New HQ DIY DAC

[Deleted]

Very tasty Jeff.

With plenty of room for add ons

[Deleted]

UPDATED PHOTO.
This coming Sunday, we hope to have a listen to 2, or perhaps even 3 SC DACs from Sydney RG members.
There is also the possibility that we may be able to do a comparison against Allan Pagan's new Buffalo 3, which the last time I spoke with Allan, was sounding very nice indeed, but needed to be tidied up a bit yet, as it was still in bread board layout.
These DACs will also be compared with david2vk's 24/192 capable W4S, and we will also be able to hear the Steve Nugent "OffRamp" in action, as well as direct playback from an Oppo 95 which has a Sabre DAC inside ,and also using it's USB input with a Corsair Voyager, and via Digital Out into the various DACs.
Alex

Attachments:

[Spirit]

Oct 13, 2011 6:37:08 GMT @sandyk said:

This coming Sunday, we hope to have a listen to 2, or perhaps even 3 SC DACs from Sydney RG members.
There is also the possibility that we may be able to do a comparison against Allan Pagan's new Buffalo 3, which the last time I spoke with Allan, was sounding very nice indeed, but needed to be tidied up a bit yet, as it was still in bread board layout.
These DACs will also be compared with david2vk's 24/192 capable W4S, and we will also be able to hear the Steve Nugent "OffRamp" in action, as well as direct playback from an Oppo 95 which has a Sabre DAC inside ,and also using it's USB input with a Corsair Voyager, and via Digital Out into the various DACs.
Alex

Looking forward to hearing of the results

[Deleted]

UPDATED PSU WIRING.

Attachments:

[Deleted]

The same transformer Primary winding colours must be connected to the mains input via the mains switch. e.g. The Blue wires to Neutral and the Brown wires to Active.

Alex,

With a transformer that has the same colour primary leads, is it safe to assume that using the same corresponding physical entry points of the leads (i.e. left & right) would suffice?

Chris

[Deleted]

Oct 21, 2011 6:46:27 GMT @cjarchez said:

The same transformer Primary winding colours must be connected to the mains input via the mains switch. e.g. The Blue wires to Neutral and the Brown wires to Active.

Alex,

With a transformer that has the same colour primary leads, is it safe to assume that using the same corresponding physical entry points of the leads (i.e. left & right) would suffice?

Chris

Hi Chris
I would imagine so. If you wanted to experiment you could try reversing one and see if you can notice any difference in "focus".
I found the way that the secondary windings were connected to the PSU PCBs made quite a notieable difference.
I replaced a toroidal with orange primary wires with one the same as the other.i.e. brown and blue wires . I was surprised by the improvement. Did you try replacing the 10K resistor in the JLH sections that feed +5V with a 3.3K to 4.7K, or a 1.5mA CCD ?
Alex

[Deleted]

Hi Alex,

to make life easy I think I'll replace the existing transformer as well. The original will always be handy elsewhere!

Aha, CCDs, I'd forgotton. I'll pop some on my next order.

Cheers,

Chris

[jeffc]

Hi Alex,

I got a batch of 2SK117 jfets a while back, might these work as a CCD for low-voltage JLHs.

www.datasheetcatalog.org/datasheet/toshiba/1025.pdf

cheers.. jeffc

[leo]

Should be ok I'd have thought Jeff, I've used 2SK174 and J309's before with a resistor across gate and source to set the required current.

[Deleted]

Jeff
In the absence of more suitable FETs, or the proper CCDs ,it may be easier, and probably just as effective to replace the 10K with a 3.3K to 4.7K resistor where the JLH is used at 5V.
Alex

[Deleted]

The J506 CCDs seem pretty tough to find, there is the ebay seller that I think Shaun found and there is also this mob;

www.utsource.net/j506.html

They supply small quantity, I've just ordered 10 x J506 (@ $1.50 ea) and we'll see how good their supply is.

[Deleted]

Oct 23, 2011 20:23:21 GMT @cjarchez said:

The J506 CCDs seem pretty tough to find, there is the ebay seller that I think Shaun found and there is also this mob;

www.utsource.net/j506.html

They supply small quantity, I've just ordered 10 x J506 (@ $1.50 ea) and we'll see how good their supply is.

They just rolled in!

Look good to me.

