Flat Phono Preamp Based on John's P10 and 2SK389

Where we discuss new analog design ideas for Pro Audio and modern spins on vintage ones.
Post Reply
ricardo
Posts: 262
Joined: Sun Apr 10, 2011 7:24 am

Re: Flat Phono Preamp Based on John's P10 and 2SK389

Post by ricardo »

Wayne, if you are doing the de-clicking digitally, why not apply the RIAA EQ digitally too?

RIAA is only a couple of bi-quads and the IEC LF roll-off can be done with another 2 to give 24dB/8ve. Save loadsa critical caps & resistors.

My most useful take from your tests is the 7-10dB extra gain for a full RIAA preamp compared to a flat preamp to achieve the same peak levels.
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Flat Phono Preamp Based on John's P10 and 2SK389

Post by mediatechnology »

ricardo wrote:Wayne, if you are doing the de-clicking digitally, why not apply the RIAA EQ digitally too?
I am.
The RIAA filter values I put into Cool Edit are as simple as a text file.
Analog RIAA de-emphasis is primarily for monitoring and preview though it can be applied during recording.

Monitoring Transfers with Analog RIAA EQ

It's pretty darn painful monitoring the transfer of an entire album side during recording without EQ.
I have analog RIAA EQ in the monitor path so I can enjoy the transfer and hear the interim results of declicking before applying software EQ.

Monitoring Declicking with Analog RIAA EQ

In Cool Edit, declicking and software RIAA EQ require two separate passes to render the file.
I find it useful to preview the declicking and check it with analog RIAA de-emphasis applied as I monitor.
Thus I don't have to apply RIAA EQ in software before making a final declick decision.
Though I can commit to software EQ and use "Undo" if I over-declick and hear it in the final result it adds a step.
If I like the declicking results I get in preview using analog RIAA monitoring I then apply the EQ in software when I render the file.

Does that make sense to anyone besides me?
The whole raison d'être is to speed workflow and be easy on the ears.
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: Flat Phono Preamp Based on John's P10 and 2SK389

Post by JR. »

I have some old vinyl that I don't ever expect to find on digital media. So have thought about this some. I know that I would be bored silly listening to the stream while remastering. I can imagine any number of non-real time tricks but we've probably discussed them before (backwards, half speed, etc).

JR
Cancel the "cancel culture", do not support mob hatred.
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Flat Phono Preamp Based on John's P10 and 2SK389

Post by mediatechnology »

I know that I would be bored silly listening to the stream while remastering.
If you can't listen to it once its probably not worth converting to bits. :lol:

I've had some musical re-discovery experiences while transferring my own stuff.
One of the the things lost moving from to CD to LP was the segue.
For example Abbey Road side 2 "The Long One/Huge Medley" segue.

Lost when moving from CD to mp3 (albums to individual songs) was the art of sequencing.
Santana Abraxis was my recent re-discovery.
It just wouldn't be the same in a different order.

And wouldn't you want to know that it skipped during transfer?

The Parasound Zphono USB (now discontinued) comes pretty close to the record section I envision.
The RCA outputs are RIAA-EQ'd but the USB output is flat: https://www.parasound.com/vintage/zphonoUSB.php
It also has two line inputs for routing the soundcard output back into the box.

Unfortuantly I've never found a USB converter that I really liked.
Virtually every USB box I've tried has rectified RF low-level spurs around 1 kHz due to USB polling.
There's a lot to be said for the transformer-coupled SPDIF/AES interface.
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: Flat Phono Preamp Based on John's P10 and 2SK389

Post by JR. »

mediatechnology wrote:
I know that I would be bored silly listening to the stream while remastering.
If you can't listen to it once its probably not worth converting to bits. :lol:
I want to fast forward to the end result.

Sometimes at night when I am saturated with commercials I will surf up to some satellite music feeds and find the low-fi mono, old big band and '60s.70s. music channels very pleasant for a couple songs.
I've had some musical re-discovery experiences while transferring my own stuff.
One of the the things lost moving from to CD to LP was the segue.
For example Abbey Road side 2 "The Long One/Huge Medley" segue.
it helps to try smoking the same ganga as the Beatles were in the studio. :lol:
Lost when moving from CD to mp3 (albums to individual songs) was the art of sequencing.
Santana Abraxis was my recent re-discovery.
It just wouldn't be the same in a different order.
some of that is an artifact of how our memories were burned in back in the '60s when our brain buckets weren't so full and we didn't just play a single track at a time because it was way too much effort.

Then the operatic theme albums from Who and others ("Tommy can you hear me... Tommy").
And wouldn't you want to know that it skipped during transfer?
I wouldn't mind it being automated enough that a bad take can be easily redone without much concern. I don't think I have enough hours left to spend them paying close attention while transcribing music that I admit I do not even play now.
The Parasound Zphono USB (now discontinued) comes pretty close to the record section I envision.
The RCA outputs are RIAA-EQ'd but the USB output is flat: https://www.parasound.com/vintage/zphonoUSB.php
It also has two line inputs for routing the soundcard output back into the box.

