Experimenting with waveguides using the LiteVNA

[coppercone2]

I thought wave guide loads consisted of a dissipator?

is that thing in the left of the picture you posted a 50 ohm dummy load (5W?) Narda 3something

I thought it was like this thing
https://testparts.com/waveline-wr90-waveguide-termination-dummy-load-x-band/

In a youtube video I think I heard something about carbon, but I am not sure what form.

[joeqsmith]

I thought wave guide loads consisted of a dissipator?

is that thing in the left of the picture you posted a 50 ohm dummy load (5W?) Narda 3something

I thought it was like this thing
https://testparts.com/waveline-wr90-waveguide-termination-dummy-load-x-band/

In a youtube video I think I heard something about carbon, but I am not sure what form.

Oh!!! Ok, I take it you didn't watch my video on calibration and waveguides are new to you.   First, there is no 50ohms.   It's closer to 500.   Like a coaxial load, the waveguide load just needs to absorb the signal.     If you want some idea how they are made, Mario had posted a link in the third post.      That Narda part I show is rated for 1W average.   I would imagine if you put 100W into it, it would change.   I am banking on the cross check I performed with the Agilent to give me some idea they are still all fine. 

Still aging of lossy ferrite or other absorbing materials is interesting.  Not just for the terminator but for attenuators, circulators, isolators...   I bet there have been some very good studies done on this.  I just spent the last several minutes with Google and no luck.   If you decide to contact the Maury Microwave or other standards manufacture, post what you find out here. 

[coppercone2]

well I see some deals on old stuff but I also wonder if they use the same material a a carbon comp resistor for their loads (forgot about the 50 ohm, that info is on first page of the wave guide book) and it is basically horribly drifted now.

it might not even be sealed up nicely like a resistor is, possibly pulled from wet wave guide (the stuff will condense water pretty good IMO). Recpie for disaster with carbon comp I would not be surprised to find 6000 ohms change

Some say silicon carbide though that should be as good as a rock.

Hehe if I am disappointed by a damaged rubber dove tail in this tuner, how will I feel when I find a carbon comp ingot inside of that thing. 

that kind of 'standard' can send you up a tree when you realize you retuned your experiments to run at 2000 ohms impedance or something

But when I hear ferrite and carbon, both can drift really bad. 20% inductor drift with age is not unheard of either for the right types, and carbon comp can be whatever IMO.

i.e. whats in this thing 🤨
https://www.ebay.com/itm/294824491288

when I think about it like that and see how it looks it reminds me of a old suit case, those never work out well. Baking might help but I got some serious suspicions.

[joeqsmith]

The dimensions of the WG should be fine, as long as it was taken care of.  Look at  the precision mismatch load I show in that last video.  Granted, remove that absorbing material and the reflection will increase but to get that mismatch, they change the dimensions of the WG.   The fixed precision Narda terminator uses a conical element that extends the length of the part.  The Narda mismatch sliding load uses several wedges.   

[coppercone2]

I wonder how hard it would be to replace the suceptor in there if it failed. I don't know enough about quality but I did have that experiment that I did trying to make the ceramic silver PCB, if you arrange silicon carbide stones like stone henge in a microwave in fire brick they will glow real bright.

I would if its as simple as getting the correct sized silicon carbide stone and gluing it in a wave guide. (the vitrified ones or whatever).

I still need to redo that experiment, I got actual glass powder extracted from glaze and I need to just mix it with silver in the correct proportion (quartz powder does not sinter well, you need like glass). If it needs ferrite in there then its not easy and you might as well buy a new load.

[joeqsmith]

That is a J910A, where I assume you would want the P910A.  Even then, these are not a reference standard. Maybe you would want the P914A.  Their standard fixed terminator was the 816 but it looks like they only offered this for the X-band.   

I don't know if HP ever published the materials they used, or even if they made these parts.   You could try a patent search.   

Realization of Wide Band Waveguide Terminations at Ku-Band
Pramendra Kumar Verma1*, Raj Kumar
https://ijamt.com/index.php/ijamt/article/download/43/22

[joeqsmith]

Just to give you some idea how the terminators I have compare.   The Narda 300C has a max VSWR of 1.015   So the return loss is  -20Log10(vswr-1/vswr+1) or 42.6dB.    The Narda 410-15 has a VSWR of 1.353 or a return loss of 16.5dB.

I used the extender (new amp and adapters, same coupler)  to sweep from 9 - 12GHz using a IFBW of 1kHz.   I calibrated using my home made standards as before but for those that don't like the error cause by the ideal model I  have used the Agilent model.   This was also taken with the latest revision of the LiteVNA hardware.

Attached plot showing all three standards and also the short.   Also shown are the two LRL507 terminators that were part of the university training kits.    These are just slightly better than the Narda offset load.    Performance suffers above 11GHz but again that coupler is spec'ed to 8.3GHz and its isolation is poor. 

My point is you may not want to buy just any old terminator, if you plan to use it as a reference.

[joeqsmith]

Member Virtualparticles video on VNA accuracy and directivity. 

[coppercone2]

I cleaned a dirty mylar foil type rotary attenuator.

