Digital Phono Pre-Amplifier- Anybody interested?

[Ravindra Desai]

Hello,

There is a lot of response here on the forums for vinyl.
It got me thinking that I could use my digital audio processor and program it for RIAA curve to make a Phono Pre-amp.
Does it sound interesting?
My processor board has 6 outputs and I could output any combination of signals on those. e.g.

1. Lt-Rt, L-R-SW
2. L-R-SW, C-S-SW

I have attached a simulated RIAA response curve for your review. Since the signal is shaped to the required response in digital domain, we don't have to worry about its stability or its accuracy.

Also, since the platform is re-programmable, we could literally customize it to suite the individuals' need.

I could add other processing to bring out the details and the sound stage, if/ as required.

Let me know what you guys feel.

Regards,

Ravindra.

 

[dr khanwelkar]

Dear Ravindra,
I have many preamps including those inbuilt with the integrated amplifiers.
I like the Tube preamp.built by Mr Viren Bakshi of Lyrita the most ,despite a constant "hum" audible through the subwoofer.
I have recently purchased Schiit Mani and it is also a good preamp.
I would like to know ,what advantages we will perceive ,if we go to digital phono preamp and why?
Thanks.

 

[Ravindra Desai]

Hello Dr.
Long time.....

The reason for going digital is because I don't see any disadvantages. I only see advantages.
The whole system will have only two electronics components: Digital pre-amp with post processor and Power amplifier which can be analog.
Here are the advantages of a digital pre-amp:

1. Better signal to noise ratio as we can now spread the gain before and after the post-processing
2. Stable and accurate RIAA curve that is temperature independent
3. Integrated filters and crossovers to suit the speaker system
4. Integrated DC controlled tone/ master volume controls for perfect tracking of all the channels
5. Perfect isolation between channels of best sound stage
6. Integrated post-processing for clear sound with silence included
7. Re-programability allows real time tweaking with laptop instead of a soldering gun and ultra high precision components

What you heard at my place was a digital pre-amp with post processor with analog power stages for both stereo and surround configurations.

Regards,

Ravindra.

P.S. Free home delivery, installation and trial for you!

 

[Ravindra Desai]

Hello again,

The lack of response confirms that people associate Vinyls with analog and more so to Tubes.
Fatigue, typically associated with 'digital' could be the reason.(!?)

I have not tried Class-D pre-amplifiers, yet, but can vouch for DSP for pre and post processing without having to worry about fatigue.
It brings flexibility and if implemented correctly, can produce warm, sweet sound.

Digital Phono Pre-amplifier will certainly be on my plate and I will persue it as time permits.

One nice application I stumbled upon on the DSP manufacturers website is scratch remover that are so annoying with over used/ old vinyls.
It is based on the nature of signals a scratch produces rather that a feed forward or predictive approach. If I can demonstrate, in an A/B test, its effectiveness, it could turn out to be a good advantage.

I'll keep you all posted. Thanks!!

Regards,

Ravindra.

 

[tuff]

In my opinion, people who use vinyl are allergic to anything digital in their signal path . Hence not much response on this thread.

 

[HormusPeston]

Agree with Tuff. Tube/vinyl fans do not like anything solid-state/digital. About twenty years ago I played a prank on a dear friend with a fake tube amp. The tubes were on display with the heaters powered up. Grids were unconnected. The amp was, in fact, powered by a small Class A (a 3W JLH) amplifier hidden inside the chassis. He loved it, but he DID say that it sounded different. Then I opened the box... Suffice to say, an expensive, liquid peace-offering was needed to earn his forgiveness. Tube-fans like the sound that tubes make. Amen.
Cheers,
~HP

 

[rikhav]

I don't think so
Forum member @jls001 has a all solid state setup and his vinyls sound wonderful

I don't know much about how a phono works but maybe people don't want a DSP in between and yeah I know people do like the cracks and pops too whchh are generally association with vinyl playback

 

[greenhorn]

I do not see how this can be done, because of limitations of the front end of most CODEC's
Not sure which codec you use,
Taking the AK4626A used in the Minidsp 2x4HD
https://www.akm.com/akm/en/file/datasheet/AK4626AVQ.pdf
the input S/N+D is around 90dB for a 3V Signal
Which means noise is around 0.2mV

A phono cart - Let's take one with a very high output, like the shure m44-7 - it has 9.5mV output.

so we are looking at some 33.5 S/N at the input stage. Now add the fact that there is going to be a 20dB boost applied at the Low frequencies, your signal to noise becomes a vanishingly low 13 dB.


