The Stax Thread III

Aug 7, 2014 at 12:40 PM Post #2,897 of 27,880
  Cavalli has a new Liquid Lightning version, the LL2T.
A major change is a digital volume control. Any comments on such a thing?  (I think Malvalve has one too.) Thanks.

 
This sounds interesting. I went to his website a few weeks ago and only saw the old LL2. I have the original LL and I really love it. Looking forward to hear from anyone who ends up getting it... 
redface.gif

 
Aug 7, 2014 at 4:08 PM Post #2,898 of 27,880
[/quote]
Cavalli has a new Liquid Lightning version, the LL2T.
A major change is a digital volume control. Any comments on such a thing?  (I think Malvalve has one too.) Thanks.


Ehh? How can we have a digital volume control with analogue signal only? It is not a DAC. Maybe you mean a circuit controlled
volume setting changing the resistance values?

Back to an Actual Digital volume, they all sound worse than analogue (shunt type resistors or old style wipers) IMO. It flattens the dynamics
and gives high frequencies an edge, as in read errors.
 
Aug 7, 2014 at 4:16 PM Post #2,899 of 27,880

Ehh? How can we have a digital volume control with analogue signal only? It is not a DAC. Maybe you mean a circuit controlled
volume setting changing the resistance values?

Back to an Actual Digital volume, they all sound worse than analogue (shunt type resistors or old style wipers) IMO. It flattens the dynamics
and gives high frequencies an edge, as in read errors.[/quote]

64 bit dithered digital volume control is very good. Heck even non dithered is fine if outputting 24 bit because it has bits to 'throw away' .
 
Aug 7, 2014 at 4:24 PM Post #2,900 of 27,880
   
This sounds interesting. I went to his website a few weeks ago and only saw the old LL2. I have the original LL and I really love it. Looking forward to hear from anyone who ends up getting it... 
redface.gif

http://cavalliaudio.com/index.php?p=morestuff&stuffId=13
 
 
Sorry, I got the attenuator description wrong.
It's this (quoting):
 
A Cavalli Audio unique processor-controlled, optically-coupled, photoresistor volume control. This entire assembly lives on a small board that sits at the rear of the amp near the input section.
You can see the cable running to the control pot on the front panel. The control is a small TKD pot that tells the opto controller where to set the volume, but since it is a nice TKD pot it has the same smooth action as all TKD pots. We have tested and listened to this setup. It sounds really good.
 
Aug 7, 2014 at 4:30 PM Post #2,902 of 27,880
There are 2 problems with Light Dependent Resistors.
The first problem is that they drift over time and need calibration.
 
These guys seem to be the king of in the field recalibration
http://www.tortugaaudio.com/product/ldr3x-passive-preamp-controller-version-2/#prettyPhoto
takes 10 minutes (20 balanced) to do a full calibration.
 
The other problem with LDR's is that they have significant THD in the range where
you would want to use them. Nelson pass did a fair bit of work testing these things
 
http://www.diyaudio.com/forums/analog-line-level/80194-lightspeed-attenuator-new-passive-preamp-134.html#post1520243
 
others have done the same
scroll down to the middle of the page here
http://www.diyhifi.org/forums/viewtopic.php?f=13&t=1892
 
constant impedance step attenuators with very high quality resistors are one way to do it right.
The twisted pear thing is NOT constant impedance. The sigma1 by amb IS constant impedance.
 
The other way it seems is something like the khozmo which is only 2 resistors in the path at
any time. But built with a switch that lasts more than a month.
 
There is a company in Brittan that makes a really nice one, pretty big and pretty expensive.
analogue devices, 42 steps.
 
and if you have a bunch of room, the shallco that is also used in the ctc blowtorch, 45 steps
http://www.hificollective.co.uk/components/shallco_stepped_attenuator.html
 
the ayre preamp also uses the shallco I believe, with a motor drive.
 
Aug 7, 2014 at 5:16 PM Post #2,904 of 27,880
Agree with this 100%.


JRiver, Amarra, Audirvana+ all those volume controls are aweful. If you think they sound good you need to upgrade
your equipment. They kill the sound, and the software makers admit it, they recommend an actual pot if possible.
 
Aug 7, 2014 at 5:25 PM Post #2,905 of 27,880
What about the stepped attenuators on a chip, are these any good? I believe my Grace Design m903 uses one like that. It has a clicky dial on the front, which operates an analog stepped attenuator on a chip in steps of 0.5 dB. Channel matching is superb on all levels, and it's a joy to use, but I'm wondering if anybody can comment on the relative sound quality of such a device.? What are the trade-offs if you compare it with a high-quality pot?
 
Aug 7, 2014 at 7:25 PM Post #2,906 of 27,880
The goldpoint uses a 23 position switch, the DACT uses a 24 position switch.
Both seem reliable for many years but in my opinion both are too coarse and
do not have enough positions.
 
For the crazy out there, there is this
http://www.hificollective.co.uk/catalog/glasshouse-ladder-stepped-stereo-charcroft-seiden-p-10088.html
 
and I do have one of these, also crazy expensive
http://www.tkd-corp.com/products/att/pdf/type-cs-e.pdf
 
The 48 step khozmo would be great, maybe the newer ones are more reliable.
 
