In the brochure "Understanding and ApplyingDigital Potentiometers" written by

Chris Hyde and Dave Lunsetter, I found a good overview of different digipots

in different applications published by Analog Devices.

3 wire links from a AD5290 to a microcontroller proto board observe me, that the

AD5290 obvious can not reach absolute zero and also can not reach absolute max.

values if used as three terminal potentiometer (as voltage divider).

My personal selection of AD5290 is based on a must of +-15V (AC/DC) range and 8bit

resolution and avail. in several kOhms ranges.

Is there an other architecture avalilable from AD with virtual absolute connection

of W to A and W to B, when applied as potentiometer to get a FULL range potentiometer?

It is fully clear to me, that a minimal resistor value of some ten ohms is left if used

as rheostat, as potmeter not relevant.

Any hints?

Koebi

Hello Koebi,

I have used good old analog pots and had the same problem where there was a little residual resistance at the ends of the pots.

I have also used this exact pot in several designs and had worked into the design some compensation for this residual resistance. If you are using the 10K value for this part then for an 8-bit pot the smallest increment is 39 ohms. This does make for a high error at the ends if you add 10 -ohms to this. If you are using the 50K or 100K values then this error becomes much smaller. The impedance of the circuit you are driving will also play an important role in this.

I will move this discussion over to the precision DAC community which is where this part is supported. There one of the applications engineers for this part should be able to help you. If you do not get a response soon then the best way to get their attention is to re-post over in the

Precision DACs part of this forum.

Thanks,

Dave T