Digital pot wiper settling time with RC load?

[treez]

Hello,
Page 7 of the MCP4013 digital pot datasheet says that the wiper settling time for the 2K1 Digital pot is 0.5us when the load is 100pF.....though what is the settling time if the load is a series RC with R = 10k, and C = 300nF?....presumably the 10K between the pot and the 300nF cap means the wiper can be moved at the fastest speed?

Also, supposing I had a 100n cap load, then if I move the wiper faster than 1 step per microsecond, then what happens, does the wiper just not respond to the up/down pulses, and the wiper just stays where it is?

https://ww1.microchip.com/downloads/en/DeviceDoc/21978c.pdf

 

[KlausST]

Hi,

ts is the time from the digital wiper change to the analog value change.

you may set your wiper with 1MHz up/dn frequency. But there is a delay ts to the analog output.

*****

If you use the external RC 10k and 300nF, this forms a low pass filter with delay of 3ms.
but also here you may change wiper with 1 MHz.

***

supposing I had a 100n cap load, then if I move the wiper faster than 1 step per microsecond, then what happens

Just the analog output is delayed...

Klaus

 

[treez]

thanks, but are you sure that's whats meant, I mean, customers would be expected to know about rc time constants, surely that needn't put that in the datasheet?
It calls it a "wiper settling time", which suggests to me there is a delay for the actual wiper movement...and if we move it faster than this, then the wiper will simply fail to move?

 

[betwixt]

No, it isn't a maximum speed for changing the wiper position, it is the delay while the internal amplifiers and their internal capacitance can adjust the output to the new level. They cannot respond instantly to a step change in input, the specified delay is how long it takes to adapt to a new output while driving a capacitive load. The output stage may have to sink or source additional current to reach the newly selected voltage so it's output impedance in conjunction with the existing charge at it's output pin will determine the speed.

Brian.

 

[treez]

so no matter what capacitance I have on the output, I can still pulse it at 1mhz and it will catch up eventually?

 

[KlausST]

Hi,

Yes.

********
I think the given 100pF capacitance may be the capacitance of the used scope jig while they measured this value.
Calculating back you may find an internal (real or not) capacitance in the same magnitude of 100pF.

Klaus

Btw. I also find the naming misleading. I'd rather call it "digital to analog delay".