Hello, got your reply. Will reply back as soon as I can confortably multi-quote on my desktop back home
Best,
That's what I use for my personal daily use.
Inputs -> Khozmo SMD relay based stepped attenuator w Vishay precision SMD resistors / 64 steps quad (balanced) with microcontroller -> UGS -> Outputs.
I personally don't think of a buffer as a voltage gain device
. It adds minimalistic parts of noise and disto and no voltage gain; almost unmesurable as you stated. Feel free to play w/wo buffers
Armand said:
Quite Ok with that.
Armand said:
I use UGS_out with 10R or so series before output jacks.
Then the XLR cable, 11m to the power amps.
Aleph J's mono blocs, input impedance 240k balanced.
Works fine with no added noise/disto/weird things
Armand said:
I wouldn't say it's A LOT of 3rd, respectively to the fundamental in level.
Besides, if the First Watt is so important, for a preamp the lowest part level of output gain is very important for sound character,
rather than absolute disto% measured.
I would say that musically, the sweep spot seems there
For the left part of curve, disto goes down, till one point.
Here what we see is output noise generated by psu / active gain devices rather than real distortion added by gain.
As it is a balanced class A gain stage, distortion is monotonic, that means it vanishes at low levels.
If the output swing disto of the right part of the curve would be lower at higher output voltages,
then it rather would be certainly higher at very low levels on left part of the curve.
And the lowest % disto point would be far further right on the graph I guess.
Best,
Thanks for your input nar.
The stepped attenuator solution is nice. I made a prototype using this also once, but both in blind tests and measured test we were not able to find any difference compared to the Muses volume chip so we left the idea. But i just loved those clicking sounds
The beauty with relays is that the output impedance is zero. That is a luxury we don't have when using the Muses chip. Anything below 1Mohm starts to degrade the performance and that is my main concern now when connecting the UGS after the volume chip. But I will try to remove the 47k resistance to ground on the UGS gain module and see what happens. I am unsure how much current the circuit will draw from the volume chip. The FET's wil certainly not cause any problems, but there might be something from the feedback..?
The stepped attenuator solution is nice. I made a prototype using this also once, but both in blind tests and measured test we were not able to find any difference compared to the Muses volume chip so we left the idea. But i just loved those clicking sounds
The beauty with relays is that the output impedance is zero. That is a luxury we don't have when using the Muses chip. Anything below 1Mohm starts to degrade the performance and that is my main concern now when connecting the UGS after the volume chip. But I will try to remove the 47k resistance to ground on the UGS gain module and see what happens. I am unsure how much current the circuit will draw from the volume chip. The FET's wil certainly not cause any problems, but there might be something from the feedback..?
Sorry, my bad, 44k2 or so balanced (sum of the input Rs to virtual ground)
Yes, I use standard values. I can tell that UGS gain module has no trouble driving this front end
To compare the MUSE UGS with my own XP20 clone with MUSE as well, I took a few measurements with my CLIO system. The system resolution is not as good as the R&S system, but you'll get the idea...
First image is the preamp noise floor (GREEN) compare to my CLIO own noise floor (RED), with input SE, terminated in 50 ohms, maximum volume. XP20 noise floor is near -130dBV, that's less than 0.4uV, very quiet...
SE measurements:
Voltage gain is measured at Vin = 200mV, Vout = 0.61V, Gain = 9.5dB
Zout = 150 ohms (Measured at 1V, 1Kz), btw there is a buffer for the SE output...
Second image is spectrum at 1V, distorsion very low at 0.005%
Than I measured the THD at 1Khz, from Vout 200mv up to 8Vrms, THD started at 0.006% and went up to no more than 0.008%
Third image is the spectrum at maximum Vout of 9.6V (measured using a 10:1 volt divider to reduce the CLIO input), and I got a maximum of 0.01%
BTW the preamp can goes up to a Vout max of 21.5V RMS before clipping. The preamp inside is shown next. BTW not shown, there is a large overkill separate power supply with it, fully dual mono with caps mutipliers, and separate digital supply.
