lw1ecp wrote:
Hi!. I need to cover the HF and VHF ranges with as few VCOs as
possible. Think of this as the varactor version of the old general
purpose RF bench generators or grid-dip-meters. I know the penalties:
high phase noise, high drift, high harmonic content. I don't care,
this won't be made into a high dynamic range receiver.
What I do need is a reliable means to keep the peak-to-peak RF voltage
across the varactors (varicaps) below 1 or 0.5Vp-p. Otherwise, even
having back-to-back diodes they rectify the RF, create a DC component
into the 100k resistor that feeds the tuning voltage, and this
dramatically raises the bottom frequency. If I reduce the R, the f
goes lower, but the RF is nastily clipped by the diodes.
The oscillator is a FET Hartley (no Colpitts capacitive tap in order
to maximize C swing). All attempts I made on Spice and in real life to
AGC the amplitude always created a low frequency relaxation
oscillation.
Can anybody tell me about a proven way to accomplish this?. I have
already googled a little with no success.
Many thanks!

See if you can find any of the old (now) op-amp cook books or general
applications manuals. I saw one with a modified Wien-bridge that had an
AGC, but that was in 1975 for a breadboard I was doing.
Good luck.
Bill Baka

I saw one with a modified Wien-bridge that had an AGC (...)
Thanks Bill for your immediate answer!. Yes, RC oscillators have
impressive f spans: you change R or C by 10:1 and f changes 10:1, not
its square root. But... my needs go into the many hundreds MHz, sorry,
I have to stick to LC oscillators.

lw1ecp wrote:
I saw one with a modified Wien-bridge that had an AGC (...)
Thanks Bill for your immediate answer!. Yes, RC oscillators have
impressive f spans: you change R or C by 10:1 and f changes 10:1, not
its square root. But... my needs go into the many hundreds MHz, sorry,
I have to stick to LC oscillators.


Many hundreds of MHz? Are you trying for D.C. to daylight? I could poke
around a bit and see if I could trip over something while I look for
parts for my own projects. Are you looking to build or buy? It seems you
want a circuit to build. Am I on the right track? Some of the older ARRL
handbooks may have a circuit or two. Lately I have seen them migrating
towards VHF and UHF and higher frequency stuff.
Good luck.
Bill Baka

In article , Bill Baka
wrote:

lw1ecp wrote:
I saw one with a modified Wien-bridge that had an AGC (...)
Thanks Bill for your immediate answer!. Yes, RC oscillators have
impressive f spans: you change R or C by 10:1 and f changes 10:1, not
its square root. But... my needs go into the many hundreds MHz, sorry,
I have to stick to LC oscillators.


Many hundreds of MHz? Are you trying for D.C. to daylight? I could poke
around a bit and see if I could trip over something while I look for
parts for my own projects. Are you looking to build or buy? It seems you
want a circuit to build. Am I on the right track? Some of the older ARRL
handbooks may have a circuit or two. Lately I have seen them migrating
towards VHF and UHF and higher frequency stuff.
Good luck.
Bill Baka


A typical way of getting this kind of range is to use a pair of
oscillators, at least one voltage controlled, a mixer, and a low pass
filter -- with the amplitude stabilizing loop described previously.

Use a pair of YIG tuned oscillators and a wideband mixer, and you can
get as wide a range as you want -- a lot of sweep generators work this
way.

You could use a pair of Mini Circuits VCOs in the 2 to 4 GHz range and
build a vco with a few hundred MHz range.

"lw1ecp" wrote in message
...
I saw one with a modified Wien-bridge that had an AGC (...)
Thanks Bill for your immediate answer!. Yes, RC oscillators have
impressive f spans: you change R or C by 10:1 and f changes 10:1, not
its square root. But... my needs go into the many hundreds MHz, sorry,
I have to stick to LC oscillators.

I remember what was known as a "seven league oscillator" which derived
its name from the folklore of "seven league boots". The design was
discussed in one of my old 1930's vintage vacuum tube books. If I
remember correctly, one such oscillator covered from a few Hertz to
several MHz. Unfortunately I am in SC and my library is back in
Tennessee.

My memory may be wrong, but I think the book was "Functional Circuits and
Oscillators" by Herbert Reich.

73, Dr. Barry L. Ornitz WA4VZQ

I'm astonished at the tsunami of answers this arose!.
- Tim: yes: detector + error amplifier (minimum gain, even a single
NPN) + FET bias shifting.
- Bill: I love old ARRL publications, 80's QSTs are my favorites, but
3:1 (continuous) coverage is not common on ham bands. I am looking to
build, and design (oh, well, guesstimate).
- Artie and Barry: I knew the heterodyne way of getting broad coverage
thanks to the old Wavetek 2000 Sweep/Signal Generator. This is also
used in tracking generators for spectrum analyzers. But keeping
unwanted mixing products low is an art. Maybe later.
But, the first answer from Tim made me think: what's the difference in
amplitude stabilizing methods on a Wien bridge and my Hartley?. In a
RC oscillator you control amplitude thru some voltage dependent
resistance (a lamp, a FET channel, etc.). But in my oscillator I vary
it thru gate bias, which varies the average transconductance of a
device working in a (nonlinear) class C. So, I Spice'd a FET damping
the coil (source) tap with its Rds being controlled by oscillations
amplitude, an voilà!, I got ~1Vp-p at the gate (and the varicaps), the
gate's 1N4148 almost does not conduct. I will try a more decent of
controlling the feedback (without ruining Q) and let you know.

lw1ecp wrote in news:a6888c2a-b312-443b-ae70-
:

snip
- Artie and Barry: I knew the heterodyne way of getting broad coverage
thanks to the old Wavetek 2000 Sweep/Signal Generator. This is also
used in tracking generators for spectrum analyzers. But keeping
unwanted mixing products low is an art. Maybe later.

