John Curl's Blowtorch preamplifier part II

Status
Not open for further replies.
No , but higher level (about 0,007-0,02%) of low order distortions (only 2. and 3. ) , frequency independent. Too much for my taste..No , but higher level (about 0,007-0,02%) of low order distortions (only 2. and 3. ) , frequency independent. Too much for my taste..Higher harmonic burried in noise background (under -120dB)
This is the same complaint used against digital volume controls, used on digital data - there's more distortion at lower levels. Well, guess what? The sound coming from the speakers is softer, but the distortion level in absolute terms has not altered one iota, it's just higher in comparison to the current maximum level that can be produced by the system. I've tried to hear this terrible behaviour on a system missing an analogue preamp able to pump up the volume to a ridiculous degree, and I can't pick it, even with my ear jammed against the tweeter cone ...

Frank
 
Last edited:
AX tech editor
Joined 2002
Paid Member
Chris, Richard is right. It is not Hawksford.

Yes there are priors, which AKSA highlighted after I originally posted it up a few months ago. However, as far as I can tell, it's never hit the mainstream.

No its not Hawksford - Hawksford involves positive feedback to increase the loop gain so there's more effective nfb.
Andy's way is a form of active feedback.

jan
 
Best from those was CMOS DAC (DAC8043) with good OA, e.g .LM4262. Catalogue data are quite nice, but try to measure e.g. PGA2310 at -6-20dB gain (not 0dB..) and input signal about 2V RMS.
Well, i had not measured, or even listened to it, but, on the paper, DACs used as stepped attenuators suffer from several problems, that the PGAs are rid of.
Resistances used inside the"R-2R"of DACs are silicon resistors, known for their measurable and audible distortions. And the signal flows across several of them and several switches as well. The attenuation curve is linear, not log.

Because we are talking about error cancellation, that is a good way to cancel the (litle) distortion of MOS FETS used as switches, adding one of them in the feedback loop of the following OPA (gain 1) .

All that is on the paper. In real life, comparing with strait wire, i found PGAs does not distroy signals is such a way that your listening pleasure can be reduced, nor have a too obvious signature. Not even sure if i do not prefer the sound with-it than without. I prefer their sound from an ALPS (more 'fluid') and, with all the advantages added (low price, simplicity, no wear, remote, perfect gain step accuracy) i think the problem of controlling gains has been addressed in a satisfying way for the long run.
Now, if snobbish audiophiles prefer to fulfill a preamp with expensive relays and a lot of crap around, they are free to do -it, but, why ? They should better use their time working on more obvious sources of deteriorations, like their loudspeakers.
 
Last edited:
Now, if snobbish audiophiles prefer to fulfill a preamp with expensive relays and a lot of crap around, they are free to do -it, but, why ? They should better use their time working on more obvious sources of deteriorations, like their loudspeakers.
It has nothing to do with that..The only reason is sound quality (minimal degradation). And differencies between PGA, DAC with OA in multiplying mode and relay attenuator are clearly audible and measurable, even with "ordinary" speakers.
 

Attachments

  • ma6-a.jpg
    ma6-a.jpg
    4.8 KB · Views: 233
It has nothing to do with that..The only reason is sound quality (minimal degradation). And differencies between PGA, DAC with OA in multiplying mode and relay attenuator are clearly audible and measurable, even with "ordinary" speakers.
Everything is clearly audible. there is a difference between any DAC, any power amp etc.
The question is: did-it change the sound in a bad way, "which can reduce your listening pleasure" as for a good potentiometer, my response is 'NOT really'. And LESS than a good potentiometer.
Does-it reduce separation between instruments ? No.
Does-it change the texture of them ? No (Pots add a little granular sound)
Does it reduce instant dynamic (transients) ? Not sure.
Did-you lose details or presence ? Not sure.
Does it add noise you can notice ? No.
Are the levels steps low enough for comfort ? yes.
Are tuning the volume agreeable enough ? Yes.

The changes that occurs are less than those that could happen if the mix of the record you listen too began half an how later. or mastered in an other place.
Of course, you'll be sure the circuit is correctly implemented, source is low impedance, grounds well separated between digital and analog, PSU low noise enough and good decoupling of the PGA analog. And no capacitances in the signal path.
Here is the sens of my 'snobbish'. there is a limit of the quest of 'absolute', because 'absolute does not exist.
 
Last edited:
What Parasound has been using for many years is a motorized Alps pot, sometimes 4 gang, for a volume control. Personally, while I have never MEASURED any problem with Alps pots, I prefer TKD, P&G, or a quality dual selected 10 turn wirewound resistor, for a volume control. Unfortunately, TKD is seemingly impossible to work with, especially motorized, although I once had a sample that worked great.
P&G is rated to only 10K, (in my experience) too low for Parasound, and while they are expensive, I suspect that the really well tracking ones are 'cherry picked' out for major manufacturers, like Levinson and Krell, in advance.
10 turn wirewound pots are VERY EXPENSIVE, almost too much for CONSTELLATION, and that is saying something.
However Alps, is greatly reducing its selection, and we may have to use an electronic volume control in future designs, because of this.
Personally, anybody who prefers a deliberately reduced bandwidth is not someone who I would rely on for subjective input, so that leaves a few inputs here in suspension.
Now, what about REAL BANDWIDTH? Normally the loudspeaker will be the ultimate limit, and reducing the bandwidth in the electronics should barely make a difference, UNLESS the playback is passing nearly ultrasonic garbage to the following amps and making them misbehave.
I don't know specifically why most tweeters have a problem passing info above 20KHz, but it appears to be so, including most horns.
Therefore the RATE OF CHANGE of the drop-off of the high frequencies must be an important contributor to the SOUND of the high frequency attenuation, more than just extended response of the electronics.
 
www.hifisonix.com
Joined 2003
Paid Member
Look attached pictures, DAC+OA measurement is about 3 years old, loopback ,unbalanced connection (so visible 50Hz), PGA in balanced connection. Same level, same measurement conditions.

Can show the circuits you used. A lot of this has to do with implementation.

That is NOT the PGA23xx I know and love! 2H -75 dB down? You should be sitting between -100 and 110 dB into a 2k load.

Thanks
 
Last edited:
Look attached pictures, DAC+OA measurement is about 3 years old, loopback ,unbalanced connection (so visible 50Hz), PGA in balanced connection. Same level, same measurement conditions.

OK, so it's not MDACs in general, it's either the specific one you used or your implementation. The measurements on the ML I linked to were outstandingly good.
 
Status
Not open for further replies.