Unfortuantly I've never found a USB converter that I really liked.
Virtually every USB box I've tried has rectified RF low-level spurs around 1 kHz due to USB polling.
There's a lot to be said for the transformer-coupled SPDIF/AES interface.
Time to roll your own...

JR
Cancel the "cancel culture", do not support mob hatred.
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Flat Phono Preamp Based on John's P10 and 2SK389

Post by mediatechnology »

Well I guess I will roll my own.
And stamp out a few more PC boards along with it.
Some real world experiences using this for transfers made me think that de-clicking flat material is the way to go.

I also have this strange habit of wanting to look at X-Y vectors as I listen.
Clicks and warp are definitely - more often than not - a Side (L-R) thing.
You see these huge clicks at right angles to the mono line.
That realization came to me late but it makes sense.
A boulder in the highway is going to deflect your car upwards as it hits the oil pan.
The stylus is no different - the microscopic boulder or pit in the groove causes vertical modulation.
We had a reader here make that point for warp.
The vertical mechanical vector is Side.

Once I started capturing some test tracks I realized I needed less gain.
Lots less gain.
The A/D has a +10 dBu dBFS.
Input to the converter was hot.
And those clicks need some extra headroom.

I took 6 dB of gain out of the preamp overall.
This gives with my cartridge and most material about 10-15 dB of headroom on audio, less on major clicks.
I also lowered the RIAA EQ path gain from +8 to about +4.

Along the way I had to figure out how to make Cool Edit's DirectX 9 plugins run under Win 7 x64 DirectX 11.
viewtopic.php?f=8&t=712

I just revisited a couple of Steely Dan tunes.
"Ricky Don't Lose that Number" had some bodacious clicks during the low-level intro.
I did a flat transfer, de-click, and then applied software-defined RIAA EQ with an FFT filter: http://www.a-reny.com/restauration/avance.html#Riaa

What I need to do for comparison is record it with RIAA EQ, de-click and then compare the files...
ricardo
Posts: 262
Joined: Sun Apr 10, 2011 7:24 am

Re: Flat Phono Preamp Based on John's P10 and 2SK389

Post by ricardo »

mediatechnology wrote:What I need to do for comparison is record it with RIAA EQ, de-click and then compare the files...
Record > cartridge > preamp has a transfer function with an associated Impulse Response (IR)

Clicks are 'nearly' impulses which will be convolved with this IR.

ANY bandwidth limitation, both LF & HF, will smear the IR making it longer so your de-clicker needs to 'remove' a longer section.

RIAA is sorta a leaky 50Hz integrator which will further smooth any rough edges left by de-clicking.

Flat Preamp > Digital de-clicker > RIAA is the way to go.

Have to admit to never considering separate de-clicking for M & S :) What a clever idea!
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Flat Phono Preamp Based on John's P10 and 2SK389

Post by mediatechnology »

RIAA is sorta a leaky 50Hz integrator which will further smooth any rough edges left by de-clicking.
Hadn't thought about it that way.
Supports the reasoning of applying RIAA de-emphasis in the monitor path while previewing/adjusting the de-clicker as it works on the flat file.
What you hear while auditioning will closely approximate the final result of RIAA de-emphasis in software.
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Flat Phono Preamp Based on John's P10 and 2SK389

Post by mediatechnology »

This is an example of a rather large click having a strong Side (L-R) vertical component.
I didn't have to look hard for this: It was the first big click in the file.

Image
Anatomy of A Phono Click, No RIAA, Flat Transfer. Note strong L-R (Side, Vertical) Component.

This is the same click after RIAA EQ:
Image
Anatomy of A Phono Click, With RIAA Equalization (FFT).

Note the time-smearing and increased difficulty of detecting the click if RIAA EQ had been used during recording.
And it's primarily a Side, L-R, vertical event.
This bad boy would have been almost at right angles to the mono line when viewed in Vectorscope X-Y mode.
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: Flat Phono Preamp Based on John's P10 and 2SK389

Post by JR. »

I am not familiar with the term leaky integrator but indeed the first pole in the RIAA starts down at 3280 uSec, there is a little wiggle (zero) in the middle before returning to the simple one pole roll-off above 75uSec.

De-clicking before the RIAA does work better and was being done by some of my kit customers back in the '80s using the stand alone de-clickers available back then.

Along the same lines playing the recording at half speed should reduce the effective click perturbation duration relative to music when the captured file gets sped back up to normal speed for processing, shifting the click energy even higher frequency for less time.

JR
Cancel the "cancel culture", do not support mob hatred.
Post Reply