Jump to content
IGNORED

JLSoundsi2soverusb USB to SPDIF adapter board and galvanic isolation


Recommended Posts

I have been using this board in conjunction with my 47Lab. Dac - it is only a 16/44.1 dac w/SPDIF input- and my Apple MB pro using Audirvana. This has been going on for a few months with the purpose of testing computer audio waters. I know the computer set up is not the best for CA, especially considering that the dozen cd's I have copied are in the computer hard disk. The usb cable is a diy using Jena Labs wire, pretty good I might add.

 

Audirvana has been optimized though to the extreme parameters and mode 1. So the device was using the computer's usb 5v feed. This morning I confectioned a battery 5v power supply, 8 - 1.5v batteries in series with a 5v regulator via a 50 uf poly cap. The board has a pin which disconnects its power feed from the 5v computer output. So it was disconnected and the battery supply went in. Results: better sound than before, but nothing that made me change my mind that my CD transport and dac sounded better. Then decided to install another identical battery power supply independent from the first one in another part of the board circuit as explained by the JLSounds instructions, which according to them it meant there was galvanic isolation in the circuit, whatever that means. Now the presentation really changed towards a smooth and warm one, very musical, I think better or par with the CD player. A very big leap in sound quality. So this galvanic isolation thing and relieving the dac from the computer's 5v really is a must. Before all this I was beginning to give up though I considered my CA set up was producing a cold presentation albeit with all the details. Thanks for reading!

Link to comment

I'm glad to hear this. I ordered one with the clock board and power supply. I will be directly feeding the pcm63 chips on my Adcom DAC-600, bypassing the DF1700 digital filter. I also use A+ and will use the Izotope for oversampling (or not) and digital filters.

 

I know the Shigaraki is NOS, so you could use Izotope for software filtering with this setup too. In case you wanted to experiment w differences in OS and NOS.

Link to comment

Hey thanks! I am planning on 2 linear power supplies because those battery pack supplies are current drained pretty soon - after 2 hours or so-

 

I am not an expert in Audirvana settings yet, and because my dac limitations there is no oversampling I can do. This Izotope thing , which I have in my settings, and the NOS, how would they correlate?

Link to comment

According to the TDA1543 datasheet, it can accept up to 4x oversampling. For a 48khz signal that's 192khz. However, the max sample rate accepted on the CS8414 is 96khz so that's likely the upper limit you could send over. For a test try these settings and see if you get a signal.

 

1. Go to Audirvana+ Preferences

2. Go to the "Audio System" Tab

3. On that tab you should see your active audio device and in Green the sample rates it will accept. Since the I2SoverUSB will be the active audio device, it should be all lit up.

4. At the very bottom of that screen, you should see "Max Sample Rate". Use the drop down to select 96khz.

5. I'd select the bit depth to be 24bit just in case.

6. Go to the Audio Filters Tab. Make sure iZotope 64bit is selected.

7. For "Forced Oversampling" Select Power of 2. This means that 44.1khz will get sampled up by 2x or 4x up to you max sample rate. In this case 88.2khz.

8. Quit and see if you get music out of your DAC. If so take a screen shot of the default settings for the iZotope SRC. Then read this thread and play around: http://www.computeraudiophile.com/f11-software/izotope-sample-rate-convertor-15352/

9. If you don't like the sound, change the Forced Oversampling back to none.

 

I'm not expert, this is just what I will be doing once I receive my I2SoverUSB.

Link to comment

Well it looks like I screwed up the board with all the iterations I did with the power supplies. I tried to implement 2 linear regulated power supplies yesterday and in the midst of a track the music stopped. Although the green led is on it does not work. Even going back to the configuration powered by the computer's USB. The issue is that Audirvana or the MacBook Pro is not recognizing the board anymore. Any ideas? So be careful with the soldering.

Link to comment

Tried the 3 power supply alternatives and even with #1 which is the simplest-usb 5v from computer- the Macbook pro dies not recognize the board. Neither Audirvana nor MIDI.... So I guess the board has gone bad. Emailed JLsounds but their suggestions have not worked either

Btw, the green led lights up

Link to comment

Not yet, I need some chassis wire to connect up to my DAC. Should be coming over the weekend. Hopefully I have time to do it. Otherwise won't be in till next week.

 

My install is a little more involved as I'll have to remove the digital receiver chip and send the data and clock signals from to board to the appropriate spots on my DAC board.

