IanCanada's Latest RPi GB Goodies Impressions... and your tweaks, mods and hints...

Jump to Latest
Terry Demol

Terry Demol

Member
Joined 2002
bisesik said:
But seems it is possible to use less expensive regenerators than especially designed for audio. Something similar to this stabilizer should make very similar work I believe: Стабилизаторы <<ИнСтаб>> 5000–12000 ВА (настенное исполнение) - купить на сайте официального производителя
Actually it is the AC REGENERATOR. But you forced to sell them below $1000 (for 12kW) if you call it as regular stabilizer instead of magic REGENERATOR :idea:

I heard the junior model with DAC (IS350) - the effect was the same as with PSAudio regenerator.
Click to expand...

I'm surprised these regenerators actually sound better as they are really just
a form of switch mode power supply with an AC OP (instead of DC) and as
such will produce a lot of RF noise if not properly filtered.

I've had best results with large balanced isolation transformers similar to
what Ian is using with some RF filtering. How much filtering depends on
application. The filtering can have a negative effect on the sound, not sure
why.

T
 
simon dart

simon dart

Member
Joined 2004
Markw4 said:
Because reality is always more complicated than the models we use to understand it?
Click to expand...

Absolutely! How often do we go down the rabbit hole and can say we have considered everything? Look at what's happening in the current domain with mains. The rectifiers in a power amp switch on and off pulling 10's of amps in spikes for milliseconds as they charge the reservoir caps. Any series components in a power conditioner has to cope with that, and that's before we even look at hysteresis or vulnerabilty to DC in the magnetic circuit or the interaction with the power transformer, or other system components connected in parallel to the mains. It is fiendishly difficult to understand fully. Nearly as bad as digital audio :D

I use a very high current DC blocker and one parallel polyprop cap on my mains for the power amp. I don't even go there with common mode chokes, one step forward, two steps back. Low current draw source stuff is a different kettle of fish.
 
Last edited:
M

Markw4

Member
Joined 2016
simon dart said:
Any series components in a power conditioner has to cope with that, and that's before we even look at hysteresis or vulnerabilty to DC in the magnetic circuit or the interaction with the power transformer, or other system components connected in parallel to the mains. It is fiendishly difficult to understand fully. Nearly as bad as digital audio :D
Click to expand...

With audio, given our ultimate goals and imperfect modeling, we have to rely on some combination of measurements and skilled listening during the process of design. Lot of measurements were taken to design and validate HTPS 7000 MkII. Although the details remain proprietary, I have seen the data (shown to me by one of the designers, Richard Marsh). I have also performed some experiments of my own to see how well it works in practice vs other options. It impressed me greatly and that is why I recommend it.
 
simon dart

simon dart

Member
Joined 2004
Terry Demol said:
I'm surprised these regenerators actually sound better as they are really just
a form of switch mode power supply with an AC OP (instead of DC) and as
such will produce a lot of RF noise if not properly filtered.


T
Click to expand...

There are a few different techniques around. Some compare the mains waveform with the "ideal" with an error amplifier and fill in or subtract the waveform to achieve "regeneration". That's the only way to achieve high power ratings in something that isn't the size of a small car.

Others do turn AC to DC and back to AC via a linear power amp and step up TX but they can't do much power for obvious reasons. The best route for source components.

And finally, yes, using switching to do the job is a more "efficient" in real estate and cost but fraught with problems noise wise.

No easy answers....!
 
pinnocchio

pinnocchio

Member
Joined 2010
quebecdiy.forumactif.com/
Wondering if series/paralleling these 3000F Graphene Supercapacitors could be a good supply... They're very small compared to graphite supercaps but holds the same capacity as their bigger siblings... This is something that could be a game changer and good size to design a PCB and circuit for it? What are your thoughts?

Ghc 3000 Farad Super Capacitor Hybrid Car Battery Car Audio Capacitor 2.7v 3000f - Buy Super Capacitor 2.7v 3000f,Hybrid Car Battery,3000 Farad Super Capacitor Product on Alibaba.com
 
Subbu K

Subbu K

Member
Joined 2010
Paid Member
Finished Ian's goodies ....

Thought of sharing the awesome DAC and LiFePO4 in a home built enclosure. Been listening to it for sometime before putting into the enclosure.

Finally repurposed one of my old Par metal Amp Chassis to build this.
 

Attachments

  • RPi DAC Front.jpg
    RPi DAC Front.jpg
    113.9 KB · Views: 645
  • RPi DAC.jpg
    RPi DAC.jpg
    361.7 KB · Views: 631
iancanada

iancanada

Member
Joined 2009
twitter.com/iancanadaTT
Paid Member
unixdeveloper said:
Thought of sharing the awesome DAC and LiFePO4 in a home built enclosure. Been listening to it for sometime before putting into the enclosure.

Finally repurposed one of my old Par metal Amp Chassis to build this.
Click to expand...

@unixdeveloper

You did great job, much better than my wood base:). Congratulations!

There still some ways to improve the sound quality. Hopefully I can post the solution today how to upgrade the LifePO4 3.3V rails to ultra capacitors hybrid power supply. It will be another improvement, especially for the FifoPi clean side. And I can guarante, very easy:).

Regards,
Ian
 
Subbu K

Subbu K

Member
Joined 2010
Paid Member
iancanada said:
@unixdeveloper

You did great job, much better than my wood base:). Congratulations!