Attachments:

[Deleted]

Alex,

I just popped a couple of J506 on the two 5vdc boards.

I could swear your correct! There seems to be a tad more clarity, granted I can't A-B but...

Plus, two new 15vac transformers have arrived today but I need to mount the whole kaboodle back into the old test chassis to wire them in, the box I had squeazed it all into won't accommadate a fag paper more, let alone another tranny!

[jeffc]

I just popped a couple of J506 on the two 5vdc boards.
I could swear your correct! There seems to be a tad more clarity, granted I can't A-B but...

Good to here Chris as I also have some on their way. Price is creaping up though, $2 ea now for a lot of 10.

cheers.. jeffc

[Deleted]

More progress;

the psu has had a second tranny (new pair) added as per Alexs configuration a few posts above.

Not so much another small step forward as a leap! Mostly a sense of width, openess and sheer power when needed.Multi layered, busy pieces are filed out with great authority.

Only one down point is that it also makes my mains borne buzz more obvious during quiet parts and silences, oh well, warts and all!

I think that only leaves the xtal mod to do, out of the current tried and tested list?
Plus a new box

Which brings me to the question;

What is the easiest/best way to implement the xtal change from the point I'm at now?

In the meantime

[Deleted]

Hi Chris
It really does give the big names a run for the money now, doesn't it ? It goes to show that it's not whether it's USB or SPDIF, but the way it is implemented , and the quality of the power supplies,
as well as the isolation between Analogue and Digital areas as far as PSU is concerned.

I am not completely convinced about the external Xtal oscillator,
as in theory it shouldn't matter.I did that after Alfred recommended it. Perhaps others who have done this too could offer comments here ? Personally, I would be doing the 3.3V Paul Hynes regulator mod to the Input PCB first, although a Leo bodge to a JLH using a 5V to 3.3V regulator in the front may also do the trick without the PH regulator.
Syd can give you info about the regulator type, as Syd very kindly sent me a couple to play with.
Regards
Alex

[Deleted]

Only one down point is that it also makes my mains borne buzz more obvious during quiet parts and silences, oh well, warts and all!

Chris
Perhaps you could provide some more info about that problem , as some of the smarter people here like Frans, may be able to suggest a fix for you.You should be able to get rid of that completely with a bit of work.
Alex

[Deleted]

Sorry Alex, a bunch of questions follow

Nov 9, 2011 8:57:28 GMT @sandyk said:

Hi Chris
It really does give the big names a run for the money now, doesn't it ?

Yep, it's pulled away from my Meridians on board DAC very convincingly now!

I am not completely convinced about the external Xtal oscillator,
as in theory it shouldn't matter.I did that after Alfred recommended it. Perhaps others who have done this too could offer comments here ?

I think most were unsure, as you are and it was only myself remaining to add the Xtal at such a late stage to see if there was any difference or not?

Personally, I would be doing the 3.3V Paul Hynes regulator mod to the Input PCB first,

I assume that's the shunt version?

although a Leo bodge to a JLH using a 5V to 3.3V regulator in the front may also do the trick without the PH regulator.

Is that any better than my current JLH->cricket set up?

ref.

Quote:Only one down point is that it also makes my mains borne buzz more obvious during quiet parts and silences, oh well, warts and all!

Q

Chris
Perhaps you could provide some more info about that problem

I know exactly what causes it Alex.

It is the SMPSs on most especially my laptop if used on the same radial circuit (no ring circuits here), amoungst other units. If they are off/running on battery or on another circuit there is no problem.
I was just pointing out how much more "stuff" was audible/more audible

Cheers,

Chris

[Deleted]

It is the SMPSs on most especially my laptop if used on the same radial circuit (no ring circuits here), amoungst other units. If they are off/running on battery or on another circuit there is no problem.I was just pointing out how much more "stuff" was audible/more audible Cheers,Chris

Hi Chris
A good mains filter in line with the audio gear MAY help here, but I can't guarantee it.
Alex

[Deleted]

Personally, I would be doing the 3.3V Paul Hynes regulator mod to the Input PCB first,I assume that's the shunt version? although a Leo bodge to a JLH using a 5V to 3.3V regulator in the front may also do the trick without the PH regulator.Is that any better than my current JLH->cricket set up?

