Date: Wed, 24 Jun 2026 10:11:58 -0400 From: David Olean <[email protected]> To: [email protected], 'Bob Abernethy' <[email protected]> Cc: [email protected], [email protected] Subject: Re: Topband: 5kW amps
Hi folks, There is a reason to use a big tube in an amateur amplifier. I have a six meter amplifier that uses a 3CX3000AA7.? It uses the shorted turn tuning method with a copper tube that moves in and out of the PI-L coil. It was used in a multi op station where operators were unfamiliar with the equipment. They also never looked at plate and grid meters! A 3 by 3 is not a huge tube as in the 160 meter amp being discussed, but it is capable of making lots of power. It has 13.0 dB gain at 50 MHz. The loading cap etc is set to run it at 1500 watts and it needs about 65 watts of drive. There is no way an operator can over drive it. There is also no way anyone can get more than 1500 watts out of it without going inside and tweaking the loading cap values etc.? In short, it is a perfect contest amp. It is clean and produces no splatter. I do the same trick with a single band 10 meter amp used on that band. It is designed to produce 1500 watts of very clean RF and is not capable of being over driven with my K3 exciter. At one point in a ten meter contest, I was called by the op at K1RX on ten meters. He said it was the cleanest amplifier that he ever heard. K1RX was seven miles away, all downhill, and there was basically no splatter.? All my old amps with smaller tubes had issues with over driving and more with ops who were great contest ops, but never watched plate and grid meters. 4CX250B, 4-400, 4CX1000 and small triodes etc all had issues.? I am temporarily off 160 now. I am hoping to be QRV again soon. >K1WHS Instead of your shorted turn method, ( 6m amp), when using tubes with a high C between anode and chassis (24 pf on a 3x3..which rises to 33 pf when tube plugged into the grid ring...due to the proximity of the lower fins to the chassis below it) I used a L-PI network, with a small value inductance inserted between the block cap assy and the C1 vac tune cap. The tube C plus the extra small coil, form a step down L network...that drops the plate load Z of the tube way down in value..... down to something a normal PI network, with a normal Q can be designed. That was all done using GM3SEK's Pi / PI-L spreadsheets, which factor in all stray L and stray C. In fact, we didn't use coils at all for the pre-coil or main coil. We used 8.3" length ( tnx Terman) of 1" wide strap, formed into a perfect half circle. Diam = 5", so 5" was the spacing between the vac tune and vac load caps. Same deal for the smaller pre-coil, but it's even smaller, and shorter. The advantage then is, RF current flows on both sides of the strap, instead of one side. With tubing coils, no RF flows inside the actual tube ( like water inside a pipe)..and no rf flows on the inside of the coil, it gets bunched up on the outside. End result was real low IMD, like -59dbc if biased correctly..... and harmonic suppression is excellent due to the L-PI network. Another option is an L-PI-L neywork. Eff was carefully measured at 72% on 50.125 MHz, using a calibrated Bird slug, calibrated HV meters and calibrated DC plate current meters..and also inserting a calibrated Fluke DVM in series with the plate current meter ( and also the multiplier string of resistors used for the HV meter). The 3x3 is 4.125" daim. The 4x5 tube is 4.94" in diameter... not much bigger. The 3x3 tube is actually 4 kw max anode diss.... moot point. The seller of that amp had access to a NIB Eimac 4x5, dirt cheap, so used what was readily available at the time. The 8877, new, was not on the short list. At the time, the 8877 was $2900.00 at Richardson's. Jim VE7RF _________________ Searchable Archives: http://www.contesting.com/_topband - Topband Reflector
