SWR Mismatch Unit?

Everything technical about radio can be discussed here, whether it's transmitting or receiving. Guides, charts, diagrams, etc. are all welcome.
Post Reply
s2000

SWR Mismatch Unit?

Post by s2000 » Wed Jan 09, 2019 6:32 pm



Hi, this video has been on here before but there's a particular instrument that the guy uses (@ 1.36), the little gold box that they have made to test the mosfets. I would like to know how to make a similiar device?

I'm not looking to blow up expensive rf devices, so it wouldn't need to go up to swr's of 125:1! (lol). Something more gentler like 1:1 to 4:1 would be fine... so I can test the accuracy of some homemade swr monitoring circuits I am making.

I did once see a small photo of inside one of these units (although I cant find it now), it just appeared to have a variable capacitor and a coil in some kind of tuned resonant circuit. Apart from that, I cant find anything on the net which relates to the subject.

Anyone ever made such a device?

radium98
proppa neck!
proppa neck!
Posts: 907
Joined: Fri Aug 26, 2016 7:01 pm

Re: SWR Mismatch Unit?

Post by radium98 » Thu Jan 10, 2019 7:26 am

https://iw1cgw.wordpress.com/2016/01/04 ... fet-e-fet/

prova mosfet good luck with the label ,i am building one seems good ,455khz osc + buffer not completed yet .

Albert H
proppa neck!
proppa neck!
Posts: 2737
Joined: Tue Apr 05, 2016 1:23 am

Re: SWR Mismatch Unit?

Post by Albert H » Thu Jan 10, 2019 10:06 am

At TRW in Bordeaux, we had a bunch of 1 kW resistive dummy loads with various LC combinations added to give mismatches at various phase angles and in various amounts.. We most often blew up the deliberate mismatching capacitors and inductors (rather than the PAs or the loads themselves). When we developed the twin TP9383 amplifier (truly broadband 300W output at 28V supply and 30W drive), we tested it into every possible mismatch and only destroyed the transistors on the prototype board when we added a persistent short to ground!

The 9383 twin was the first "pallet" amplifier that I ever designed, and is still quite widely used today - almost 40 years later. It's gain wasn't spectacular, but the robust nature of the devices used made it close to indestructible in normal use. It was ideal to be paired for higher power, but the price of the devices was rather high.

Ten years after its design, I worked with my first "big" VHF FET - the MRF150 - which was also developed at TRW. That device was a real "game-changer" since we could develop valve-like gain and the device had a negative temperature coefficient which meant that (unlike conventional transistors) as it got hotter, it drew less current, providing a self-limiting protection. These were also difficult to destroy, unless you put too much DC or RF into the gate!

I have the drawings for the mismatched dummy loads here somewhere - I'll see if I can find them and put a couple up on here.
"Why is my rig humming?"
"Because it doesn't know the words!"
;)

s2000

Re: SWR Mismatch Unit?

Post by s2000 » Fri Jan 11, 2019 7:33 pm

Thanks for the input guys, I did manage to find the document I was mentioning earlier with the mystery box in...

https://www.microsemi.com/document-port ... er-mosfets

They also got a video



Something interesting in that video, they've managed to somehow mount the mosfets with hardly any soldering on the tabs. Must be for easy replacement when they eventually blow up (off camera of course) :lol:

I did manage to homebrew a mini swr mismatch unit in the end, although it took a lot of experimenting. Mine goes from about 1.06:1 all the way up to well over 10:1. Perfect for any low power testing. I'd imagine it could be quite useful for checking the accuracy of homemade swr meters too (when inline with a pro swr meter for reference), something I plan to do one day. Here it is (with a few mr blobbys)... :lol:
20190111_192135.jpg
It's been interesting making this unit, it's a good example of how you can have a perfect impedance of 50 ohm (well 47 in this case) but due to reactive L and C elements, the swr can be bad! I'd imagine there are an infinite number of ways to get a given swr reading! :shock: I am sure a vna would be the best way to illustrate the loads characteristics , shame they are so expensive! :ugeek:

I will see if I can make a high powered version of this now I get the drift. I was thinking it'd probably be possible to use a high powered dummy load/attenuator as the main resistor. I have a bird 8325 30db attenuator, that can be a dummy load with a 50 ohm termination on the output end. At the input, a "n type" tee piece could be fitted with one end going to a screened box with a high voltage adjustable L-C combination. The other free end of the tee, straight to the transmitter.

