• The Vendor's Bazaar forum is for commercial offers and transactions. Only unmoderated members can post here.

    diyAudio provides this forum for the convenience of our members, but makes no warranty nor assumes any responsibility. We do not vet any members. Use of this facility is at your own risk. Customers can post any issues in those threads as long as it is done in a civil manner. All diyAudio rules about conduct apply and will be enforced.

Output transformers for DACs

Dear Ivan, have you tried any of your transformers with DDAC1794? I have that DAC with 4 boards and it might be one of the options.

Unfortunately some guys who purchased my trafos and wanted to try them with DDDAC didn't give any response yet. I do not see anything extraordinary and simply know that it will give for DDDAC "a new life":sing:

Nothing extraordinary in terms of my transformers usage possibility I do not see, but there is someting strange in the described by you connection measurements and output voltage. If the i/v resistor is 34 Ohm for all the boards, that means you should have (7.8mApp*4)/2*1.41=~11mArms. 0.011*34=0.37Vrms. Then to get 2.3Vrms after your ISKRAs the turns ratio should be 2.3/0.37=~7 (taking trafos losses in to account), not 1:1 (600:600). Can you make some bigger photos of your DDDAC setup to understand what we dealing with?
As all I can understand from previous photos is only that you have "something" that looks like a transfomers, very small resolution of pics.
:cheers:
 
Last edited:
Each of the differential outputs is connected to the GND with a 34 ohm resistor.

I was not attentive. Seems there are 8x1794A chips in total (2 pcs on every PCB), right? Every 1794A is in mono mode, right?

OK. Then calculations again: single chip gives us: 7.8mApp*2paralleled_internally_channels=15.6mApp=5.5mArms
4 boards gives us 5.5*4=22mArms.

So one (left or right) output from all the 4 boards will gives us: 68*0.022=1.5Vrms.
Hmm... Is your Iskra trafos are 1:1.6?

I think, I need to take a look at the description of DDDAC...

p.s. But anyway, nothing extraordinary for my trafos.
 
I have take a look at the DDDAC description, and yes it is like I thought. Dual Mono mode per one board means 2x1794A on every board. OK.

One only point - you will need to use trafos with the CT at GND, as the better result will be given if to move all that small i/v resistors and to place their increased (with less resistors material influences as well) values after the trafos to allow the trafos to form the sound by themselves in a greater extent than inside the resistors. I mean it is more correct way, than to make i/v conversion before transformers.
 
This DAC is specific, the quiescent current is 80mA (4boards, 8DAC chips in mono mode). The DC output voltage at each output (pos & neg) is + 2.72V + -2mV (0.08Ax34ohm). The transformer is not used for I/U conversion but only for eliminating DC voltage (common mode). The transformer is connected between the pos and neg outputs. The DC offset is very smal, about 2mV. The DAC outputs about 2.4VRMS balanced and 1.2VRMS unbalanced. When I add a 1:1 transformer (ISKRA and others), then the output is around 2.3VRMS balanced and unbalanced. Some use the Sowter transformer 1:2 to raise the output voltage, but I don't need it, it's even undesirable.


It may be best to wait for a response from those who have tried DDDAC with your transformers. Anyway, thank you Ivan for the answer.
 
Last edited:
The transformer is not used for I/U conversion but only for eliminating DC voltage (common mode).

I have doubts, as seems the DC eliminating relates to I/U resistors, not to transformer. I mean the values matching of i/v resistors will help to self-compensate. What if to lift up resistors from GND?🙂

It may be best to wait for a response from those who have tried DDDAC with your transformers. Anyway, thank you Ivan for the answer.

Of course! It is your DIY project:cheers:
 
I'm not sure you can raise the resistors from the GND. I don't think PCM1794 will work that way? If the transformer were to be used for I/U it would need CT. But the DC current is 80mA without signal, that is the problem.
I paired these I/U resistors so that all my DC output voltages were practically the same, so I could connect the transformers without saturating the core. One channel is 2mV difference, the other less than 1mV. So I need a 1:1 transformer, I tried various ones, 600:600, 2k4:2k4, 10k:10k ... They all sound good.
I'm currently on 2k4:2k4 transformers taken out of some measuring device, the cores are permalloy extended E sheets. I will use them until I find something better. Input impedance of the preamp is 100klog pot.
 
