Violectric HPA V281 - Vorsprung durch Balanced
May 3, 2018 at 4:05 PM Post #3,631 of 5,166

jerick70

Headphoneus Supremus
Joined
Jun 25, 2014
Posts
6,181
Likes
7,723
Location
Utah, USA
Anything is possible. The upgrade was just unplugging the old ribbon cables and plugging in the new cables, and of course replacing the faceplate and old volume control. The difference is definitly there. I asked Violectric about it and they didn't think it was an issue with the unit. They said that some people see this change with the upgrade. I'm glad the money was well spent.
 
May 3, 2018 at 4:11 PM Post #3,632 of 5,166

ra990

Headphoneus Supremus
Joined
Jun 13, 2017
Posts
1,783
Likes
2,632
Location
Virginia, USA
Anything is possible. The upgrade was just unplugging the old ribbon cables and plugging in the new cables, and of course replacing the faceplate and old volume control. The difference is definitly there. I asked Violectric about it and they didn't think it was an issue with the unit. They said that some people see this change with the upgrade. I'm glad the money was well spent.
Cool, I'm glad it was a worthy upgrade for you. Was it difficult to do by yourself or pretty straight forward?
 
May 3, 2018 at 4:45 PM Post #3,633 of 5,166

jerick70

Headphoneus Supremus
Joined
Jun 25, 2014
Posts
6,181
Likes
7,723
Location
Utah, USA
Cool, I'm glad it was a worthy upgrade for you. Was it difficult to do by yourself or pretty straight forward?
The most difficult part for my install was getting the volume and balance knobs centered so the indicator cut matched the correct adjustment. The other thing that throws a curve ball into the install is drilling out the volume control hole on the face plate to fit the new volume control. I didn't have to do that because I upgraded the face plate and it was already done. Other than those two items it is very simple and strait forward.

Remove the top of the unit -> Remove ribbon cable connections from volume control to mainboard on main board end-> Remove the face plate -> Remove volume and balance knobs -> Detach old volume control (Just a few screws) from face plate -> Drill the volume control hole on face plate larger -> Install new volume control (Pretty much the same as when removing old control but with a few more ribbon cables) -> Test remote control

I would say this is a low skill DIY project.
 
Last edited:
May 3, 2018 at 5:05 PM Post #3,634 of 5,166

zhgutov

100+ Head-Fier
Joined
Dec 17, 2014
Posts
285
Likes
89
I would expect better stereo image due to less crosstalk
Standard volume control has excellent crosstalk (taken through line out post):

(this measurements are limited by the noise floor primarily)

I think the crosstalk (which is worse under the load, and this is not because of the volume control)
does not affect the stereo image when it is below -60 dB (probably even -40 dB).

Better stereo image is caused by something else.
 
May 3, 2018 at 6:26 PM Post #3,635 of 5,166

jerick70

Headphoneus Supremus
Joined
Jun 25, 2014
Posts
6,181
Likes
7,723
Location
Utah, USA
Standard volume control has excellent crosstalk (taken through line out post):

(this measurements are limited by the noise floor primarily)

I think the crosstalk (which is worse under the load, and this is not because of the volume control)
does not affect the stereo image when it is below -60 dB (probably even -40 dB).

Better stereo image is caused by something else.
Intriguing. I'm not an expert with audio measurements, but would like to be. What do you think we can measure to track down the changes I'm hearing between volume controls? I imagine this being a really interesting experiment.
 
May 3, 2018 at 6:51 PM Post #3,636 of 5,166

Justin_Time

500+ Head-Fier
Joined
May 4, 2009
Posts
862
Likes
296
Location
Houston/Dallas, Texas
Standard volume control has excellent crosstalk (taken through line out post):

(this measurements are limited by the noise floor primarily)

I think the crosstalk (which is worse under the load, and this is not because of the volume control)
does not affect the stereo image when it is below -60 dB (probably even -40 dB).

Better stereo image is caused by something else.

If as you said crosstalk is not part of the benefit, would you care to speculate as to why these Lake People went through the trouble of offering a rather expensive option with the potential—now a reality for some—of bothering customers with irritating noises?

From a purely a layman viewpoint, I thought that intuitively the small and variable differences between the left and right would cause the center image to slightly shift, which would be audible. But I fully realize that intuition can be completely wrong.

Any explanation to shed more light on the benefit or lack thereof these relay attenuators would be much appreciated.
 
