ST2 Scantenna tested with NanoVNA

[Ubbe]

Just test with that 50 ohm calibration you have. The scanner are more or less 50 ohm so you should test with that impedance.

In reality a scanner sway a lot off from that 50 ohm down to 20 ohm up to 200 ohm and I think that it hasn't gone too much though into the antennas impedance, it's several different dipoles in parallell and each one should have a 75 ohm impedance more or less. But outside of it's narrow natural tuned frequency it also sways a lot off from that 75 ohm. And where that balun connects are very suspisous and feels out of place, it should have been connected to the two open wires nearest the mast pole but probably needed to be moved due to unforseen issues.

The balun used where probably only for easy of availability and had nothing to do with impedance match but only as a balun to connect dipoles to a coax. You probably would see much better reception with a 1:1 50 ohm - 50 ohm balun as a 300/75 will reduce the signal voltage 4 times.

It's no rocket sience. It's just a couple of dipoles tuned for different bands and connected together in the simplest way and designed to have the lowest cost possible when manufactured in batches of hundreds.

/Ubbe

 

[prcguy]

The problem with the ST2 type antenna is the elements are not 1/2 wavelength resonant, they are something else and the antenna appears to be designed to have 300 ohm impedance at its feedpoint. This is not a native 50 or 75 ohm antenna, its 300 ohm plus or minus ?? Directly connecting 50 or 75 ohm coax would be a mistake. Using a 300 ohm to 75 ohm matching transformer gets you to 75 ohm coax and a cheap way to feed the radio. The original antenna was designed and manufactured by a TV antenna company. There was some mild rocket science in the design, which nobody on this forum seems to understand including me.

Just test with that 50 ohm calibration you have. The scanner are more or less 50 ohm so you should test with that impedance.

In reality a scanner sway a lot off from that 50 ohm down to 20 ohm up to 200 ohm and I think that it hasn't gone too much though into the antennas impedance, it's several different dipoles in parallell and each one should have a 75 ohm impedance more or less. But outside of it's narrow natural tuned frequency it also sways a lot off from that 75 ohm. And where that balun connects are very suspisous and feels out of place, it should have been connected to the two open wires nearest the mast pole but probably needed to be moved due to unforseen issues.

The balun used where probably only for easy of availability and had nothing to do with impedance match but only as a balun to connect dipoles to a coax. You probably would see much better reception with a 1:1 50 ohm - 50 ohm balun as a 300/75 will reduce the signal voltage 4 times.

It's no rocket sience. It's just a couple of dipoles tuned for different bands and connected together in the simplest way and designed to have the lowest cost possible when manufactured in batches of hundreds.

/Ubbe

 

[digitalanalog]

ok, so today I start testing again, everythings the same same setup as before.
Recalibrate the Nano, and look back at the reading I got last time to see if they were the same and guess what something changed and changed big time.

The reflection IE LogMag db numbers are much much higher then before, here is just a small sample of the 30-54MHz range i tested before.

original test: start - 30.000Mhz - 7.84swr - LogMag - 2.22db
new test - 30.000Mhz - 7.10 swr - Logmag - 87.81db
original test Dip - 34.560Mhz - 2.99swr - LogMag-6.03db
new test - 34.560Mhz - 5.65swr - LogMag-87.56db
original test Peak - 39.840Mhz - 8.99swr - LogMag-1.93db
new test- 39.840Mhz - 9.35 swr - LogMag-92.70db
Original test Peak - 45.360Mhz - 9.75swr - LogMag-1.78db
new test 45.360Mhz - 10.74swr - LogMag-86.69db
Original test Dip - 49.440Mhz - 4.55swr - LogMag-3.87db
New Test - 49.440Mhz - 4.69swr - LogMag-97.38db
Original test End-54.000Mhz - 6.65swr - LogMag-2.63db
New test-54.000Mhz - 6.54swr - LogMag-89.58db

and these high db numbers are happing all accross the bands not just in the 30-54mhz rang, everything
is way higher in the db and the swr is off as well however not such a big change.

SO as you can see, the "db" numbers are way way higher and the swr showing slight changes.

Did I have a loose cable last time? or is this VNA a POS?

What is going on, This is very very very confusing???????????????????????????????????????????????

