Maker Pro
Maker Pro

group delay equalisation of high-pass filter

P

Phil Newman

Jan 1, 1970
0
Anyone know of any handy free programs that I can use to equalise the
group delay of a high-pass elliptic filter (fc approx 1GHz), or a
band-pass filter?

Cheers,

Phil
 
D

Don Pearce

Jan 1, 1970
0
Anyone know of any handy free programs that I can use to equalise the
group delay of a high-pass elliptic filter (fc approx 1GHz), or a
band-pass filter?

Cheers,

Phil

Just write yourself the calculation for an all pass network in Excel -
it is only half a dozen lines.

d

Pearce Consulting
http://www.pearce.uk.com
 
P

Phil Newman

Jan 1, 1970
0
I have, and it's not working, which is really annoying. it works fine
for a low-pass filter, but with a high-pass network, with different
frequency (higher) the networks do not even act as an all-pass filter.

So, does anyone know of any programs that can do this for me with
correct answers??

Phil
 
D

Don Pearce

Jan 1, 1970
0
I have, and it's not working, which is really annoying. it works fine
for a low-pass filter, but with a high-pass network, with different
frequency (higher) the networks do not even act as an all-pass filter.

So, does anyone know of any programs that can do this for me with
correct answers??

Phil

Well, I'm not in the design business any more, but I once did this for
a filter at about 1.5GHz on FR4. I ended up with a small network of
fine tracks and chip capacitors. I used the theoretical numbers from
my programme and estimated the track lengths that would make an
equivalent inductance. A couple of iterations of boards was all it
took.

d

Pearce Consulting
http://www.pearce.uk.com
 
P

Phil Newman

Jan 1, 1970
0
Oh right,

I'll probably be doing the same. I got the spreadsheet to work, I was
copying the numbers into ADS to simulate, and I found that the numbers
were so sensitive to change. So the circuit I got is impractiable with
normal inductors, so I guess tracks to replace the inductors is needed?

Cheers,

Phil
 
D

Don Pearce

Jan 1, 1970
0
Oh right,

I'll probably be doing the same. I got the spreadsheet to work, I was
copying the numbers into ADS to simulate, and I found that the numbers
were so sensitive to change. So the circuit I got is impractiable with
normal inductors, so I guess tracks to replace the inductors is needed?

Cheers,

Phil

Absolutely - use thinnish single strand copper wire flush against the
board to breadboard, then measure and replace with etched copper track
for the real thing.

d

Pearce Consulting
http://www.pearce.uk.com
 
P

Phil Newman

Jan 1, 1970
0
Cheers,

is there a mathematical method of transforming my inductor values to
microstrip length/width rather than empirically guessing at wire
lengths?
 
D

Don Pearce

Jan 1, 1970
0
Cheers,

is there a mathematical method of transforming my inductor values to
microstrip length/width rather than empirically guessing at wire
lengths?

Rough rule of thumb would be about a nanohenry to a millimetre for
reasonably thin wire.

d

Pearce Consulting
http://www.pearce.uk.com
 
D

Don Pearce

Jan 1, 1970
0
so "no" then....?

The answer is yes, but it is complex, and depends on all sorts of
factors like substrate material, distance to the cover, copper
thickness etc. MDS should be able to help you.

d

Pearce Consulting
http://www.pearce.uk.com
 
P

Phil Newman

Jan 1, 1970
0
MDS? I have all the information necessary to calculate this.
 
Top