Buffalo II & transformers

[Joe Rasmussen]

So there is no doubt, I have amended your diagram, so it should look like this:

[rolls]

Originally Posted by Joe Rasmussen

Thanks.
I suspect that using a step-up transformer with 1K on the Primary is not so good because the source Z of the DAC is way too high into the Primary.
Cheers, Joe

Who is the blame? I am using 1.2 Ohm, probably you misread my post??!!

[rolls]

Originally Posted by Joe Rasmussen

Yes, in my opinion that is what will get you the max results. It does two things, converts DC voltage offset to current offset, and creates the lowest impedance possible (our topic is Txs) into the Primary.
Cheers, Joe

There must be something in this grounding schema, that prevents people from doing/believing it, so they still ask about it.
I have thought about and would explain it this way:
Without grounding the mid point, it is for me like a class B amplifier. There is no current, when there is no sound. More loudness means more current, so the dac bounces between more voltage mode (loudness low) and more current mode (loudness high).
When grounding the midpoint, we have a class a situation, there is always a current flow.
I want to stress again, that the dac in current and hopefully passive mode sounds extremly well and can only be compared to the very best, i.e. Meitner. Best power supply is mandatory, I use LipoFe/NimH cells without any regs.

[Ciu]

Originally Posted by rolls

There must be something in this grounding schema, that prevents people from doing/believing it, so they still ask about it.

When grounding the midpoint, we have a class a situation, there is always a current flow.
.

Hello Rolls !

If there is always current in primary, can use any transformer without gap ?
Untill now, I was using Jensen JT-11-emcf without grounding , with 6 ohms I/V from +- ess9018 output and tube stage 50x gain.
Thanks

[wktk_smile]

rolls means current flow that appears on DAC-resister-GND (not on primary)
Right?

[rolls]

Originally Posted by wktk_smile

rolls means current flow that appears on DAC-resister-GND (not on primary)
Right?

Yes, current can just flow to ground, because there is the same tiny voltage on either side. Very thin wire is the best I/V resistor.

[chaos]

i also want to thank mr. rasmussen! i listened for some months with a s&b tx103, 1:10, 2x 25R to ground. sounded very nice, but last week i changed to 1:5, 2x 1R to ground followed by a choke loaded 5842(130v, 15ma), choke input psu, superreg and i could not believe it. just fantastic:-))))))
will change the resistors to tx2575, cap to duelund.....nothing will beat this;-)

[Joe Rasmussen]

Originally Posted by rolls

There must be something in this grounding schema, that prevents people from doing/believing it, so they still ask about it.

Maybe because this is a trick you can only do with Sabre DAC where you have a DAC with a 'voltage' offset. Try that with any of the other 'voltage' DACs or even 'current' DACs, and you will likely cause near instant damage.
You can do it with 'current' DACs like Burr-Brown 1792,1794,1796 DACs, as these have only a current offset (not a problem to deal with) and no voltage offset (much more tricky) and can be brought straight to Ground anyway.

Originally Posted by rolls

When grounding the midpoint, we have a class a situation, there is always a current flow.

When you have an output device, whether Fet or BiPolar, if you increase the current or 'bias' you also get higher transconductance and that should improve linearity. This is also a fact in Class A circuits, whether gain stages or buffers. So in this case it affects an internal change within the DAC by pulling out external current (again trading offset voltage to create offset current). But I believe there are a number of factors here, but this is an important one IMO.

Cheers, Joe

[Joe Rasmussen]

Originally Posted by chaos

i also want to thank mr. rasmussen! i listened for some months with a s&b tx103, 1:10, 2x 25R to ground. sounded very nice, but last week i changed to 1:5, 2x 1R to ground... and i could not believe it. just fantastic:-))))))

I have been preaching this for near two years, and In think the penny is starting to sink in. I would not want to run the ES9018 any other way, it's needs to be pulled to a literal ground rather than a virtual one using the lowest value possible, as low as you can get away with.

Cheers, Joe

[chaos]

Originally Posted by Joe Rasmussen

.....using the lowest value possible, as low as you can get away with.

Cheers, Joe

so you think it would make sense to use the transformer in 1:20 and just take a wirewrap wire which would give me around 0,25R with a lenght of about 2,5"?
at least the transformer was made for voltages like this

cheers,klaus