Experimenting with waveguides using the LiteVNA

[joeqsmith]

Adjusting the attenuator from 24dB and normalizing the data with the attenuator set full open.   

[joeqsmith]

Microwaves Part VI - Test Equipment & Frequency Conversion
October 1949 Radio-Electronics Article
https://www.rfcafe.com/references/radio-electronics/microwaves-part-vi-radio-electronics-october-1949.htm

Microwaves Part IV - How Waveguides Are Joined and Tuned
August 1949 Radio-Electronics Article
https://www.rfcafe.com/references/radio-electronics/microwaves-part-iv-radio-electronics-august-1949.htm

***
Another interesting article after bands were provided for amateur use.   The attached photo is from the article showing a home made waveguide.   I thought mine were bad.

Microwaves for the "Ham"
August 1952 Radio & Television News Article
https://www.rfcafe.com/references/radio-news/microwaves-ham-august-1952-radio-television-news.htm

[joeqsmith]

As I continue to search for used waveguide parts, many in poor condition, I wonder about lapping their mating surfaces.  Seems like with a flat surface and a bit of wet paper, you could restore their flatness.   

For coaxial adapter storage, I have a wooden box that I drilled out with various sizes to keep them separate.   For the smaller ones, I use ammunition cases (very low cost).   With the waveguides, I see most new parts are supplied with protective caps.  I am trying to locate a source for these caps but so far my Google foo is letting me down.   

https://edadocs.software.keysight.com/kkbopen/how-can-i-order-replacement-protective-caps-for-waveguide-flanges-577940541.html

[coppercone2]

well since its soft

[joeqsmith]

I haven't checked for a flatness spec yet to see what would be required.  Profilometer? Shinny != flatness. 

I did put a video together showing some of what I have been doing with them.   

[coppercone2]

thats looking really nice, I want to play but I need to finish a few repairs because I am annoyed at having too many equipments that have partial function and or are on the fritz . Feel so much better working in the lab after the equipment has been gone over instead of trying to remember which displays are lying, which knobs don't work right  And I need to make a wooden case for some RF stuff. Made like 4 storage boxes last year but I need more.

[joeqsmith]

Not bad when you consider that my investment for all the waveguides shown so far cost me maybe $100 and that was mostly connectors.   Nothing yet deserving of a nice home made wooden box.  Instead they all fit in a Digikey cardboard box with the original paper packing.      

[coppercone2]

I make them out of a enduring stockpile of old cabinet doors etc

[joeqsmith]

All the better when the materials are free.   

Post a few photos of your work. 

[joeqsmith]

Applications of waveguide and circuit theory to the development of accurate microwave measurement methods and standards
Beatty, R. W.

https://nvlpubs.nist.gov/nistpubs/Legacy/MONO/nbsmonograph137.pdf

[joeqsmith]

I have ordered new matching pads to try and improve the dynamic range or our extender along with a new sheet of brass to make one last waveguide for the next set of experiments.   

Someone had posted a comment about The Signal Path channel making a video on 3D printed waveguides.  The attached link takes you to what I believe they were referring to.   He made a few horns and a filter.   He's experimenting at much higher frequencies than I am and surface finish.... I would suspect will be much more problematic.   Where I had posted a fair bit of data showing the effects of adding my PCB horns,  we don't get see anything like that with his experiments.   
 
https://youtu.be/bo8pQmKgwgw?t=1966

I would think in the X-band where I am experimenting, plastic 3-D printed parts could be an option with a bit of prep work before coating.  Then again, I'm not sure about tuning them.  I couldn't find where he made a follow on video but certainly an interesting topic. 

[joeqsmith]

Interesting article on 3D printing horns.   They are using solvents, epoxy and files to try and improve the surfaces.  Shown with a professional 3D rotor and chamber.   

https://antennatestlab.com/3dprinting

Most of the articles I have found on printing waveguides, use metal.   

Looking at designs people have uploaded (I was surprised to find any!),  this one is a student.  Maybe we will see some test data from them in the future.
https://www.thingiverse.com/thing:1290578

[joeqsmith]

Here's the last homemade waveguide I plan to construct using these techniques.  Not a bad use of a c-note.   

[joeqsmith]

Another possible coating for printed parts.

https://sandstromproducts.com/products/conductive-coatings/

[joeqsmith]

Test horn #1.  I may need to do a part 3 where we have a horn shootout.   Expensive production vs homemade PCB vs 3D printed with various coatings....

[coppercone2]

oh they have a bunch of prints for this one now
https://www.thingiverse.com/search?q=horn+antenna&page=1&type=things&sort=relevant

I have the MG chemicals nickel spray paint that I was going to do this with one day.

