UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > General Vintage Technology > Components and Circuits

Notices

Components and Circuits For discussions about component types, alternatives and availability, circuit configurations and modifications etc. Discussions here should be of a general nature and not about specific sets.

Reply
 
Thread Tools
Old 27th Nov 2017, 3:45 pm   #1
cmjones01
Octode
 
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 1,711
Default Directional coupler - terminated and unterminated

Inspired by Wendy's recent thread, I have a query about the use of a directional coupler.

The attached fragment of circuit diagram is from the early version of the Tektronix 7L12 spectrum analyser plugin. On the middle left is a connection labelled 'Aux output', J35. This is an output of one of the local oscillators (input via the DC block at the left), used to drive a separate tracking generator.

There is no termination shown on J35, and sure enough, inside my 7L12, J35 is present but unterminated. However, later versions of the 7L12 bring J35 out to the front panel, where it's terminated, and shown as such on the circuit diagram.

My question is, what is the effect of this addition or removal of termination on the other outputs of the directional coupler? Presumably adding a terminator has to absorb energy from somewhere, so the level at the coupler's other outputs will reduce? There appear to be no other changes in the later circuit diagram to take account of this, so perhaps the change in level doesn't matter - it's feeding a mixer anyway?

Chris
Attached Thumbnails
Click image for larger version

Name:	7l12-dir-coupler.jpg
Views:	46
Size:	17.1 KB
ID:	152966  
__________________
What's going on in the workshop? http://martin-jones.com/
cmjones01 is offline   Reply With Quote
Old 27th Nov 2017, 5:18 pm   #2
G8HQP Dave
Dekatron
 
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,017
Default Re: Directional coupler - terminated and unterminated

Changing the termination on the coupler output will affect signal levels and impedances on the other ports. How much will depend on details such as how much power is coupled to the aux output.
G8HQP Dave is offline   Reply With Quote
Old 27th Nov 2017, 11:01 pm   #3
G0HZU_JMR
Octode
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 1,281
Default Re: Directional coupler - terminated and unterminated

Yes, it's hard to say much without knowing more info. But I'd guess the coupling factor will be around 10dB as I think it has to couple out about +5dBm and the LO level at the coupler is probably in the region of +15dBm. I also think a lot depends on the type of coupler and how much effort went into getting good performance.

But I can offer a few ballpark numbers for a typical PCB based 10dB coupler made from coupled transmission lines. Assuming the coupler has been optimised for through VSWR and a 10dB coupling factor then (without a termination at port P3) I'd expect the input return loss at port P1 in the image below to degrade to approx double the coupling factor. So the input return loss would degrade to about 20dB. This is because I'd expect the input reflection coefficient to degrade to 0.1 in this case. This is a fairly insignificant change. This happens because of the reflected signal (at port P3) that is coupled back through the coupler to port P1. These numbers assume the coupler was optimised for input match during the design phase. So the optimised return loss would degrade because of the degradation in the reflection coefficient at port P1 when P3 was left unterminated.

When used in an ideal 50R system I think any change in mainline loss would be tiny/negligible with a typical 10dB coupler. I also think the match at port P2 should be virtually unaffected when port 3 is left unterminated. So the overall impact would be negligible on a typical printed coupler. But this coupler is presumably a wideband design as it has to work over the full range of the first LO and it may be that the coupling factor won't be that consistent over its full range. But it may still hold true that the input return loss could degrade to twice the coupling factor.
Of course, in the real system the coupler won't be in a perfect 50R environment so it will be harder to predict what will really happen without a a termination at port P3. But I doubt it will be anything significant?

But the Tek coupler could be completely different because we don't know much about it? Not all couplers will behave like the example I quoted above, but quite a few will, I think.
Attached Thumbnails
Click image for larger version

Name:	tekcoupler.gif
Views:	22
Size:	29.0 KB
ID:	152981  
__________________
Regards, Jeremy G0HZU

Last edited by G0HZU_JMR; 27th Nov 2017 at 11:13 pm.
G0HZU_JMR is offline   Reply With Quote
Old 28th Nov 2017, 12:23 am   #4
Radio Wrangler
Dekatron
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 9,439
Default Re: Directional coupler - terminated and unterminated

Ah, I just spotted most of "2nd LO" so the LO is probably fixed and the sampler is for a harmonic lock PLL to stabilise it. The printed couplers are cheap, so tek have used a couple more instead of transformers for the sampling diode, so they may well be maximally-coupled 3dB hybrids. As could be the two couplers used to peel off some LO2 for the sampler and to split the drive to the mixer and to the aux output (part of the LO set provided for a tracking generator most likely so the same drive level as the analyser mixer would seem useful.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now   Reply With Quote
Old 28th Nov 2017, 1:57 am   #5
G0HZU_JMR
Octode
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 1,281
Default Re: Directional coupler - terminated and unterminated

I think the coupler/sampler section is there for locking the first LO. I think the 2nd LO stuff will be off to the right of what we can see in the small snippet of the overall circuit.

