2.4mm 50GHz coax cables

I have an HP 70820A with V connectors. I just bought a few V-to-SMA adapters to use as connector savers.

Cheers

Phil Hobbs

I'm not a GHz type, but would semi rigid coax work for you? George H.

You can get awesome used microwave cables cheaply on eBay.

--
 Thanks, 
    - Win

Buy cable and connectors and make up your own.

Element 14 - was Farnell and Newark - have an extensive ranges of cables and connectors.

You will probably need to buy a crimp tool to put the connectors on the cables. They aren't cheap but can get quite a lot of cable and quite a few connectors for $2000.

And why 2.4mm OD? The characteristic impedance matters, but the diameter is usually determined by the connectors you want to mate with.

You really do need to tell us what you are trying to do with the cable.

And George Herold is right - semi-rigid coax gives you the best quality cable, but it is expensive.

--
Bill Sloman, Sydney

Thanks for the info.

The 70820 is a 40 GHz instrument, so you need 2.92mm cables. These will be a bit less expansive than 2.4mm cables. Where do you get them?

Thanks. I've tried but no luck. EBay lists anything that has 2.4 in the title, but nothing to do with 50 GHz coax cables. Same problem with Amazon, and a lot of cruft with Google.

I wondered about the 50 GHz in the subject line...

Now I understand where you got it, apparently the text written by some marketing people :-).

While 50 GHz might be the upper cut-off frequency for a 2.4 mm coaxial cable, since at higher frequencies only some waveguide propagation modes might occur. But what is the dielectric losses at 50 GHz ?

Steve Wilson wrote in news:XnsAB4D1BD83358idtokenpost@

144.76.35.198:

That operate at 50GHz? Wow.

Out to 20GHz is common. But 50?

Won't be cheap.

Winfield Hill wrote in news: snipped-for-privacy@drn.newsguy.com:

A lot of mil surplus stores have spools of it too. At least the one in El Cajon did.

If you want to do your own terminations. Depends on what your tool library is comprised of. Crimpers are not cheap. If you are making a bunch though, might be cheaper. Getting 50GHz performance is a lat to ask. The places that sell them to you that meet that spec likely culled it out of a batch of ten with only a few from each batch making the grade. Until you form or bend it into shape. Then all bets are not only off, but they went out the window.

Spurious.

Bill Sloman wrote in news: snipped-for-privacy@googlegroups.com:

No shit. All us dummies here in sed thought they went together like snap together model cars from the seventies.

Don't forget to mention how hard it will be to get 50GHz acceptable performance. That is what he asked for.

Bill Sloman wrote in news: snipped-for-privacy@googlegroups.com:

There are standard form factors for coaxial cables. That way actual engineers can incorporate them into a chassis and know what displacement they and their connectors will have.

.141 .250... Those are Imperial of course. The .141 is common in high frequency applications. That is 3.6mm. 2.5 is .1 imperial. The closest the industry makes for that sizing is .085 inch. or 2.16mm.

No, it actually has APC-2.4 connectors. V connectors fit.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC / Hobbs ElectroOptics 
Optics, Electro-optics, Photonics, Analog Electronics 
Briarcliff Manor NY 10510 

http://electrooptical.net 
http://hobbs-eo.com

Where do you get the cables?

I just leave the adapters on the box and use SMA cables. I don't do anything above 26 GHz, so they work fine. (The 70820A doesn't get that much use anyway.) I forget where I got the adapters, but they weren't cheap either--$120 for the pair or thereabouts. Pasternack wants $300 apiece. I only paid $900 for the whole instrument including the mainframe, so I wasn't about to shell out as much again for connectors.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC / Hobbs ElectroOptics 
Optics, Electro-optics, Photonics, Analog Electronics 
Briarcliff Manor NY 10510 

http://electrooptical.net 
http://hobbs-eo.com

Thanks. I have the same problem, but I need to go to 50GHz. Maybe CentricRF is the solution, but I need to find out what grade they are offering.

Next problem is a 50GHz prescaler. Pacific Microchip sells a macro for a

50GHz ASIC, but that sounds like a risky way to go. InPhi used to sell a 50GHz D-flop and prescaler, but I can't find them on their website any more.

They do design a high speed TIA that might interest you.

Thanks for your help.

Seems like a good deal, except for the part about being 4 inches long. :(

For the cables, you could search on eBay for the specific HP part # that has the fittings you want. E.g., one good part number I use with an

8517B / 8510C is 85134-60003. These are convenient because the 3.5mm hardware at the other end will mate with an SMA F-F barrel.

On my 70820A I use 2.4mm / 3.5mm adapters. $100 will get you good-quality adapters from various sellers. If you insist on a 50 GHz-rated cable, as opposed to adapters that you can "get by" with, it may hard to find anything new or used for less than a kilobuck.

-- john, KE5FX

A 50 GHz signal won't travel far over coax!

I've wondered how useful a 100 GHz sampling oscilloscope can be. It's hard to get a signal to it.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

(

That depends a bit on the insulator. Vacuum insulated coaxial cable should offer relatively low losses, and if you used a high temperature super-condu ctors for the conductors, they'd be even lower.

Portable oscilloscopes can be taken fairly close to the signal. I once ende d up using a time domain reflectomenter halfway up the side of a large supe rconducting magnet. Some of it's wiring had gone open circuit, and we were trying to work out exactly where. The guy who had had the idea had one leg blown off by a land-mine in 1946 (when he was about nine) and wasn't up to doing the clambering required to test it.

--
Bill Sloman, Sydney

Thanks for the reply.

Here's 6, 12 and 18 inch cables. All 50 GHz, all $150:

Yes, that's a problem. There are different grades of cable. The Metrology Grade is the most expensive, followed by the Instrument Grade. These are probably the kilobuck cables:

Some versions cannot be mated with others without damage:

I don't need Metrology Grade or Instrument Grade, but vendors rarely tell you what grade they are offering.

Fortunately, I only need short lengths, and because they are short, ordinary Production Grade will be fine. I suspect this is what CentricRF is selling, but I'll still email them and ask.