EMRFD Message Archive 976
Message Date From Subject 976 2007-08-25 07:52:01 Allison Parent mosfet wide band amps Has any one exprimented with DMOS fets (VN10kn or 2n7000)
as wide band amps? I usually see them in switch applications
or as QRP finals for under a watt but rare is the use in
linear aps. I suspect these are capable of good gain
Another application I'd like to know is if the 2n7000
would be useful in a H-mode mixer for HF?
Just considering applying the 2n7000s I have to apps where
I've used more exotic or scarce devices for.
977 2007-08-25 12:42:15 cfwa2unn Re: mosfet wide band amps 978 2007-08-25 13:25:29 Allison Parent Re: mosfet wide band amps 979 2007-08-26 16:01:49 Giancarlo Re: mosfet wide band amps 980 2007-08-26 19:14:32 Allison Parent Re: mosfet wide band amps 982 2007-08-26 21:48:41 Wes Hayward Re: mosfet wide band amps Hi Allison, and group.
Just a short comment. First, re the MOSFET feedback amplifiers:
Doug DeMaw and I built some several years ago and they worked just
fine. However, you do not have the high available gain with most
FETs that you do with a bipolar, so it is a little harder. You do,
of course, have the high available gain with the power FETs. I
think that Doug published something on it, but not sure.
The same basic formulas or software can be used for design.
Essentially you evaluate the transconductance of the FET you are
going to use, and at the bias conditions you plan to use. Then you
ask what the current would be in a bipolar to get the same gm. Use
that current and you are there.
Regarding mixers: I did some measurements on some mixers that I
published in RF Design in November, 1990. Several mixer types were
considered, some with just 1 FET, but others with 2 or 4 in a variety
of topologies. This data was used as the basis for several designs
that we did at TriQuint Semiconductor. We used both the single FET
version of what we called the Maas Mixer, and then a two FET version
that was singly balanced. These became quite popular and I'm told
that there are now about half a billion of them out there in cell
phones. Who knows -- that number probably came from marketing.
In looking at the various circuits, I investigated both the TriQuint
FETs that we used to build the integrated parts and discrete parts.
Most of the measurements reported in the RF Design paper were done at
home at 14 MHz, for I had strong two tone measurement gear set up at
the time. Among the parts investigated was the 2N7000. A single
passive 2N7000 produced a conversion loss of 5 dB with IIP3 of +11
dBm. A J310 did better for IMD.
Although it was never reported, I also investigated bipolars and
found that they would also do a good job. This is similar to the
singly balanced circuit that Ulrich published in QST in the early 90s.
Some additional work was sort of published in that era in that it was
presented as part of a class that we gave once or twice per year at
TriQuint that we called the GaAs Class. Much of that mixer work was
included in one of Ulrich's books: Rohde and Newkirk, "RF/Microwave
Circuit Design for Wireless Applications," Wiley, 2000.
The basic Maas mixer is kicking around in EMRFD, but is not
"featured." Rick and I are now well away from TQS, and they seem to
have no interest in mixers, so we may be taking things further with
this circuit. It is really a good one that is deserving of more
I doubt that you can push it to +50 dBm for IIP3 though.
And while we are on the subject of super intercept mixers, I want to
chime in to applaud the work that Marteen has done. It is on his
web page. But let us not forget the ground work that Colin, Bill,
and Harold (a.k.a., the Triad) did. All of this work was the basis
for what became the front end of the PicaStar and similar
transceivers. It predated those publications by a decade, with the
earliest publications (that I know of) being the notes that Colin
sent to Pat Hawker for inclusion in RadCom's Technical Topics.
Anyway, I hope this is useful. Give this mixer topology a try
--- In firstname.lastname@example.org, "Allis
983 2007-08-27 09:28:43 Allison Parent Re: mosfet wide band amps 988 2007-08-28 03:27:53 Mike Brainard Re: mosfet wide band amps --- In email@example.com, "Allis 989 2007-08-28 05:30:37 Allison Parent Re: mosfet wide band amps 990 2007-08-28 06:45:05 Giancarlo Re: mosfet wide band amps Hi Allison,
I hope that following m suggestions you do not get lost in W1-land...
Wes is always giving good comments; there is always something to
learn. I have one of the first EMRFD books, a gift from another very
good Ham Bill Carver, W7AAZ (ex K6OLG).
I tried one of Ulrich's 2x BJT mixers when I modifed my FT726R 50 MHz
converter. It was very noisy so I had to replace the original 1st
mixer with a classic SBL1. Maybe it was me not good for that mixer as
I had no test equipment, except a good scope.
Too bad we are too far away, I cannot come to "sample" someof your
If you like you may try the I7SWX 1 Transformer double balanced
mixer/demodulator using the FST3125 (half or 2+2 switches in
parallel).It is like a half H-Mode Mixer, it is used in the GXJP's
STAR transceiver as second conversion mixer IF to 15kHz. Insertion
loss <-6dB, IP3 >+25dBm, squarer is as for the H-Mode Mixer with
74AC86, I have also a version with passive squarer, it needs more
power.I will send you the two .pdf files.
Not everywhere we do need >+50dBm IP3, it is important to have the
possibility of selection and repeatibility.
--- In firstname.lastname@example.org, "Allis
991 2007-08-28 07:20:44 Allison Parent Re: mosfet wide band amps 993 2007-08-28 08:51:22 Giancarlo Re: mosfet wide band amps 994 2007-08-28 12:35:06 brainerd@wildblue... Re: mosfet wide band amps On 28 Aug 2007 at 10:27, Mike Brainard wrote:
> --- In email@example.com, "Allis