There still some ways to improve the sound quality. Hopefully I can post the solution today how to upgrade the LifePO4 3.3V rails to ultra capacitors hybrid power supply. It will be another improvement, especially for the FifoPi clean side. And I can guarante, very easy:).

Regards,
Ian
Click to expand...

Ian, initially the setup was on a wooden block like yours to make it easy to experiment with different options.

I ran into one issue when I used a CCHD-575 22/24 pair, FiFoPI LED indicators near the OSC were flashing and there was no lock. I assumed the problem was with the clocks. Once I put back the CCHD-950 45/49 pair it started working again . At what situations the LEDs near the clocks flash on FiFoPi?

I didn't have much time to debug that further. I will reflow the clocks and check if they work properly before installing them again.

Regards,
Subbu (unixdeveloper@diyaudio)
 
Y

yunyun

Banned
Joined 2019
iancanada said:
I’m working on this comparison test for a while. I'd like to share my results here with you. Sorry for the delay. I could be wrong. Just for you to reference. I picked up these three because they have the same configuration. I could have more reports for others down the road.

#3 eBay 600:10K audio transformer

I bought this pair of 600:10K transformers off eBay. They are not expensive but look pretty decent. I didn’t expect too much from them. However the real listening test did give me some surprise. They have all the sound signature of a typical transformer I/v stage. Nature, analog, peaceful and without any additional flavor. I would stack with them if didn’t experience the other two. I ran them at my LL1544A transformer I/V PCB with same schematic and same configuration. The only thing is that the footprints don’t fit very well. If I have chance to design the transformer I/V II PCB I’d like to make it compatible with them.

If you have never tried the transformer I/V, I would highly recommend this one as starting point.

#2 LL1544A

My transformer I/V PCB was originally designed for them. I have been using them for a long time before I had the BisesiK transformers. I have to say I’m very familiar with the sound. For me, it’s my first transformer I/V that could touch the ‘high end’ in my heart. Besides all the sound signature of a typical transformer I/V stage, LL1544A has higher sound density and plays music more dynamically. Especially it has a very beautiful mid-high range which makes the music a little sweet feeling. I enjoy listening to my favorite vocal jazz with them very much. I could listen to music for a long time almost forgot I was doing comparison test.

LL1544A I/V stage will suitable for all kinds of music and all kinds of system

#1 Bisesik transformers

I must put them in the first position though they are not the kind of transformer I/V that you can fall in love with them immediately at the first time you use them. However, after two weeks of run in, I’m getting in love with them more and more and for now I almost can’t live without them.

If I had only one word to express them I would use the word ‘perfect’. They are not good at certain frequency range but good at all frequency ranges. The low rang could be extended much lower (why they are so big?) than other transformers which feature could be looked upon as common weakness of all transformer I/V stages. So I’m considering maybe I could degrade my speakers from B&W802 to B&W803 because of that. Just kidding! They sound not only with higher density at full frequency range but also play music much vividly. The bigger stage and bigger band you are listening, the stronger you feel about that. If I had another word to use, I would use the word ‘musical’. They play all music with emotion. You can feel the whole atmosphere surrounding you. When I’m listening violin with them, I almost couldn’t help cry. The Bisesik transformers are suitable for all kinds of music. But if you really like the classical music, my advice is that sticks with this one no need to consider any others.

I had a pair of Japanese Oyaide AR-910 pure silver XLR cables. They are pretty decent but I always feel the high range sounds a bit strange when I use them with other I/V stage. However, with Bisesik transformer, everything becomes correct and precision. When I use this configuration listening the sound track of film Pirates of the Caribbean, The Curse of the Black Pearl (I like the big band), I was almost shocked. The huge band and the whole stage were surrounding me and I can hear each hidden details. All instruments are playing vividly together at their positing. I can feel each of them. This feeling could be even better than I was in a real theater.

Though I put Bisesik transformers in the first position, but I wouldn’t recommend them for you if your system is not good enough. The better system you have, the more advantage of these transformers you can feel.

My system for listening test

Transformer I/V for test

1. ES9038Q2M DM DAC HAT (with three independent 3.3V voltage rails)
2. ShiledPi
3. FifpPi with Pulsar 90.xxx MHz clock
4. ReceiverPi
5. RPi3
6. LiefPO4 power supply with ultra capacitor conditioners at both 5V RPi rail and the 3.3V FifoPi clock rail.

The rest of the system
1. B&W 802 D2 speakers
2. PASS XA160.5 amplifiers
3. PASS XP-10 per-amplifiers


Ian
Click to expand...

Maybe just because Bissik transformers are sold in pairs, pairing has been done beforehand, but LL1544A is bought randomly so that the sound is poor.

In the coupling coefficient, bandwidth, etc., the 600:10K transformer is not as good as 10K: 10k. 600:10K transformer is not suitable for the analog output of the DA chip.
 
M

Markw4

Member
Joined 2016
yunyun said:
...LL1544A is bought randomly so that the sound is poor...
Click to expand...

Don't know about that, but reportedly LL1544A amorphous core and other models using that core type have audible 4th-order harmonic distortion. It has been said that despite the manufacturer's claims of superiority and the high cost, the core material isn't the best sounding.
I personally haven no opinion on the transformers, just reporting what someone else described to me.
 
Last edited:
You must log in or register to reply here.
Top Bottom