Building AD1865 DAC

[Ken Newton]

I am thinking if I can parallel multiple DAC chips together, I am able to increase the current output and hence allow me to use resistor with smaller value. Don't know if it works or not.

I understand what you are thinking. Paralleling the current outputs of multiple DACs increases the signal current magnitude, thus allowing use of a lower value I/V resistor - but all that does is to lower the output impedance of the passive I/V circuit. The signal voltage amplitude, however, remains the same. The amplitude does NOT increase. This is because the increase in signal current is exactly countered by the decrease in I/V resistance. By Ohms Law, the resulting signal voltage remains the same.

 

[flowerpot]

Parallel as many DACs as you can and use the smallest resistor I/V possible... the smaller the better....use a transformer as part of the I/V and add a tube output stage to bump up the signal...

 

[edwtsui]

I understand what you are thinking. Paralleling the current outputs of multiple DACs increases the signal current magnitude, thus allowing use of a lower value I/V resistor - but all that does is to lower the output impedance of the passive I/V circuit. The signal voltage amplitude, however, remains the same. The amplitude does NOT increase. This is because the increase in signal current is exactly countered by the decrease in I/V resistance. By Ohms Law, the resulting signal voltage remains the same.

HI Ken,

I just checked from datasheet, there is an internal 3k ohm feedback resistor Rf, it will give a 3V output for 1mA current, does it mean that the compliance voltage is at least 3V? If so, I have no concern to use large value of resistor.

 

[edwtsui]

You can certainly use a smaller valued resistor when you parallel DACs - but didn't you want a larger output amplitude? If so then as @Ken just said you're constrained by the DAC's compliance voltage. A common-base connected transistor will help you get beyond that.

That's another good solution, thanks.

 

[Ken Newton]

HI Ken,

I just checked from datasheet, there is an internal 3k ohm feedback resistor Rf, it will give a 3V output for 1mA current, does it mean that the compliance voltage is at least 3V? If so, I have no concern to use large value of resistor.

That internal 3k feedback resistor is intended for use with the AD1865's built in op-amps, forming an active I/V stage. The 3V peak output then comes about from the 1mA peak signal current output from the AD1865, routed through the 3k internal resistors, thereby giving a 3V peak output from the active I/V op-amps. It has nothing to do with the voltage output compliance of the AD1865's current pins. Rather than active I/V, you may instead utilize a simple external passive resistor for I/V. The passive resistor approach is what we've been discussing.

 

[edwtsui]

That internal 3k feedback resistor is intended for use with the AD1865's built in op-amps, forming an active I/V stage. The 3V peak output then comes about from the 1mA peak signal current output from the AD1865, routed through the 3k internal resistors, thereby giving a 3V peak output from the active I/V op-amps. It has nothing to do with the voltage output compliance of the AD1865's current pins. Rather than active I/V, you may instead utilize a simple external passive resistor for I/V. The passive resistor approach is what we've been discussing.

OK, thanks. I will do a research on how large the resistor value can be without causing a serious distortion.

 

[Ken Newton]

OK, thanks. I will do a research on how large the resistor value can be without causing a serious distortion.

I suggest that you begin by determining the target maximum signal voltage you require. Let's assume, for the sake of illustration, it is the industry norm of 2VRMS. If you are planning to directly connect the voltage output of the passive I/V circuit to your power amp, then the I/V resistor will need be large enough to produce that amplitude.

So: 2.83V peak / 1mA peak = 2.83K ohms. Therefore, begin your evaluations with an passive l/V resistor of near that value. If the distortion is audible (which, I suspect it will be), then you will need to use a lower value resistor, along with some form of voltage amplification stage somewhere between your DAC and your power amp.

 

[edwtsui]

You can certainly use a smaller valued resistor when you parallel DACs - but didn't you want a larger output amplitude? If so then as @Ken just said you're constrained by the DAC's compliance voltage. A common-base connected transistor will help you get beyond that.

Hi abraxalito,

Could you share the schematic of the common base output stage for my reference? Thanks

 

[abraxalito]

You can find one example schematic here : Please help: nOOb building a 9028 DAC