1197	2007-11-28 09:24:04	cwfingertalker	Mitsubishi RD New Series Mosfet RF Power Transistors
Message	Date	From	Subject
Greetings to the Group I was doing some research on RF Power Transistors for a future transmitter project and came across these "new" RD types. I wonder if anyone has an opinion on them. What interests me about them are the economical single lot prices and power out levels that are in the range I am interested in. Please take a look and comment. http://www.rfparts.com/mitstran.html Regards Bill Martin N7EU Sam's Valley Oregon.
1198	2007-11-28 09:42:29	Dan Rae	Re: Mitsubishi RD New Series Mosfet RF Power Transistors
cwfingertalker wrote: >Greetings to the Group > >I was doing some research on RF Power Transistors for a future transmitter project >and came across these "new" RD types. I wonder if anyone has an opinion on them. > > I and several (hundred?) others have used these (06 and 16) in the PIC-A-STAR transceiver in a 20 Watt three stage design by G6ALU. http://www.radio-kits.co.uk/ and go to the radio related items page I have bought a pair of the 100 Watt ones to try in a higher power version, not yet built. The 100 Watt ones seem to be used exclusively by Yaesu in their current 12 Volt 100 Watt PAs. Good enough? 73 Dan ac6ao
1199	2007-11-28 18:16:59	Graham Haddock	Re: Mitsubishi RD New Series Mosfet RF Power Transistors
Bill: I have used the 0.3 watt, 6 watt and 16 watt transistors in several HF transmitter designs. Multi-band PAs for 1.8 to 30 MHz. They are very well behaved, and are flat in gain up to above 30 MHz, and don't roll off above 10 MHz like IRF510 switching transistors. And, they are real RF transistors, which means the heat sink locations want to be connected to ground; they are not connected to the collector like switching transistors, with all the problems that insulating a RF and DC voltage point has, while trying to get the heat to ground.
1200	2007-11-28 21:59:35	cwfingertalker	Re: Mitsubishi RD New Series Mosfet RF Power Transistors
Graham and Dan thanks for taking the time to comment. I had a feeling these transistors have been prototyped into ham use. Thanks for the website information too. This gives me food for thought. I have looked on the internet before but did not find any examples of amplifiers with these transistors. I am a long way off in building one but in the "gathering" stage for information. My initial idea would be to use a DDS-VFO to drive a chain of buffers and amplifiers to a 15 to 30 watt level for CW. A direct frequency method. Any others in the group have a comment or suggestion please don't be shy. Regards Bill Martin N7EU Sam's Valley Oregon.
1201	2007-11-29 04:21:42	Ted Bruce	Mitsubishi RD New Series Mosfet RF Power Transistors
Based on discussions awhile back in this group, I ordered 30 of the RD16's from RF Parts for members of the NoGa QRP club. I was quoted a decent price break when I initially inquired about purchasing 10 of them, and I was surprised that the price was automatically dropped even more by the company when I actually ordered the lot of 30. I haven't had a chance to try them yet, though, as I've been under the weather for awhile. 73, Ted KX4OM ____________________________________________________________________________________ Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs
1202	2007-11-29 12:44:13	Graham	Re: Mitsubishi RD New Series Mosfet RF Power Transistors
Bill: I have built a CW-only, HF all band 20 watt transmitter that is much as you described, using the Mitsubishi RF transistors, a PIC for control, and a AD9951 DDS for the VFO. The DDS drives a single Mitsubishi 0.3 watt transistor, which drives two more in push pull, which drives a pair of the 16 watt transistors in push pull. Output through a harmonic filter is 20 watts on all HF bands. The PIC is a 16F877A that is the iambic keyer as well as directly controls the AD9951 DDS that is used as the VFO for the transmitter. Being a later generation DDS, it also has output level control and rise and fall time shaping, so it also controls drive level for the transmitter power amplifier, as well as rise and fall time shaping for the individual CW bits, all under software control. I have not finished the mechanical packaging of the transmitter yet, so it it three PC boards sitting in the open, but it works well. So, what you are thinking about is certainly a "do-able" project.
1203	2007-11-29 20:57:10	cwfingertalker	Re: Mitsubishi RD New Series Mosfet RF Power Transistors
Graham, Excellent. Any chance you could take a few digital pictures of your transmitter and or a schematic and upload to the photo/file folders for the group? These transistors sound like a good basis for a transmitter project to me. Thanks for all your interest. Regards. Bill Martin N7EU Sam's Valley Oregon.
1204	2007-11-29 21:53:56	Baruch Zilbershat...	Re: Mitsubishi RD New Series Mosfet RF Power Transistors

