Hi Ron:
Follows are a few general suggestions to follow along after
the excellent comments of others.
>From my prospective, I would try:
1.) Optimum Ic for reasonable IMD is 100 ma. or so per device.
Amp stability seems a bit easier to obtain at lower
quiescent currents.
2.) Unless an active regulator is used,
I find a 6A diode to ground with 20 Ohms 10W
switched by input relay to +12 as a suitable voltage regulator.
Use Caps & RFC's to get on & off the RF relay.
Use voltage divider from diode to set Vbe to about 0.62 at base
with no drive. A 1.0 Ohm to common junction and somewhere
around 4.7 Ohm to gnd will get you into the ballpark.
3.) Use a couple "Z" beads or a VK200 choke to connect bias
voltage to center of input transformer's back plate. Use 3-4
turns on input & output transformers.. No capacitor at CT of
input transformer, let it float RF wise. Bypass here sometimes
makes a quasi stable circuit oscillate. I like 10 Ohm 1/2W base
to ground R's. A Base to Base capacitor helps stability.
2.) Your oscillations are most likely from ground loop.
Heatsink is most useful as common grounding point of small circuit
boards, which often have ground loops by themselves.
Attach Connectors to common grounding with Lo Z connections.
Use a series cap of .001 at RF input, 0.01 in series with
output connector.
3.) Normally, this 10m amp will have only ~ -20 db rolloff of
2nd & 3rd harmonics. 20m down sees virtually no harmonic
reduction without the band switched lowpass network.
Obviously, lowering of drive will help IMD immensely.
4.) It is easiest to use broadband network on input side
with no tuning. As mentioned, a pad at the input is a good
idea to swamp out variations seen by both driver & amp.
Base-Base R is a good way, also. You will have to add a gain
compensation network (at input side of RF Amp) due to lower
drive requirements on 80-40m.
5.) Negative Feedback should consist of a 2W metal oxide res.
(33-47 Ohms) in series with a 0.1 disk.
6. Let's see what happens--
I have a few more tricks to try for taming quasi stable designs.
I seem to have deleted earlier communications.
How much drive do you have available? Is it for a QRP rig?
..........Merit W6NQ
----- Original Message -----
From: "Ron" <w8ron@stratos.net>
To: "skipp isaham" <nospam4me@juno.com>
Cc: <amps@contesting.com>
Sent: Wednesday, January 16, 2002 6:03 PM
Subject: Re: [AMPS] amplifier help
>
>
>
> Hi Skip, Merit, Ian , Steve , Sam and Pete.
>
> Thanks for your time to reply and I will study your comments.
> I like your reply Skip in that it was targeted at the questions of
> structure of these PA specifically about the RC feedback to limit 100Khz
> - 1 Mhz , the diode mirror /biasing and the emitter by-pass components.
> These were the kind of information that I was seeking that are of a
> general nature and useful for other than just this project.
>
> I did reduce the zero signal bias to about 3 amps before this all
> started and was concerned that the amp required such high bias to be
> stable and had caused my own problems. I would say that all of you
> comments indicate that that is not the case and that 3 amps is about
> right. THere are two silicon diodes in the emitter circuit thermally
> coupled to each device and I suspect that it would stabalize the bias
> should one device get hotter than the other.
> My first recall to the group was from memory in that I am writing during
> lunch and the rig is at home. THe devices are 2sc2290 and are the same
> used in my FT-107.
> I don't have a service manual but do have a good schematic (with
> errors) that has been the only hints I get .
>
> THe LPF I have is a small unit and I had several . We took one apart
> and measured the components and plugged the values into "Radio Designer"
> circuit analysis software to see what it theoritically would do. It
> rolled off at 30 and was 80dB down at 52 Mhz ...then up to -60db from
> there on. This network was fed and terminated at 50 ohms.
>
> Armed with new ideas , I'll go back and take another stab at it and see
> what I discover. Any and all further comments welcomed.
> If some one has a good reference book/article/app note for this subject
> let me know.
