Build thread - building the Subbu DAC V3 SE

FWIW, I have some of the 4 pin headers in my cart and they will come along when I add enough to warrant shipping costs. Hope whoever is reviewing/correcting the DAC BOM adds those numbers. Both X2 and X6 on the version I'm using have no vendor info on those two lines. There may be a newer BOM version available but don't know if the "final" has been developed or distributed yet.

Once that has been established, and if someone hasn't already done so, I'd be happy to build a one-click purchase shared shopping cart for a "standard" build for those who don't have parts on order yet. That's been helpful on some other projects.
 
Last edited:
Disabled Account
Joined 2002
After Gary stated that a larger value for C22 would improve matters I only could get Panasonic FM 220/10 in time to try out. No time for measuring but to my ears it sounds a tad better. Not a very large difference but still... This is still with 1 µf ceramic cap in parallel.

So feel free to experiment with larger values than 100 µF for C22. Gary said 470 µF was his choice but the ones I could get hold of don't fit on the PCB. The Pana FM 220/10 does fit exactly.
 
Last edited:
Member
Joined 2002
Paid Member
Perhaps this part then: 6SEPC470MW (polymer) or this one EEU-FR0J471 (electrolytic)
It does not need to be 10v. Right?

I need to read the polymer cap apps note for the ceramic cap.

I prefer the polymer caps to conventional electrolytic caps for this location. I tried some 220uf electrolytics first and similar to JP's observation, found it to be a small improvement. Changing to a 220uf polymer cap gave a much bigger improvement. And there was yet another big improvement moving up to 470uf polymer. All of this with the 1uf ceramic bypass cap. I want to try a 10uf bypass cap but haven't done that yet.

---Gary
 
Disabled Account
Joined 2002
Perhaps this part then: 6SEPC470MW (polymer) or this one EEU-FR0J471 (electrolytic)
It does not need to be 10v. Right?

I need to read the polymer cap apps note for the ceramic cap.

Both would fit but I could not get the SEPC cap this week. Farnell says it will be available 30th of December. I would try polymer as Gary has the best results with those.

Remember nothing has been measured with the polymer caps so you are on your own if you like to try. I can tell the Panasonic FM 220/10 is an improvement but a small one. The reg does not seem to oscillate but I can't verify that now. You can use caps with a voltage rating above 3.6 V so even 4 V types would be suitable although I think 6.3 V as a minimum would be wiser.
 
Last edited:
Sadly, this is turning out to be my hard luck (read stupid) build. I've dropped out because I missed getting R4 on the PS - twice.:mad: My big resistor package includes 1K, 1.5K and 1.8K pieces. Will any of those work and let me avoid waiting for another order?
 
Last edited:
The polymer caps have such an incredible low ESR that they might cause a resonance circuit when paralleled with small value ceramic caps. I read all this yesterday evening and think polymer caps are very exciting but should be used with attention. Solid polymer caps can not just be used as a replacement in many many cases ! Manufacturers advise to use larger value bypass ceramic caps and explain the whole matter in white papers.

hi jean-paul,
on a similar subject, the united chemicon polymer caps specified for C2, C3, and C18 were OOS at both of my usual suppliers, so i took the liberty of making a substitution with a panasonic SEPC part.
APSA160ELL101MFA5G 100uF/16v 25mOhm esr
16SEPC100MW 100uF/16v 10mOhm esr

would the lower esr here cause potential resonance even when paralleled with the larger 4.7uF ceramics at C5, C7 and C19?

thanks,
-matt
 
finished psu

here's the psu. just finished soldering the last components tonight (2200uF caps, bridge, led header, and transformer) and cleaning it. the picture doesn't do it justice; it's much shinier in real life :D
-matt
 

Attachments

  • psu-3.jpg
    psu-3.jpg
    124.5 KB · Views: 507
Disabled Account
Joined 2002
hi jean-paul,
on a similar subject, the united chemicon polymer caps specified for C2, C3, and C18 were OOS at both of my usual suppliers, so i took the liberty of making a substitution with a panasonic SEPC part.
APSA160ELL101MFA5G 100uF/16v 25mOhm esr
16SEPC100MW 100uF/16v 10mOhm esr

would the lower esr here cause potential resonance even when paralleled with the larger 4.7uF ceramics at C5, C7 and C19?

thanks,
-matt

Hi, I do use polymer 100/10 myself for C2, C3 and C18. No problems, also with C5, C7 and C19 in 4.7 µF tantalum. The ones I got fit exactly (3.5 mm pitch) and the SEPC 100/16 are smaller with a smaller pitch (2.5 mm). They don't fit if you like shortest connection and a tidy built neat looking DAC. The 100/16 SEPC fit on the footprint for C22. I have the SEPC 100/16 but I don't use them on the V3.