May 3, 2018 at 7:15 PM Post #3,637 of 5,166

zhgutov

100+ Head-Fier
Joined
Dec 17, 2014
Posts
285
Likes
89
Intriguing. I'm not an expert with audio measurements, but would like to be. What do you think we can measure to track down the changes I'm hearing between volume controls? I imagine this being a really interesting experiment.
Maybe the difference in THD and IMD between the measurement setup itself (!) and the whole setup with the V281 (in my case it makes distortion very slightly higher, 2nd harmonic primarily).
Maybe the difference in phase over frequency, but in the case of relay volume control this may be different for line out and phones out.
Maybe something else... Everything which may show the difference, I suppose. It is possible to produce more interesting custom test signals.

All this may show the difference, but it is possible, that you will not find anything meaningful in such measurements.

Maybe it is even better to make "loopback" recording with the different music, match the volume and compare the following using the different playback systems:
- original
- measurement system only
- standard line out
- relay line out
- standard phones out with no load
- relay phones out with no load

This should be working, because the volume control is situated before the output stages (and this is easy to use in A/B comparison when you have only one V281).
But probably you need very transparent setup to estimate the difference, not only to hear it is different.

But I don't have relay volume control option to do such experiment.

From a purely a layman viewpoint, I thought that intuitively the small and variable differences between the left and right would cause the center image to slightly shift, which would be audible. But I fully realize that intuition can be completely wrong.
This is not about crosstalk, this is about channel imbalance. This may be useful in mastering tasks.
This can make you happy because you know for sure that you have 0.1 dB tolerance.
This can make me happy, I want US4+, not US4. And I don't care about possible noise during the volume adjustment. But this is me.
This is what I think does not have a negative effect on the sound quality in the given range (1 dB), but I don't know about other people, maybe somebody can hear that and consider the effect as negative.

If as you said crosstalk is not part of the benefit, would you care to speculate as to why these Lake People went through the trouble of offering a rather expensive option with the potential—now a reality for some—of bothering customers with irritating noises?

Any explanation to shed more light on the benefit or lack thereof these relay attenuators would be much appreciated.

It is possible that the relay volume control has even better crosstalk. Why not?
But I think the crosstalk importance is overestimated (by the community) when we are talking about such numbers, when the crosstalk is flat over the frequency range, and when its phase is flat too.

The crosstalk is not the only thing!

Just for example, there is RS05 (clock) which makes my V800 (DAC) better sounding in terms of imaging/depth.
This is easy to hear, but it is really hard to see in measurements clearly. Nothing wrong with that.
P.S. I'm talking not about the effect of the resampling vs bad source. Particularly this is very easy to see in measurements.
 
Last edited:
May 4, 2018 at 3:42 PM Post #3,638 of 5,166

zhgutov

100+ Head-Fier
Joined
Dec 17, 2014
Posts
285
Likes
89
Well, after this discussion I found myself measuring this amp :)

Looks like the crosstalk becomes worse when the volume (standard RK27) is turned down.


The whole graph for 12 o'clock is higher because I reached the limits of my ADC (noise floor).
The most interesting thing is the 3 kHz .. 20 kHz range. This is where some degradation happens.
Pre and post look pretty similar here when the volume set to maximum.

Another thing is... let it be SMPTE (IMD).


Pre accurately repeats the harmonics.
Post shows slightly higher second harmonic.
The difference in THD looks very similar.

So, there is some room for improvements, at least according to that measurements.
The most interesting thing is to compare that (Post) with the relay volume control.

Maybe there is something else, what can be measured too.
And this is still interesting, what exactly we can hear.
 
Last edited:
May 4, 2018 at 8:00 PM Post #3,639 of 5,166

ra990

Headphoneus Supremus
Joined
Jun 13, 2017
Posts
1,783
Likes
2,632
Location
Virginia, USA
Well, after this discussion I found myself measuring this amp
Really interesting zhgutov! That makes a lot of sense and is what I would expect. Otherwise, they wouldn't offer a superior volume control option that would improve on this at very low volumes. The relay stepped attenuation always has that benefit, but at what point on the base volume control do these issues basically become negligible? I normally listen to it around the 12 o'clock position and at that point the volume control should result in pretty good numbers, I would expect.
 