 

[vagrant]

If you are using 50 Ohm coaxial cable to your AT197, connect your VNA to that and see what should be a normal plot. The result should not look erratic. Set your VNA to look at 200-500 Mhz. (I use LMR-400 with my AT197)

I have connected a VNA to 75 Ohm coax to my ST2 and the result was an erratic mess. I was not surprised considering the mismatch. The result was absolutely not useful. That is why I tested the ST2 by dialing in various steady TX frequencies morning, noon and night, and then again with the discone to note and compare the results. I used an SDR for that so I could get some workable numbers.

Some fine advice was provided in this thread on how to use a 50 Ohm VNA with a 75 Ohm feed, while understanding there will be some introduced loss. I have not looked at the cost for that, as I am pleased with the results I have of the ST2 over discone for a certain range of frequencies. Still, if I could put up only one RX antenna, it would be a discone as I enjoy/prefer monitoring Mil Air and fire service aircraft. Terrestrial public service stuff is blasting away with repeaters, so it‘s easy to RX that at my location.

 

[digitalanalog]

If you are using 50 Ohm coaxial cable to your AT197, connect your VNA to that and see what should be a normal plot. The result should not look erratic. Set your VNA to look at 200-500 Mhz. (I use LMR-400 with my AT197)

I have connected a VNA to 75 Ohm coax to my ST2 and the result was an erratic mess. I was not surprised considering the mismatch. The result was absolutely not useful. That is why I tested the ST2 by dialing in various steady TX frequencies morning, noon and night, and then again with the discone to note and compare the results. I used an SDR for that so I could get some workable numbers.

Some fine advice was provided in this thread on how to use a 50 Ohm VNA with a 75 Ohm feed, while understanding there will be some introduced loss. I have not looked at the cost for that, as I am pleased with the results I have of the ST2 over discone for a certain range of frequencies. Still, if I could put up only one RX antenna, it would be a discone as I enjoy/prefer monitoring Mil Air and fire service aircraft. Terrestrial public service stuff is blasting away with repeaters, so it‘s easy to RX that at my location.

The results in question are from the ST-2, have not tested the 197 yet.

 

[ScubaJungle]

I had the same issues, if you see my post regarding the NanoVNA.
I wouldnt call it a POS, but I wouldnt call it an accurate tool. It will give you a result that will give you the gist, but probably not the exact correct values. So, it is useful for quick field testing or just messing around to see if there are any major issues.

Those logmag #s are way too high.
From my mistakes, make sure youre connecting to the right port (Ch0), calibrating correctly and saving the calibration/recalling the correct calibration. The first numbers seemed much more reasonable as far as the return loss.
Check my thread and see if that helps, and follow what PRCGuy and Ubbe say, they are extremely knowledgeable about the technicals regarding antennas and radios/receivers.

 

[vagrant]

Exactly - I have connected a VNA to my 75 Ohm coax which is connected to the ST2 300/75 Ohm transformer and the results were a mess due to the mismatch of using a 50 Ohm VNA.

A search provided this matching pad, but I am unfamiliar with it. There could he others for more/less. Are you already using one? I may have missed that in one of your posts.

Matching Pad 50 Ohm SMA Male To 75 Ohm F Female Operating From 0.009 MHz to 3 GHz RoHS Compliant

Matching Pad 50 Ohm SMA Male To 75 Ohm F Female Operating From 0.009 MHz to 3 GHz RoHS Compliant from Fairview Microwave will ship same day. Fairview Microwave Matching Pad 50 Ohm SMA Male To 75 Ohm F Female Operating From 0.009 MHz to 3 GHz RoHS Compliant is part of over one million RF products...

www.fairviewmicrowave.com

 

[digitalanalog]

Exactly - I have connected a VNA to my 75 Ohm coax which is connected to the ST2 300/75 Ohm transformer and the results were a mess due to the mismatch of using a 50 Ohm VNA.

A search provided this matching pad, but I am unfamiliar with it. There could he others for more/less. Are you already using one? I may have missed that in one of your posts.

Matching Pad 50 Ohm SMA Male To 75 Ohm F Female Operating From 0.009 MHz to 3 GHz RoHS Compliant

Matching Pad 50 Ohm SMA Male To 75 Ohm F Female Operating From 0.009 MHz to 3 GHz RoHS Compliant from Fairview Microwave will ship same day. Fairview Microwave Matching Pad 50 Ohm SMA Male To 75 Ohm F Female Operating From 0.009 MHz to 3 GHz RoHS Compliant is part of over one million RF products...

www.fairviewmicrowave.com

I am aware of the miss match, but that does not answer the question of why the numbers have changed so drastically from one
Test to another.