Link to comment

Glad to hear it worked! Hoping i have some time on Sunday to do a test install on mine.

 

Have you tried limiting your bit depth to 24bit? Also did you check out the thread I linked above? Has a ton of information on the Izotope settings.

Link to comment

Hey Sev, things I have found out: The board really works much better with the 2 independent power supplies as exposed by me previously. However, I achieved that using 2 banks of 8 AA batteries in series with a 7805 regulator and an ASC 40 uf cap and although it gave the aforementioned results the USB ps did not last long. The USB section draws a lot of current and the batteries were drained pretty fast, and they started to emit a liquid...

 

After confectioning a pair of linear ps's it seems that the USB part is very sensitive to the ps and I could never make it work; had to use the computer's usb 5v -with a Audioquest Jitterbug- voltage for that section and a linear ps for the reclocking section. And that is giving fine results. I am not an electronics expert so I don't know what the issue might be with the usb section power s.

Link to comment

Something was wrong then as the USB power ought to be easy to replicate. I have never, ever had issues creating a PS for the USB side of things.

Hey Sev, things I have found out: The board really works much better with the 2 independent power supplies as exposed by me previously. However, I achieved that using 2 banks of 8 AA batteries in series with a 7805 regulator and an ASC 40 uf cap and although it gave the aforementioned results the USB ps did not last long. The USB section draws a lot of current and the batteries were drained pretty fast, and they started to emit a liquid...

 

After confectioning a pair of linear ps's it seems that the USB part is very sensitive to the ps and I could never make it work; had to use the computer's usb 5v -with a Audioquest Jitterbug- voltage for that section and a linear ps for the reclocking section. And that is giving fine results. I am not an electronics expert so I don't know what the issue might be with the usb section power s.

Forrest:

Win10 i9 9900KS/GTX1060 HQPlayer4>Win10 NAA

DSD>Pavel's DSC2.6>Bent Audio TAP>

Parasound JC1>"Naked" Quad ESL63/Tannoy PS350B subs<100Hz

Link to comment

This I2S over USB with the Clock board is simply amazing!

I have a Lampizator Big7 DAC for just over 2 years now and replacing the Amanero board by adding the combination above has made a massive difference to the Big7

 

I had been using a Singxer Audio F-1 XMOS USB Digital Interface Module XU208 U8 upgraded version for the past few months and this was a significant improvement over the Amanero but this board is so much better.

 

I am powering both the boards off the clean 5v off the Big7 which was on the Amanero and it is working flawlessly.

PCM and DSD never sounded so good, it sounds like my Big7 was in hiding in the long grass for the past 2 years with the Amanero holding it back.

 

I have just gone and ordered both again to see if a Meitner MA1 can benefit from it but this could be a real challenge to figure out the digital side.

 

Thanks to JL Sounds for a wonderful product, Pearse.

Audio PC - Gigabyte H97M-D3H, i7 at 800Mhz, RAM at 800Mhz & PPA OCXO Mobo, Teradak ATX Linear for 20 pin ATX on Mobo, Paul Hynes SR7EHD 12v, 5v & 5v supply on Mobo, Stammheim 12x LT3045's for 1.3v to RAM direct supply, JCat V2 USB Card, WTFPlay Linux Audiophile Player control by MELE F10, Startech LEX to REX on 12v Paul Hynes with 2x SLC cards and out by POE to ISO/Regen, PPA Red USB Cable, Lampizator Big7, Nige design Lifepo4 powered amp, Raidho C1s.

Link to comment

I agree that this card is a step up from the Amanero, even without the clock board. May I ask what clock you are using, and if yuo have gotten the device to do DSD256 or 512?

This I2S over USB with the Clock board is simply amazing!

I have a Lampizator Big7 DAC for just over 2 years now and replacing the Amanero board by adding the combination above has made a massive difference to the Big7

 

I had been using a Singxer Audio F-1 XMOS USB Digital Interface Module XU208 U8 upgraded version for the past few months and this was a significant improvement over the Amanero but this board is so much better.

 

I am powering both the boards off the clean 5v off the Big7 which was on the Amanero and it is working flawlessly.

PCM and DSD never sounded so good, it sounds like my Big7 was in hiding in the long grass for the past 2 years with the Amanero holding it back.

 

I have just gone and ordered both again to see if a Meitner MA1 can benefit from it but this could be a real challenge to figure out the digital side.

 

Thanks to JL Sounds for a wonderful product, Pearse.

