John Curl's Blowtorch preamplifier part II

Status
Not open for further replies.
Richard- Application is really important for these concepts. If you are using this with an early non-oversampling DAC setting the FC low is important but 20 KHz may be too low since the sample rate is 44.1 KHz and the images are all above 44.1 KHz.

If the player is truly non-oversampling the images start around 24kHz i.e. the audio bandwidth below the sample rate.

However I have had really positive results using a passive LC filter to remove the image before the external circuitry. I have always thought that an active filter to remove the HF stuff that opamps and analog circuitry has problems with was naive.
Me too - I've experimented with elliptic, Chebyshev and quasi-elliptic passive filters to very good effect.

Richard - you'll need a bit more than just +1dB to overcome the NOS droop at 20kHz. Ah, re-reading I see your CD player probably isn't NOS so ignore this.
 
Last edited:
It will be slightly better after I build from bread-board to something usable. But even if I could hear it... it would sound a tiny bit brighter with anything that high up. But bright sound not sound the same as distortion which I am listening for.
I was joking, Richard. I don't care for a bump of 2dB at 20Khz. For txwo reasons: too old to hear so high, and my horn has a bandwidth limited at 15KHz, it could help-it.
 
Member
Joined 2004
Paid Member
You are going over old ground... i am well past that. Didnt you see the output spectrum I showed here? There are also idle tones seen. I would like to think none of us have to listen to such gear... but apparently, we do. My CD player tested for extraneous HF noises is the Alesis Mastering CD recorder.


THx-RNMarsh

Link? I obviously missed it. I have not seen issues in the DAC's I have looked at but its not a large set. My feeling would be to replace the DAC with a decent external one. They can be had on eBay for peanuts. Screwing around trying to fix a bad implementation seems like a low ROI effort.
 
I was joking, Richard.
I think there is no need to filter so close to the bandwidth upper limit of 20KHz, as long as your system is able to deal with high slew-rates. The high frequencies noises will be amplified as-it, without other artifacts, don't you think ?
If your amp is (internally) flat up to 1MHz, i believe a margin of 10 is high enough, and i would chose to filter at 100KHz.
 
no gremlins
 

Attachments

  • 900x900px-LL-a91b2ae5_rega5.jpeg
    900x900px-LL-a91b2ae5_rega5.jpeg
    38 KB · Views: 256
Disabled Account
Joined 2012
NOW you tell me.

But that isnt the real point here.... but do some amps do better dealing with out of band HF junk than others.... notably the VFA when the feedback starts to decrease and non-linearity increases above audio range.


I have BenchMark ADC and DAC-2 here - BTW.


THx-RNMarsh
 
Disabled Account
Joined 2012
The idea fits into a lot of places..... why many designers think it sounds worse if the amp with nfb has narrow or much limited freq range --- high distortion above the band-pass. Why wide open-loop design sound better and why high SR sound better. All most readily and easy to dealt with (including stable) with CFA design. > Why many designs do not sound good with CD as source. Why some DAC sound better than others. When equipment is tested individually, the audio freq range distortion is very low. ETC. ETC.


THx-RNMarsh
 
Member
Joined 2004
Paid Member
http://www.diyaudio.com/forums/loun...ch-preamplifier-part-ii-7635.html#post4527009

Richard wants good noise attenuation by 90KHz, so starting the roll off just above 20KHz seems a very logical starting point. Plus there is an interesting idea he is testing.

Now I see what the fuss is about. That spectrum is completely typical of a DAC without an image filter (that's really cutting corners). Somewhere I have the same spectrum photo before and after a simple LC filter. Without much effort all the junk was reduced by 70+ dB.

What is the utility of that "mastering" box today?
 
DF96 said:
Images start from around 22.052kHz (assuming the input rolls off at 20Hz).
Sorry, I got that wrong. The first set of images is inverted, so they start from 44.1-f, where f is the cutoff frequency of the anti-aliasing filter used before the ADC.

abraxalito said:
If the player is truly non-oversampling the images start around 24kHz i.e. the audio bandwidth below the sample rate.
That is correct.

Esperado said:
I think there is no need to filter so close to the bandwidth upper limit of 20KHz, as long as your system is able to deal with high slew-rates. The high frequencies noises will be amplified as-it, without other artifacts, don't you think ?
If your amp is (internally) flat up to 1MHz, i believe a margin of 10 is high enough, and i would chose to filter at 100KHz.
That would depend on whether you wish to achieve signal reconstruction, or have images masquerading as percussive musical content.
 
That would depend on whether you wish to achieve signal reconstruction, or have images masquerading as percussive musical content.
yes, but...
The inertia of your tweeters are high enough to smooth the signal. Your DAC was supposed to do-it long time before, with acurate oversampling and low pass filters. Son an analog low pass filter at the input of any amp is supposed to be here to filter only the evils introduced by the analog connecting cables, with various noises due to HF, AC ground leakeages etc...
 
Not really much utility in the Masterlink today. Many other ways to capture and store 24bit material are handy, which wasn't necessarily the case earlier.

In the studio it is often handy to have a device that will capture the mix in real time external to all other equipment. In the "old days" that was the two track tape recorder, in the digital age it can be a separate device such as the Masterlink, an independent computer, or just another input on the playback computer/DAW. The one thing the Masterlink is missing as a "modern" device is the ability to go in and move the file out of the hard drive without playing it out. Not desirable in todays world.

Alan
 
Disabled Account
Joined 2012
And, if you take that same CD/DVD/3D player and run it thru an 1800-2000 USD$ converter via coax/digital I/O ports, you get No HF trash to cause increased distortion.



DSC02089.JPG



The Alesis was 800 USD$ new and the CD/DVD/3D player is about 300-350 USD. Used ones shown on eBay are about half the new price.

I'll also check some other players but there might be a general price point you dont want to fall below to get the max performance; I wish we knew which were the exceptions but magazines and equipment mfr do not test above 20-30KHz. You are on your own. But might ask for such info before you buy from mfr. If budget is tight, an external LPF is cheap enough to make.



THx-RNMarsh
 
Last edited:
Status
Not open for further replies.