 

[prcguy]

If you use a matching pad you must calibrate the VNA with a 75 ohm open/short/load at the F connector. Otherwise it will fool the instrument and make things look better than they actually are.

Exactly - I have connected a VNA to my 75 Ohm coax which is connected to the ST2 300/75 Ohm transformer and the results were a mess due to the mismatch of using a 50 Ohm VNA.

A search provided this matching pad, but I am unfamiliar with it. There could he others for more/less. Are you already using one? I may have missed that in one of your posts.

Matching Pad 50 Ohm SMA Male To 75 Ohm F Female Operating From 0.009 MHz to 3 GHz RoHS Compliant

Matching Pad 50 Ohm SMA Male To 75 Ohm F Female Operating From 0.009 MHz to 3 GHz RoHS Compliant from Fairview Microwave will ship same day. Fairview Microwave Matching Pad 50 Ohm SMA Male To 75 Ohm F Female Operating From 0.009 MHz to 3 GHz RoHS Compliant is part of over one million RF products...

www.fairviewmicrowave.com

 

[digitalanalog]

Can I use this VNA as it is using my 75 ohm coax and 75/300ohm transformer to check the swr for tuning and trimming my antenna, will that get me where I need to be just using the swr portion.?

 

[vagrant]

If you use a matching pad you must calibrate the VNA with a 75 ohm open/short/load at the F connector. Otherwise it will fool the instrument and make things look better than they actually are.

Yes definitely and thank you for pointing that out. I should have noted that and not assumed performing a calibration was understood step after attaching the pad. I thought it was already mentioned in this thread, but reminders are good.

Can I use this VNA as it is using my 75 ohm coax and 75/300ohm transformer to check the swr for tuning and trimming my antenna, will that get me where I need to be just using the swr portion.?

I would never use a 50 Ohm VNA with 75 Ohm coax connected to a 75/300 Ohm transformer to check anything without the correct pad...And calibrating the VNA after connecting the pad with a 75 Ohm O/S/L as prcguy reminded us.

I'll bet they want a pretty penny for an 75 O/S/L kit. Oh holy cats! Yeah, I just found one that exceeds my hobby funds for one tool. Well, open is open and a short is a short, but you'll need a load. When it comes to gear like this, one gets what one pays for. Still, you can probably find a 10 pack of 75 ohm F connector load for cheap, even if you only need one. For a scanner hobby I am not sure how "good" those results would be. We are definitely on the hobby side of things price wise here, which is where we are.

Edit: You may have a 75 Ohm F terminator already in a drawer, or using one on something. Just double check it with a multimeter.

 

[prcguy]

Last time I bought a 75 OHM VNA from HP/Agilent it was around $37k, which is expected, but the cal kit with 75 ohm F connectors and adapters was another $7k or so. If you want to test stuff for real it really costs.

Yes definitely and thank you for pointing that out. I should have noted that and not assumed performing a calibration was understood step after attaching the pad. I thought it was already mentioned in this thread, but reminders are good.

I would never use a 50 Ohm VNA with 75 Ohm coax connected to a 75/300 Ohm transformer to check anything without the correct pad...And calibrating the VNA after connecting the pad with a 75 Ohm O/S/L as prcguy reminded us.

I'll bet they want a pretty penny for an 75 O/S/L kit. Oh holy cats! Yeah, I just found one that exceeds my hobby funds for one tool. Well, open is open and a short is a short, but you'll need a load. When it comes to gear like this, one gets what one pays for. Still, you can probably find a 10 pack of 75 ohm F connector load for cheap, even if you only need one. For a scanner hobby I am not sure how "good" those results would be. We are definitely on the hobby side of things price wise here, which is where we are.

 

[vagrant]

Thank you for sharing that professional cost to keep things in perspective. Still, these low cost VNA's are reasonably useful, but always knowing the results may be incorrect. Fortunately, that lesson only cost me $40 in microfarad capacitors, time, emails and probably lots of head shaking by those that knew better. hahaha we live we learn. Fortunately, an HP 8935 took care of things easily. Oh...there's one on eBay for $500. I think the 8935 I have now is on indefinite loan, as it was buried in a friends closet and he prefers his small and quite portable Rohde & Schwarz FSH6.