With the V2.6 I once had oscillating MIC regs with ceramic caps at their outputs. in the hope that others read this: don't use ceramic caps straight after the regs !
 
Last edited:
After Gary stated that a larger value for C22 would improve matters I only could get Panasonic FM 220/10 in time to try out. No time for measuring but to my ears it sounds a tad better. Not a very large difference but still... This is still with 1 µf ceramic cap in parallel.

So feel free to experiment with larger values than 100 µF for C22. Gary said 470 µF was his choice but the ones I could get hold of don't fit on the PCB. The Pana FM 220/10 does fit exactly.

OK, I'm confused now - I ordered 68uF SAL-RPM caps for C22 (at some cost) - Do I fit these or 100uF Polymer, or 220uF polymer or 470uF Elect. What, if anything will I lose by using the SAL-RPM's.
That said, I am very sceptical when it comes to swapping capacitors as in the past I have seen very little, if any, improvements.

Ken
 
Disabled Account
Joined 2002
Just use the SAL caps and you won't be confused. Exchanging Avccc/Vpos caps for polymer caps is done by Gary but it is not what we used ourselves. We do find that caps make a difference like for example C22 but we made a choice back then. With SAL you will have a good result and when you're brave you could try out polymer caps at your own risk. You don't "loose" something when using prescribed parts and yes they come at a cost. I still have to find free parts :) Wait till you have built around 25 of them, then add up what you spent !

Exactly avoiding this confusion is why you all wanted a BOM. Any deviation and chaos arises....Only the more adventurous builder tries stuff out and it will only be possible to compare if 2 or more are built. In development stage we do also try out parts , how else can we do otherwise ? So we stick to certain parts and don't bother again afterwards. Many af us don't take things for granted and want to go beyond the standard. That is fine but I now see that such info can work confusing for those that want to follow a guideline and/or are not used to choosing themselves.

With the BOM parts you will have a good working good (if not excellent) sounding DAC. For those that want to press the last % out of it there is some room for experimentation. If there are any preconceptions, beliefs etc. I would not even bother to read any info regarding modifications.

This is not going a direction I like and any statements done for improving the V3 are entirely for the person making them and I include myself in that. Any of you that likes to try out parts please do so but do not complain if results are negative. Maybe this should not be in the building thread at all ?
 
Last edited:
Just use the SAL caps and you won't be confused. Exchanging caps for polymer is done by Gary but it is not what we used ourselves. With SAL you will have a good result and when you're brave you coul try out polymer at your own risk. You don't "loose" something when using prescribed parts and yes they come at a cost. I still have to find free parts :)

Exactly this is why you all wanted a BOM. Any deviation and chaos arises....

With the BOM parts you will have a good working good (if not excellent) sounding DAC. For those that want to press the last % out of it there is room for experimentation. If there are any preconceptions, beliefs etc. I would not even bother to read any info regarding modiciations.

Many thanks for the sanity check JP - Excellent advice - Even to us with a high paranoia factor!

Ken
 
Disabled Account
Joined 2002
Oh I really need to weigh my words before saying them. I don't know of any new GB. My guess is that Subbu needs some rest now. Of course we are busy with something new. I doubt an I2S version still is valid with so many offerings for little money. Just check Curryman's I2S DAC.... I will discuss this with Subbu.

I have confidence in our V3 to be one of the better SPDIF ES9023 devices. Just modified an O2 Joggler with Squeezeplay OS to have an SPDIF output and it is amazing how this combo sounds. Must admit that adding a 75 Ohm BNC in the Joggler is a nerve racking task but it works and the device still is unscratched etc.

Tip: O2 Jogglers are excellent replacements for Squeezeboxes ! You need to be experienced to add RCA or BNC SPDIF and to have a fearless approach to Linux...
 
Last edited:
JP, Didn't mean to give the wrong impression. I was referring to your comment about building a couple DACs for comparison purposes. I was asking if there had been any over production of just the PCB, or if everything has been sold, is there enough interest to warrant a GB/production run for the boards only. I really like your idea of multiple builds as an alternate to pulling and replacing parts.
 
Last edited:
Disabled Account
Joined 2002
Hi Bob I did exactly understand what you are saying but I think Subbu has had his share of Group Buys for the moment, that also counts for extra production of boards. It was a one time only affair from the beginning.

Of course I am willing to let my spare PCBs go for enormous amounts of money ;) No I won't, not everything in life is about money !
 
Last edited: