From tractorb at ihug.co.nz Mon Oct 31 05:35:15 2016 From: tractorb at ihug.co.nz (Dave Brown) Date: Mon, 31 Oct 2016 22:35:15 +1300 Subject: [Microwave] testing waveguide coax transitions Message-ID: <98B47FC136EA4D1F98055ABE932D7F5C@athlon3200> I have several w/g coax transitions (WR75, in this case) that I want to test for possible use on 10 GHz. They are ex 13GHz equipment. As I see it, I need a well matched load on the w/g port to sweep the transition, while looking into the coax port. I have to test them this way as I dont have anything I could use to test them looking into the w/g port. Question is- what should I use for the low SWR load on the w/g port? To properly characterise the transition alone, a known low SWR load in WR75 (at 10.3 GHz) is what I need. I have one or two w/g loads in WR75 that I could use but they are ex 13 GHz equipment and I dont know if they would have low SWR at 10.3 GHz. But I need a transition with known characteristics at 10.3 GHz to check the w/g loads....This seems like a chicken and egg situation! Similarly, a small horn with WR75 feed might have good enough return loss to be used as the load on the transition, but again, its a matter of getting a known transition in WR75 to check the horn first! Suggestions etc welcomed. 73 Dave, ZL3FJ From geraldj at netins.net Mon Oct 31 15:50:52 2016 From: geraldj at netins.net (Dr. Gerald N. Johnson) Date: Mon, 31 Oct 2016 14:50:52 -0500 Subject: [Microwave] testing waveguide coax transitions In-Reply-To: <98B47FC136EA4D1F98055ABE932D7F5C@athlon3200> References: <98B47FC136EA4D1F98055ABE932D7F5C@athlon3200> Message-ID: <5817A09C.3000801@netins.net> It is hard to predict how a WR75 transition and load made specifically for 13 GHz will work at 10 GHz though 10 GHz is in the standard working range of WR75. The match depends so much on the probe location and dimensions, and its easier to make it narrow band that full waveguide band. Some of the WR75 transitions I have acquired came with a couple tuning screws. Since most of our use is very narrow band, even bad mismatches can be corrected with tuning screws without imposing a bandwidth limitation. Those I have used for experiments so far like taking a feed horn to MUD or CSVHF for antenna range testing and I have just used a directional coupler as a load bridge (in coax or waveguide depending on the part being tested) and with a power meter of the return power or a detector and HP415E SWR meter and a 1kHz modulated signal source, tuned the screws for minimum reflected power, or greatest return loss (same result different names). The directivity of the directional coupler can affect the final result to some extent but for ham purposes most are good enough. On those transitions the screws appear to be about #4-40 (about .110" diameter or 2.7mm) diameter and are spaced 1/2" and 3/4" from the probe. They have Farinon part numbers. A WR90 load would probably match well, especially if there was a tapered guide 3 or more wavelengths long between the WR90 and the WR75 port. i want to do some experiments butting WR62 to WR90 and have the same load quality problem, I have the parts to build a WR62 slotted line so I can check the load quality at 10.368 GHz in WR62 and then change to my WR90 slotted line to use the old fashioned vector circuit analyzer to look at the effects of the butted connection and the alignment of the guides, centered or corners aligned or faces aligned to see if any is workable or better than the other possibilities. I learned using slotted lines a long time ago and they still work, just don't make 1000 measurements per second over a 7 GHz frequency span like a modern vector network analyzer but the good modern ones cost as much as a house in the country and I'd rather have the house. 73, Jerry, K0CQ On 10/31/2016 4:35 AM, Dave Brown wrote: > I have several w/g coax transitions (WR75, in this case) that I want to > test for possible use on 10 GHz. They are ex 13GHz equipment. > As I see it, I need a well matched load on the w/g port to sweep the > transition, while looking into the coax port. I have to test them this > way as I dont have anything I could use to test them looking into the > w/g port. Question is- what should I use for the low SWR load on the w/g > port? To properly characterise the transition alone, a known low SWR > load in WR75 (at 10.3 GHz) is what I need. > I have one or two w/g loads in WR75 that I could use but they are ex 13 > GHz equipment and I dont know if they would have low SWR at 10.3 GHz. > But I need a transition with known characteristics at 10.3 GHz to check > the w/g loads....This seems like a chicken and egg situation! Similarly, > a small horn with WR75 feed might have good enough return loss to be > used as the load on the transition, but again, its a matter of getting a > known transition in WR75 to