Hi Chris
It is a small shunt type on a small ,but fairly tall PCB with 3 leads.
The leads are as for a LM317, so a couple need to crossover to fit the original 3.3V reg place on the PCB. You can see it bent over a little in my most recent photo.The other idea may not be feasible due to the different pinout of the LM317 as used by Leo in his bodge, and the pinout of the 3.3V reg which is the same as a 78XX.
Added to that you would need another single JLH and find space to fit it. I shouldn't have mentioned that without fully thinking it through.

The PH reg gives a further small improvement, but whether it is worth the additional cost over the dual Crickets is debatable.I am still using half of the original dual Cricket to supply the DAC PCB via the cable from the Input PCB.
Alex

Attachments:

[Deleted]

The PH reg gives a further small improvement, but whether it is worth the additional cost over the dual Crickets is debatable.I am still using half of the original dual Cricket to supply the DAC PCB via the cable from the Input PCB.Alex

Alex,

Thanks for the info.

The PH regs look pretty crude for £30!

Now that we're all SMD pros ;D maybe we should do some aka RG regs

[Deleted]

Chris
The main reason for the success of the PH regs is the speed of some of their opamps. I understand that they may be 1GHZ types.
P.H. obliterates the type numbers.We are paying for his knowledge and personal assembly and testing time.You will have noticed that it is a 150mA shunt type. IIRC, it should have no more than around +6V at the input, so +5V from a JLH supply is ideal.
Alex

[Deleted]

I pulled my part-completed project out of storage yesterday. Its had, I think, all of the cap mods carried out but I refrained from putting the 40910 metal cans in just yet.

I had a few Jaycar PSU kits so built one up and used that for a trial run. As expected I have a few blips.

I think i must have initially put the third channel selector button upsidedown, it selects the SPDIF channel OK but no light. The other two light up fine. I'm getting yellow and green lights as I should.

However, , I'm getting low sound in the left channel and none in the right. (I have since discovered there's a minute sound from the right if I listen to just it). If I advance the volume control slowly volume increases OK until there's a click in the right channel and all sound stops. Sounds like a relay clicking. Lower the volume and the left channel returns. It never gets near a listenable level.

I'm working my way through everything so early days. I've checked the continuity of the ribbon cables and all is well. One strange thing, when I was examining the board with a loupe I noticed an SMD had a tiny piece of solder across two legs. I really don't think I did it but removng it changed nothing.

I'm just wondering if this peculiar sympton might point to one particular area?

I didn't mount the Toslink devices but I assume that wouldn't affect anything.

Undaunted I'll carry on......

Syd

[Deleted]

Nov 11, 2011 18:32:11 GMT @juke said:

I pulled my part-completed project out of storage yesterday. Its had, I think, all of the cap mods carried out but I refrained from putting the 40910 metal cans in just yet.

I had a few Jaycar PSU kits so built one up and used that for a trial run. As expected I have a few blips.

I think i must have initially put the third channel selector button upsidedown, it selects the SPDIF channel OK but no light. The other two light up fine. I'm getting yellow and green lights as I should.

However, , I'm getting low sound in the left channel and none in the right. (I have since discovered there's a minute sound from the right if I listen to just it). If I advance the volume control slowly volume increases OK until there's a click in the right channel and all sound stops. Sounds like a relay clicking. Lower the volume and the left channel returns. It never gets near a listenable level.

I'm working my way through everything so early days. I've checked the continuity of the ribbon cables and all is well. One strange thing, when I was examining the board with a loupe I noticed an SMD had a tiny piece of solder across two legs. I really don't think I did it but removng it changed nothing.

I'm just wondering if this peculiar sympton might point to one particular area?

I didn't mount the Toslink devices but I assume that wouldn't affect anything.

Undaunted I'll carry on......

Syd

Hi Syd
Getting the yellow and green LEDs is a good sign as it indicates the DSD1796 has locked to a signal.I had a problem initially with the blue LEDs due to the very fine wires, and solder not flowing around them properly due to the gap between the wire and the outside of the hole.Have you checked the voltages around the output I.C at the output of the DAC PCB ?
Alex

[Deleted]

Syd
Something else to check.
With at least some of the earlier Jaycar DAC boards, there is a missing short track between pin 3 of IC12 and the 180R resistor.
Regards
Alex