I tried to measure the distortion at Vout 20V, but my generator signal at this amplitude has already 0.134% of distorsion. The preamp added not THD that I can measure, meaning it is still probably less than 0.01% at Vout 20V!
This is stellar THD performance...
First image is the preamp noise floor (GREEN) compare to my CLIO own noise floor (RED), with input SE, terminated in 50 ohms, maximum volume. XP20 noise floor is near -130dBV, that's less than 0.4uV, very quiet...
SE measurements:
Voltage gain is measured at Vin = 200mV, Vout = 0.61V, Gain = 9.5dB
Zout = 150 ohms (Measured at 1V, 1Kz), btw there is a buffer for the SE output...
Second image is spectrum at 1V, distorsion very low at 0.005%
Than I measured the THD at 1Khz, from Vout 200mv up to 8Vrms, THD started at 0.006% and went up to no more than 0.008%
Third image is the spectrum at maximum Vout of 9.6V (measured using a 10:1 volt divider to reduce the CLIO input), and I got a maximum of 0.01%
BTW the preamp can goes up to a Vout max of 21.5V RMS before clipping. The preamp inside is shown next. BTW not shown, there is a large overkill separate power supply with it, fully dual mono with caps mutipliers, and separate digital supply.
I tried to measure the distortion at Vout 20V, but my generator signal at this amplitude has already 0.134% of distorsion. The preamp added not THD that I can measure, meaning it is still probably less than 0.01% at Vout 20V!
This is stellar THD performance...
Attachments
Last edited:
Here similar results from my UGS ''All-Inclusive'' preamp, with switching relay volume control.
Gain 3.5dB (Vin 1V, Vout 1.5V)
Vout max before clipping 9Vrms
First image, noise floor, again very good (SE input, terminated in 50 ohms, Max volume)
1V, THD 0.034% (1Khz, max volume, SE input), second image, mostly 2nd & 3rd harmonics
3V = 0.064%
5V = 0.121% (Last image)
7V = 0.17% (but I had to used an other generator that is already 0.134% by itself...)
8V = 0.2%
9V = 0.3%
Considering that I never used this preamp with 2V output, since it would had blown my hearing and my speakers, I was never concern about this very low distortion below 2V... And it is still pretty low at 3V. All in all pretty good preamp...
Gain 3.5dB (Vin 1V, Vout 1.5V)
Vout max before clipping 9Vrms
First image, noise floor, again very good (SE input, terminated in 50 ohms, Max volume)
1V, THD 0.034% (1Khz, max volume, SE input), second image, mostly 2nd & 3rd harmonics
3V = 0.064%
5V = 0.121% (Last image)
7V = 0.17% (but I had to used an other generator that is already 0.134% by itself...)
8V = 0.2%
9V = 0.3%
Considering that I never used this preamp with 2V output, since it would had blown my hearing and my speakers, I was never concern about this very low distortion below 2V... And it is still pretty low at 3V. All in all pretty good preamp...
Attachments
Last edited:
I have finally got some time to work on finishing my Muses UGS. I am currently in the stage of putting together and run all modules on the bench so still a bit to go.
Notice one issue so far - I can not get balance to work. I 'learn' uC balance buttons of my RC5 remote exactly the same way as all other buttons/commands e.g. standby, volume, etc but when I try to use this function once programming in finished there is no reaction except for power led blinking.
From what I could google translate from french forum the same question was raised some time ago but there was no follow up.
I tried to re-flash again but no succues. I am on 2.9 version with dual OLEDs.
Can someone of you who has working version with firmware 2.9 confirm if balance works as expected?
Appreciate you help!
Notice one issue so far - I can not get balance to work. I 'learn' uC balance buttons of my RC5 remote exactly the same way as all other buttons/commands e.g. standby, volume, etc but when I try to use this function once programming in finished there is no reaction except for power led blinking.
From what I could google translate from french forum the same question was raised some time ago but there was no follow up.
I tried to re-flash again but no succues. I am on 2.9 version with dual OLEDs.
Can someone of you who has working version with firmware 2.9 confirm if balance works as expected?
Appreciate you help!
