|
Post by lschropp on Dec 21, 2013 20:00:37 GMT -5
I have a VNA 2180 as well as a Power Aim. I am curious how I would use the 2180 to measure phase delay across a network such as a T network. A way to calculate the delay in degrees across an existing network is needed. My application requires that I measure existing networks and I do not have the individual component values available without disassembling and measuring components.
Any ideas how to do this?
Thanks
|
|
|
Post by Bob on Dec 22, 2013 16:25:48 GMT -5
The VNA2180 can measure the phase delay of a network by connecting the input of the network to Port A and the network output to Port B. Scan Port B and display S21 and the Phase of S21. This phase plot is the phase delay through the network in degrees. You can also display the Group Delay to get an indication of distortion through the filter.
-- 73/Bob
|
|
|
Post by lschropp on May 29, 2015 8:45:39 GMT -5
Bob,
I'm still having some trouble getting a correct measurement of delay across a "T" network. I set things up to scan port B and read S21. This provides a reading for phase delay through the network. However, the resultant phase reading I get is not correct. For example, I set up a leading network for test @ 1mHz and placed an appropriate load on the output, measured each leg of the "T", measured the impedance transformation across the network at the input - everything correct. The predicted phase for the network and values for each leg of the "T" was determined by a broadcast "T" network design computer program and double checked with a calculator.
In the above example I should have gotten a positive phase reading across the "T" network that was reasonably close to the design value. The delay was far off from the predicted by more than 100 degrees and was negative for a leading network instead of positive.
I have tried various combinations for calibration of the VNA test leads to compensate for lead induced errors. ( The leads should only be able to account for a few degrees of error at the 1 MHz test frequency anyway)
So I must be missing something. Any idea what I might be doing wrong?
Larry
|
|
|
Post by Bob on May 29, 2015 16:02:58 GMT -5
Hi Larry, That sounds like an interesting application. Send me a screen shot of the S21 mag and phase and the scan file for port B. Also I'd like to get the parameters for the legs of the T-network. I'll take a look at it.
--73/Bob
|
|
g3txq
New Member
Posts: 34
|
Post by g3txq on Jun 3, 2015 14:10:16 GMT -5
Forgive me for suggesting the obvious, but could it simply be a convention issue?
By that I mean for example +200 degrees = -160 degrees.
Steve G3TXQ
|
|
phil
New Member
Posts: 14
|
Post by phil on Jun 1, 2016 17:36:51 GMT -5
I asked this question yesterday: It looks like this is a question that others have concern.
I am sorry if this has already been answered on this forum. I am a Power Aim 120 user of long standing. However I am considering the VNA 2180 (2 port) in order to work with AM directional antenna systems specifically to check not only the usual matching parameters in say an ATU (the power AIM does this well) but I also need to know the calculated phasing through such networks. Looking through the literature indicates the VNA 2180 will perform S 21 measurements and should fill the bill to quickly set up the T network's match as well as the phasing through it. Any comments would be appreciated.
Phil
One of the things that has been suggested outside of the forum for an accurate display of S21 phase through a T-Network is to reverse the input and output. In other words feed port A to the antenna side of the T and port B to the transmitter (or phasor side in a DA). I am concerned as I want to purchase a VNA 2180 but looking at reasonably accurate phase response through a T match is VERY important for me.
Phil
|
|
phil
New Member
Posts: 14
|
Post by phil on Jun 2, 2016 8:00:53 GMT -5
I really just guessing here as I do not have a VNA2180 yet but to measure the correct phase through a matching network requires the correct termination impedance. Since the VNA 2180 port B is 50 ohms J 0 it stands to some reason that if the T network were measured backwards that the correct termination at the 50 ohm side would create the correct impedance transformation when testing by inserting the test signal at the T output. The displayed resistance portion of impedance would be that of the antenna and the conjugate of the reactance would show up if the T network is properly adjusted. I have used the Power AIM many times in this manner since I did not have a dummy load with the same characteristics of the antenna. I terminate the T input with 50 ohms and measure at the output side. It is a great trouble shooting tool. We need to take this a step further and see if we can see the S21 phase response of the network at the design F. That is what I am hoping the VNA2180 would in fact do.