If you are interested, the mylar is fine, and the glue is also fine, after 20 seconds in the ultrasonic for the sensitive assembly. There was alot of crud in there and it looks good now after a DI rinse. The other parts I cleaned between 1 minute to 5 minutes. I rinsed with alcohol spray and after it dries up I will relube the bearings and other stuff with silicone oil or wd40 specialist grease or teflon grease. Drip the fluid silicone grease into the stacked rotating stuff, and wd40 grease in the bearings so it can be cleaned out if I make it/see this thing again in a few decades, and silicone teflon grease on the worm because its easy to clean with a brush.

Looks like they used q-dope or something to hold the mylar sheets in the circular wave guide. Same stuff as inductor wires in other RF equipment that I see 'tacked' on the chassis.

It looked kinda clean but the ultrasonic pot is dirrtttyyyy, and the bearing grease gelatinized a bit.

This one looks like boba fetts ship

and never - dull polishing wadding did a wonderful job on the highly tarnished wave guide flange for the twisted segment.

[coppercone2]

damn, chipped a small amount of mica (0.5mm triangle) (i think its silvered mica, I thought it was mylar initial) from the corner of one of those plates inside the wave guide. I wonder how well that will work  just could not get some scrap of something out of there 

I will wait for the stand to come in before I do testing so I don't damage anything more.

[coppercone2]

Hooking this up is not easy, this will take a while. That attenuator is just a weird object. I need to make some kind of shelf thing that is compatible with lab jacks, make some adapter for the attenuator out of wood so it sits on the lab jack, etc

I need to invest and solve a few problems before I end up in shit creek and have to post a thread about how to fix a snapped APC-7 connector and panel beating a TR set

the rack setup I had was fine for coaxial, but for wave guide its a whole different ball game.

I almost did it, and then I remembered executive decision. And the intro sequence to aliens colonial marines. And I decided to hold off. The worst shit in the world seems to start from shoddy umbilical practices. 


it looks like the damn mounting nuts on my rack mount structures is blocking the shelf from being put on, I need to make some kind of offset hook thing so I can clamp it on a place where there is a shelf nut sticking out.

[joeqsmith]

I'm surprised you have not 3D printed custom stands by now.   

[coppercone2]

the stand is not the problem, the shelf is.

I was confident I could put a rack half shelf on the front of the unit because I forgot that the nuts were in the way. The plan was to hang the rack shelf off the holes in the front of the rack so it can be adjusted to any height, but I am not getting a good face to face connection on the flats because there is shelf nuts in the way, so basically I need like a extender. with hooks on it. It needs to be like a comb that fits between the nuts and makes a true connection

it would look like a lower case pi shape foot thing

And if I make more stands, its SUPER tempting to copy the HP stand that i Need to take pictures of. It looks like a cone. It has enough space in there to put a commercial collet in there for a small adjustable post. Then you just put different forks on top for different wave guides with a universal stand.



I got one, can't find more, but I HIGHLY recommend that design. Its just a shaft that gets tightened into a cone, and the shaft has a hole on top that you tighten a fork into. Easy to make out of brass but my 3d printer is too small to print the base in one piece. I want to do it with some collets I found once that you thread into place, instead of a side screw, but even so that one is easy enough to make with a hole saw, some metal plate aluminum, drill bits and brass rod for the center post, and you can even put it on a turned wood base no problem.


But the uh, attenuator, I think it was pulled from the side wall of a B52 or something, very irregular mount.

[coppercone2]

I made some plasma cutter mods to get the shelf to install nicely (I also will advertise the NOGA brand sheet deburring tools (looks like a rapier handle) as useful here, but unfortunately the ceramic 'non ferrous' one is not good enough to debur the nasty flash from the plasma cut, so you need to use the hard metal (carbide) one then the ceramic one and then light sanding, but its nice)....

but I am still scared to just put the wave guide there resting on a offset because its too hard to tell WTF is going on with the racks. Maybe if they were really sturdy rackss, but these I can tell are bowing etc...

So I bought a few sections of flex wr62. Because I was thinking I need to glue alignment flats and maybe a compass on the parts and start doing some weird shit, and something is gonna break on a assumption of something being true if I do that.

I made this after having a good look at what a transmission reflection test set on a rack mount really is, geometrically

[joeqsmith]

The cast pie shaped ones I have I'm sure would hold a 100+ lbs ea.   I think I have maybe five or six of them.  I like that they have a low profile.  Keeps the WG bits very stable.   

Post a photo of your vintage attenuator. 

[coppercone2]

this is what I managed to cobble together that I am scared to try to plug into the analyzer without a flex relief

[joeqsmith]

WOW!!!  That thing is huge.   Thanks for posting.   

It does sort of look like your photo on the right.

[joeqsmith]

Did you take photos when it was apart? 

Looks like there are both WR62 and WR90 versions of it on ebay.
https://www.ebay.com/itm/295049543429
https://www.ebay.com/itm/294103708332

So your intending to use this as part of a bridge?  How does it fit in to the picture?  You have the 8510.  Is there something you can't directly measure with it?