May 5, 2018 at 1:10 AM Post #3,640 of 5,166

zhgutov

100+ Head-Fier
Joined
Dec 17, 2014
Posts
285
Likes
89
Really interesting zhgutov! That makes a lot of sense and is what I would expect. Otherwise, they wouldn't offer a superior volume control option that would improve on this at very low volumes. The relay stepped attenuation always has that benefit, but at what point on the base volume control do these issues basically become negligible? I normally listen to it around the 12 o'clock position and at that point the volume control should result in pretty good numbers, I would expect.
Ha! Thank you for asking :)

This looks like the ALPS pot nonlinearity. I will show my case here, but you can have different results.
This nonlinearity exists in my case between 11 and 4 o'clock.
This is my own test signal, but it is very similar to the signal which is used in RMAA for the crosstalk test.
Green is signal, blue is crosstalk. It is slightly different for Left and Right channels, but the whole picture remains the same.

Max level:


4 o'clock:


12 o'clock (your favorite):


11 o'clock:


10 o'clock:


There is even more interesting thing. The graphs shown above are not using the whole digital range.
I will show three volume pot positions which are close to each other.
Normally, the crosstalk of V800 > RS04 is below -120 dB at 20 kHz.
Blue is signal, green is crosstalk here.

Max level:


4.5 o'clock (two detents lower):


4 o'clock (another two detents):


What else can we expect from the ALPS RK27?

But actually I think, that the levels of this crosstalk (in my case) are negligible.
Despite that, now I want to upgrade to the relay option more.
 
Last edited:
May 5, 2018 at 1:18 AM Post #3,641 of 5,166

ra990

Headphoneus Supremus
Joined
Jun 13, 2017
Posts
1,783
Likes
2,632
Location
Virginia, USA
Ha! Thank you for asking :)
That's amazing, thanks for running these tests! Can you elaborate on the results? It looks to me like there is the least amount of crosstalk at the 4.5-max volume setting, is that correct?
 
Last edited:
May 5, 2018 at 1:32 AM Post #3,642 of 5,166

zhgutov

100+ Head-Fier
Joined
Dec 17, 2014
Posts
285
Likes
89
Can you elaborate on the results? It looks to me like there is the least amount of crosstalk at the max volume setting, is that correct?
Yes, this is true in my case. This is likely to be the same in your case too.
But the crosstalk becomes lower not only at the maximum level.
It becomes lower when the volume is below 11 o'clock (again, in my case).
 
Last edited:
May 5, 2018 at 3:54 AM Post #3,643 of 5,166

fdg

Member of the Trade: Lake People
Joined
Oct 29, 2009
Posts
178
Likes
175
@ zhgutov (Dimitry) and the measurements.

Crosstalk is always measured by every manufacturer with the volume potentiometer fully open (right or CW position).
As a matter of facts and by design a potentiometer will drive the signal to high(er) impedance state
– thus makes the signals more sensitive to hum, noise, the other channel(s) - when turned lower (to the left, CCW).
Also, the signal traces to the next stage will suffer from these higher impedances.

This is one of the reasons why passive attenuators will not make anything better!

There are some measures to lower these negative side effects but you will never totally eliminate them:
1. Chose a potentiometer as big as possible (as long as you can afford it) to achieve some space between the signals.
This is the advantage of the Alps potentiometers RK27, RK40 or RK50.
They are big and have separate chambers for the channels and so have lower side effects compared to
miniature potentiometers like RK16, RK14, RK11 or RK9 and of course the makes of other manufacturers.
2. Design an op-amp as close as possible to the potentiometer acting as a buffer amp to achieve low impedances after the buffer.
3. Chose electronic attenuators instead of mechanical potentiometers. Those will have less crosstalk issues but the sound will suffer.
4. Invest in alternative electronic designs like relay controlled attenuators :smile_phones:
 
May 5, 2018 at 4:39 AM Post #3,644 of 5,166

zhgutov

100+ Head-Fier
Joined
Dec 17, 2014
Posts
285
Likes
89
As a matter of facts and by design a potentiometer will drive the signal to high(er) impedance state
Always wanted to ask about the balance control.
This is another potentiometer (RK27) with the different parameters. This potentiometer exists in the relay-based implementation too.
Is my understanding correct, and it is not in the signal path directly, but its impedance controls the corresponding op-amp which is responsible to make low impedance after the volume control?
 
Last edited:
May 5, 2018 at 4:56 AM Post #3,645 of 5,166

zhgutov

100+ Head-Fier
Joined
Dec 17, 2014
Posts
285
Likes
89
It looks to me like there is the least amount of crosstalk at the 4.5-max volume setting, is that correct?
No, I don't think so, not 4.5 o'clock. This looks like there is another source of the crosstalk (maybe this is V800 and RS04, maybe something else), and it has different phase, so they can compensate each other partially.
 
Last edited:

Users who are viewing this thread

Top