Forrest:

Win10 i9 9900KS/GTX1060 HQPlayer4>Win10 NAA

DSD>Pavel's DSC2.6>Bent Audio TAP>

Parasound JC1>"Naked" Quad ESL63/Tannoy PS350B subs<100Hz

Link to comment

I ordered the Oscillator board and both the Crystek CCHD-957 Oscillators which Lyuben kindly offered to solder on the board free of charge. My Big 7 still has the 1x & 2x DSD board so I have not tried higher DSD rates yet. Thought about sending the DAC back for the upgrade a few times but it is sounding excellent as it is.

 

USBtoI2S with Crystek Clocks.jpg

Audio PC - Gigabyte H97M-D3H, i7 at 800Mhz, RAM at 800Mhz & PPA OCXO Mobo, Teradak ATX Linear for 20 pin ATX on Mobo, Paul Hynes SR7EHD 12v, 5v & 5v supply on Mobo, Stammheim 12x LT3045's for 1.3v to RAM direct supply, JCat V2 USB Card, WTFPlay Linux Audiophile Player control by MELE F10, Startech LEX to REX on 12v Paul Hynes with 2x SLC cards and out by POE to ISO/Regen, PPA Red USB Cable, Lampizator Big7, Nige design Lifepo4 powered amp, Raidho C1s.

Link to comment

Thank you for the reply. I may have to revisit the board soon. I am hoping that they worked out the issues with the higher rates I need for my Signylist DSC1. I purchased the clock board, but will not expend the energy unless I can get those higher rates. In their specs I just noticed that they no longer claim DSD512, but I'd take DSD256 if it could work.

I ordered the Oscillator board and both the Crystek CCHD-957 Oscillators which Lyuben kindly offered to solder on the board free of charge. My Big 7 still has the 1x & 2x DSD board so I have not tried higher DSD rates yet. Thought about sending the DAC back for the upgrade a few times but it is sounding excellent as it is.

 

[ATTACH=CONFIG]27690[/ATTACH]

Forrest:

Win10 i9 9900KS/GTX1060 HQPlayer4>Win10 NAA

DSD>Pavel's DSC2.6>Bent Audio TAP>

Parasound JC1>"Naked" Quad ESL63/Tannoy PS350B subs<100Hz

Link to comment

Unfortunately my DAC does not do DSD 256 so cannot test.

I have been testing the Diyinhk XMOS 768Khz USB card today and again great sound on PCM, have not got it working on DSD yet.

No isolation like the JL Card but both have linear Reg on board which is power hungry and gets hot and the Diyinhk XMOS 768Khz USB card is fed by LiPo 3.3v battery.

 

I am going to post some pics for the 5v in which I take from the Big7 for the JL Cards.

I connected both the 5v for the Clock Card and the USB card together and works great.

 

5v in on JL Sounds.JPG

 

I soldered a 2 pin connector as below and connect the Lampi 5v here.

 

5v in on JL Sounds 4.JPG

 

For testing I lifted the I2S lines off the DSD board which sits just below the top panel, drilled holes in the top panel for short legs with pins soldered on each line so I could plug and play with ease as i tested each card.

 

When I am done with testing and find a clear winner I will install inside.

 

3 Cards 2.JPG

 

For the JL Sounds combo on my Big7:

FSCLK Line goes to Pin 13

DATA goes to Pin 14

BCLK goes to Pin 15

MCLK goes to Pin 17 (Pin 18 with higher clock rates did not work right)

There is a Ground connection to the right of the MCLK on the Big7 DSD board and I took this up through the top panel for Ground and connected this to a free ground on the JL Card, there are several to choose from.

 

For DSD

DSDOE goes to Pin 8

F1 goes to Pin 5 (2x DSD enable)

 

The Mute line was of no use.

 

Hope this helps.

Cheers, Pearse.

Audio PC - Gigabyte H97M-D3H, i7 at 800Mhz, RAM at 800Mhz & PPA OCXO Mobo, Teradak ATX Linear for 20 pin ATX on Mobo, Paul Hynes SR7EHD 12v, 5v & 5v supply on Mobo, Stammheim 12x LT3045's for 1.3v to RAM direct supply, JCat V2 USB Card, WTFPlay Linux Audiophile Player control by MELE F10, Startech LEX to REX on 12v Paul Hynes with 2x SLC cards and out by POE to ISO/Regen, PPA Red USB Cable, Lampizator Big7, Nige design Lifepo4 powered amp, Raidho C1s.

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...