Opamp Active Low Pass Filter Weirdness

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Curtis

Well-known member
Joined
Aug 24, 2006
Messages
305
Location
Australia
Chaps,

I've been tinkering with a simple opamp-based 2 pole Salen-Key low pass filter, like this one.  Ra and Rb are 10K, Ca is 10n ceramic, Cb is 1n5 MKT (just what I had handy in the parts drawer).  I'm using an LM833 as the opamp.

The problem I have is that I'm seeing a rising DC voltage at the output, which starts at 0V at switch-on and rises all the way up to the positive supply rail over the course of a few minutes.  I'm assuming what I'm seeing is one or both of the capacitors being gradually charged up (input bias current maybe?).  Is there a simple way to prevent this charging from occurring?  I did a quick Google and couldn't find anything that described the effect I'm seeing.

Thanks.
 
Curtis said:
I've been tinkering with a simple opamp-based 2 pole Salen-Key low pass filter, like this one.  Ra and Rb are 10K, Ca is 10n ceramic, Cb is 1n5 MKT (just what I had handy in the parts drawer).  I'm using an LM833 as the opamp.

The problem I have is that I'm seeing a rising DC voltage at the output, which starts at 0V at switch-on and rises all the way up to the positive supply rail over the course of a few minutes. 

It is absolutely necessary to have DC path to the ground for + input bias current. My guess is that what ever you are driving your filter with is capacitive coupled. If Vs impedance is low, just drop 10K or more between input and ground.

cheerz
urosh
 
The R there will change the tuning of your the filter and scrub off a little level. Instead positioning the R across the input should have an inconsequential effect on any DC blocking pole frequency.

JR


 
Sorry, I wasn't clear enough, you should put resistor from input of whole filter to ground (that is, from
junction of Vs and Ra). That resistance is insignificant as long as Vs impedance (that is output impedance of previous stage) is low (which should be the case anyway, high source Z will mess with filter response).

cheerz
urosh
 

Latest posts

Back
Top