[coppercone2]

I dunno I just read the lists for various projects and it usually mentions an attenuator or two, really big work in progress. I was just gonna calibrate it when its 0db and see what the wave forms look like when I turn it, and just see how it behaves on 'thru'.

I can take a picture inside after I break it down later but its just a worm gear and a big gear that has the center tube in the middle that is attached to two bearings. I don't want to take it apart until I decouple it, tired.

It has 3 glass? metalized plates inside, shaped like machinist fish tails that nest into each other. All in a line at 0 db and the middle is orthogonal at 55db

I wanted like the basic wave guide kits to do anything even if I can measure it with better equipment, with whatever I will eventually do (Honestly I don't have a solid plan, just scrounging the stuff up till some kind of critical mass occurs). I find usually when there is a.. complete set of parts around, just getting to do something is more spontaneous. I don't expect miracles, just mild amusement and interest.

[coppercone2]

Solid state attenuators are also interesting. Not sure where the state of the art is, but for someone with a bigger budget then me that wants to be more modern, with out worm gears and mechanical computers,

I need to get the picture of the 'computer' there too, its like a helix cut into the indexed aluminum cylinder.

http://dbmcorp.com/broadband-solid-state-attenuators/
https://www.ebay.com/itm/144877536115 (don't think they go that far, think its a passive accessory, but its still a nice idea for a wave guide stand or something)

But that might have cylon problems

I feel like it might be capable of doing AM voice communication on MW with only attenuation. That would be a interesting experiment, similar to that old 'water cooled carbon microphone' that was supposedly used to modulate that big transmitter.

[joeqsmith]

This was my last purchase.  It's a 40mm section of WR90 with UG-39/U flanges.  She has some heft and feels like brass.   I shaved off some coating and metal on the backside and no plating.   Shown next to aluminum with a drop of bleach on each.  It has a few marks I need to lap out but overall in very good condition.   Looks like current scrap price is $25.84 per oz.  My balance is not able to measure it so at least 4 oz. 

[coppercone2]

Try never dull on it, it made my old wave guide look great.

I need a very short section to test it on with the granite plate  to see how much material is being lost through that process though. Its like wet cotton in a can, you tear off a bit and buff it gently. I feel alot better using it them messing with liquid polish + rag.

I think its a mechano-chemical polishing process that uses very fine abrasive and also chemicals. I think they use similar for wafer polishing.. it might be a more advanced polishing process that does less damage. It leaves a wet clear residue, not a milky one, so its something different then just mixing brasso and cotton. I think the abrasive remains stuck in the cotton. If you try to wring it out, you just get clear liquid coming off it, there is no 'mud'.  And no strong odor, brasso smells super pungent in comparison.

I had some silver plated stuff, I found that the only thing that quickly wore through the plating to show copper was the brown (coarse) silicon carbide rubber cups. The green (fine) one had a high level of control, you could definitely stop before you wore through the tarnished plating, with a feather touch. I don't like using the buffs BTW, too messy and hard to deal with the polishing wax sticks, cotton cleaning, etc... the rubber ones are much better if you can afford them.

https://www.amazon.com/Nevr-Dull-NEVER-DULL-POLISH/dp/B00097CRY6

[joeqsmith]

The plan is to resurface the flanges.   I did remove a fair bit of material and repeated the bleach test and once again, black.  Appears there is no plating. 

Using my larger balance, I measured the part at 170 grams.  Density of Silver is 10.49 g/cm^3.   I used water to measure the volume and measured roughly 20cc.  So roughly 210 grams.  A fair bit of error but does seem to point towards some sort of silver alloy.    Would need to run it on a mass spec to see what it actually is. 

6oz at $25.68 or about $150.   Paid about $20 for this one but want it for a sample holder not for scrap.   But that video I showed where the ham was talking about where all the WG materials have gone.  It makes perfect sense that salvage yards would want to maximize profits and scrap makes perfect sense.    I'm glad the part was saved. 

[coppercone2]

they should just stick to scrapping cars, should be plenty with EV roll out, then waste time on esoteric scientific equipment

 I suspect alot of 'road trash' will not be in service anymore since there will be reasonable automobiles for sale

[coppercone2]

how do you plan on holding a tall object up on a lapping plate?

Like, the part is difficult to hold on to without tilting. I often wondered if you could make a 'collar' that fits the object in the middle so you have a big thing to grip for the lapping motion. The taller it is, the more ridiculous it is. I also imagined some kind of stabilizer spider looking thing to go on really big objects, with like loose dampers or something that keep it mostly aligned (like the mechanism of a hone kind of).


I want like those hands from the old andromeda strain movie for this.
https://en.wikipedia.org/wiki/Remote_manipulator

Like imagine trying to do a 2 foot long brass one.

You think if I managed to pick up a MSM-8 at a government auction I could hand lap big stuff with it? Amusing to think about what would happen if we actually got good haptic feedback powered exoskeleton armor suits one day, what kind of ridiculous will result. Like a big ass space marine resurfacing a tractor engine on some crap made out of old sewer grates and 'frontier guard' gear being mis used to make giant gauge blocks and such for sale on the black market. rust belt of the milky way. You think the emperor knows what we are doing with his shit?