dac I/V convertion with very low distortion

[smms73]

Originally Posted by Ken Newton

Okay, today I took the time to enter the circuit in my simulator. I must say, I'm impressed. I randomly substituted all manner of various transistors throughout the circuit, and found the THD indeed remains in the vicinty of -140dB, although the specific profile of hamonics vary. This is highly unusal behavior for an open-loop bipolar circuit in my experience. Well done.

thanks Ken. what is the simulator that you use?

[smms73]

Originally Posted by Calvin

THD seems really low, but input impedance rises above 1MHz with some weird up-and downs, especially around 66MHz, where seemingly a resonance builds up.

I have been analysing this, and the conclusions is
the rising of the impedance is due to the lost of gain of the input transistors as the frequency goes up.
The dips and ups around 66Mhz are due to the interaction of the dac output capacitance (22pf) with the input.

The best soluction is to use a snubber (zobel) in the input.

In the pictures below is the output and input gain and phase with the snubber and without it.

Its very important to use the snubber in the circuit.

[dirkwright]

Originally Posted by Ken Newton

The dynamic signal current swing from EACH output pin is 3.9mA peak single-ended. Therfore, the peak-peak signal current swing from EACH output pin is 7.8mA single-ended. The DIFFERENTIAL output signal net current change for a given channel then is 7.8mA peak, and 15.6mA peak-peak.

I'm sorry this does not make sense, particularly in light of the chart in post #155.

[Ken Newton]

Originally Posted by dirkwright

I'm sorry this does not make sense, particularly in light of the chart in post #155.

Let's walk this through. The table in that chart you refer to shows two anti-phase output signals, IoutN and IoutP. You can read the peak-peak current magnitude for either output phase directly from the table, which is 10.1mA minus 2.3mA = 7.8mA. The signal current swing is centered around a BPZ (Bipolar Zero) bias current of 6.2mA. So, 10.1mA minus 6.2mA = 3.9mA peak, from BPZ, and then 6.2mA minus 2.3mA = 3.9mA for the opposite peak, from BPZ - from either output pin. The resulting 7.8mA pk-pk signal current swing is abailable from BOTH IoutP and IoutN. Now, since the two output signal currents move in anti-phase (they are differential), the differential peak-peak net signal current change is 2 x 7.8mA = 15.6mA.

[smms73]

Is time to present to you guys, the almost finished differential version that i resolve to call TAGUS.
It has a new current mirror that is also a voltage mirror.

The distortion is measure with a 500ohm/22pf input
and is 0.000130% at 7.8ma p-p at each input.
and only 0.000055% at 3.9ma p-p.

Below are the schema and the fft for 7.8ma and 3.9ma

I will talk about it later because today going to be a busy day, hope you enjoy.

[smms73]

Here are the I/V of the texas datasheet and the FFT just for comparing.

[Joachim Gerhard]

Cool, i like that mirror.

[smms73]

I am thinking in building this:
Modular audio DAC
and change only the I/V convertor, i like the modular idea.

What you guys think, if you have or know were to find a better pcb for a dir9001 and pcm1794a please let me know. I really dont have the time to design my one, thanks in advance.

[smms73]

Originally Posted by Joachim Gerhard

Cool, i like that mirror.

have you ever seen one like this?

[CeeVee]

Originally Posted by smms73

I am thinking in building this:
Modular audio DAC
and change only the I/V convertor, i like the modular idea.

What you guys think, if you have or know were to find a better pcb for a dir9001 and pcm1794a please let me know. I really dont have the time to design my one, thanks in advance.

Are you aware that DIR 9001 only does 96k ?

The CS 8416 module does 192k.

I'm thinking you want to do the PCM1794A justice, so 192k!?
...and be prepared for a huge hunt for the TORX thingies....very scarce .

CV