[joeqsmith]

The Signal Path video shows them using MG's copper silver.   They offer an even more conductive coating. 

Hone#1 was from that first link.  One was a WR90 which is what I show.   I saw some on thingverse where they had printed the transition with the horn.   I'm not sure how you would coat something like that as one solid part.   Even that WR90 horn I show isn't the best solution for coating.   You have to somehow apply a coating inside the waveguide. 

[coppercone2]

thin it down and dip coat, or print as 2 pieces and coat them and then glue together with conductive epoxy over the coated flange., it should be OK with the high conductivity MG chemicals epoxy maybe. And give it some lashings on the outside to hold it together with fiberglass or something

https://www.amazon.com/MG-Chemicals-Conductive-Extreme-Conductivity/dp/B018AAT4VC

you got a gallon of that paint right? And my experiance with dip coating is just bad, like with varnish for transformers, it ends up looking like crinkly ass. Maybe if you made a hoist that uses a slow motor to actually slowly remove it from the liquid... I think the idea is that the layer of the paint in the dip coating tank is like a... accumulator of excess paint on the surface of the material, so if you take it out really slowly, it 'wicks' the excess off and keeps it even. Very hard to do with just your hands even for a small object, you need at least I think a pully and a spool to slowly lift it up. I think various dip coating guidelines say like.. 6 inches per minute (thats for red insulating varnish). If you can some how thin the paint down to be like laquer maybe you can use that as a base line. IDK how well conductive coating thin or if they even work after thinning though. And I think you want to align the object so its not pendalum motion during retraction and its very parallel to the liquid layer.

need the pizza guy to figure this one out

[coppercone2]

actually for dip coating a small object, gluing a dowel to it and putting it in a full sized drill press might do the business. I mean it won't be perfect but you can retract the quill rather slow without building anything, and it has a adjustable table. If you had a very slow power feed on the drill press that might be close to ideal?

Maybe putting a wall clock against the handle to follow the second hand would work. Or a projection of a clock or something, like tie a laser pointer to the handle so it aims at a wall clock and rotate the handle so the laser follows the seconds. 

Or maybe there is no upper limit and 1 minute per 6 inches is just a reasonably fast number, so go as slow as possible.

I tried with pliers and plasti dip and transformers + varnish and usually it came out like total garbage, with transformers I don't care too much though, I just textured it a little with a towel. I associate the process with mega bad smells, a mess, general failure and disappointment.

[coppercone2]

btw does that stuff even paint well? I had bad luck with painting 3d prints, it was kind of dodgy. I ordered a special plastic primer once but they said it was out of stock after a month and I gave up

and maybe a good way to resurface the object is instead of lifting it, to keep it suspended, and then drain the container with a small hole, the level should drop consistently .

[joeqsmith]

Some of the printed WGs I looked at snapped together allowing you to apply your coating inside the tube.  The epoxy you listed may be a better solution to joining them.   

The datasheet I linked has a section on application.  It's thinned with water and applied with a gun.   That's a one pint sample container.  Doubt it's any good now with the limited shelf life.  I plan to check it today.  Looking at production parts, some are plated copper.  In some cases with silver.  I image if you attempt to hone the mating surfaces, you may remove the plating.   

After playing around the last few weeks, I think at these lower frequencies, we can get away with fairly crude practices.  You're not going to get much worse than what I have shown so far.   Then again, I've currently got the grinder out.  Not a very scientific approach.      

[joeqsmith]

Waveguide flange dies:   
https://www.ebay.com/itm/144459117195?hash=item21a26f328b:g:kDAAAOSwJoViLOIg

Horn:
https://www.ebay.com/itm/144362628892?hash=item219caee71c:g:RyIAAOSwMMZgkyPR

Looks like they have several components that would have been supplied in the university training kits.  Here's a cavity, brand new, plated but not complete.
https://www.ebay.com/itm/144364961251?hash=item219cd27de3:g:IccAAOSwY45UL~nr

[joeqsmith]

Coppercone2 = coppercone? 

Doing a search on EEVBLOG for other threads on the topic:
https://www.eevblog.com/forum/rf-microwave/waveguide-high-frequency-(past-rating)-behavior/

[coppercone2]

yeah changed my email or something, i forgot

[joeqsmith]

Wanted to provide a little more data for my poor man's frequency extender for the LiteVNA.   I started out measuring the LO with my counter but then a homemade slotted line to measure the extended Port 1.  I also repeated my stepped attenuator test to prove out the input stage.  There are a few other experiments along the way.