These analyser plug ins are old school and (relatively) low cost and I'd expect the 2nd LO to be a free running oscillator with a narrow sweep range and it presumably locks the first LO with the sampler on narrow spans and then sweeps the second LO.

I'm going to guess the first mixer runs with about a 10mW drive level so the LO level at the aux coupler will be 10dBm plus 3dB for the T attenuator plus maybe 1.5-2dB for the loss in the two couplers and the cables/connections. So it could be up at about +15dBm at the input to the aux coupler? With a 10dB coupler here it would manage +5dBm at connector J35. But that is all just a guess.

If the coupling factor for all the couplers is as low as 3dB then I'd expect the input VSWR of the aux coupler at the P1 mainline input to be up near 3:1 if J35 was left unterminated. That would be a bit poor and may cause lots of mismatch ripple across the LO1 range.

Also the input to the aux coupler would be +10 +3 + 3 +3 + maybe another 1.5dB = +20.5dBm. The output at the aux connector would then be +17.5dBm when an external connection was made here. That seems quite a high level.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline   Reply With Quote
Old 28th Nov 2017, 7:23 am   #6
Radio Wrangler
Dekatron
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 9,439
Default Re: Directional coupler - terminated and unterminated

I know my way around HP analysers much more than Tek, and theirs with LO outputs rather firmly state that terminations must be fitted when they're not in use.

Economy models like the 8558 plug-in analyser had no tuning stabiliser and a free-running cavity oscillator for the 2nd LO. Better ones had samplers on the YIG and VCXO-multiplier chains for LO2.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now   Reply With Quote
Old 28th Nov 2017, 11:56 am   #7
cmjones01
Octode
 
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 1,711
Default Re: Directional coupler - terminated and unterminated

Thank you for the very comprehensive analysis. Jeremy, your hypotheses about the analyser's architecture are spot on. The 1st LO is a 2.095-3.9GHz YIG, which has phase locking at spans of 100kHz/div and less. The 2nd LO is a 2.2GHz +/-1MHz and is swept only on narrow spans.

I'll try some experiments. Firstly, fitting a terminator to the 1st LO output and seeing if it makes any qualitative difference to the analyser's behaviour. Then I'll try and have a look at the amplitude of the 1st LO output, though the only tool I have for that is a sampling scope so there may not be enough amplitude to get it to trigger. If it works, though, that might give us some idea of the coupling factor of the mysterious component.

Chris
__________________
What's going on in the workshop? http://martin-jones.com/
cmjones01 is offline   Reply With Quote
Old 28th Nov 2017, 2:54 pm   #8
cmjones01
Octode
 
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 1,711
Default Re: Directional coupler - terminated and unterminated

OK, experiments done. Firstly, adding a terminator to J35. 'Before' and 'after' traces with the analyser's 50MHz calibration signal coupled to the RF input. I can't see any difference.
Attached Thumbnails
Click image for larger version

Name:	IMG_20171128_144506.jpg
Views:	9
Size:	66.9 KB
ID:	153014   Click image for larger version

Name:	IMG_20171128_144733.jpg
Views:	9
Size:	66.0 KB
ID:	153015  
__________________
What's going on in the workshop? http://martin-jones.com/
cmjones01 is offline   Reply With Quote
Old 28th Nov 2017, 3:03 pm   #9
cmjones01
Octode
 
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 1,711
Default Re: Directional coupler - terminated and unterminated

Then I tried looking at the output from the 1st LO using my sampling scope. I couldn't get a connector on to J35 because there's only room for a right-angled SMA connector and I don't have one in my armoury of cables. I couldn't get an elbow to fit. I could, however, get the connector on to J38, the output to the first mixer, with the attached result (after an initial schoolboy error: sampling scope doesn't work usefully if the oscillator frequency is sweeping...). I'm not sure if it's legible in the photo but the scope says the signal is 297mV RMS, which I make to be 1.8mW at the scope input, minus 3dB for the splitter to the scope's trigger input. So a 1st mixer drive level of about 5 to 6 dBm, I reckon.

This is without a terminator on J35. Putting a terminator on J35 made no discernible difference to the signal on the scope.

Having spent ages wrestling with SMA connectors on hardline coax in confined spaces to put it all back together, I now find that the analyser has lost 10dB in sensitivity and there's loads of breakthrough from the 50MHz calibration source, even with nothing connected. Aargh.

Chris
Attached Thumbnails
Click image for larger version

Name:	IMG_20171128_150449.jpg
Views:	13
Size:	47.6 KB
ID:	153016  
__________________
What's going on in the workshop? http://martin-jones.com/
cmjones01 is offline   Reply With Quote
Reply

Thread Tools



All times are GMT. The time now is 8:54 am.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2018, vBulletin Solutions, Inc.
Copyright ©2002 - 2018, Paul Stenning.