 

[Mike_G_D]

For what it's worth, here's what I would try:

1) Don't use the 75 ohm cable for the testing, instead get a usable length of 50 ohm low loss (as low as possible up to the highest test frequencies) which allows you to attach to the antenna from the vna with enough length to allow the antenna to be away from the vna and you as far as possible;

2) Calibrate the vna to the end of the test cable (the 50 ohm cable); so use the provided open, short, and load standards attached at the end of the test 50 ohm cable where it will be connected to the antenna and run the full S11 calibration with those standards attached to the test 50 ohm cable; the idea is to calibrate out the cable, or, put another way, move the measurement reference to the end of the test cable;

3) For initial tests, try using specific test frequencies at specific points in the common LMR and air bands, like 127 MHZ for the middle of the civilian air band, maybe 162 MHz for VHF-Hi, 460 MHz and 480MHz for UHF, etc.; otherwise you could run separate tests for full band for each specific band; I think trying to run a super wide sweep with that vna across the whole 30MHz to 1GHz range is too much and hurts accuracy and perhaps reliability; this takes more time but, I think, will produce better and more readable and reliable results; if the vna allows it, you could run separate calibration routines for each test frequency or band sweep and store the calibration data so that you wouldn't need to re-run them again each time you run the test as long as you use the exact same setup (e.g. one calibration set for VHF-Hi, another for UHF, etc. and then call up whichever one you need at the time);

4) Attach the antenna to a non-conductive mast like a pvc pipe or wooden pole as far away from any metal or foliage, people as possible in as open an area as possible; it would be interesting to run the tests that way and then try again with the metal mast just to see the effect at a later time.

You will have to use an adapter to attach the 50 ohm cable end (presumably a BNC, SMA, or N connector - don't use a PL259 connector) to the F connector on the balun attached to the antenna which will, of course, introduce some error but should be adequate for these types of measurements.

Get some 75 ohm resistive terminators as Ubbe said - just those cheap cable TV ones you can get a pack of at nearly every where would be ok;

5) Before testing the antenna first attach the 50 ohm cable test end to a 75 ohm resistive terminator (will need a F female-to-female barrel to connect to the F male connector on the end of the test cable) and run through the whole series of tests at all test frequencies and store the results; when looking at a Smith Chart normalized to 50 ohms you want to see just a dot or series of dots scattered closely around the resistive line at 75 ohms with little to no reactive component; if you don't see this then re-calibrate and try again; if you can't get the readings to look as close to 75+/- j0 as possible or about 1.5:1 VSWR then something is wrong and not ready for testing the actual antenna until it is fixed; you want to make sure that the cable and VNA are "viewing" the test end of the cable correctly so what you hope to see is a purely or mostly resistive 75 ohm resistive response from the resistive load with the cable, if calibrated for correctly, not coloring the results too much;

6) Once everything looks good with the 75 ohm resistive loads then remove them and attach the test cable to the antenna and run all of the calibrated for tests at each designated test frequency or band sweeps and store those results;

7) You might, if possible, want to re-run the tests with the antenna swiveled around as you have done before but with a non-conductive mast to see the effect, if any and compare to doing the same with a metal mast but making sure to have the height of the antenna above ground the same for both test cases.

-Mike

 

[Ubbe]

Professional stuff are made to be accurate with less than 1% tolerance from low KHz to many GHz. For hobby use you can tolerate a dB or 2 difference at lowest and highest frequencies and impdance doesn't need to be 75,00 Ohm but 73 and 77 might be ok.

That 50/75 pad costs $100 and attenuates 6dB and works from 7KHz to 3GHz so must be a resistor net with that huge attenuation. Having one that do 25Mhz to 1000MHz with 2% tolerance must be much cheaper, maybe solder together yourself and put in a metal box. With a multimeter you can measure the resistor values and select the ones that are closets to it's marked value. Look at the web how such a resistor net are built to give 50 ohm at one end and 75 ohm in the other. When you do calibration, with a measured 75 ohm resistor in a metal can, everything is calibrated with that 50/75 pad in place and its eventual anomolies are caibrated out.

/Ubbe

 

[prcguy]

If you get a 50 to 75 ohm resistive matching pad you must get a 75 ohm F type short, open and good quality load to cal the VNA, otherwise the matching pad is useless.

For Mike_G_D, the antenna being tested is a copy of the ST2 and that was designed to have a 300 to 75 ohm TV matching transformer attached and 75 ohm coax to the radio. You will find resonant points using 50 ohm coax but you need to have a feedline matched to the antenna to really find out how the antenna is working. For initial tests to see if it resonates anywhere near where its supposed to the antenna with the TV transformer and 75 ohm coax to the 50 ohm VNA is a start and will tell you important stuff, its just not the bottom line stuff.

