Filter for audio analyzer

[jfetter]

Has anyone used the TI simple filter for audio analyzer?

SLOA107 app note


http://www.ti.com/lit/an/sloa107/sloa107.pdf


I don't want to invest in a AES-17 adapter.

thanks

-

 

[jfetter]

Here is filter to use, plugging it in now.
Half bridge, irs2092.
The UR22 has 20k input Z.


`

 

[jfetter]

Here is update in tube forum since the pre is a hybrid. Don't think there is a forum for hybrids.

I used a first order filter and added overload protect for the ur22.

http://www.diyaudio.com/forums/tubes-valves/291660-hybrid-output-8.html#post4762339



-

 

[jfetter]

so making some real progress on general purpose passive filter. I used as a test case a filterless TI BTL part , isolated the *input* side with EDCOR isolations and also isolated the balanced analyzer tap going to the 1k 1k .0047 .0047 Mica filter.

The freq response looks excellent with the transformers looped back on each other. The analyzer internal THD is higher at low frequencies as a result of using transformers.

The filterless test part is now 'clean' enough for measuring.

Scaling can use the transformer taps and/or resistor pad.

Tomorrow checking it out with half bridge.

Here is pic of the test plot.

 

[doctormord]

Nice one.

 

[jfetter]

I know now the simple reliable passive solution for diy is transformer isolation.

The EDCORS work great but have a 10v max input. So I tried using a few different power transformers and they all worked. Here is some data on the common F-41X. I think 1% THD can be used based on the measurements. So if analyzer said 1.04% THD, the actual is 0.04%. We'll see if it works in practice.

Now can go to 2.5KW without using a pad.


One transformer is needed for input to BAL line input to the UR22.

Another is needed to isolate the line out to D-Class under test.


This worked using the EDCORs. Half and full bridge.

Frequency response is super, down 1db at 20khz

The bal 1k,1k,0047,.0047 filter not used here or shown. It adds about -1db 15khz up.

-bruce