Now this is not usable, unless you have a low noise analog stage before the DSP to boost the signal to 3V, and then let the DSP apply the eq, but then, you have a hybrid phono stage now, because your prea amp work is being done by the analog stage, and your dsp is doing just the eq.

most of the Digital Phono stages i know of use an analog phono stage followed by a standard ADC, because a gain stage needs to be built anyway before the ADC, and its far cheaper to build the eq by providing some negative feedback to that gain stage, than to design DSP to apply it in digital.

It's worth a shot doing it from an academic point of view, not sure what benefits it would yield.

And then there is the philosophy of analog, which most vinyl purists are trying to embrace, which considers digital blasphemy

 

[Ravindra Desai]

@tuff , @HormusPeston , @rikhav , @greenhorn : Thanks!
This is forum at its best!

I tried to reason out why digital normally sounds bad here: https://www.hifivision.com/threads/what-changed-with-digital-audio.69388/

So if you know the modern way of recording sound, you should be able to get tube like sound, even with semiconductors! But for that you need DSP!!

@greenhorn : This is the first time that I have come across math in my thread and I welcome it! Thanks!!
Why the need for 3V? Most power amps are designed to deliver rated power at 1V single ended input.
I'll have my calculations in a spreadsheet for 9.5mv input and share it here for review.

Thanks again you guys, let's keep it up!

Regards,

Ravindra.

 

[greenhorn]

3V because that's the voltage at which the noise spec in the datasheet has been measured. 3v* -90db = your noise signal voltage
I agree with you on Digital sounding as good as or better than analog, but analog has become more of a cult, so logic may not often work

 

[jls001]

@Ravindra Desai : do you propose to do the RIAA equalisation in the digital domain? What level of accuracy to the RIAA curve do you anticipate can be achieved?

How about the gain needed - how do you plan to achieve the 40/60+ odd dBs needed for most MM/MCs?

 

[Ravindra Desai]

Hello,

Detailing the digital phono pre-amp did bring out the devils. It is still preliminary.....

It seems that to use ADAU1701 DSP I would need an external op-amp or a diff-amp of the low-noise and high input impedance type. I have not shortlisted the devices yet. I will however not use a dual version to keep the Lt and Rt as isolated as possible.
The ADAU1701 ADC has a full scale input of 1V. Interfacing the phono transducer directly to the ADC will result in loss of resolution.

If we want to keep it 'single-chip' then the ADAU1761 is another option to consider. It has a built-in PGA with a gain of 37db that will pre-ampli the signal to the level required by its ADC.

Once the phono signal is digitized, I would apply the RIAA eq which will give me a flat signal @ 0db reference.
Then after that I can play in the digital domain to implement any features that I want.

Any other suggestions are welcome.

@jls001 : Yes, the RIAA eq will be implemented in digital domain. The buil-in CODEC operates @ 24bit/192KHz.
The DSP uses a 28bits for operation. The 4-bit headroom can accommodate 24db of additional headroom without clipping and uses a standard 5.23 numeric format.

This translates to a very accurate, near ideal, temperature independent RIAA curve. Kindly check the simulated graph attached in post # 1 of this thread for more details.

Regards,

Ravindra.

 

[jls001]

Thanks for the clarification.

