I'm designing a 40-70 MHz synthesizer using the Mini-Circuits POS-75 VCO.
The VCO output varies from 9 dBm at 40 MHz down to 7 dBm at 70 MHz. My
circuit needs to provide isolation, tolerate open and short-circuit loads,
and deliver 15 dBm into 50 ohms across the tuning range. What do folks
think of the solution below? In particular, what about C2? This "hack"
seems to compensate for the input level variation; but I don't know if it
will work for real. This is for very low volume production and C2 could be
set-on-test.

Well, it's not exactly the way I'd do it, but you must have a reason for
doing it the way you are doing it, so why not explain to us what was on your
mind, especially after considering the magnitude of C2.

Jim

--
"It is the mark of an educated mind to be able to entertain a thought
without accepting it."
--Aristotle


"Andrew Holme" wrote in message
...
I'm designing a 40-70 MHz synthesizer using the Mini-Circuits POS-75 VCO.
The VCO output varies from 9 dBm at 40 MHz down to 7 dBm at 70 MHz. My
circuit needs to provide isolation, tolerate open and short-circuit loads,
and deliver 15 dBm into 50 ohms across the tuning range. What do folks
think of the solution below? In particular, what about C2? This "hack"
seems to compensate for the input level variation; but I don't know if it
will work for real. This is for very low volume production and C2 could
be
set-on-test.

Andrew Holme wrote:
I'm designing a 40-70 MHz synthesizer using the Mini-Circuits POS-75 VCO.
The VCO output varies from 9 dBm at 40 MHz down to 7 dBm at 70 MHz. My
circuit needs to provide isolation, tolerate open and short-circuit loads,
and deliver 15 dBm into 50 ohms across the tuning range. What do folks
think of the solution below? In particular, what about C2? This "hack"
seems to compensate for the input level variation; but I don't know if it
will work for real. This is for very low volume production and C2 could be
set-on-test.


Well, you're running 2.6V/100 ohms or 26 mA through that poor
transformer, which may not like it much...and the low-frequency input
impedance of the transistor is about 1 ohm, plus whatever the extrinsic
emitter resistance of a 2N3904 is. At 70 MHz, the capacitor's reactance
is about 15 ohms, so it won't do a whole lot on its own if the input
resistance is really low.

On the other hand, since the beta of the transistor is dropping like a
rock in the low VHF, the base resistance will be making the emitter
inductive-looking, so you might be parallel-resonating the emitter
inductance.

I'd be wanting to use a better circuit for this, myself--something along
the lines of a faster transistor running common-emitter, with a lead-lag
emitter degeneration network (i.e. R in parallel with a series RC).

Cheers,

Phil Hobbs

"RST Engineering (jw)" wrote in message
m...
"Andrew Holme" wrote in message
...
I'm designing a 40-70 MHz synthesizer using the Mini-Circuits POS-75 VCO.
The VCO output varies from 9 dBm at 40 MHz down to 7 dBm at 70 MHz. My
circuit needs to provide isolation, tolerate open and short-circuit
loads,
and deliver 15 dBm into 50 ohms across the tuning range. What do folks
think of the solution below? In particular, what about C2? This "hack"
seems to compensate for the input level variation; but I don't know if it
will work for real. This is for very low volume production and C2 could
be
set-on-test.


Well, it's not exactly the way I'd do it, but you must have a reason for
doing it the way you are doing it, so why not explain to us what was on
your mind, especially after considering the magnitude of C2.


I tried and failed to find a suitable IC. Perhaps someone could suggest
one.

The AC voltage across C2 is quite small.

I'd be fascinated to hear how you would do it.

"Phil Hobbs" wrote in message
...
Andrew Holme wrote:
I'm designing a 40-70 MHz synthesizer using the Mini-Circuits POS-75 VCO.
The VCO output varies from 9 dBm at 40 MHz down to 7 dBm at 70 MHz. My
circuit needs to provide isolation, tolerate open and short-circuit
loads,
and deliver 15 dBm into 50 ohms across the tuning range. What do folks
think of the solution below? In particular, what about C2? This "hack"
seems to compensate for the input level variation; but I don't know if it
will work for real. This is for very low volume production and C2 could
be
set-on-test.

Well, you're running 2.6V/100 ohms or 26 mA through that poor transformer,
which may not like it much...and the low-frequency input impedance of the
transistor is about 1 ohm, plus whatever the extrinsic emitter resistance
of a 2N3904 is. At 70 MHz, the capacitor's reactance is about 15 ohms, so
it won't do a whole lot on its own if the input resistance is really low.

The transformer datasheet says RF power 0.25W and DC current 30mA maximum -
so I'm within spec - but is this still a bad idea?


On the other hand, since the beta of the transistor is dropping like a
rock in the low VHF, the base resistance will be making the emitter
inductive-looking, so you might be parallel-resonating the emitter
inductance.

I'd be wanting to use a better circuit for this, myself--something along
the lines of a faster transistor running common-emitter, with a lead-lag
emitter degeneration network (i.e. R in parallel with a series RC).

Thanks.

I tried and failed to find a suitable IC. Perhaps someone could suggest
one.

The AC voltage across C2 is quite small.

And this matters how?



I'd be fascinated to hear how you would do it.


The easy way to do it if you aren't all that good at RF is to use one of the
MiniCircuits small amplifiers. You can set up a matching circuit at the
input that throws some of the gain away in return for a gain slope that
compensates for the slope in the output of the VCO.

Jim

"RST Engineering (jw)" wrote in message
m...


I tried and failed to find a suitable IC. Perhaps someone could suggest
one.

The AC voltage across C2 is quite small.

And this matters how?

I didn't understand your earlier remark about the magnitude of C2. I
assumed you meant it was large for the frequency. I just wanted to point
out that its effect is small because of the small AC voltage / low impedance
at that point in the circuit. The circuit also presents a resistive 50 ohm
input termination at all frequencies because of this.




I'd be fascinated to hear how you would do it.


The easy way to do it if you aren't all that good at RF is to use one of
the MiniCircuits small amplifiers. You can set up a matching circuit at
the input that throws some of the gain away in return for a gain slope
that compensates for the slope in the output of the VCO.

Andrew Holme wrote:

I'm designing a 40-70 MHz synthesizer using the Mini-Circuits POS-75 VCO.
The VCO output varies from 9 dBm at 40 MHz down to 7 dBm at 70 MHz. My
circuit needs to provide isolation, tolerate open and short-circuit loads,
and deliver 15 dBm into 50 ohms across the tuning range. What do folks
think of the solution below? In particular, what about C2? This "hack"
seems to compensate for the input level variation; but I don't know if it
will work for real. This is for very low volume production and C2 could
be set-on-test.

In the past, for 26MHz I have used a 74HC04 (input signal AC-coupled with
1nF capacitor to the input of one gate with a 100k resistor from the input
to the output of that gate to set the DC bias, then that gate drives a
second gate, then the second gate drives the other four gates in the
package which have their inputs all in parallel and their outputs all in
parallel). The observant reader will notice that this circuit produces a
square wave, so I low-pass filtered the output and notched out the 3rd
harmonic, and it provided a little over +10dBm sinewave into 50 Ohms. At
70MHz you might have to use a 74AC04 which is a bit faster, and it might
get a little too warm if you are unlucky. You can adjust the supply
voltage for a trade-off between too much dissipation and not enough output
power. The phase noise was pretty good, and the cost is very low.

Chris