The Best DAC is no DAC

Thanks for the positive comments and sharing your project with us Hazard.

Your comment above is quite timely as I've been thinking about the effort that will be required to get this group buy progressed; I never expected anything like this level of interest when I offered the spare PCBs. I certainly don't want to appear as though I'm treating this as a money making exercise but equally I don't want to be 'out of pocket'. How would people feel if I was to revise the price to £3 and £4.50 per SE and flip-flop boards respectively - shipping would still be at cost?

Ray

No problem for me, Ray
 
Thanks for the support guys.

Attached is an up to date spreadsheet listing what boards are required by whom.

Any further requests for flip-flop boards will now go onto a waiting list and will be met either when the list has 10 boards required or if anyone currently on the list changes their mind.

I can accept a few more requests for single-ended boards before the same rule as for flip-flops applies.

I already have 8 or 9 of each type of board from the initial batch that I ordered and I will aim to start posting them out towards the end of this week. I will offer them out in list order.

I will order the additional boards this week, however, as it is the Chinese New Year there may be some delay before the order is progressed.

Ray
 

Attachments

why do you don't had used for power supply something of better? You could to use Lt3080 low dropout, also if I don't see all thing around the regolator, filter and others...What trasformer ratio did you have used?
Hi Emilio,

The JL Sounds USBtoI2S has on-board regulation (low noise LP5900 @ 3.3V) so the 7805 is merely dropping voltage to a reasonable level - a better reg won't make much difference here. The flip flop is powered from a 6V SLA battery, which is better than any active reg IMHO.

The transformer is 1:1. The flip flop has 6V rail (actually measures at 6.4V) and therefore has higher output compared to straight off the board (3.3V voltage rail). And the flip flop has complementary outputs so voltage swing is doubled again. So there is no need to use transformer for voltage gain.

Regards, Hazard
 
Thanks for the positive comments and sharing your project with us Hazard.

Your comment above is quite timely as I've been thinking about the effort that will be required to get this group buy progressed; I never expected anything like this level of interest when I offered the spare PCBs. I certainly don't want to appear as though I'm treating this as a money making exercise but equally I don't want to be 'out of pocket'. How would people feel if I was to revise the price to £3 and £4.50 per SE and flip-flop boards respectively - shipping would still be at cost?

Ray
Ray, a good coupling cap would cost much more than this. I think this is still very cheap for a well designed PCB.
 
Thanks for the positive comments and sharing your project with us Hazard.

Your comment above is quite timely as I've been thinking about the effort that will be required to get this group buy progressed; I never expected anything like this level of interest when I offered the spare PCBs. I certainly don't want to appear as though I'm treating this as a money making exercise but equally I don't want to be 'out of pocket'. How would people feel if I was to revise the price to £3 and £4.50 per SE and flip-flop boards respectively - shipping would still be at cost?

Ray

No problem for me Ray, it's still dirt cheap[but not in quality)
 
Hi Emilio,

The JL Sounds USBtoI2S has on-board regulation (low noise LP5900 @ 3.3V) so the 7805 is merely dropping voltage to a reasonable level - a better reg won't make much difference here. The flip flop is powered from a 6V SLA battery, which is better than any active reg IMHO.

The transformer is 1:1. The flip flop has 6V rail (actually measures at 6.4V) and therefore has higher output compared to straight off the board (3.3V voltage rail). And the flip flop has complementary outputs so voltage swing is doubled again. So there is no need to use transformer for voltage gain.

Regards, Hazard

Hi Hazard,

I would understanding how many volt you have in output? It necessary an pcm file 0db to converting in dsd and read the output voltage. Because Ray, have said that voltage after lp filter in the flip-flop is +- 0,6 V..

Thanks
Emilio
 
Cheers chaps.

I've just been assembling the chassis for the flip-flop/BCF project ready for a good listen when the household is up and about....

An externally hosted image should be here but it was not working when we last tested it.


A little more information about the ingredients other than the JLSounds and flip-flop board:

Power supplies for the JLSounds/Flip-flop are TPS7A4700 low noise LDO regulators (see here: http://www.diyaudio.com/forums/vendors-bazaar/263618-tps7a4700-low-noise-ldo-regulator-pcb.html)

Info about the Broskie BCF buffer is available at the Glassware store and in the TubeCad Journal. (see here: BCF PCB and Kit and here: BCF-2 & Aikido Single-Ended Output Stage). My BCF is configured for 6DJ8 tubes running at 200V.

The 200V B+ is supplied by a Morgan Jones inspired Statistical Regulator circuit (the small board with a big red film cap. This was bought via a group buy on one of the UK audio forums. (MJ Statistical Regulator PCBs - audio-talk). Upstream of the regulator board is a CRC filter.

The filament supply is from another Audio-Talk group buy (6.3v indirectly heater valve PCB (possible group buy?) - audio-talk).

Tranformer was custom wound by Canterbury Windings here in the UK (Canterbury Windings Home)

Chassis is from China, purchased via ebay.

Ray
 
And a better picture of the updated original project, now using the revised SE board;

An externally hosted image should be here but it was not working when we last tested it.


The output from the SE board goes to offboard DC blocking caps (Mundorf 250V MCap MKP Capacitor), then an Alps Black Beauty pot and finally to a JLSounds JG Buffer.

Now that I have the Flip-flop/BCF project up and running and work is progressing with the SE/Aikido project I don't really need this initial project anymore so I'll probably list it on Swap Meet sometime soon; if anyone is interested in a ready built no-DAC DSD decoder for a good price drop me a PM (toroid primary windings are 230V only).

