John Curl's Blowtorch preamplifier part II

[1audio]

Breakdown voltage is not the only issue related to supply voltage. I asked a national guy (before it was swalled up by TI) why they had so many opamps in the (soon to be defunct) LME family. He told me that they were sorted by leakage and breakdown form the same die in a number of cases. Probably a 50% overvoltage opamp won't meed many of its specs, everything from offset, input bias current, freedom from latchup etc. It may survive and amplify but all manner of strange things could happen. By the time you reach the smoke point a lot of other specs may have gone by the wayside.

 

[31697B]

I dont think too much will change dramatically until near destruction.... Thats why to test and see which audio favorites are affected and how... but first some limit finding on breakdown voltage for a few types.
. Then below the breakdown limits. There are many specs that dont really matter much for line level audio amp apps, for example..... changes in leakage currents if small. We'll see. no sense wondering and maybe's. I would check the thd and the noise and BW and see if they are affected. Long term life should be done as well. but I am not the only curious one... am I? To find the middle of the distribution curve and use it there might be perfectly fine and reliable for audio when selected properly. That is what I want to find out.


THx-RNMarsh

 

[31697B]

Last year I picked up a jitter analyzer with 100ps scale. Today I bought a used CD-DVD jitter analyzer from Leader (LE1850). It has 3ns to 60ns fs ranges.... and Go-NO-GO limits for QA/QC. I'm wondering where those set point limits are deemed acceptable in production. [Then measure a few in various price ranges old and new. And interfaces - coax and FO jitter affect and with lengths]

This may be a good starting point on audibility numbers for jitter then compare with what we actually get for low price and high price... and esp with longer lengths of cables. I would hope the best now are pretty good in all respects.... but I' dont really know. esp interested in mid price CD-DVD. The price range many DIY would buy.

-- http://www.nanophon.com/audio/jitter92.pdf


THx-RNMarsh

 

[milkshake]

I would hope the best now are pretty good in all respects....

Even the very cheap ones are pretty good in all respects, see your posted pdf.

 

[31697B]

There are always surprises and fall-out issues to be found it seems from experience. I am also most curious about FO vs coax vs thier length as far as jitter is concerned. I will see. I have no reason to trust marketing and limited tests in magazines. Going to involve the Mastering CD recorder too. But I have several cheap Blue Ray/CD players to check also. Not all have DAC's in them as good as the BenchMark DAC 2. After I see what jitter shows, I'll put it to a spectrum analyzer. This is following BenchMarks designer who said jitter was a reason for audible differences between units.
Of course, when I compare to the original master recording directly... I have no jitter issues (analog).


THx-RNMarsh

 

[SY]

This is following BenchMarks designer who said jitter was a reason for audible differences between units.

In the links you provided before, he did not say that. Where can I find this statement?

It appears that you're conflating several different aspects of jitter. The only one that affects the analog output (assuming that the problem isn't so gross that there are pops and dropouts) is the clock at the DAC. And unless you know what you're doing, the measurement of that will affect what's going on there to the extent that the derived numbers are meaningless. Just owning equipment doesn't make a measurement valid.

It's simple- measure the analog output. That's the only thing that truly matters.

 

[31697B]

You wont find it .... It was from personal communications with him.... telephone conversation.

I also want to know what DVD-CD makers use for a max jitter limiit in production. The 3ns range is there for some reason.... What variations exist between low to high price models. And what issues might arise with interfacing.

And, finally, what ever the jitter numbers are.... do they exceed the audible limits ?

After my curiosity is appeased, I'll sell the gear and move on to other curiosities. I dont need to know a lot about the details and the guts and gore involved. Just compare the numbers found with latest research into what amount and type is audible.
And NO SY I am not going to do a DBLT for you. Just find models and interfaces which are the best I can get in My home system.


THx-RNMarsh


THx-RNMarsh

 

[SY]

And, finally, what ever the jitter numbers are.... do they exceed the audible limits ?

The ones that count are hundreds of times lower than audible thresholds, even in cheap players, as has been pointed out and referenced about a jillion times.

I dont need to know a lot about the details and the guts and gore involved.

If you want a valid measurement of something sensible, then yes, you do. Otherwise, it's just putting on a white coat and playing make-believe scientist.

 

[31697B]

Thanks for the advice and predictions, SY. Maybe you will get to say... I Told You So.

I'll see what I have and compare and then see if there is room for improvement.

http://www.digido.com/index.php?opt...icle&id=15:jitter&catid=13:bob-katz&Itemid=90

http://benchmarkmedia.com/blogs/app...evils-of-digital-audio-and-how-to-defeat-them


THx-RNMarsh

 

[john curl]

Hi Richard, SY is right in this quibble. Both the 1 and the 2 have essentially the same JITTER. However, compared to MOST inexpensive DAC's, I suspect that measurable jitter in most other DAC's will be much worse than either the Benchmark 1 or 2. It would be great to measure this and show everybody.
Again, thanks for the links.