> ---
> Ron
>
>
>
>
> > Reads like a common HF emitter balasted push pull
> > design. First, put the driver into a 50 ohm termination
> > and look at it's output. If you don't have a clean
> > driver section... throw it away and get on with your
> > life.
> > -
> > On with the circuit...
> > The first input transformer provides a step down
> > impedance ratio, separate drive signals for
> > each transistor base and a method to provide a
> > the base bias current which should be voltage
> > regulated. One often sees active voltage regulator
> > circuits, but I do have a diode current mirror
> > circuit that works well. It ends up being some large
> > resistors, bypass caps and a silicon diode.
> > -
> > In your repair work on/with the amplifier, you can
> > make up a simple half or 1 dB pad to install/insert
> > at the amplifier stage input. It will provide a
> > relative measure of stability while you trouble shoot
> > the amplifier circuit.
> > -
> > The zero signal bias current for the transistors are
> > set using the bias supply and is set to a proper
> > mfgrs data sheet value. Going the real extra mile
> > includes adjustments of the bias supply in operation
> > for best practical working performance (lowest
> > distortion). The best bias supply setting is often
> > different for static, low drive and active full tilt
> > operation.
> > -
> > Your other post mentions large values of zero
> > signal collector current. Something is wrong, it's
> > probably the base bias circuit or it's adjustment
> > or an oscillation problem.
> > Most likely the circuit has a problem which
> > might have been the reason for the previous
> > device(s) failure(s).
> > -
> > Oscillation can also cause large (excessive) Collector
> > current (Ic) values.. obviously this is an unwanted signal.
> > The output of the amplifier should be stable into a
> > termination and slightly reactive loads as found in
> > most common antenna systems.
> > -
> > A series RC network is often used to provide feedback
> > in broad band amplifiers. The gain of broad band
> > 3-30 stages is often much higher at the 3 Mhz end
> > of operation. It's common to see feedback to reduce
> > gain a lower frequencies and provide an overall "flat"
> > amplifier gain to the end user.
> > -
> > The "chip caps" at or near the base of most
> > transistor amplifiers need careful attention for
> > proper type, value and condition. This is a very very
> > big deal. Many amplifiers I've worked on have
> > serious short comings in this very important area.
> > Capacitors have RF resistance, dilectric Q and
> > dissipation factor (DF)which should always be
> > carefully considered. Shortcuts can end up as
> > excessive heat... I have seen more than one
> > amplifier with self de soldering caps around the
> > transistor.
> > -
> > The output transformer provides step up back
> > towards 50 ohms impedance and an easy
> > method to Apply the collector voltage. One will
> > often see silver mica caps or arco style trimmer
> > caps on one or more of the transformer. From
> > my experience, the arco type trimmer caps are
> > common problem generators.
> > -
> > I've got a nice little 2 pole low pass filter for use
> > at 10 meters. It fits into a small space and will
> > handle 150 watts no problem... drop me a note
> > direct if you need details.
> > -
> > So you have amplifier problems from IMO
> > poor design and problematic parts. Service
> > manuals for that unit can be had easy enough.
> > If you can find one and share a copy, I'll walk you
> > through some of the repair issues as time
> > allows.
> > -
> > Your results will probably vary..
> > -
> > cheers
> > skipp
> > -
> > ________________________________________________________________
> > GET INTERNET ACCESS FROM JUNO!
> > Juno offers FREE or PREMIUM Internet access for less!
> > Join Juno today! For your FREE software, visit:
> > http://dl.www.juno.com/get/web/.
> >
> > --
> > FAQ on WWW: http://www.contesting.com/FAQ/amps
> > Submissions: amps@contesting.com
> > Administrative requests: amps-REQUEST@contesting.com
> > Problems: owner-amps@contesting.com
>
> --
> FAQ on WWW: http://www.contesting.com/FAQ/amps
> Submissions: amps@contesting.com
> Administrative requests: amps-REQUEST@contesting.com
> Problems: owner-amps@contesting.com
>
>
>
--
FAQ on WWW: http://www.contesting.com/FAQ/amps
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
|