Does this aerials chassis need grounding?

[Simon]

I'm doing some RF measurements on some equipment that failed EMC. We failed at around 65MHz, most of the EMC measurement aerials available are not too good down to this frequency or very expensive. I reasoned that given that this frequency and the results I have been getting from any mods are a bit higher I may as well just get an off the shelf FM radio aerial as my measurements are relative rather than absolute.

I got this one: https://uk.farnell.com/stellar-labs/30-2460/four-element-directional-outdoor/dp/2802329

The ground pin of the connector is not connected to the chassis, inside the box is a balun, while the specs state 75ohm impedance which is a simple dipole I suspect this is not 75 ohms hence the balun as this is not a simple dipole. So the supporting bar seems to have everything electrically connected to it. Does this mean that this is the earth input to the antenna and must be earthed? I am in a building so not necessarily putting it on a steel pole driven into the ground which would make it earthed, presumably I must therefore earth this antenna at it's mount point.

[Geoff-AU]

I think an FM Yagi is not going to work well for you, it won't have the directivity or the impedance match you expect at 65MHz.  You're better off making a regular dipole out of 2 pieces of wire, or even better making a small loop of wire as a nearfield probe.  Of course, if all you want is a relative measurement, any random piece of wire at a suitable distance will probably do as long as you don't move anything between measurements.

As for the yagi, baluns can have electrical isolation (or not, depending on their type) and impedance transformation.  The characteristic of a dipole at its design frequency is 72 ohms, but when you put parasitic elements near it to form a yagi the impedance drops.  Often, designers then move to a "loop-fed" yagi (the loop is actually a folded dipole at 300 ohms), then when parasitic elements drop the impedance you can end up closer to 72 or 75 ohms again and have less work to do in the balun.

The other role of a balun is to decouple the feedline from the antenna, and that's a whole different rabbithole because of skin effect.  Coaxial cable should be thought of as three conductors - the outer surface of the centre conductor, the inner surface of the shield, and the outer surface of the shield.  The outer surface of the shield can certainly become part of your antenna if you're not careful.

The parasitic elements of the yagi can either be electrically connected to the boom or isolated.  The design changes slightly, but a DC short circuit is very different to what happens at radio frequencies.   

Earthing is easy to do for DC (or for lightning protection) but much more challenging at RF.  The key aspect as it matters to you is that any DC electrical connection is also an opportunity to inject noise into your measurement.

[geggi1]

You should be looking into Logperiodic antenna for hamradio you can get some that goes from about 50 Mhz and up to around 500 Mhz or even higher.
https://moonrakeronline.com/eu/clp-5130-1n-50mhz-to-1300mhz-log-periodic?___store=base_eu&___from_store=default
Another option is a band II Tv antenna.

[nctnico]

Or make a bowtie dipole antenna.  An 80-ish cm high one worked ok for me to look at very specific frequencies.

[Wallace Gasiewicz]

Or maybe a discone?

[Simon]

I think an FM Yagi is not going to work well for you, it won't have the directivity or the impedance match you expect at 65MHz.  You're better off making a regular dipole out of 2 pieces of wire, or even better making a small loop of wire as a nearfield probe.  Of course, if all you want is a relative measurement, any random piece of wire at a suitable distance will probably do as long as you don't move anything between measurements.

Well I need to be high enough above the system noise floor to see the 20+dB drop when I achieve that. My biconical antenna that the idiots at telonic recommended is a total lie in it's advertised range as at 30MHz it has a gain of -30dB, at 65MHz -25dB, the whip aerial that came thrown in with the spectrum analyser did a better job, the biconical antenna really is for higher frequencies.

So I thought that an FM radio antenna would be better than a simple dipole and at £50 not worth knocking up something myself. Else I can just make my own log periodic as I am simply comparing a failed machine to a good one so actual spec is irrelevant, I just need gain and directivity so that I can point into my 5 sided faraday cage and the cage will block outside signals (mostly) but given the directivity of the aerial stuff coming from behind will not come up so strong and I am well above the noise floor.

The 50 quid aerial is 5dB so not much and less at my frequency but has 10-15dB front to back ratio that hopefully will still be relevant at a lower frequency. Experience so far has been that any filtering while slightly reducing amplitude does push the frequency up so I could find myself quite close to the 88MHz that FM starts at.

[nctnico]

I tried a large discone antenna for a similar frequency but I was not impressed. From what I have read a discone is not very sensitive for nearby sources.

A biconical antenna should do the trick but you need a really large one for -say- below 100MHz which is expensive.

[Simon]

Well I'm not sure that this aerial has helped much but I only got  a half hour to run some tests.

[Geoff-AU]

The 50 quid aerial is 5dB so not much and less at my frequency but has 10-15dB front to back ratio that hopefully will still be relevant at a lower frequency.

Not a chance.

[Simon]

The 50 quid aerial is 5dB so not much and less at my frequency but has 10-15dB front to back ratio that hopefully will still be relevant at a lower frequency.

Not a chance.

But versus a 1/4 wave whip antenna that is about 0.6m long (125Mhz) ?

[Geoff-AU]

I'm not sure what the question is.

A Yagi operated well outside its design bandwidth is about the same amount of awful as a dipole or a 1/4 wavelength operated well outside its bandwidth.

[Simon]

After a little more reading I discovered that what I actually have is a folded dipole Yagi which are very good at their intended frequency and not much use at anything else as I was warned.

I finally found a telescopic whip antenna that is 1.3m long. I bought several as I could make my own adjustable Yagi, but for now just used the one. Indeed, if extended to 1.3m the actual frequency at which the emissions were supposed to be the strongest was actually the weakest. But then 1.3m was not 1/4 wavelength of 70.9MHz, it's 1.05m, so I shortened the antenna to 1.05m and bingo, I then got a pattern similar to the one on the labs report and the worse frequency came out load and clear, this meant that it was high enough above the noise floor to see my required 20dB reductions in the mods I made which seem to work.