PDA

View Full Version : Measuring Sander style UHF antennas



Hucker
27th January 2012, 07:45 PM
Sander:

I'm trying to measure/tune some antennas for Scotttu. He sent me a couple and they look pretty good. One came in at 434MHz and the other at 438MHz. Interestingly the one that was slightly longer is the one the measured out at 438.

I'm using an Anritsu siteMaster tester in a lab (not a range!). We calibrated for the frequency range we were looking at. When measuring reflectance we we got a very flat wavy line without the characteristic valley at the resonant frequency. If I touched the sma connector on the Sander UHF antenna while performing the test it locked in at the expected frequencies with a nice sharp valley. (I forgot to get screen shots).

What is strange is we tested just like we did when we tested some 2.4GHz SPW/CL antennas in terms of setup and did not have any issues like this. It just worked.

We did go over the test setup and believe the connections were good but we could still have something AFU. We may need to test this differently than we expect, the guy helping me hasn't tested this style of antenna before. Is there anything you can think of that might be causing this apparent ground issue?

Thanks!

sircana
27th January 2012, 08:12 PM
I had exactly the same issue. The monopole needs a ground plane. The EzUHF itself is supposed to act as a ground plane, but for testing and fine tuning purposes I am not sure what the best procedure is. I tried placing the antenna vertical over a big metal sheet, and it work quite well, but honestly I don't know which one is the right method for testing monopoles.

Sander, How did you tests yours?

Hucker
27th January 2012, 09:23 PM
Thanks sircana, I feel a little better. RF is humbling, every time I think I get it there is something that confuses my intuition.

When you say to put it over a big metal sheet, is that sheet connected to the sma electrically (sort of like me touching the connector)? When you say big, how big? It seems that it shouldn't be bigger than the EZRx. Since I don't know the math it is hard to see what is going on...

Since I can now measure the performance should I tune the two antennas to be the same or should I put one at the low end of the band and the other at the high end?

I told the guy I was testing with that this setup on 1W was used in trappy's long flights. He didn't believe the distances. Was a directional antenna used or was it a stock EZUHF with a standard tx whip and monopole RX?

sircana
28th January 2012, 03:48 AM
I used something like 50x50cm metal sheet, with the SMA touching it. Again, I don't believe this is the right method, but it was the only way to get some resonance from the monopole and be able to measure the frequency.

Both antennas should be tuned to the EzUHF channel you are planning to use.

Trappy used these ones on the plane, and a Yagi on the Tx.

Mictronics
28th January 2012, 04:09 AM
The Sander style antenna will only work when connected directly to the EZUHF RX. Any extension by coax cable will shift the resonance frequency.
Like sircana said, a ground plane is needed, at least 1/4 wavelength in radius around the monopole.

Whenever possible you should choose a half wave dipole instead a monopole. Performance is much better and no ground plane is needed.

ssassen
28th January 2012, 06:16 AM
Pretty much what Juan and Michael said, the Sander-style antennas work best when connected directly to the EzUHF as it (and the wiring attached to it) form the counter pole, i.e. the ground plane. Measuring them is done using a hacked up EzUHF Rx with the antenna attached, which is a close representation of what actually happens in practice.

However Michael is right, a half wave dipole offers much better performance, but is physically quite a bit larger of course (twice the size).

On trappy's flights (but for the 87km flight, he used a 4.5dBi yagi on the Tx there) the stock setup was used, i.e. Sander style antennas on the Rx and a Diamond SRH-771 on the Tx.

Hucker
28th January 2012, 11:07 AM
Does the size of the ground plane impact the frequency or is it just the gain? In my crude experiments (using me as the ground plane) I didn't see a frequency shift but I did see the valley get deeper depending on how the sma was touched.

For for test purposes should I just cut out a copper clad pcb or a piece of copper foil in the dimensions of a ezuhf rx (I'm guessing the PCB on that guy has a ground plane on one of the layers)?

ssassen
28th January 2012, 11:11 AM
I would just stick to 164mm and leave it at that, any tweaking will be offset by your cabling to the EzUHF Rx anyway. And yes, the EzUHF itself is basically one GND block, as both PCBs (it is a stacked PCB design) have a full groundplane on one side.

Hucker
30th January 2012, 01:04 PM
Not one to leave well enough alone...I did some experiments (that could be flawed) varying the size of ground plane and found that the resonant frequency does move around a bit...

This leads me to believe (agree?) that this style of antenna cannot be tuned as an independent component, which I think is what Sander meant.

So, if you want to tune your system you need to measure the whole system in the airplane with all of the wiring installed and running. It seems to me a way to do this would be to use an attenuator on the two antennas (possibly f'ing up the measurement) and then looking at RSSI using the ezosd, modify the length of the antenna and check RSSI and repeat. You could skip the attenuators if you had a long distance test range. I'm assuming that connecting the RSSI to a scope and skipping the ezosd would also screw up the measurements.

Dionysos
26th October 2012, 01:22 AM
now I get what you mean :)

Dio

Hucker
26th October 2012, 10:42 AM
Yeah, frustrating, this was my first foray into tuning.. 164 works pretty well. Resolution off of rssi isn't great.

I finally got an ezosd so I may revisit this as I trust what the ezosd is doing to display the rssi value more than I do skylark. I *think* the way to micro tune this is to fly with a stock monopole and the one you are tuning. Do some checks to make sure both sides work about the same by swapping antennas on the ground and verifying (sircanna says they are the same). Orient both antennas upright and then go fly. Get "pretty far away" and fly N/S/E/W (away/towards/left/right?) and record the RSSI, come back cut off .25mm and repeat. you need to have your tx fixed BTW... Onerous and painfull but if the real answer on your plane is 162.5 then you can smile.

BTW I fly at 438MHz* and mine are not 164mm :) I suspect though that the resolution of the rssi output will not be good enough unless you go from really bad to really good and do a curve fit and pick off your length from zero slope of that curve.

* I considered writing mhz just to get Sander going, but I like the 4ch rx's so much I'm giving him a break.