Professional stuff are made to be accurate with less than 1% tolerance from low KHz to many GHz. For hobby use you can tolerate a dB or 2 difference at lowest and highest frequencies and impdance doesn't need to be 75,00 Ohm but 73 and 77 might be ok.

That 50/75 pad costs $100 and attenuates 6dB and works from 7KHz to 3GHz so must be a resistor net with that huge attenuation. Having one that do 25Mhz to 1000MHz with 2% tolerance must be much cheaper, maybe solder together yourself and put in a metal box. With a multimeter you can measure the resistor values and select the ones that are closets to it's marked value. Look at the web how such a resistor net are built to give 50 ohm at one end and 75 ohm in the other. When you do calibration, with a measured 75 ohm resistor in a metal can, everything is calibrated with that 50/75 pad in place and its eventual anomolies are caibrated out.

/Ubbe

 

[Mike_G_D]

If you get a 50 to 75 ohm resistive matching pad you must get a 75 ohm F type short, open and good quality load to cal the VNA, otherwise the matching pad is useless.

For Mike_G_D, the antenna being tested is a copy of the ST2 and that was designed to have a 300 to 75 ohm TV matching transformer attached and 75 ohm coax to the radio. You will find resonant points using 50 ohm coax but you need to have a feedline matched to the antenna to really find out how the antenna is working. For initial tests to see if it resonates anywhere near where its supposed to the antenna with the TV transformer and 75 ohm coax to the 50 ohm VNA is a start and will tell you important stuff, its just not the bottom line stuff.

Well, I get that it was designed for a 75 ohm feedline (through the balun) but in the end it is still just another "black box" to look at relative to the VNA so you still want to make the feedline "invisible" (calibrate it out) so why not use the test line that the VNA was designed for and then just factor in the 75 ohm nature of the antenna (plus balun transformer) when looking at the test results?

But it's an antenna and not an active or passive circuit lumped or distributed "exactly" so I bow to your expertise here.

In any case, building a 50 to 75 ohm minimum loss pad isn't that hard. Take a look here for a guide: CATV Minimum Loss Pad for 75Ω Measur - Maxim Integrated.

Keeping the caveats of differences between voltage loss and power loss as stated in the guide linked it should be fairly easy to build this with two resistors (or more when trying to get the values as close as possible with what you have available using series and parallel combinations).

But then, as you've pointed out, you have to have some decent 75 ohm standards and calibrate at the end of the 75 ohm port of the pad and the VNA has to be able to handle the nearly 6dB loss (in power). For the "standards" I would think you could get away with using some regular cheap 75 ohm terminators for the load and build a short with a simple shorted F female connector and same for the open (or even just leave the test end of the coax disconnected from anything as an open in a pinch). Not "lab standard" by any means but maybe close enough for this measurement.

-Mike

 

[Ubbe]

A calibration with a short isn't 75 ohm related, it's 0 ohm. An open calibration are just that, as high impedance as possible. You probably already have those. Only the 75 ohm load are important. Take one 154 ohm and one 147 ohm resistor in parallell and you get 75 ohm.

/Ubbe

 

[digitalanalog]

ok, so I have the "Open" I have the "Short" [0 ohms] from the NanoVNA
I also was looking into 75ohm load terminations and remembered I had
bought some to terminate unused splitter ports, I was understanding that
unused ports should be terminated.

So I looked through my plastic storage drawers and look what I found.


I checked these with my meter and they read 74.8ohm
and they have the "F" connector I need.

Is there any reason these will not work as a 75 ohm load?
I have the adapter (female-female "F") to put this on the end of the coax to do
a calibration or I have the SMA Male to "F" female to put it on the VNA, so
either way I could put this 74.8 ohm load in the calibration.

 

[vagrant]

Which 50 to 75 Ohm resistive matching pad did you buy or build? The one I linked with SMA male to F female, or something else? That will tell us all if an adapter is needed. From your note, it seems like you are trying to take a shortcut and not include the 50 to 75 Ohm matching pad. I could be wrong, but asking about an adapter to connect the VNA to the 75 Ohm terminator is telling me otherwise.

Just to be clear, you will need that matching pad inline during calibration, as well as when you are sweeping the antenna. You should also understand there will be losses with the pad. This must be taken into consideration when looking at the results. Others have posted about that and you must keep it in mind when analyzing.

The three others posting feedback are much more up to speed on this than I am. My intention was to provide affordable hobby level pricing information on how to accomplish your task. Hobby level pricing gets hobby level results, but that may be just fine. We do not know until we try.