Go Down

Topic: Active Lowpass Filter for PWM to DC to control a Photomultiplier (Read 2412 times) previous topic - next topic


Dear all,

I need a lowpass filter to turn an Arduino PWM output into a proper DC with MINIMUM ripple and drift.

I have to control a photomultiplier-tube (PMT) via a 0-5V DC control voltage that generates the high voltage (HV) needed to drive the PMT.
As the HV follows the 0-5V DC control voltage (CV) according to the equation HV = CV * 250 any fluctuations are scaled up by a factor of 250 and hence are introducing noise and variability in the PMT output.
As the voltage - once set - is not changed frequently, settling time of the DC/filter is uncritical (maybe in the order of 0.5 sec or so).

I was considering an active 2-3 pole filter employing a LM358 OpAmp (I do not have symmetrical power available in the present circuitry hence this consideration).

Has anyone done this already and can help me with a schematic or other hints, dos, donts?

Many thanks


How about using a digital pot or a digital to analog converter.



Thanks Mark, I have no experience with a digital poti, can you pls expand a bit on this?



Feb 15, 2013, 02:41 pm Last Edit: Feb 15, 2013, 02:44 pm by 0AlphaOmega Reason: 1
Does this solution help?

This IIC version of the pot may be nicer, with a 10bit resolution
For whom does the clock pulse? It pulses for you!


Thanks, these pot solutions definitely look very interesting, I'll have a closer look.



If you want to be able to design good active (op-amp) filters, download TI's Filter Pro software.  It's great software and it's free. 

You kinda need to be more specific on ripple numbers as to what is acceptable.  Is 1mV too much or is 1uV too much?
Experience, it's what you get when you were expecting something else.


Why bother with the op amps and passive components you need to make a good low-pass filter, when a DAC is much simpler? For example, http://uk.farnell.com/microchip/mcp4921-e-p/12bit-dac-singl-spi-i-f-mcp4921/dp/1332112 if you want a through=hole part.
Formal verification of safety-critical software, software development, and electronic design and prototyping. See http://www.eschertech.com. Please do not ask for unpaid help via PM, use the forum.


Thanks afremont and dc42,

I would like to keep the ripple below 1 mV, as said, settling time is not critical, so I think Rs in the range of 100k and Cs at about 0,47 to 1 µF should work, I'll check out the filter calculator!

I don't know how to add a DAC to the arduino, so I think I go for an OpAmp active 2-pole lopass.
Again, many thanks.



I don't know how to add a DAC to the arduino, so I think I go for an OpAmp active 2-pole lopass.

"Who is like unto the beast? who is able to make war with him?"
When all else fails, check your wiring!

Go Up