RF Antenna Design - Homebrew SWR measurement this time

[icon]

Hi

I asked a couple of questions about making antennas for 5.8gHz in this thread: https://www.eevblog.com/forum/beginners/any-rf-antenna-design-boffins-around/.

I got some useful info from that. Now, I've bought myself a directional coupler, which happened to come with a 'negative detector'. This is the coupler:



And this is the detector:



The data sheet for the coupler is here: http://krytar.com/pdf/2618.pdf. The headline figures are -16dB coupling, directivity > 15dB. I've got a handle on what those two mean from watching W2AEW's videos on couplers. A metric I don't understand is 'Maximum VSWR: 1.35'; what is the implication for that, for trying to measure the SWR of an antenna? I have no idea whether these specifications are suitable for what I'm trying to do, but it wasn't hugely expensive s feel free to tell me!

Assuming I haven't bought the wrong thing entirely, here's what I think I'm going to do - stop me when I go wrong. I think the negative detector is a damn fast diode. I can't find a data sheet - perhaps it's too simple to need one? If I get an SMC connector (for the output of the negative detector), and solder a resistor//capacitor from output to ground, I can measure the (negative) peaks of the rectified signal. By measuring this voltage with the coupler connected one way round, and then swapping it end-for-end and measuring again, I can get figures for forward and reverse voltage and hence calculate SWR. Sound plausible? How does the 'Maximum VSWR' figure affect this, I wonder? I don't *think* I need to worry about coupling, directivity, insertion loss and that stuff, because I'm only interested in the ratio between the two numbers, rather than absolute figures.

Feel free to pile in if you can shed light on any of the above. Incidentally, I flew with a skew planar antenna (four 'lobes' arranged in a circle) of my manufacture on my receiver, in a brief lull in the howling gales - the reception was absolutely rock solid, out to what must be nearly 1000 metres. Beginners luck, perhaps!

Thanks
John

[rfeecs]

The detector is most likely a zero-bias schottky detector similar to this:
http://cp.literature.agilent.com/litweb/pdf/5952-8299.pdf
Negative just means it puts out a negative voltage.

There a a couple of problems using that coupler for SWR measurement.  The main one is the directivity.
With 15dB of directivity, you might have perfect 1:1 SWR from your antenna, but still measure only 15dB return loss.

The maximum VSWR spec is the VSWR of the ports of the coupler.  This would cause you problems from the output VSWR of the coupler is not exactly 50 ohms, so this mismatch will interact with the mismatch of your antenna.  This is "source match" error.  It will add or subtract from your antenna VSWR.  So it is an additional source of error.

The three traditional error vectors for a 1 port network analyzer are Directivity, Source Match, and Reflection Tracking.  You can learn more about these with Google.

So there are errors.  But it should give you a fair idea of the SWR.  Worth a try.

A high directivity bridge is more ideal for measuring SWR.  Wiltron (now Anritsu) used to make them.  Here is some more info on directivity:
http://www.anritsu.com/en-us/test-measurement/solutions/en-us/Understanding-directivity

[TheSteve]

For initial testing you might want to test it with a known working antenna from a dual band wifi router. Mating connectors will be a pain though and without a signal gen I think you're just shooting in the dark - better then nothing though. To give you an idea of what you'd ideally like to see this is a scan of the dual band antenna from my Netgear R7000. You can see the two expected dips in SWR for 2.4 and 5.8 GHz.

[icon]

Well, I've got the necessary connectors on order, so it'll be interesting to see if I can get anything useful out of it! SMC connectors are harder to get hold of than SMA, that's for sure.

Regards
John

[Ivan7enych]

What is the source of 5.8GHz signal? If it's a simple video transmitter, I suggest you to add 10dB attenuator between Tx and coupler input, to normalize Tx output impedance.

And get a hi quality terminator (DC-18GHz) as a reference, to check your coupler directivity.

[icon]

Funnily enough, I've just ordered an attenuator - 6dB as it happens - and a 50 Ohm terminator. My thinking was - what would an open end look like, and what would a 50 Ohm terminator like. In the former case, I presumed that the reflected energy might damage the video transmitter so the attenuator was to prevent that. Armed with those two data points, I can at least see if any useful information can be extracted.

Thanks
John

[T3sl4co1l]

The main downside to using a directional coupler and detector is, it looks the same whether open or short circuited.  So you can't tell which direction the impedance is (too high or too low, capacitive or inductive) without varying some component values.

Also, that your signal level needs to be enough (after the attenuator and coupler) to read with the detector, but I suppose that shouldn't be too bad.

Tim

[rfeecs]

A note on the detector, assuming it is a zero bias Schottky or similar:

Typically it has the filter capacitor built in.  It has an output impedance in the kohms and gets highest sensitivity with a high impedance load.

The output is going to be in mV or uV for lower input power.  See the plot on the datasheet.

It can be blown by too much input power and is probably ESD sensitive as well.

[icon]

A note on the detector, assuming it is a zero bias Schottky or similar:

Typically it has the filter capacitor built in.  It has an output impedance in the kohms and gets highest sensitivity with a high impedance load.

The output is going to be in mV or uV for lower input power.  See the plot on the datasheet.

It can be blown by too much input power and is probably ESD sensitive as well.

Ah, OK. So just hook it to a multimeter on the millivolt range?

[rfeecs]

That should work.  Or an oscilloscope, DC coupled of course.

The output will continue to go up as you increase the input power.  Until it blows the diode.  So be careful not to put too much rf power into it.