An externally hosted image should be here but it was not working when we last tested it.



Ray
 
Hi Hazard,

I would understanding how many volt you have in output? It necessary an pcm file 0db to converting in dsd and read the output voltage. Because Ray, have said that voltage after lp filter in the flip-flop is +- 0,6 V..

Thanks
Emilio
OK I will see if I can create a 0dB test signal in Audacity and test as you suggest. I'm not working this week so should get time to try this out.

Regards, Hazard
 
Would it be possible to use this with computer based crossovers feeding a exa21 multichannel usb to i2s converter in a 4 way active system? I have been told the exa will output dsd....

Or, would I be better off with 4 usb to i2s 2 channel boards with 4 of these for a multichannel setup?

Last question...anyone have the Soekris r2r dac and planning a comparison? I bought 4 of the Soekris dacs when they first came out but have yet to do all the mods and set them up....This seems so much simpler, just wondering if they will offer comparable performance...

Thanks for any input.
 
Still work to do...

I've had a good session trying out both of my DSD playback projects as sources in my main audio system; this is running 300B OTL amps (modified versions of the Transcendent 300B OTL amp) and Lowther BLH speakers so great speed and resolution.

For input data I'm running HQPlayer, upsampling my CD rips to DSD256, then streamed to an NAA. Obviously I'm only playing DSD.

After a couple of hours auditioning I have a bit of a mixed bag of results. Sound quality is really good but I have two issues to resolve.

The first is that I now have loud pops whenever I use the 'Play' or 'Stop' buttons in HQPlayer with both the flip-flop/BCF project and the original project updated with the revised SE board. There are no such issues transitioning from one track in the playlist to the next whether that is the natural transition at the end of a track, using the 'Play Next' button or double-clicking on a different track in the playlist. I didn't have this pop noise with my original build. The problem is that a few things have changed apart from the LP filter/mute boards (which are essentially identical with respect to the mute circuit), for example I've just updated my version of HQPlayer. With the amplifiers turned off I can hear the mute relay clicking and I've checked the relay coil voltages; it's as if the mute trigger no longer gets enough notice so cannot activate the relay quickly enough.

The second issue was a mushy, white noise that I could hear in the background, particularly during quiet passages. The noise was only present during playback and didn't seem to change in level. After some experimentation I found that the noise is being introduced by the HQPlayer digital volume control.

I've posted observations on the two issues for Jussi at HQPlayer as the next step.

I'm not to concerned by the HQPlayer volume control issue as that ought to be resolvable and could be circumvented by using an alternative volume control arrangement if necessary. The loud pops are more of a concern.

I'll keep you upated.

Ray
 
Still work to do...

I've had a good session trying out both of my DSD playback projects as sources in my main audio system; this is running 300B OTL amps (modified versions of the Transcendent 300B OTL amp) and Lowther BLH speakers so great speed and resolution.

For input data I'm running HQPlayer, upsampling my CD rips to DSD256, then streamed to an NAA. Obviously I'm only playing DSD.

After a couple of hours auditioning I have a bit of a mixed bag of results. Sound quality is really good but I have two issues to resolve.

The first is that I now have loud pops whenever I use the 'Play' or 'Stop' buttons in HQPlayer with both the flip-flop/BCF project and the original project updated with the revised SE board. There are no such issues transitioning from one track in the playlist to the next whether that is the natural transition at the end of a track, using the 'Play Next' button or double-clicking on a different track in the playlist. I didn't have this pop noise with my original build. The problem is that a few things have changed apart from the LP filter/mute boards (which are essentially identical with respect to the mute circuit), for example I've just updated my version of HQPlayer. With the amplifiers turned off I can hear the mute relay clicking and I've checked the relay coil voltages; it's as if the mute trigger no longer gets enough notice so cannot activate the relay quickly enough.

The second issue was a mushy, white noise that I could hear in the background, particularly during quiet passages. The noise was only present during playback and didn't seem to change in level. After some experimentation I found that the noise is being introduced by the HQPlayer digital volume control.

I've posted observations on the two issues for Jussi at HQPlayer as the next step.

I'm not to concerned by the HQPlayer volume control issue as that ought to be resolvable and could be circumvented by using an alternative volume control arrangement if necessary. The loud pops are more of a concern.

I'll keep you upated.

Ray

I am not sure the mute triggers from the jlsound helps... I try to implement the mute using the amanero once also does not help... The mute has to come in before the signal cuts of before the signal start...

The pops will be there as it's the nature of pure signal bit DSD... A LPF averages the signal so during the beginning or stopping... There is an abrupt start or stop to the signal... During pause or next track there is no pop as the signal is not "stop" in a sense... I have 2 pure dac less single bit DSD dac also behaves the same way... It has to be dealt with on the sources believe...During start/stop the player has to fade off to silence I think rather than end it abruptly... How I overcome this is by turning the volume down at the start of the first song and press pause immediately... Bring up the volume and press play there will be no pop... And from there never press stop... Just pause or next track and add more songs to the playlist as I listen... When I'm done listening I will turn down the volume and press stop... Quiting the NAA with the volume up will give u a loud screeching sound.... Using HQplayer as well btw... Same for jriver which has a even louder pop.... To add furthe... This pop also is system dependent... I have tried this on another setup using HQplayer ... The pop is still there but bearly noticeable...

If anyone prefer daphile....This player is able to handle this pop better than any other player... The pop is at a much tolerable level...