1205	2007-11-29 23:11:47	brainerd@wildblue...	Re: Mitsubishi RD New Series Mosfet RF Power Transistors

1206	2007-11-30 09:58:27	Graham	Re: Mitsubishi RD New Series Mosfet RF Power Transistors
Bill: Yes, I will post an info package with a few pictures, schematics, and summary of operation to the files section. It will probably be a few days.
1212	2007-12-07 08:56:36	Graham Haddock	Re: Mitsubishi RD New Series Mosfet RF Power Transistors
All: I just uploaded a file with some descriptions, schematics, and pictures of the 20 Watt DDS based CW transmitter I am developing, which uses the Mitsubishi "RD" series RF power FETS. Look in the folder "KE9H Projects."
1213	2007-12-08 06:43:29	Allison Parent	Re: Mitsubishi RD New Series Mosfet RF Power Transistors

1214	2007-12-08 07:44:42	Graham Haddock	Re: Mitsubishi RD New Series Mosfet RF Power Transistors
Allison: Thanks for the comments. I'll add some information on the transformers, as well as a picture of the transmitter running at the ARRL recommended 5 ms rise and fall times. As far as writing it up for publication, I will consider that after I finish the mechanical packaging. With respect to the power out, the output stages saturate first although the push-pull drivers are not too far behind (about 3 dB). So output is not drive limited. The limit in total output power is more a compromise between what you can do with a broadband match with the output transformer, versus an optimum match which would require frequency specific reactive matching components for each frequency band to get more power/full rated power out of the output transistors. But then you would not have a broad band PA. For single sideband linear use, you would want to have the drivers running in a more linear range. More like the G6ALU design where he used a pair of 6W transistors as the drivers. The end to end PA gain varies about 2 dB across 1.8 to 30 MHz, and output stage efficiency (including transformer losses) runs between 43 and 53 percent. Since drive level is under software control, the gain variation can be hidden by assigning a different drive level for each frequency band. --- Graham / KE9H Allis
1215	2007-12-08 10:43:01	cwfingertalker	Re: Mitsubishi RD New Series Mosfet RF Power Transistors
Graham, Very nice work and I agree you should work up an article and publish this project. Thanks for sharing with the group. Bill N7EU Sam's Valley Oregon
1216	2007-12-08 13:28:33	Graham Haddock	Re: Mitsubishi RD New Series Mosfet RF Power Transistors
Allison: I added two more files. One showing 5 ms rise and fall time, and another with construction and experiment notes on the four transformers.
1217	2007-12-08 15:46:18	Allison Parent	Re: Mitsubishi RD New Series Mosfet RF Power Transistors

1218	2007-12-08 16:38:14	Allison Parent	Re: Mitsubishi RD New Series Mosfet RF Power Transistors

1219	2007-12-08 17:39:39	Graham Haddock	Re: Mitsubishi RD New Series Mosfet RF Power Transistors
Allison Parent wrote: > > On T4: In the past for similar transformers I've used wire such > as #20 or as heavy as #14 (TFE stranded which is also silver > plated) for the secondary with good results. It's an area where > the high currents require a thick conductor. It is not so much about thick conductors above 10 MHz as it is about the surface area of the conductors. If you look at the current penetration depths, the center of the wire is not doing you any good. At the high end of the bands, larger wires are good because they have more surface area. (They could be hollow and work just as well.) But numerous smaller wires have even greater surface area. 8 each #30 wires have about the same total cross section are as a single #22, but the same surface area as a single #12 wire. 8 each #30 wires have twice the surface area as a #18 wire and therefore half the skin resistance, and four times the surface area of a #24 wire, and therefore one fourth the skin resistance. All this only makes a difference above 10 MHz, but is quite noticable above 20 MHz. > Using multiple > strands of wirewrap wire is one to keep in mind when I need > a more flexible wire or have have to fit it in a different > core. It's also a solution that avoids a specialized wire > need. It is more flexible and easier to get through a given size hole, but that is not why I did it. --- Graham / KE9H