The 63 step shallco are very VERY nice. Also large.
 
way back when, Collins radio made 79 step attenuators designed for mixing boards
for radio stations and the like. The most fantastic thing I have ever used. Also large.
Had the best feel of anything else and lived forever. Silver contacts...
 
The TI pga2320 is a well known and fairly decent chip based thing with a thd of .0003%
but I have never seen a high end preamp use them.
 
Way back when Levinson had two different solutions both of which evidently people
really liked, and both were expensive. The first one was a 12 bit multiplying dac
(actually 4 of them) where the audio input was the voltage reference. So you can
consider this IC based. The other one was a standard R2R thing that used
datageneral dg501 (now Vishay) cmos switches as the switching elements
all on a Teflon circuit board.
 
The nelson pass high end preamp does the same R2R attenuator with hand
selected fets used as the switching elements, biased in a way such that they
produce ultra low distortion.
 
I like my 256 position .25db per step gold crosspoint relay attenuator better
than all the above. (well if I could get the Collins radio things...)
 
Aug 8, 2014 at 5:17 AM Post #2,907 of 27,880
That is very detailed information Kevin. Thank you. This subject is huge and has sonic impacts more than people realise. It might be better under it's own thread. Back my the Khozmo I have just bought one of their Passive 'pre amps', the stereo model with selector. I choose the Takman z foils version. I will post my findings in the next few weeks. It is actually going i my speaker rig, fed by a low output impadance of 10 ohms from my Tubed DAC which has 2v output and transformers on the output.

Short interconnects then the Khozmo 10k passive, feeding my Plinius power amp which is 47k input impedance + an active sub which is 10k input impedance. So I was thinking my DAC source would see a load of around 37K? Is that more or less correct Kevin? If the passive sounds awefull due to impedance mismatch, I will need to revert back to using an active pre-amp. The Plinius can reach it's rated power on .8v input so hopefully there will be enough current.

Any thoughts on this subject Kevin. Sorry, off topic on this thread guys.
 
Aug 9, 2014 at 2:42 AM Post #2,908 of 27,880
Hello,
 
Here are the latest pics available of the new Energizer AudioValve Verto, who seems wanting to go further in terms of versatility and the sound quality (coupled to the amplifier of the same brand Audiovalve RKV-II or III) than its direct competitor, the Woo-Audio Wee.
 
These photographs, taken this time from inside the case, show only two (big) transformers of the brand Pikatron, as well a series of... (?) participating in the creation of a Self Biasing (SB) from the input source (somewhat after the manner of Energizer Stax SRD-7 SB of yesteryear).
 
The supposed superiority of the Verto compared to the Wee would come from both of the best intrinsic quality of the Pikatron transformers, as well as a lower ratio of amplification of the voltage transformers (ratio 1:5 for the Verto versus 1:50 for the Wee)  (NB: transformers optimized for the RKV-Verto combo), that which would give greater transparency and clarity to the sound, as well as less sound distortion.
 
The Self-Biasing (SB) 580 Volt DC charging time of the Verto (if headphone Stax Pro plugged, taking as a source input Jack of Verto (connected to the RKV)) would be of the order of two minutes.
 
A question: what is the advantage of a Self-Biasing (SB) compared to a transformer directly dedicated as 580 Volt DC bias generator?
 
Another question, related to the first question: Spritzer seemed to say that the generator (transformers) of bias voltage (580 V DC) of the Woo-Audio Wee Energizer was a 'diaphragm killer' (because of the instantaneous voltage of diaphragm implementation?, which would not be the case of loading much more progressive and long provided by a SB device). Is this true?
If this criticism is justified, would it not also the case of all Stax amplifiers when they are plugged and unplugged on Stax headphones on the output (outlet Stax) of the amplifier already lit, same potentiometer to zero from the amplifier?
 

 

 
" the high-voltage cascade VERTO produced up to 1.5 kV, and is generated only by the signal voltage. It takes about 2 minutes to the charge for the bias voltage established and stable, you can see this at the white glow of theLED, it only serves to control. To establish the charge an adequate volume is required "
 
 

 
http://www.audiovalve.info/to/verto.php
 
Aug 9, 2014 at 7:25 AM Post #2,909 of 27,880
  The goldpoint uses a 23 position switch, the DACT uses a 24 position switch.
Both seem reliable for many years but in my opinion both are too coarse and
do not have enough positions.

 
Hahaha, not enough positions.  OK.
 
Reminds me of a girl I used to date.  
 
Aug 9, 2014 at 7:58 AM Post #2,910 of 27,880
   
 
Another question, related to the first question: Spritzer seemed to say that the generator (transformers) of bias voltage (580 V DC) of the Woo-Audio Wee Energizer was a 'diaphragm killer' (because of the instantaneous voltage of diaphragm implementation?, which would not be the case of loading much more progressive and long provided by a SB device). Is this true?
If this criticism is justified, would it not also the case of all Stax amplifiers when they are plugged and unplugged on Stax headphones on the output (outlet Stax) of the amplifier already lit, same potentiometer to zero from the amplifier?
 

 
All stax made amplifiers and in fact most other amplifiers have 5 Megohm series resistors to the diaphrams from the bias supply limiting the charging current
and thus controlling the charging time. Anything that puts any amount of capacitance after this series resistor can damage the headphones.
 

Users who are viewing this thread

Back
Top