No big deal. Use two high freqeuncy oscillators (one fixed, one
tunable), say around 2 GHz. When you mix them, the difference freqeuncy
gives you the low freqeuncy you want, and the sum freqeuncy will be
around 4 GHz & easily filtered out. The higher order terms that might be
an issue will be things like three times one minus twice the other. A
little care with your frequency plan & a good mixer & you should be fine.

There is a free DOS executable mixer spur calculator buried in with the
MARTHA/LLAMA APL CAD softwa http://www.marthallama.org/
The program is MIXSPUR.EXE, and is included in the LLAMA distribution. It
is documented in the LLAMA manual. No APL is required.

If folks are interested, I could put together standalone documentation on
MIXSPUR & park it and the executable on my personal website.

Doug White

Hey OM:

Maybe you should looky at an old NTSC tv tuner. The brothers at RCA,
made sum tuners for the cable tv bands that went from 5Mhz all the way
up to 900Mhz

And they did it with quad matched varicaps, dual gate FET's, and a
PLL.

Not only did they control the VCO, but the TRF front end also.

Me thinks your OSC is way to efficient and your Q is way to high. Roll
the Q down with the right LC ratio or swamp it with sum resistance.

If you series varicaps you would increase the the amount of RF voltage
they could handle before starting to rectifry the RF.

73 OM
de n8zu

Raypsi:
- Thank you for the hint, the CATV tuners cover 54 to 900 MHz (5 to
~40 is for the upstream), an impressive 16:1 range, but the L.O. is
much narrower in relative values. Suppose 1st IF at 1200MHz, the L.O.
would sweep 1254 to 2100, 1.67:1, not enough for me.
- "The TRF front end also". Are you sure?. I dismantled a Jerrold and
the incoming CATV hit directly the balanced modulator thru a fixed
high-pass. But even if it really tuned 54-900, it would be easier than
my needs because the CATV amplitude is much lower than in an
oscillator.
- "Roll the Q down"?. It wouldn't help, I just need to vary 9:1 the LC
product.
- "Series varicaps"?. Sorry... they are back-to-back in series
already!.
Doug:
Thank you for encouraging me about the heterodyne way, and the links.
Yes, I think I will go this way for my gen-purp lab generator, but for
the dipper I still need the coil to oscillate right at the measured
frequency.
Brian:
I'm afraid this is more complicated than choosing a long time
constant, I have already tried in the simulation. The problem seems to
stem from the fact that the effective FET gm is dependent not only on
applied bias but also on RF amplitude. The error amplifier senses the
amplitude is high, applies more -Vgs, gm decreases, amplitude
decreases, but this lower amplitude makes effective gm to fall even
more, and as a result oscillation ceases and then starts again, and
again, and again. I think I will have to prevent the FET from working
in class C and forget about Vgs as a control means.
Again, many thanks to all of you. As soon as I can come back to this
subject I will let you know the results.

Bill Baka wrote:
lw1ecp wrote:
Hi!. I need to cover the HF and VHF ranges with as few VCOs as
possible. Think of this as the varactor version of the old general
purpose RF bench generators or grid-dip-meters. I know the penalties:
high phase noise, high drift, high harmonic content. I don't care,
this won't be made into a high dynamic range receiver.
What I do need is a reliable means to keep the peak-to-peak RF voltage
across the varactors (varicaps) below 1 or 0.5Vp-p. Otherwise, even
having back-to-back diodes they rectify the RF, create a DC component
into the 100k resistor that feeds the tuning voltage, and this
dramatically raises the bottom frequency. If I reduce the R, the f
goes lower, but the RF is nastily clipped by the diodes.
The oscillator is a FET Hartley (no Colpitts capacitive tap in order
to maximize C swing). All attempts I made on Spice and in real life to
AGC the amplitude always created a low frequency relaxation
oscillation.
Can anybody tell me about a proven way to accomplish this?. I have
already googled a little with no success.
Many thanks!

See if you can find any of the old (now) op-amp cook books or general
applications manuals. I saw one with a modified Wien-bridge that had an
AGC, but that was in 1975 for a breadboard I was doing.
Good luck.
Bill Baka

The old school method - for a wien bridge for example - was to place a
small filament bulb in the feedback path. This had a non linear response
to amplitude - and a low natural bandwidth. It's that slow response
that you have been lacking, apparently. Putting together an op amp
with variable amplitude output that is smoothed on a five+ second time
constant is what you need.

Brian W