adjust gains on my amp as well to match the preamp output to prevent the amp from going into clipping or produce distortion at higher outputs. I do this setting at about 80% of the clean preamp output level at 0db gain on the amp. The target AC voltage across the speaker terminal on the amp as will be measured with the multi meter will be 'square root of the rated impedance x the amps nominal power rating'. After this approximation I then fine tune the amp gains with my ears with tracks that I am familiar with.
I have to do this for each channel as my power amp is a dual mono design with individual gain pots for each channel.
Sadly I do not process any load resister to do this, so the final setting is essentially by listening.

It is quite a painful process, but patience pays rich dividends.
 
@jls001 yes sir, Oscilloscope is the most scientific approach, also ensures accuracy.
But since I don't process one, I follow the multi meter way as a guide to help me set me bias, DC offset and amp gains.
Though bias and DC offset are straight forward, the power amp gain setting is a tough job with multi meter.

Since multi meters are usually reasonably accurate for 60hz sine wave, I use that tone to adjust gains on my amp as well to match the preamp output to prevent the amp from going into clipping or produce distortion at higher outputs. I do this setting at about 80% of the clean preamp output level at 0db gain on the amp. The target AC voltage across the speaker terminal on the amp as will be measured with the multi meter will be 'square root of the rated impedance x the amps nominal power rating'. After this approximation I then fine tune the amp gains with my ears with tracks that I am familiar with.
I have to do this for each channel as my power amp is a dual mono design with individual gain pots for each channel.
Sadly I do not process any load resister to do this, so the final setting is essentially by listening.

It is quite a painful process, but patience pays rich dividends.

How do you ‘adjust’ the gain on your amp?
 
Ok friends, I could take the Valhalla 2 tube preamp (as well as the Bifrost DAC) to an highly experienced repairing enthusiast. He checked the DC voltage and found the following:

Valhalla 2 tube preamp:
Left channel around 30-50 mV
Right channel around 200-500 mV (0.2-0.5 V)... pics attached for this channel
Note that there was no load and these readings on the multimeter keep fluctuating around the above values.

Bifrost DAC: around 20-30 mV both channels

So it seems the DAC doesn’t have much of a problem . But the tube preamp is passing high enough DC voltage that makes the 8200A poweramp’s protection circuit cut off.

Both the DAC and preamp are not leaking any AC voltage.

Would you agree with the inference based on the above readings? If so, I’d think the next step for me is to call up the distributor (headphonezone) and arrange for sending my Valhalla2 back to them for repair. It’s likely some capacitor in the output circuitry is malfunctioning. Am I right? 12ABCAB4-AD7C-426F-9A8F-0B1384B0F051.jpeg1261A311-53D5-4764-B5C0-11A8358FD306.jpeg
 
When you say:

Quote
He checked the DC voltage and found the following:
Unquote

Measured where?
Under what condition - was input given or no input given?

Quote
Valhalla 2 tube preamp:
Left channel around 30-50 mV
Right channel around 200-500 mV (0.2-0.5 V)... pics attached for this channel
Note that there was no load and these readings on the multimeter keep fluctuating around the above values.
Unquote

I feel even 30 mV is high even if it doesn't trigger protection circuitry.

Quote
Bifrost DAC: around 20-30 mV both channels
Unquote

This is also high.
 
When you say:

Quote
He checked the DC voltage and found the following:
Unquote

Measured where?
Under what condition - was input given or no input given?

Quote
Valhalla 2 tube preamp:
Left channel around 30-50 mV
Right channel around 200-500 mV (0.2-0.5 V)... pics attached for this channel
Note that there was no load and these readings on the multimeter keep fluctuating around the above values.
Unquote

I feel even 30 mV is high even if it doesn't trigger protection circuitry.

Quote
Bifrost DAC: around 20-30 mV both channels
Unquote

This is also high.

Measured without input. The first pic in my post above shows where he measured.
 
IMG_20180723_022933_HHT.jpg

This is the preamp output without any input at zero volume.
IIMG_20180723_022826_HHT.jpg

This is the preamp output without any input at max volume.
 
Aren't these method of measuring not correct ? You would need to feed 60Hz sine wave or use an oscilloscope ?

You should have followed methods to measure DC leaks as suggested by JLS.

However you can single out the troublesome channel by connecting one at a time.


JLS also said “However, a multimeter is perfectly adequate to measure DC offset as it is a fixed voltage”. And I have done the same. Measured DC offset. Why is the method not correct then?

The DC leakage was also checked (only on the body) and there was none.
 
Aren't these method of measuring not correct ? You would need to feed 60Hz sine wave or use an oscilloscope ?

You should have followed methods to measure DC leaks as suggested by JLS.

However you can single out the troublesome channel by connecting one at a time.

You are correct, but this is just to check if there is any extraneous DC flowing out of the output on switching on without any other connectivity.
In fact this is the first time I checked this way, after seeing the pics posted by Sachin.
 
How do you ‘adjust’ the gain on your amp?

The gain adjustment in only needed if your power amplifier has a gain pot. These gain pots are not volume controls, but a boost to the input signal if it is low.
For example, most modern sources and DACS have high output voltage upto 2 volts and commonly 4 volts for balanced outputs. Similarly the output from a pocket media player or your mobile phone is in milli volts to essentially drive loudspeakers.
So if you do not have a gain pot on your power amp, the music output from speakers will vary according to the input level at its source. A regular preamp will make the speakers sound dynamic, and a mobile phone output direct to the power amp will make it sound dull and lifeless. At the same time, if a preamp output is too high, it can drive the amp to distortion.

The power amp I use is Adcom GFA-4302 which has a gain pot for each channel that I can adjust from 400mv to 2v.
The TEAC UD-301 that I use serves both as DAC and preamp.
In pure DAC mode, it outputs 2 volt fixed and in preamp mode it maxes out at around 3.2 volts. This means that even in preamp mode, the output with the least THD is at 2 volts which is more than enough to drive the Adcom.

Step 1: Setting volume control on preamp
I put the TEac in preamp mode and set the volume control position where it measured 2 volts. To have a bit of headroom, I put the volume control on TEAC to around 85% of the 2 volt point.

Step 2: Finding max voltage at amp output
Finding the intended AC volate on speaker output of the amp. It follows the simple formula Voltage = square root of watts x ohms. The Adcom is rated for 30 watts per channel at 4 ohms. So the max AC voltage without being over driven at peak input DC voltage for Adcom is square root of 30x4 - around 11 volts.

Step 3: Setting the gains:
Since we now have the values, the ideal way is to put a 4 ohm load (resister) at the speaker terminals to measure, but load resisters are expensive, so I go without it and fine tune with my ears.
Download a long playing (atleast 5 minutes) 60hz, 0db sine wave and play it. The volume on the preamp should be at the position as in Step 1.
Now fix the multimeter prongs to the power amp speaker terminals and set the gain for each channel so that it is at around 11 volts, a little lesser is always better.
In my case, I needed to set only around 7% gain on the pot for the optimum voltage at the amp output.
After this, rewire the speakers, and play familiar tracks and fine tune the gain.
 
Follow HiFiMART on Instagram for offers, deals and FREE giveaways!
Back
Top