Phil
|
|
phil
New Member
Posts: 14
|
Post by phil on Jun 10, 2016 10:55:04 GMT -5
I received my new VNA 2180 yesterday. I immediately realized I don't know how I got along without this previously. Setting up AM broadcast matching networks does in fact produce S21 results of the phase through the matching network. As mentioned above the input to the T matching network must be connected to port B. Port B offers a 50 ohm termination with relatively good return loss. Port A is connected to the T matching network output. You must do a port B calibration using the clip leads you intend to use in the measurements first. The VNA port B is a 50 ohm system so it is impossible to get correct results connecting port B to the matching network unless it is designed as a 50 ohm output network.
The S21 results are very close to setting up the T matching network the old traditional way, that is measuring the reactance of each component separately then putting it all together and tweaking. Strays are always an issue and reduce accuracy. Displayed S21 phase is very close to the calculated values. The A port displayed scan of impedance at the network output is the expected R of the antenna driving point impedance and the conjugate of the expected antenna driving point reactance X. Of course in many AM directional antennas measuring the drive point impedance of a particular tower is sometimes the hardest part. OIB's inserted in the antenna J plug create issues that must be overcome. Subject of a different matter. So far I am very pleased with the VNA 2180. Also possessing a Power Aim 120 begs the question if this is still needed? It looks like connecting a VNA 2180 to a tower in an RF environment could be somewhat dangerous to the VNA 2180 front end. The Power AIM front end appears to be protected from this to some degree. There is also a power output difference during measurements. I am not sure what that is and if it would affect accuracy in an RF interference environment. The VNA 2180 has the the same DSP capability built into the Power AIM so they are equal on that front. There are other features in the same family of single port network analyzers that the Power AIM is in. I have not checked to see if all of those features and recent updates to the single port analyzers apply to the VNA 2180 as well.
There is a whole lot the VNA 1280 can do. I'm just scratching the surface.
|
|
phil
New Member
Posts: 14
|
Post by phil on Jun 14, 2016 8:02:28 GMT -5
Follow up: I have checked the S21 phase delay measurements on the VNA 2180 on several mocked up T-Networks here in the shop with different values of output impedance transformations from a 50 ohms J 0 input. Using the method in the above posting, that is connecting the network 50 ohm input to Port B and the output of the network to port A the S21 numbers for phase response through the network are very accurate as compared to an oscilloscope display of the time delays. This is for lagging phase T-networks. Other S21 numbers are not accurate and not expected to be. Network losses due to port A connected to the output of the network severe mismatch is the reason for this. The S21 phase number is correct and that is what we are interested in at this point. As a side note a T-Network designed for 50 ohms J 0 in and out will show all S21 numbers accurately.
What has not been tested is the phase indication on a leading phase T-network (One with the series arms as capacitors and the shunt arm as an inductor.) That is the one Larry tried above. Hams use this T design more than broadcast in most cases. I have to mock up one of these with parts on hand and test the leading network S21 phase indication on the VNA 2180. I will post the findings once I get to this. WB2IQG.
|
|
phil
New Member
Posts: 14
|
Post by phil on Jun 20, 2016 18:37:25 GMT -5
In continuation. I put together some typical T-networks with leading phase (high pass configuration) of different impedance transformations and phase responses by the formulas. This is the typical ham use of the T-network although we do see it in AM broadcasting antenna matching sometimes. Again the S21 phase indication of the VNA 2180 is right on when compared to a scope display. Connected as port B to the T-network input and port A to the T-network output. The S21 numbers for losses is not meaningful due to the mismatch of the VNA port A to the output of the T-net unless it is a 1 to 1 50 ohm J 0 network.
What we are looking for is an accurate phase indication through the T-network so an ATU can be set up for the correct impedance transformation and phase response which is critical in AM directional antenna systems. This is a relatively inexpensive and very portable way to do this compared to other methods we have employed. Plus all of the other stuff this unit can do. WB2IQG
|
|