I am looking for good books describing DAC I/V designs, specifically for dual mono DACs. If you have any suggestions, please share! I prefer books that are available as e-books.
Jan Didden has an interesting version you could experiment with/elaborate: here. It uses a matched-pair PNP. The particular one he used may not be available now but there still are a few dual PNP devices out there.
The generic term for an I/V converter is a transimpedance amplifier (A.K.A. TIA). There is quite a bit of information out there regarding them, most-frequently in the context of photodetectors -- but issues like frequency response and stability should be useful w/regard to any application. IIRC Texas Instruments has some good info on design tradeoffs w/regard to TIAs.
One of the tricky aspects of TIAs is that their stability depends on the capacitance presented to their summing node. If purely-resistive feedback is used the capacitance introduces a phase shift -- if too much, the TIA can oscillate, worst-case. If it doesn't oscillate it will likely have a peaky FR. The way to solve this is to add a capacitor across the feedback resistor, but if it's too large you get the opposite problem -- gain falloff at high frequencies.
The DAC output WILL have some capacitance associated with it, so this not a theoretical issue.
The generic term for an I/V converter is a transimpedance amplifier (A.K.A. TIA). There is quite a bit of information out there regarding them, most-frequently in the context of photodetectors -- but issues like frequency response and stability should be useful w/regard to any application. IIRC Texas Instruments has some good info on design tradeoffs w/regard to TIAs.
One of the tricky aspects of TIAs is that their stability depends on the capacitance presented to their summing node. If purely-resistive feedback is used the capacitance introduces a phase shift -- if too much, the TIA can oscillate, worst-case. If it doesn't oscillate it will likely have a peaky FR. The way to solve this is to add a capacitor across the feedback resistor, but if it's too large you get the opposite problem -- gain falloff at high frequencies.
The DAC output WILL have some capacitance associated with it, so this not a theoretical issue.
Not sure if there are any books on I/V designs for dacs. A lot depends on the particular dac output, its impedance, its RF content, its peak current, if there is a DC offset, its voltage linearity, and perhaps other factors. It may also depend to some extent on what the dac's intended use will be. Measurement dacs may require one type of optimization, whereas dacs intended for human listening enjoyment may in some cases benefit from some type of human factors optimization.
Also, I/V feedback capacitors can be used to help facilitates dac output filtering requirements. Often they are larger value than needed for stability. Sometimes a slightly more complex feedback network may used if stability becomes becomes an issue.
That said, I would agree the literature on photodetector I/V is a good starting point.
EDIT: It may be that questions on I/V design for dacs will get more responses if posted in the "digital line level" sub-forum. That's where most of the dac guys tend to hang out.
Also, I/V feedback capacitors can be used to help facilitates dac output filtering requirements. Often they are larger value than needed for stability. Sometimes a slightly more complex feedback network may used if stability becomes becomes an issue.
That said, I would agree the literature on photodetector I/V is a good starting point.
EDIT: It may be that questions on I/V design for dacs will get more responses if posted in the "digital line level" sub-forum. That's where most of the dac guys tend to hang out.
Last edited: