Asynchronous I2S FIFO project, an ultimate weapon to fight the jitter

Hi Ian,

Ok, I understand I can do this way, but I would know what parts I have to remove (LDOs and other components on the power path).
I would get pure battery powered way.

Please, let me know.

Thanks
Andrea

Only Need to remove the three LDO boards on the McDualXO and the LDO chip on th PCM board.

Please refer to corresponding user's manual for details.

Regards,
Ian
 
- I2S to PCM converter, powered with the same 3V3 LiFePo4 batteries

I would exclude all LDOs and all other unuseful parts on the power path of the devices that will be batteries powered.
Andrea
Hi Andrea,
There was some recent discussion about bypassing the ldo on the I2S to PCM converter. It is covered in the documentation on point 9, but effectively jump 3.3v in to 3.3v out and connect the battery to the J6 5v in connector. It will short bypass the ldo. I doubt that removing the ldo will make an appreciable difference.
 
Disabled Account
Joined 2019
Hi Ian,

Ok, I understand I can do this way, but I would know what parts I have to remove (LDOs and other components on the power path).
I would get pure battery powered way.

Please, let me know.

Thanks
Andrea


+ 1


We need to know : what to remove and also what tokeep in each board ! I explain myself : for instance it is not the same thing to feed the I2S to PCM board from its side where the three vias are (were made for three legs regs in the manual) than the main 5V input connector because it is bypassed by a large polymer cap // with a ceramic class II cap !


If removing the ldo what are the caps and amount of capacitance involved after the three vias area is used, etc, etc !


@iancanada : your products are great but the manuals are sometimes too lite... or with contradiction as I pointed out in some posts above . I called for a manual revisions in order to use at the best LiFePo4 tech and you don't answer as well :eek: ! Greg is very helpfull as some others but I doubt they know more than you about your products anf how it is designed !

An other one for instance : lot of free area for adding caps at the back or top of the PCM boards but despite those free areas have a number we can't know on the manual which loads or areas of the pcb they decouple and what values to solder in order not to waste your design and measurements (few of us have high speed circuitry knowledge and skill to measure it, more often we don't have always a scope at home.

Your products deserve more attention from you about manuals revisions :) , and less silence when we ask for something not already said in the threads;)

That said you are always helpfull when our devices have sign of failures and thanks for that.
 
Last edited:
@diyiggy,

If Ian tells you what parts to remove, and you find that it isn't easy to do without damaging the board or something more or less like that, something you don't like, then would you complain that if he is going to tell you which parts to remove, he should have known all along that if using batteries and eschewing LDOs then people would want to remove any LDOs and he should have designed the boards to make it easier and safer. You know, like some people don't even have scopes at home for troubleshooting...

Ian tries to make things that are useful and fun, but not necessarily intended to be modified by people who don't know how to figure it out themselves.
 
Last edited:
Disabled Account
Joined 2019
No I wouldn't complain and you should not think people are silly and/ot not accountable for what they do ! Can we imagine we are free to risk our own pcbs ay our own risks ?

Removing LDOs and feeding with battery is mentionned in the manual but with very few indicators (and subtle contradiction on how to proceed... which make it indeed risky !)

And I'm reaaly not sure you can include me or Andrea Mori in the "People" you're talking about... did I tell you your Wima MKP caps stacks are not the proper way to improve dacs power supplies ? No I didn't and you have a scope ! So please no lessons !
 
I own a Dual XO, not a MCDualXO.

Andrea

For PCM board, you only need to remove U4 and then short TP4 and TP5. Please figure it out from the attached schematics.

For Dual XO II, nothing needs to be removed. You just need to short the IN(TP3) and OUT(TP5) pin of the external LDO connector.


@diyiggy,
I'll try my best to support, though modifying circuit could be out of range of a user's manual :).


If Ian tells you what parts to remove, and you find that it isn't easy to do without damaging the board or something more or less like that, something you don't like, then would you complain that if he is going to tell you which parts to remove, he should have known all along that if using batteries and eschewing LDOs then people would want to remove any LDOs and he should have designed the boards to make it easier and safer. You know, like some people don't even have scopes at home for troubleshooting...

Ian tries to make things that are useful and fun, but not necessarily intended to be modified by people who don't know how to figure it out themselves.

Thank Markw4 for your kind understood.

Regards,
Ian
 

Attachments

  • PowerSupplyPCM.png
    PowerSupplyPCM.png
    91 KB · Views: 893
Disabled Account
Joined 2019
Thanks Ian, I know you will.

It was not about modyfing the circuit, it's about to profit the best options you gave already in he manual allowed by the great versatility of your designs.

I just enlighted it could be less risky to have a more detailled way to do.