 

[Max Headroom]

The ones that count are hundreds of times lower than audible thresholds, even in cheap players, as has been pointed out and referenced about a jillion times.

In cheap/midrange CD players, the oscillator crystal can is floating.
Earthing the crystal can (ime always) change the sound of the CDP.
Whether the jitter absolute level or jitter spectrum changes I do not know, but the change is readily audible.
Richard, this might be an experiment for you to try.

Dan.

 

[31697B]

Hi Richard, SY is right in this quibble. Both the 1 and the 2 have essentially the same JITTER. However, compared to MOST inexpensive DAC's, I suspect that measurable jitter in most other DAC's will be much worse than either the Benchmark 1 or 2. It would be great to measure this and show everybody.
Again, thanks for the links.

I have no idea about his quibble ..... usually his 'spin' gets in the way and I dont listen further. But thx for the comment as it gives me more time to expand or connect dots ----

I said he told me two things were responsible for the sound differences between dacs-- Jitter and analog. We were not talking about 1 vs 2. Neither am i ... i just said i hear an improvement in the -2. [if not jitter, maybe it is the changes related to analog or PS?] But talking in general, as per my questioning in regard to direct HD down loads of masters vs CD sound. . But your other comments are more where I am focused at this time...... not individual dac's per se.... CD and DVD players which play CD's.

At the moment I am using either the Mastering CD record/player'w digital output to a BMark -2 DAC or new OPPO player. [ I also use their ADC to record]. What the average guy gets and interfacing is a curiosity for me. I had average CD players for many years until recently when I heard HD downloads via Auraliti and BMark-1. Now I am upgrading my digital. and I aint going back to CD except as a legacy system/backup for my CD collection.

MAX.... I wasnt intending to go inside.... but that sounds easy enough to try.

THx-RNMarsh

 

[31697B]

Yes... I will look at outputs as well for jitter artifacts.

I have to find a really jitter free signal source.


THx-RNMarsh

 

[31697B]

This might do the job: Ultra-Small Any-To-Any Clock Management ICs From Microsemi Boost Jitter Performance | Communications content from Electronic Design

for 12KHz only, though.


-RNM

 

[Jakob2]

This example of course is fabricated, 3dB only occurs if the entire passage is an exact 11025Hz sine wave. I hope we're actually worried about recorded music. The problem I assume was normalizing a CD to exactly the maximum data point and then running it through a reconstruction filter.

Of course it is fabricated?!
Afair George wanted to investigate the behaviour of various DACs/CD Players wrt intersample overs and he has choosen the 11025 sine wave at various levels.
Up to now he did not find any clipped outputs and the reason might have been the described synchronisation effect.

A 11025 Hz sine wave is special because a level meter only monitoring the digital values will never realize that clipping occurs; i think TC electronics made the people aware of this problem.

As said a couple of years before, while discussing certain measurement waveforms, nobody guarantees that digital made/enhanced/manipulated audio signals do not contain waveforms that could never occur during a normal recording process of real acoustical events.

The ones that count are hundreds of times lower than audible thresholds, even in cheap players, as has been pointed out and referenced about a jillion times.
<snip>

Simply untrue.
Measurements at analog outputs of CD players have shown jitter levels of up to 1 - 2ns and that surely isn´t "hundreds of times lower than audible thresholds"

<snip>

for 12KHz only, though.


-RNM

Afair these very neat ics are offering oscillator output frequencies up to at least 250 Mhz (from ~< 10Hz? ) and provide some jitter attenuation in this range too.

 

[SY]

Measurements at analog outputs of CD players have shown jitter levels of up to 1 - 2ns and that surely isn´t "hundreds of times lower than audible thresholds"

Cite?

edit: Looking through recent test reviews, the only players I can find that seem to have jitter issues fall into the Fashion Audio niche.
Here is a typical cheap mass market unit.

 

[billshurv]

I hope this is not against the rules, but here is the jitter spectrum for a $300 DAC with 16/44.1 data. Image from stereophile Musical Fidelity V90-DAC D/A processor Measurements | Stereophile.com

The green line is the expected spectrum if things are working as they should. Benchmark gives the same results. What is noticable with this unit is an oddity in the right channel with some odd noise modulation going on. Certainly an area where the benchmark improved things but literally down in the noise for most of us. So question is how much cheaper do you have to get before you can see these synthetic artifacts?

EDIT: http://www.stereophile.com/content/lector-strumenti-audio-digitube-s-192-da-converter-measurements you can spend $3500 and get something truly awful though.

 

[scott wurcer]

EDIT: Lector Strumenti Audio Digitube S-192 D/A converter Measurements | Stereophile.com you can spend $3500 and get something truly awful though.

IME you have to work at making things this bad, typically some very bad engineering is viewed as overlooked "genious".

 

[31697B]

And if we do not just play back a single tone? and that even two channels are recordings of multi- channels each with an ADC thru to mix down......

NwAvGuy: Jitter Does it Matter?


THx-RNMarsh

 

[SY]

J-test is not a single tone.