Build thread - building the Subbu DAC V3 SE

Disabled Account
Joined 2002
That was already tried out by a member but I forgot who it was. The Salas was only very slightly better but considering the size/cost/heat the original PSU was preferred. Feel free to experiment, I think our LM723 :) (still makes me laugh) text book design performs very good. There are only a few regulators that beat the LM723. No 3 legged device can beat LM723.

This japanese gentleman also found out TPS7A4700 is not better than LM723... interesting read.

????????????: Switching & Linear Regulator_
 

Attachments

  • 20130516_31926.png
    20130516_31926.png
    253.1 KB · Views: 461
Last edited:
That was already tried out by a member but I forgot who it was. The Salas was only very slightly better but considering the size/cost/heat the original PSU was preferred. Feel free to experiment, I think our LM723 :) (still makes me laugh) text book design performs very good. There are only a few regulators that beat the LM723. No 3 legged device can beat LM723.
It could be me, actually. I think the LM723 based reg is still excellent with respect to the simplicity, ease of housing the whole unit in a case, and bang for the bucks. It was indeed very close to my very first p2p Salas shunt reg build.
Lately (last couple months), I have tried a few implementation of the Salas Shunt reg, I have to say that the difference is now very noticeabe when I switch the 2 reg back and forth. Since I don't have the right equipment to scope, it is my subjective opinion that the shunt reg is more suitable to my preference. The dark side of the shunt reg is that it adds complexity such as
1. One must follow the "cook book" to figure out how to get 5V.
2. the housing of the dac becomes less flexible as the shunt requires significant heat sinking,
3. a separate transformer
Since it is DIY, one can decide whether he wants to explore or not. I have all the parts (new or used) to build the Salas shunt reg, so it cost me nothing to explore to unleash the max perf of the Dac :)
 
Disabled Account
Joined 2002
OK. I like the Salas and I have them in my DCB1 but IMO they are not really suitable for the Subbu V3 in terms of size and dissipated heat. Subbu V3 should be build tight with short connections for best results. If you have a large case it is possible of course.

Although very similar I did not copy our PSU schematic from the link I gave in post #601. Just saw that when I posted it and thought I was looking at our PSU. Textbook schematics all are alike.
 
Last edited:
Working

Well another Dac working,:D This was my first go at smd soldering and I enjoyed it very much bit of a challenge .I am very pleased with the sound and it will get better with time.
Its a pity I only got one kit as I promised this to my son fancy keeping it for myself ,Thanks to all involved in this GB cheers john
 
Hello,
where can i get the Tamura 3FS-316 (if this is the one recommended) except on Mouser?
because on Mouser, it's arround 3€ the unit, but there's 20€ for shipment ...
or do you have another way to reduce the shipment on Mouser?
On the other side, what else need to be ordered (this could reduce the per-unit price of shipping taxes)?
thanks in advance
 
You can use other transformers than the Tamura... pretty much any print transformer 9V with the good foot print should work.
I built mine with a Hahn transformer I found for little money on 'schuro.de'. But the common Block print transformers (~2,5VA)
should work. The pcb has multiple foot print options (--> is quite versatile).
 
Been off the grid for couple weeks but back on :) Glad to see more successful builds. My stock build still sounds as good as ever. In fact I am still hearing things I never heard after initial power up. btw JP I sold my Schiit Bitfrost dac - yes thats how good this Subbu dac is :)

The tinkerer in me wants to start tweaking but I must fight the urge....if it was non smd I would feel more comfortable but I am leaving this one alone.
 
Hi all,

I heard lot of good comments on this V3. I have it over 30 hours on after finished but still couldn't find any merit out of it. This is to compare with my Audionote 2.1. I have the C22 changed with a OS-Con 470uf polymer, C17 with ceramic cap and without any other mod. Did all of you have buffer on it ??? Any advice would be appreciated.
Regards

Albert
 
albertli - Please don't take any offense, but often when it's reported that components sound similar it is due to the speakers (or other component) used not being capable of letting everything through. Can you give a little info on your speakers? :)

The improvements in sound that I hear with the V3 are not bold. They are more subtle (i.e. cleaner signal) but clearly beneficial.
 
Last edited:
albertli - Please don't take any offense, but often when it's reported that components sound similar it is due to the speakers (or other component) used not being capable of letting everything through. Can you give a little info on your speakers? :)

The improvements in sound that I hear with the V3 are not bold. They are more subtle (i.e. cleaner signal) but clearly beneficial.

I did post couple pictures at 587, they are OB n been playing for me for quite a while. I don't think it doesn't sound good because of my speakers n or my other equipment. I know its not fair to compare the v3 with my Audionote 2.1n this is why I would like to try some more efforts to make it shine.
 
Thanks, looks like a collection of some very nice equipment. I'll have to watch along with you for mods that have an impact.

BTW: I know the chances are slim to none, but I'll put my vote in for another run of at least the DAC boards. Having a stock build and a second for experiments would be a great convenience.
 
IMHO, what sounds better to one is very subjective, and in between, there are many factors affect one's perception/opinion. As Bob pointed out, the synergy of the whole system also play an important role of such.

From my experience of tweaking the Subbu/JP dac (v2.6 and v3) using the not so reliable 55 years old ears to measure :) the difference between switching the parts have been very subtle. I played the dac in 2 radically different audio sets - one is a Push/Pull EL34 tube amp and triode mode that drive a pair of full range, and the other is a Mini-Aleph heavily biased drive a 3 way. But for the Mini-A there is a preamp, where the tube amp as a 50K attentuator at the input. In either case, switching parts didn't change the Dac's characteristics drastically. **Bob, yup, I always buy 2 boards to comparison and learning reason. Now I have 2 x V2.6 and 2 x V3 for comparison :)**

I don't own and haven't listened to any AN dac, so it is really an internet folkore. From what I read, AN dacs tend to be tuned to be lushed which suggest that AN probably adjust the lpf to produce such effect. The Subbu/JP Dac in stock form is on the neutral side which is probably very different from AN. Of course, the Dac chips can also play an important role as well - notice that the ES9023 is less than $5, and the whole Dac cost well below $100 (without chassis). Even though it doesn't meet the hoping outcome, it doesn't cost an arm and a leg, and more important, have some fun to build it !!

Have fun!!

PS, Albert, I am sure there are others willing to buy your dac from you if it doesn't work out.