| Author |
Message |
brmadhukar
Joined: 21 Jun 2002
Posts: 844
Helped: 29
|
 21 Jan 2003 4:59 Unknown Impedance |
|
|
|
|
Hi,
I have the new GPS down convertor IC from Maxim, and unfortunately the true differential output impedances of mixer outputs and input impedances of mixer outputs are not known. Maxim tried to help on this but they require more time for the correct answer. The IF2 output is at 15.42MHz. I need a bandwidth of +/-8MHz. Should I use the approximate impedance provided by Maxim and design a passive filter or use an active filter with larger input impedance and lower output imp.
TIA
PS: I am beggining to like these forums
|
|
| Back to top |
|
 |
flatulent
Joined: 19 Jul 2002
Posts: 4926
Helped: 328
Location: Middle Earth
|
21 Jan 2003 5:07 try the passive |
|
|
|
|
| Your quickest way would be to try a passive and measure the system response from the RF input to the filter output. Another way would be to do a roofing filter, then an amplifier, then the filter using the amplifier known parameters. The roofing filter would be so wide band that any ripple in it would be minimal in the IF frequency range. By roofing I mean one that cuts out the sum mixer product at 3 GHz or so and the LO feed through at 1.5 GHz or so but lets through your IF.
|
|
| Back to top |
|
|
moron
Guest
|
| 22 Jan 2003 17:07 |
|
|
|
|
| Have U measured by Vector Network analyser ?
|
|
| Back to top |
|
|
Google
AdSense
|
22 Jan 2003 17:07 Ads |
|
|
|
|
|
|
| Back to top |
|
|
toonafishy
Joined: 18 Jun 2002
Posts: 462
Helped: 25
Location: Earth
|
| 22 Jan 2003 17:14 |
|
|
|
|
| I would measure the diff mix impedance using a network analyzer. Then design a filter to match. Its hard to say what kind of filter because I would need to know other things such as how much attenuation is needed and at what offset. Just saying the bandwidth is only part of the story.
|
|
| Back to top |
|
|
multanova51
Joined: 03 Mar 2002
Posts: 75
Helped: 3
Location: Vatican
|
| 23 Jan 2003 17:24 |
|
|
|
|
Hi Brmadhukar
There is a trick to measure Rf impedances even if they are presented differentially.
Simply use a LC circuit (if you need diff then use a coil with a secondary winding) and tune to the wanted freq. without connecting the DUT. Measure by any means the 3dB bw (netw analyser, voltmeter etc). of the LC coil.
Then connect to your DUT.
Again remeasure the 3dB bw. There will be increased loading due to your DUT. With formula Q = Rp/Lw and 3dB bw = Fo/Q you can calculate back the additional Rp the DUT is imposing on your tuned circuit.
I used it a long time ago and you can even measure negative (!) impedances with it e.g. from a one pin osc. circuit. You might also estimate the reactive loading which is there by the slight detuning of your LC circuit.
Perhaps useful for you. It is a bit tedious but gives accurate results.
rgds
Mike
|
|
| Back to top |
|
|
