Experience with this DIY DAC ?

Error in LM337 routing

For those of you using this DAC with opamp output i have to say that i found an error in my board with the LM337 routing; ADJ, Vin and Vout was routed in a wrong way.
The attached images show the modified (and corrected) board and the pinout of LM337.
I also desoldered the two trimmers and two resistors (one R and one trim for LM317 and one R and one trim for LM337 rispectively), used to set the +/- Vanalog for the opamp, from top and soldered on bottom (using fixed R in place of variable).
Another mod visible in the photo are the capacitors soldered in parallel with the resitor going from ADJ pin to ground to improve ripple rejection as stated on the LM317/LM337 datasheet.

Other important hints for using the DAC with opamp are in posts #212, #229, #233, #241, #258 and nearby posts. The argument of this posts is the optimal values for the LPF opamp, simulations and measurements for the original (not so good) and proposed (way better) values.

Ciao
Andrea
 

Attachments

  • back.jpg
    back.jpg
    327.3 KB · Views: 590
  • LM337.jpg
    LM337.jpg
    14.4 KB · Views: 576
Jensen JT-11P1 vs OPA627

Hi all, I need your expert advice. My transformer mod using the Jensen JT-11P-1 performs poorly, compared to my single OPA627 op-amp mod on the same dac, in the bass area. The secondaries are with 13k resistors and 620pf capacitors in parallel. For the primaries, I tried unfiltered,with 220r and 510r resistors in series, with and without 10nf capacitors in parallel but still get the same poor bass response. I tried 10nf because thats what i have in hand. I just ordered some wima 1nf and 2.2nf capacitors and would only get them after a week.

Attached is the frequency response chart between the 2 modifications. Any suggestions for improvement would be greatly appreciated.

Thanks,
Sonny

OPA627op-ampvsJensenJT-11p-1.jpg
 
Hi Bill,

Here is the actual photograph of my dac with the transformers installed. The photo shows a direct link from the dac to the primaries as this setting was recommended by Jensen. The secondaries are filtered by 12k+1k resistors and 620pf silver mica capacitors. The white and black wires are connected to the orange wire going to the negative out, and the yellow wire to the positive. I also tried placing 510r series and 220r series on the primaries but get the same results. I haven't tried placing 1 nf or 2.2 nf capacitors accross the primaries yet. The other modification i did was to bypass the second op-amp. I still get the same frequency response even by removing the 12 volt source. Could this be the typical response from the transformers? :confused:

Thank you...


_MG_1849-1.jpg
 
Here's the diagram that I followed. I also inquired and got an instruction from Jensen to remove any load and connect the output from the dac chip directly to the transformer primary. But i also tried with 510r in series and 10nf across the primary since I only have 10nf in hand. Im still waiting for the 1nf and 2.2nf capacitors to arrive for me to try.

Connection.jpg
 
Is it just my eyes, or do you still have the op-amp circuit attached? (IE still have wire links across the caps near the DAC) That means that you still have an RC circuit at the DAC output. This shouldn't effect your low frequency rolloff though.

I didn't remove the wire links for me to compare the op-amp with the transformer. And the op-amp seems to be the winner.
 
That's odd. The only Jensen transformers I've used were with my low output moving coil cartridge back before I dumped my turntable, and they were outstanding. I've got a set of Lundahl transformers and they produce fantastic bass. I do find it odd that your transformers are down 2dB or so in the bass region. Do you have some sort of odd input that the DAC plugs into?

The Jensen transformers are interesting to me. If you are thinking of dumping them, I might be interested in buying them if the price was right.

Sheldon
 
I'm using the Jensen transformers and I didn't notice a drop in the bass though I didn't measure it. On the primary side I have 249 ohms in series and a 2.2nF cap across the inputs. My secondary is the same as yours.

I do remember reading somewhere that the dac likes a small load on its outputs, if that makes any sense.

To me the transformers were a significant improvement over the op-amp.
 
Hi Ryssen, for 3.3v reg, you have changed both IRFP9240 to MJE350 (or other transistors that Salas suggested), or just the one on the output? And if you have photo of this modified reg, would you please post it?

Thank You.

Here is a picture:

33v-Salas.jpg

Observe that the MJE350 got one leg around another.One Led is gone and I put a 100 ohms pot there instead.

The DIR9001 installed:

DIR9001.jpg


Haven´t listened so much yet,but it don´t sound worse anyway.
 
Hi all, I need your expert advice. My transformer mod using the Jensen JT-11P-1 performs poorly, compared to my single OPA627 op-amp mod on the same dac, in the bass area. The secondaries are with 13k resistors and 620pf capacitors in parallel. For the primaries, I tried unfiltered,with 220r and 510r resistors in series, with and without 10nf capacitors in parallel but still get the same poor bass response. I tried 10nf because thats what i have in hand. I just ordered some wima 1nf and 2.2nf capacitors and would only get them after a week.

Attached is the frequency response chart between the 2 modifications. Any suggestions for improvement would be greatly appreciated.

Thanks,
Sonny

OPA627op-ampvsJensenJT-11p-1.jpg

Did you take the measurements without any load connected? I would try a 20k load resistor. Everything I have read about high impedance trafos says you need a load.

I haven't heard anything about lack of bass with any of the trafos tried. Still, you do have the opamp section in the circuit. Who knows what that could be doing to the signal with the feedback circuits and such. Try snipping your links at the cap positions, you can always reconnect them with a drop of solder.