It may be possible to go one better and take the 50 ohm termination off the attenuator and feed it into the analyser. Hey presto, a swr mismatch unit to test rigs (on the fly) with various unfriendly loads :mrgreen:

That TRW was an advanced company, I have some old TRW TP9382's here somewhere. I have never managed to find much specs about them but as far as I know they're good for 175W out (hopefully at 100Mhz)?. I might try building that amp one day, it was featured heavily on that pirate documentary, so it must be a classic! :tup
You do not have the required permissions to view the files attached to this post.

Albert H
proppa neck!
proppa neck!
Posts: 2737
Joined: Tue Apr 05, 2016 1:23 am

Re: SWR Mismatch Unit?

Post by Albert H » Sat Jan 12, 2019 12:13 am

S2000 The TP9382 was rated at 175 Watts RF out and was designed for Band II. It's a 28V device (you can get more gain at 30V, but be sure of the output match). I built a rig for a north-west London station in the late 80s with a 9382 running at 31V and delivering slightly over 200W. They had a well-installed colinear aerial up the Harrow Road, and the signal was huge over most of the city. That rig lasted about 4 months, until Eric collected it off the block.
"Why is my rig humming?"
"Because it doesn't know the words!"
;)

s2000

Re: SWR Mismatch Unit?

Post by s2000 » Sun Jan 13, 2019 8:34 pm

Thanks for the clarification Albert, that's what I was hoping! :D

I did see one of those amplifiers on Ebay (a while back). The guy had used two BLW77's for 250W out. Here it is...
$_3-1.jpeg
Can you tell me, what is the correct dimensions of the pcb outline? I will have a go and build one sometime. :tup
You do not have the required permissions to view the files attached to this post.

Albert H
proppa neck!
proppa neck!
Posts: 2737
Joined: Tue Apr 05, 2016 1:23 am

Re: SWR Mismatch Unit?

Post by Albert H » Sun Jan 13, 2019 11:49 pm

Shall we start a new PA designs thread?
"Why is my rig humming?"
"Because it doesn't know the words!"
;)

s2000

Re: SWR Mismatch Unit?

Post by s2000 » Mon Jan 14, 2019 5:22 pm

Good idea, I will start one when I get this project finished :tup

User avatar
sinus trouble
proppa neck!
proppa neck!
Posts: 1410
Joined: Fri Aug 22, 2014 11:34 pm

Re: SWR Mismatch Unit?

Post by sinus trouble » Mon Jan 14, 2019 11:12 pm

Speaking of designs, I just downloaded the latest LTSPICE!
Only done audio amplification simulations as yet, Not sure if it will be suitable for RF?
I am as stupid as I look! :|

nrgkits.nz
Neckmin
Neckmin
Posts: 337
Joined: Fri Oct 17, 2014 10:35 am

Re: SWR Mismatch Unit?

Post by nrgkits.nz » Tue Jan 15, 2019 9:25 am

sinus trouble wrote: Mon Jan 14, 2019 11:12 pm Speaking of designs, I just downloaded the latest LTSPICE!
Only done audio amplification simulations as yet, Not sure if it will be suitable for RF?
It eats up a lot of CPU the higher you go in frequency and simulation time.

nrgkits.nz
Neckmin
Neckmin
Posts: 337
Joined: Fri Oct 17, 2014 10:35 am

Re: SWR Mismatch Unit?

Post by nrgkits.nz » Thu Feb 21, 2019 3:16 am

Looking again at the first video, I don’t believe them that their LDMOS transistor is as rugged as they say it is, it’s likely they have had a few blow up off camera - look at position 2:29 in that video where the guy is trying into his mismatch test box, you can see the supply voltage is 50 and the current is fluctuating from about 1.5amps to no more than about 10amps, that’s only a maximum of 500watts and doesn’t match up anywhere near what that LDMOS transistor was delivering into the 50ohm load at just over 1kW.

Post Reply