>-----Original Message----- >From: Ken Javor [ mailto:[email protected]] >Sent: Tuesday, November 04, 2003 4:03 PM >To: Bill Stumpf; '[email protected]' >Subject: Re: RS-103 help > > > >I am not sure if you are asking for specific equipment model >numbers, or >general guidance. I will leave the specifics to others. I >would start with >a high gain horn and a calculation of how far away the far field is >developed (2D^2/wavelength). You can use the equation for >power density as >a function of gain, transmitted power and distance to figure >out how much >power you need at 1 meter or 2D^2/wavelength. Then you go >looking for that >TWT power amp. The reason for this approach is that those amps are >expensive and hard to find. I don't know if you could rent >one at all. If >you could find a 1 Watt TWTA, that would work as long as you >are satisfied >with 2D^2/lambda and not aiming for 1 meter. Also the horn >bolts directly >to the TWTA waveguide output. You probably can't stand the loss of >interconnect between signal generator and amp, unless you >plumb the whole >thing with rigid waveguide, which would be painful. I >recommend the test >operator be right in the room with the test sample. I bet >I'll draw some >flack for that, but that beam is so directional I wouldn't >worry about it. >If you are worried about it, build a bunker house out of spare foam >absorber. Also if you are very close to the EUT there are a >couple issues. >One is VSWR, you don't want reflected power splashed back into >the horn, so >you cant the angle slightly (angle of incidence equals angle >of reflection) >so that the reflected beam is diverted from the horn aperture. A second >issue is that with a small illumination spot size you have to >have a lot of >antenna positions to paint the entire test set-up. >MIL-STD-5461E covers >that. if you were going to do this all the time it would be >worthwhile to >plunk down $150K for the 40 Watt amp and get a lower gain horn >back at 1 >meter and spray the whole test set-up to save time. Like >everything else, >that decision comes down to time or money. > >> From: Bill Stumpf <[email protected]> >> Reply-To: Bill Stumpf <[email protected]> >> Date: Tue, 4 Nov 2003 14:21:20 -0800 >> To: "'[email protected]'" <[email protected]> >> Subject: RS-103 help >> >> >> Hello all. I need help locating equipment for rent that can >achieve the >> following goals. RS-103 (MIL-STD-461E) testing from 18 GHz >to 40 GHz @ >> 200V/m. If anyone could point me in the right direction I >would be most >> grateful. >> >> Bill Stumpf >> >> William M Stumpf >> DLS Electronics >> 166 South Carter St. >> Genoa City WI 53128 >> ph: 262-279-0210 >> fx: 262-279-3630 >> email: [email protected]
Ken hits all the important points, so let me just amplify <g> on a few of them. I use a 20W TWT amplifier from 8GHz to 18GHz, with an EMCO 3115 double-ridged horn antenna. Since I only go up to 18GHz, I stay in coax. However, I mount the antenna to the coax output of the TWT, to minimize cable loss. I use a long coax (low-power rating, cheap, but very low loss, to run from my external signal generator to the TWT input. This makes best use of the available power, even if the amplifier is within the chamber. My amps are 20W Hughes (not available any more), but check out IFI and Amplifier Research. But once over 18GHz, the best way to go is waveguide. Build up a collection by watching eBay! And a couple of short sections of flex waveguide greatly simplifies your setups. If you have the time, you can easily build a pair (18GHz to 26GHz & 26GHz to 40GHz) of pyramidal horns from a section of waveguide and a few sheets of copper (to make the flare). I think you might be able to do the job with 5W TWT's, but 10W units would give you a better margin. I'd guess amps would be in the $12,000 range. And you'll also need a signal source, and maybe PIN diode modulators. You might be able to find an old Gigatronix signal generator, with its external frequency doublers, on the surplus market. Signal sources to 40GHz are either very expensive or science projects. I don't like to use the calculated method of predicting RF exposure; I prefer to stick a bolometer sensor right into the real field. For a sensor, I use the IFI EFS-5 with an external Narda probe (covers 300MHz to 40GHz). New cost is about $4000, but I bought 2 of my EFS-5's off of eBay ($65 & $95 each, hah!). Add another $1000 for a remote indicator and some fiberoptic cable. All the trade-offs come down to coverage area. As a rough rule, the 3dB down points of a horn antenna's field are about the same as the visual projection of it's flare; just sight along the antenna edges to see your coverage "footprint." It wouldn't be too far off to say that your area of reasonable exposure might only be 6" long! This doesn't pose too much of a problem if you're testing a hand-held gadget, but if you have two 7-foot tall equipment racks, you'll have to make a lot of exposures. Unless you plan to do a lot of work in this region, you would be better off to work with a test lab that already has these assets. Chances are the assets are quite under-utilized, as most military immunity testing stops at 18GHz these days. The closest lab to you is probably Elite, in Chicago. Beyond that, there's Cincinnati Electronics (guess where). Regards, Ed Ed Price [email protected] WB6WSN NARTE Certified EMC Engineer & Technician Electromagnetic Compatibility Lab Cubic Defense Applications San Diego, CA USA 858-505-2780 (Voice) 858-505-1583 (Fax) Military & Avionics EMC Is Our Specialty
