Very interesting looking design, thanks for sharing. I take it its NOS and the clump of logic in the middle creates the digital inversion. Why did you opt for separate common-mode chokes on the secondary sides of the transformers rather than one choke on the primary side? Also what's the I/V stage - those discrete TO220 transistors?
Yes, as its playing now its run in NOS but with passive sinx/x. Its much easier to dampen the Q of the filters when they can be loaded with large electrolytics.
The I/V stage is pretty much that of the "D1" Only I've changed it to draw current from the power supplies more equally and not load the sinx/x filters excessively at low volume settings.
Its similar to the i2s splitter posted by pedja rogic some years back but since I had a CPLD implementation in mind first Its implemented in another way using less flip-flops.
The rest is a clock divider and a couple of XOR gate providing a truly differential signal for the DEM pins.
No, its my PCB.
I am about to start on a dual differential 1541a dac.
I am buffering and inverting the data to one dac and using a bifilar wound dual primary I/V transformer.
The plan is to use 3 pairs (in series) of 74hct14 hex inverters to buffer all three I2S lines borrowed from the inputs of the 7220. One dac will receive its data after only one inversion.
The transformer is here:Sowter Type 9545 DAC INTERFACE TRANSFORMER
For the time being the transformer supplies the valve output stage of an Audio Note DAC1 (I'll also try the transformers on this as well)
I
I am buffering and inverting the data to one dac and using a bifilar wound dual primary I/V transformer.
The plan is to use 3 pairs (in series) of 74hct14 hex inverters to buffer all three I2S lines borrowed from the inputs of the 7220. One dac will receive its data after only one inversion.
The transformer is here:Sowter Type 9545 DAC INTERFACE TRANSFORMER
For the time being the transformer supplies the valve output stage of an Audio Note DAC1 (I'll also try the transformers on this as well)
I
Yes, I was thinking that way might be worth trying. I might try both ways and see if I can hear the difference although I can't quite see why splitting them as you suggest gives a benefit.
Perhaps you could explain in more detail.
Doing it with one dac for each channel simplifies layout issues pertaining to making the signals equal and opposite and it allows one to resolve current loops locally and/or make opposite currents sum to zero which also helps minimizing unwanted potential differences from currents flowing where they should not.
Also if you look at a -60db 1Khz spectra of a 1541 you will notice that one channel usually measures worse and that this is usually tied to the same channel even between different dacs. Running one dac with inverted data would compound this error.
And there is also that that any time difference between when the different dacs latch out a new sample would translate into a level error at the output.
Also if you look at a -60db 1Khz spectra of a 1541 you will notice that one channel usually measures worse and that this is usually tied to the same channel even between different dacs. Running one dac with inverted data would compound this error.
And there is also that that any time difference between when the different dacs latch out a new sample would translate into a level error at the output.
Last edited:
It cancels out even order harmonics the 1541a benefits greatly from this. With a differential implementation the -60db 1Khz spectra really looks very very good, even compared to the previous 1541a based dac I had which was not bad.
Listening with a good 1541a and the gain turned up that -60db signal still sounds like pure sine.
Listening with a good 1541a and the gain turned up that -60db signal still sounds like pure sine.
The Audio Note Dacs* split the data into L & R for the AD1865 dac which doesn't have I2S input, using a 74xx02 and they aren't known for having jitter problems.
(*nos design - Cs8414 input )
In my case, the Sowter I/V transformer is designed for differential input so there is not really any extra complexity other than a 74hct02 and the 2nd 1541.
A simplistic view would be (using a 1541 per channel as Tazzz recommends) and considering one channel:
Inputs: L & -L (-L = L data inverted)
Ouputs L + noise and -L + noise.
Subtract in a bifilar wound I/V transformer (swap start and end of one winding to achieve this) results in:
L- -L = 2L and because the noise was added within the dac,
noise - noise = no noise (well, less noise!)
On a separate subject:
What are the consequences of inserting an LC filter with an f-3 of about 15-16kHz* between the 1541 outputs and I/V transformer?
As the impedance needs to kept very low, use of an inductor should be practicable. (Sowter suggest a 200R load resistor on the dac output)
*my ancient ears probably don't work above 12-15KHz!
I
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.