Here's a quick reality/sanity check: the equalisation curve is usually achieved by a small RC network in the gain stage feedback loop. So at most one resistor and one (or two or three paralleled caps to achieve the closest value to the calculated C) are needed to achieve the equalisation. Depending on the grade/quality of components used, that's just a few rupees worth of components. So we need to ask ourselves if using a DSP chip for achieving essentially the same curve would be tantamount to using the proverbial cannon against the mosquito. See how simple is the RIAA part of the circuitry in the MM phono preamp below:

http://www.diyaudio.com/forums/analogue-source/301116-isotope-opamp-based-mm-phono.html

With careful hand picked selection of capacitor values, the above equalisation is accurate to +/-0.2 dB in the audio band.

Secondly, it's true that a DSP chip could be used for further sound shaping, but we need to ask ourselves if that's desirable or in vogue.

PS: this is not to discourage your endeavour but to try and see the overall picture better.

 

[reachkalyan.kr]

Thanks for the clarification.

Here's a quick reality/sanity check: the equalisation curve is usually achieved by a small RC network in the gain stage feedback loop. So at most one resistor and one (or two or three paralleled caps to achieve the closest value to the calculated C) are needed to achieve the equalisation. Depending on the grade/quality of components used, that's just a few rupees worth of components. So we need to ask ourselves if using a DSP chip for achieving essentially the same curve would be tantamount to using the proverbial cannon against the mosquito. See how simple is the RIAA part of the circuitry in the MM phono preamp below:

http://www.diyaudio.com/forums/analogue-source/301116-isotope-opamp-based-mm-phono.html

With careful hand picked selection of capacitor values, the above equalisation is accurate to +/-0.2 dB in the audio band.

Secondly, it's true that a DSP chip could be used for further sound shaping, but we need to ask ourselves if that's desirable or in vogue.

PS: this is not to discourage your endeavour but to try and see the overall picture better.

Myself Diy guy,
i understood what you meant.
Its real pain.
All the best.Mr Ravindra

 

[Ravindra Desai]

Good morning everyone!

I put all the replies in this thread together and input it to my DSP. Processed it using a very complex algorithm that I will never disclose, checked the output on all the channels, it was the same. Compared it with my spreadsheet, it was the same.
It read something like this: No demand = No supply. Not worth the efforts!

I agree with all of you and appreciate that you'll came forward to share it.

I am suspending any practical activity related to this topic until further notice.

People are, however, encouraged to comment and share their thoughts.

With regards,

Ravindra.

 

[jls001]

Regarding demand and supply, one of the most popular DIY projects ever on this forum was the CNC MM phono stage. Many members here have built it and many are still using it as their daily phono stage.

For MC, the Hypnotoad phono preamp was built by a number of members.

Another phono stage built here was the Pass Pearl 2 (complex and costly, not for the first timer). There were others too, including tubed ones.

So I would say there is demand for a good phono stage, though admittedly not a huge one.

 

[Ravindra Desai]

Regarding demand and supply, one of the most popular DIY projects ever on this forum was the CNC MM phono stage. Many members here have built it and many are still using it as their daily phono stage.

For MC, the Hypnotoad phono preamp was built by a number of members.

Another phono stage built here was the Pass Pearl 2 (complex and costly, not for the first timer). There were others too, including tubed ones.

So I would say there is demand for a good phono stage, though admittedly not a huge one.

Got it!

I am a technologist and like to design and solve real life problems.
So its my job to look for opportunities.
On this particular topic, it is nice that you and others responded to convey the reality and I really appreciate that.

Regards,

Ravindra.

 

[Hari Iyer]

Can a parametric equalizer be used as a phono pre-amp with the correct EQ? Curious to know.

 

[diyaudio]

Fatigue, typically associated with 'digital' could be the reason.(!?)

How did you reach this conclusion? Isnt it a perception? Objectively there is no debate, digital surpassed analogue long time back. Film cameras, audio or video etc. Its not about personal preferences, its about facts. There were problems with digital earlier, agreed, but the current state of digital makes analog pointless. Though people may prefer analog but thats a different thing altogether.

 

[Hari Iyer]

IMO film based analogue projection in theaters had much richer colour, contrat, brightness and resolution compared to the current digital screening no mater whatever the pixel size. Period.