Last edited:
This is OK (2ohm). I have better sound with the preamp (tube buffer Aikido Cathode Follower, no gain - no pain). The amplifier is quite strong, about 2x140W RMS / 8ohm, I am safer with the preamp than with directly connected DAC to the amplifier.
 
Last edited:
I don't know if it's a good idea. Maybe it's better to first ask those who have DDDAC and have taken your transformers what the results are. I have already tried Cinemag CMLI 600: 600 (I was fixing a friend DDDAC similar to mine). They are expensive, the measurements are great, and they don't sound good. What is the price for a couple of your transformers and how much is the postage for Serbia (Belgrade)?
 
Thanks.
$ 170 for a couple of top transformers is not much, I only need 2 coils 600: 600. I'll let you know soon what I've decided. I have a technical question, what is the inductance of the primary / secondary? Or, what did others order for DDDAC?

I also asked a question about your transformers on the DDDAC forum, maybe someone will answer soon. Greeting from Belgrade.
 
I have a counter question: what inductance do you need? Because all of these abbreviations, such as 600:600, 1k:1k, 2.4k:2.4k - whoever how wants and sees, those writes so.
For those types of transformers that you have mentioned, the inductance on the midrange (1kHz for example), unlike the linear inductance in my transformers, is several times lower than for the low frequency (core material properties).

I can make any version from 1 to 60H (100Hz/1kHz/7.8kHz). DCRs will be different.
Regarding trafos, which I have sent for DDDACs there was step-up versions only and with two primary coils in accordance with what I have wrote in previous messages.

If you want to try 1:1, then I would recommend to try 1+1:2, it can be configured as 1:1, but will give a possibility for trying to move i/v resistors from primaries to secondaries. Just a suggestion.
 
I don't know what is optimal. I tried everything, from 1.5H / 20ohm to 16H / 750ohm. I need to think a little more. That it can be used as the constructor predicted, and that it fits 4-8 pcs. PCM1794 in parallel for both channels. Data for the Sowter transformer are known. I need some time to think and look at what others have used.

Basics of the Sowter transformer
Ratio: 2.000
Turns: Prim 240 / Sec 480
Core: Gapped Mumetal, Package: Mumetal Can, Termination: Color coded leads
GENLOAD @20 Hz Sec. SC: 10 ohms Max
GENLOAD @20 Hz Sec. OC: 429 ohms Min
LEVEL for THD=0.5% @ 20 Hz: 10.2 dBu Min
Primary inductance (H): 3.4 H Min
Primary resistance: 1.6 ohms Max
Secondary resistance: 35 ohms Max


Sowter is not recommended for less than 8pcs. PCM1794 (4 DAC boards), means for 4 pcs. (2 DAC boards) should probably have twice the inductance. Unlike Sowter, I want a 1:1 ratio of primary and secondary.
 
Last edited:
I agree. So I would like to take someone's proven solution with the same device as mine. Even the measurement doesn't interest me much (and I have some equipment), just the end sound result. Besides, I'm a mechanical engineer, electronics is just my hobby. I don't have enough knowledge to tell you exactly what kind of transformer I need.

The transformer I am currently using has 132ohm DCR and 15.7H. Core is made of Permalloy extended E sheets. It shouldn't sound good, but it does.
 
Last edited:
New V6 Ian's DAC I/V

Two years ago I bought version 1 of the I/V converter produced by Ivan for the Ian ES9038Q2M Dual mono DAC. Up to now I have used this output stage with great satisfaction.

I recently asked Ivan to send me version 6 of the I/V converter with detachable transformers.

After about 2 weeks of listening I can say that version 6 is much more transparent than the previous one.

This has resulted in significant improvements on many parameters:
the high range is even more refined while the low is even more controlled and deep. The scene has solidified further, the outlines of the instruments are even sharper, the voices are more natural. Even the timbre seems more coherent to me.

It is certainly by far the best sound I have heard in my system.

Excellent work Ivan, congratulations!