You wrote in the manual we need to desolder the ldo on the pcm (I don't see it as a modification of the circuit as it's reversible) while Wlowes and Greg say it doesn't need ! It just ask clarifications: Andrea has a doubt and the way the manual tips section is writed feeds this doubt. Imho a simple manual revision could be greatly appreciated and bring even more concistency to your design work.
Don't see it as a critic, it isn't. I do think your work deserve such minor manuals correction for the fans we are. IMHO as you mentioned this way to tweak, but again explanations are not so clear on this tip hence Andrea Mori's or my asks about the best option to not waste LiFePo4 power option.

And maybe it's simply cool to enrich the manual content by the finds of users here, we are all motivated to extract the best of your design with the same envy for the best sound and sharing :)

I appologize if what I wrote sounded not positive, just wanted to be constructive instead but I'm very limited by my poor english skills. So I say again we ask because we love your products :)
 
Disabled Account
Joined 2019
You did by beginning to answer to Andrea, I had the same needs. Think correcting or increase the I2S to PCM manuals should help some other hard tweakers as well. Up to you if time enough.


The only doubt maybe if the role of ldo on the path of the current coming from cells : is it bypassed because there is no voltage drop for regulation due to the low voltage of LiFePo4 cells
Also is there a difference despite no voltage drop if this regulator stays on the path! I think that's what Andrea and I want to express/ask. Wlowes and Greg gave precious informations but bypassing term is hard to be understood in that context : is it bypassing regulation inside the ldo reg because no regulation or does the use of the vias as stated wlowes bypass the ldo because the drawing circuit is made to avoid this ldo reg when you input power from the free vias on the side of the PCM boards !


As i wrote above : that's why point 8 & 9 of the manual tips section introduce doubts and it should be corrected !


So yes if you can precise to us please.
 
Last edited:
@iancanada
I'm using McFIFO and after various experiment I add simple fiber link (based on HP receivers/transmitters and S/PDIF decoding to I2S) - it works great - there is no audible differences between transports ( computers, CDs) after using fiber isolation but this solution is limited to 192kHz, so I think that major key to good sound from computer setup is isolation (only complete isolation is fiber solution, ICs based isolation always have parasitic capacitance). Do you consider design and built ultra isolation for higher bitrate and for DSD : I2S->fiber->I2S->McFIFO ?
 
Sorry, I don't mean to be a bother but I was hoping for some input on the above question. Perhaps I should have given less background.


"Does anyone know if I can hook up dual mono DAC hat via GPIO to FIFOPi, and still use the u.fl, and mute and gnd vias on the FIFOPi board to supply another DAC?"


Thanks for any help!


Third times a charm? Anybody know if this is possible?
 
Disabled Account
Joined 2019
Hi,

It s ok imo. I don't think there are logical gates which close one outputt when the other is connected first, it doesn't make sense btw. So it's surely stream at all the outputs but if a jumper forces you to choose between the gpio or the uf.l outputt.(reading the manual lets you know if such jumpers exist)
In the real world, I m not sure it s a good idea (for instance bi amp purpose), could create long ground loops and also a mismatch at impedances outputts, just the uf.l is 50 ohms proof... Gpio ???, while distance is very short between the gpio and its dac hat ( well the highest the clock frequency the morz difficult, you should stay on low speeds with old pcm chips).... At least theoricatly for this last!

Discussion is open, EEs will inputt if I'm wrong but your question is excellent indeed. Can we have a dual dac outputt with the fifopi hence not using the MCBOARD if just needing bi amp or dual rooms streaming.
 
Last edited:
hi guys,

I have a fifo with spdif interface board and i need assistance on how to extend the input selector on the spdif board so i can mount the selector switches on the faceplate of the enclosure. the external control panel interface has header pins but i do not know what switches to use and how to implement it.

your assistance will be greatly appreciated.

tubo
 
Hi all,

Did somebody try to play DSD512 files on ES9028Q2M Rpi DAC Hat?
When I started to play DSD512 file, it began to play, then paused, then began to play again. And so on to the end of the file.
The system is RPI 3 b+ , FiFo PI Ultimate, 90.3168 & 98.3040 TCXO clocks, ES9028Q2M Rpi DAC Hat, standard I/V board, Volumio image.

Does anybody have any suggestions?

Thanks!
 
It's from volumio.org site:

"Yes, finally DSD Direct is available on Volumio, your favourite audiophile music Player. The formats supported in the new Direct DSD mode are:

DSD64 (Single-rate DSD)
DSD128 (Double-rate DSD)
DSD256 (Quad-rate DSD)
DSD512 (Octuple-rate DSD)

Previously, DSD Playback in Volumio was only possible via DoP (DSD over PCM) mode, which was encapsulating DSD Audio into PCM samples. You can see DoP as a “compatibility mode” for non direct DSD capable DACs: files would still play, but losing the supposed advantages of DSD. Instead, direct DSD mode allows DSD Playback the way its meant to be played, on compatible DACs.

Direct DSD playback is available from version 2.163
If you have a direct DSD DAC, just select DoP off and Volumio will play your DSD files natively up to the resolution supported by your DAC"

So Volumio supports native DSD512 and ES9028Q2M also supports DSD512.

Any other suggestions please.