Low pass filter at high voltage

[seyyah]

I'm trying to do isolated voltage measurement of a pwm type signal. For this, i'm using LEM transducer. But LEM has a bandwidth of <10KHz. So it forces me to make the filtering of pwm signal in the high voltage side. I have some problems with this measurement. Lem has a resistance of ~250 ohm. With the current limiting resistor it has ~33k input resistance. Due to this, for a small error of measurement i have to choose a small value resistor and a big value capacitor for RC low pass filter. But due to frequency, nearly all bus voltage can be seen on resistor filter R2. This causes overheating on this resistor. And requires a big power resistor. So what else can i do? Is there a better way to do this?

 

[instruite]

Cant you use two resistors in place of R1 for 33Kohm resistance

Like you can use R2=10K and R1 = 23K (or any combination which is possible)
and then use R2 resistance as filter resistance also
which will reduce your capacitance value also

 

[IanP]

The solution is not RC filter but LC filter.
Instead of R2=1kΩ use an inductor .. (2ΠfL) .. of, say, 10mH, or more ..
Regards,
IanP

 

[seyyah]

I think i will need a big mass inductor for this.So doesn't seem practical and cost effective. Anyway, I think i can reverse the resistors' places. Since the amount of current, which is determined by the total resistor, is important for LEM so there shouldn't be problem when i reverse them. Actually if LEM's internal resistor was a bit higher may be i wouldn't need the 1k resistor too. I must try and confirm this. Actually i may use an opamp and a resistor divider. Then i can interface the output of the opamp to the lem. This will scale down the actual value. Since the output will go to a cpu then i can calculate the actual value. I'll try this. If you have any other ideas please tell. Thank you for your replies.