Experience with this DIY DAC ?

Disabled Account
Joined 2007
$320 for that thing is insane....

Even the shipment at $60 is a rip-off.

Here's the Audio-GD NFB 5 with much better design, case, power supply, no OP's.... and the shipment is cheaper as well!

ºÍ§Ó*µ响
http://www.audio-gd.com/Pro/Headphoneamp/NFB5/NFB56.JPG

Boky
Shipping for the NFB3 seems to be similar @ $50.50usd? That's to Australia.

Granted, I do like the power supply and discrete output on the NFB3.

One strange thing I have found on my gigawork dac using transformer output, is that the RCA outs have negative and signal common :confused: some amps do not like this.
 
Last edited:
Shipping for the NFB3 seems to be similar @ $50.50usd? That's to Australia.

Granted, I do like the power supply and discrete output on the NFB3.

One strange thing I have found on my gigawork dac using transformer output, is that the RCA outs have negative and signal common :confused: some amps do not like this.

hmm can you explain this further? are you saying that you have continuity between the - and + terminals on the L/R RCA outputs?
 
op amps for better sound

any recommended Op-amp that can give more bass?
hey,:D
if you go to several manufactures they will send you samples of op amps to try,I tried 6 different ones and some recomended sounded dry in my system and others that where common were best so get a few and try'em . adding caps from vcc to gnd and vee to gnd and across the + and - power pins at the op amp, make a difference too I've used different kinds of caps around 10 mfd and a low value across the + and- pins,and if you want a seperate power supply to your audio section helps too.
I did remove all that and have edcors twin matched transformers running balanced out 600 to 600 First thing I noticed was tighter quicker bass,,,,,,The transformers where about the cost of 2 premium op amps........
Good luck to you!:)
 
Disabled Account
Joined 2007
It seems that using just the "+" connections from the chip degrades performance, even using transformers.
This makes sense when the differential dac is configured the way it is, in the "-" being a mirror of the "+" connections, so the signal travels via the "-" and the "+' in a push pull kind of way, if I'm understanding it correctly?
 
Disabled Account
Joined 2007
geez, I wish I could edit instead of doing a new post each time.

Anyway, changed back to the transformers in normal config' as it doesn't upset my valve amp, I'll just need to say no if someone with a SS amp asks to hear my dac on their amp, just so as not to make their amp squeal like a stuck pig :p
 
I have a rather strange problem.. well, at least to me.

If I use the on-board USB input I get sound, but one channel is 50/50 music and static. The other channel is fine. (Left Right CLK issues?)
With coax or toslink i get no sound, but the relay kicks inn, which should mean that the receiver is getting a valid signal.

The board is a bit shabby after trying lots of different mods, so im not surprised. I just wish I could figure out what the problem is.

Only thing I can think of, is clock error. The USB-input has its own clock, så maybe thats why it works, sort of.
 
On the DAC-chip VREF and VA measures 2.45v .. should be 5v. The VREF decoupling cap has lots of goo under it. I had to use force to remove it, and scrape of the brown layer. Might as well have been glue...

Its a 10uF 50V elna cap. What would be the best replacement? 10uF or 100uF tantalum?


I also found out the CS8412 is getting 4v ... With the chip out tho, I get 3.3v from the socket.

This dac is starting to **** me off.
 
It seems that using just the "+" connections from the chip degrades performance, even using transformers.
This makes sense when the differential dac is configured the way it is, in the "-" being a mirror of the "+" connections, so the signal travels via the "-" and the "+' in a push pull kind of way, if I'm understanding it correctly?

Think of balanced as single ended and forget about ground. Consider single ended (RCA) to be “+” and “-“ (at half the swing of XLR / balanced) and you'll be ok.

The secondary and the floating RCA's (“-" not connected to ground) are suspect to oscillations, in particular if the interconnect cables are long and of high capacitance.

Get the test CD with frequency sweep and at least 50MHz CRO and find the best components' values that give clean, flat audio frequency response AND no ringing. Once tuned, do not change interconnects.

Good luck,
Boky
 
The RCA negative at DAC end is referenced to return (PCB ground) at the amplifier end. Amplifier PCB ground is usually broad spectrum AC coupled (multiple parallel capacitors of 0.1uF and 0.01uF) to the amplifier case.

There should not be any connection at DAC end between RCA negative and anything else, only to the audio transformer secondary. The audio transformers' case should be DC coupled (bolted) to DAC metal case.

The mains cable ground may be AC coupled to the metal case; 10 ohms / 1W and 0.1uF in series to connect mains ground to metal case is good start. DAC PCB ground should be AC coupled at multiple points to the metal case; surface mount ceramic caps are excellent for this job, always use 0.1 and 0.01 uF in parallel, this requires adequate PCB layout to make implementation possible….

I use AD8066 single stage (balanced to single ended “stage”) at the output, whit special attention to ground topology, power supply rail decoupling (treat it as RF design) and R snubbers between IC output and RCA hot pin. This OP is outstanding if applied correctly. However, if it is just used as drop-in replacement for standard OP’s -> it will give mediocre results due to oscillations at very high frequencies (and its very revealing nature that requires full attention “upstream” of analog stage). It absolutely shines as I to V as well with 1704’s. Max Vcc / Vee is only + / - 12V DC.


Boky