check the horn first! Suggestions etc welcomed. > 73 > Dave, ZL3FJ > > ______________________________________________________________ > Microwave mailing list > Home: http://mailman.qth.net/mailman/listinfo/microwave > Help: http://mailman.qth.net/mmfaq.htm > Post: mailto:Microwave at mailman.qth.net > > This list hosted by: http://www.qsl.net > Please help support this email list: http://www.qsl.net/donate.html > > From k0xp at k0xp.com Mon Oct 31 22:11:14 2016 From: k0xp at k0xp.com (Steve) Date: Mon, 31 Oct 2016 22:11:14 -0400 Subject: [Microwave] testing waveguide coax transitions In-Reply-To: <5817A09C.3000801@netins.net> References: <98B47FC136EA4D1F98055ABE932D7F5C@athlon3200> <5817A09C.3000801@netins.net> Message-ID: <0M6UVZ-1cow1129ql-00yTa8@mrelay.perfora.net> At 03:50 PM 10/31/2016, Dr. Gerald N. Johnson wrote: >It is hard to predict how a WR75 transition and load made >specifically for 13 GHz will work at 10 GHz though 10 GHz is in the >standard working range of WR75. The match depends so much on the >probe location and dimensions, and its easier to make it narrow band >that full waveguide band. You could bolt several of the WR75 transitions together then sweep them through 10 GHz. If they show relatively low loss, you could assume they most likely have fairly low loss... at least, lower than, say, -6 dB at 10 GHz. Personally, I would not trust a transition showing more than perhaps 0.3 dB loss (individually) at 10 GHz. In fact, I would not trust a WR-75 transition at 10 GHz without full characterization. I would assume it'd present a few dB, at the minimum, additional loss down at 10 GHz, and I'd plan accordingly. Instead, I would look for WR-90 transitions, which are rated 8 - 12 GHz and perform as such. Here in the States, such are easy to find, but in New Zealand, I suppose they're much more difficult to find and acquire. Transfer of Technology issues may apply, I dunno. SteveH From geraldj at netins.net Mon Oct 31 22:39:27 2016 From: geraldj at netins.net (Dr. Gerald N. Johnson) Date: Mon, 31 Oct 2016 21:39:27 -0500 Subject: [Microwave] testing waveguide coax transitions In-Reply-To: <0M6UVZ-1cow1129ql-00yTa8@mrelay.perfora.net> References: <98B47FC136EA4D1F98055ABE932D7F5C@athlon3200> <5817A09C.3000801@netins.net> <0M6UVZ-1cow1129ql-00yTa8@mrelay.perfora.net> Message-ID: <5818005F.5080309@netins.net> On 10/31/2016 9:11 PM, Steve wrote: > At 03:50 PM 10/31/2016, Dr. Gerald N. Johnson wrote: >> It is hard to predict how a WR75 transition and load made specifically >> for 13 GHz will work at 10 GHz though 10 GHz is in the standard >> working range of WR75. The match depends so much on the probe location >> and dimensions, and its easier to make it narrow band that full >> waveguide band. > > You could bolt several of the WR75 transitions together then sweep them > through 10 GHz. If they show relatively low loss, you could assume they > most likely have fairly low loss... at least, lower than, say, -6 dB at > 10 GHz. Personally, I would not trust a transition showing more than > perhaps 0.3 dB loss (individually) at 10 GHz. In fact, I would not trust > a WR-75 transition at 10 GHz without full characterization. I would > assume it'd present a few dB, at the minimum, additional loss down at 10 > GHz, and I'd plan accordingly. A curve here in a Mcrowave Journal handbook and buyer's guide from 1966 shows WR90 oxygen free copper wave guide has 3.4 dB loss per 100 feet at 10.0 GHz. WR75 it shows at 5.2 dB loss per 100 feet. .034 and .052 per foot, or about .003 and .0043 dB per inch. Not much waveguide loss in the typical inch or two inch long transition. The big question is how the probe is tuned whether its a waveguide bandwidth probe or unique to 13 GHz. The recommended frequency range for WR75 is 10 to 15 GHz and the cut off frequency is 7.868 GHz so 10 is fairly far from cutoff. Tuning screws won't introduce much loss and could help the match and the loss. > > Instead, I would look for WR-90 transitions, which are rated 8 - 12 GHz > and perform as such. Here in the States, such are easy to find, but in > New Zealand, I suppose they're much more difficult to find and acquire. > Transfer of Technology issues may apply, I dunno. WR90 has been in use a very long time but it has potential mode problems at 13 GHz as used for satellite uplinks, hence the popular use of WR75. WR75 hardware like transitions, feed horns, LNBs, and switches is easier to find today than WR90 though I have some of both. I'm thinking of mixing them to use pieces of a 12 GHz LNB retuned as a 10.4 GHz LNA with WR75 waveguide while using WR90 on the transmite and antenna sides and a WR90 switch for the TR relay. So far my coaxial LNAs have not proven to work well. Most of the 11 to 13 GHz satellite hardware is WR75 and its known to work decently at 10 GHz. 73, Jerry, K0CQ > > SteveH > > >