isolation from one port to another

[mcsquare]

mixer isolation definition

Isolation is a measure of the circuit balance within the mixer. when the isolation is high, the amount of "leakage" or" feed thru" between the mixer ports will be very small.
looking at the fact above, i need someone to explain to me what do u mean by "when the isolation is high", "leakage", "feed thru"? I am new in rf, i am going to do a master research on mixers. All the while I am just familiar with optical transceiver, I am not involving in wireless design. hope some one could help me in it? or do u have a good book or good article abt isolation?
thanks

 

[vfone]

Here you can find the answer:
**broken link removed**

 

[DonJ]

Isolation is a measure of the circuit balance within the mixer. when the isolation is high, the amount of "leakage" or" feed thru" between the mixer ports will be very small.
looking at the fact above, i need someone to explain to me what do u mean by "when the isolation is high", "leakage", "feed thru"? I am new in rf, i am going to do a master research on mixers. All the while I am just familiar with optical transceiver, I am not involving in wireless design. hope some one could help me in it? or do u have a good book or good article abt isolation?
thanks

Here is a short version:
A mixer has three ports, two inputs and one output, by definition. The isolation is a measure of how well the ports are separated from each other in terms of unwanted signal coupling. 30dB isolation is in most cases considered 'high isolation'. Leakage or feed thru is the terminology of the unwanted signal coupling.

A.T.B

 

[evoplus]

Hi DonJ and Vfone,

Just like to check with you all. Suppose I'm doing a RF isolation for a mixer. I connect the RF port to channel 1 of a VNA and LO port to channel2. I terminated the IF with a 50Ω. If I set the power to be -10dbm, my ioslation reading for mixer which span fr 1GHz -2GHz on the VNA shows a range of -10dbm to ard -24dbm.

My question is could I deduce that my mixer has a poor isolation loss as by right my isolation loss should be hitting ard -30 to -40dbm , supposing I am looking for the 20 dB loss.


Thanks

 

[flyhigh]

Hi,

just to add my opinion. Mixer operation is highly dependent on LO power. LO port should be excited at high power lever to ensure proper operation, RF level should be small enough to keep the small signal mixing condition (10-15 dB below LO power) to preserve linearity. So, your test on NA is not telling you the right performance. I would use the following sequence on spectrum analyzer.

1. measure LO power level nLO using spectrum analyzer
2. connect LO to mixer, terminate IF port
3. measure level of LO frequency sigal on RF port nRF_LO using spectrum analyzer
4. terminate RF port
5. measure level of LO frequency sigal on IF port nIF_LO using spectrum analyzer

Then LO-RF isolation is nLO - nRF_LO
LO-IF isolation is nLO - nIF_LO

The other isolations usually are of no importance.

Hope this gives you an idea.

flyhigh

 

[evoplus]

Hi flyhigh,

1.measure LO power level nLO using spectrum analyzer
2. connect LO to mixer, terminate IF port

Why is that the LO power level need to be measured , Could i just straightaway connect the LO input to the signal generator and key in the respective power level that I want to measure?

3. measure level of LO frequency sigal on RF port nRF_LO using spectrum analyzer
4. terminate RF port
5. measure level of LO frequency sigal on IF port nIF_LO using spectrum analyzer

Then LO-RF isolation is nLO - nRF_LO
LO-IF isolation is nLO - nI
flyhigh[/quote]

For your step 3-5, does it mean connect the LO to the signal generator, RF to the spectrum analyser and terminate the IF port if I am looking for a LO-RF isolation? Is the LO-Rf isolation equvalent to the RF-LO isolation?


Thanks

 

[flyhigh]

Hi flyhigh,

1.measure LO power level nLO using spectrum analyzer
2. connect LO to mixer, terminate IF port

Why is that the LO power level need to be measured , Could i just straightaway connect the LO input to the signal generator and key in the respective power level that I want to measure?

Yes, you can in principle, for the first test. Just make sure your power can drive mixer to reasonable extent. Almost all mixer characteristics, e.g. coversion gain-loss, noise, intermodulations are very dependent on LO power. For final test you should connect the real LO to be sure about pulling of its frequency by the mixer, real power due to LO signal reflection etc.

3. measure level of LO frequency sigal on RF port nRF_LO using spectrum analyzer
4. terminate RF port
5. measure level of LO frequency sigal on IF port nIF_LO using spectrum analyzer

Then LO-RF isolation is nLO - nRF_LO
LO-IF isolation is nLO - nI
flyhigh

For your step 3-5, does it mean connect the LO to the signal generator, RF to the spectrum analyser and terminate the IF port if I am looking for a LO-RF isolation? Is the LO-Rf isolation equvalent to the RF-LO isolation?
Thanks[/quote]

Yes, connect generator to LO port of the mixer, the rest as written. LO leakage to both, RF and IF ports are of interest. Leakage in oposite directions is minor interest and is not equal to the ones discussed here. The point is, LO signal is considerably stronger than other two and its leakage can have negative effect to the circuits connected at IF and RF port. For example, if LO-RF isolation is poor, LO signal can propagate trough finite S12 of LNA connected to RF port and then be radiated by receiving antenna. This is highly undesirable as it makes EM enviromnent "poluted". Many regulations are giving this sporious emissions from receiver mainly to address mixer isolation.