I recently picked up an HP 8640B Opt. 323 (8640B in a can) and wonder if
anyone has information on the external frequency doubler so that it can
use it's upper frequency position (512-1024 MHz). I want to get 902 MHz
out of it to test the feed for my parabolic dish's return loss (SWR).

I can build a diode doubler to generate harmonics and then put a 1/4
wavelength of shorted-end coax across its output to select the 902 MHz
harmonic, but I'd like to be able to tune all freqs between the 512 and
1024 MHz without building a lot of shorted stubs. How does HP do it
with their doubler? Bandpass filtering? Could one bandpass filter
cover that much range?? Maybe something as simple as a low pass filter
with an Fco of 1050 MHz or so, but with this, how would I keep the
fundamental freq (256-512 MHz) out of the output??

Thanks for any leads!

--
Scott N0EDV
http://corbenflyer.tripod.com/
Gotta Fly or Gonna Die
Building RV-4 (Super Slow Build Version)

Searching Google, I was able to find a copy of the manual for the
frequency doubler unit (HP 11690A) which conveniently describes the
operation and has a schematic. I could make one, except there is one
part that is just labeled as a transformer (T1) with three windings.
That probably wouldn't be too hard to make, but no idea how many turns
for each winding...

The manual is at ftp://bama.sbc.edu/downloads/hp/1169..._service_6.pdf

Ideas??

Scott
N0EDV

Scott wrote:

I recently picked up an HP 8640B Opt. 323 (8640B in a can) and wonder if
anyone has information on the external frequency doubler so that it can
use it's upper frequency position (512-1024 MHz). I want to get 902 MHz
out of it to test the feed for my parabolic dish's return loss (SWR).

I can build a diode doubler to generate harmonics and then put a 1/4
wavelength of shorted-end coax across its output to select the 902 MHz
harmonic, but I'd like to be able to tune all freqs between the 512 and
1024 MHz without building a lot of shorted stubs. How does HP do it
with their doubler? Bandpass filtering? Could one bandpass filter
cover that much range?? Maybe something as simple as a low pass filter
with an Fco of 1050 MHz or so, but with this, how would I keep the
fundamental freq (256-512 MHz) out of the output??

Thanks for any leads!


--
Scott
http://corbenflyer.tripod.com/
Gotta Fly or Gonna Die
Building RV-4 (Super Slow Build Version)

"Scott" wrote in message
...
Searching Google, I was able to find a copy of the manual for the
frequency doubler unit (HP 11690A) which conveniently describes the
operation and has a schematic. I could make one, except there is one part
that is just labeled as a transformer (T1) with three windings. That
probably wouldn't be too hard to make, but no idea how many turns for each
winding...


The purpose of the transformer is to split the incoming signal in-phase and
apply to the 2 input port of the double balanced mixer. The number of turns
will depend a bit on the ferrite that you use for the core, I suspect that a
2 hole bead would be best, but a toroid would probably do as well.

73
Jeff

In message , Jeff
writes

"Scott" wrote in message
...
Searching Google, I was able to find a copy of the manual for the
frequency doubler unit (HP 11690A) which conveniently describes the
operation and has a schematic. I could make one, except there is one part
that is just labeled as a transformer (T1) with three windings. That
probably wouldn't be too hard to make, but no idea how many turns for each
winding...


The purpose of the transformer is to split the incoming signal in-phase and
apply to the 2 input port of the double balanced mixer. The number of turns
will depend a bit on the ferrite that you use for the core, I suspect that a
2 hole bead would be best, but a toroid would probably do as well.

73
Jeff

An alternative (but similar) circuit is essentially the same as the
well-known two-diode fullwave (bi-phase) rectifier.

Single ferrite bead input transformer, trifilar wound with 2 or 3 turns
(say 32 -34 gauge). Bead typically 5mm long, 4mm dia, small hole, mu 100
- 200. One winding is the input (primary). The other two connected as
centre-tapped secondary. CT to ground. Diodes VHF/UHF switching/mixer
diodes. Output 'DC polarity' - either +ve or -ve (doesn't matter).
Output must have DC return to ground (if output is DC blocked, use RFC
or resistor - say 50 - 330 ohms).

Physical balance affects suppression of fundamental and odd harmonics.
Helped (maybe) if input feed is via series bifilar balun choke (2 or 3
turns on same type of bead as transformer).

Ian.
--

On Jun 6, 4:24 am, Scott wrote:
Searching Google, I was able to find a copy of the manual for the
frequency doubler unit (HP 11690A) which conveniently describes the
operation and has a schematic. I could make one, except there is one
part that is just labeled as a transformer (T1) with three windings.
That probably wouldn't be too hard to make, but no idea how many turns
for each winding...

The manual is atftp://bama.sbc.edu/downloads/hp/11690a/11690a_service_6.pdf

Ideas??

Scott
N0EDV



Scott wrote:
I recently picked up an HP 8640B Opt. 323 (8640B in a can) and wonder if
anyone has information on the external frequency doubler so that it can
use it's upper frequency position (512-1024 MHz). I want to get 902 MHz
out of it to test the feed for my parabolic dish's return loss (SWR).

I can build a diode doubler to generate harmonics and then put a 1/4
wavelength of shorted-end coax across its output to select the 902 MHz
harmonic, but I'd like to be able to tune all freqs between the 512 and
1024 MHz without building a lot of shorted stubs. How does HP do it
with their doubler? Bandpass filtering? Could one bandpass filter
cover that much range?? Maybe something as simple as a low pass filter
with an Fco of 1050 MHz or so, but with this, how would I keep the
fundamental freq (256-512 MHz) out of the output??

Thanks for any leads!

--
Scotthttp://corbenflyer.tripod.com/
Gotta Fly or Gonna Die
Building RV-4 (Super Slow Build Version)

Beware!! The schematic is a little misleading. The top winding is
not coupled to the other two. Consider in a perfect transformer, if
it were, the output (right) ends of the upper and lower windings would
be identical. The purpose of the top winding is to improve the
balance at the output. In the bridges we've built, we normally make
two (as nearly as practical) identical transformers, both with bifilar
windings, but the one used for the "upper" winding in the schematic
shown just shorts the two windings together, so it's just an inductor,
but it matches the inductance of the other one, used as a current-mode
balun transformer, and also matches the parasitic resistances and
capacitances. The circuit for the balun is easy to simulate (e.g. in
LTSpice) and demonstrate how the added inductor dramatically improves
the balance at low frequencies.

As others have pointed out, you really don't need much filtering; the
circuit suppresses the fundamental and odd harmonics, so if you've
built it carefully, just a LPF that cuts off at 1.2GHz or so should be
about all the help it needs, unless you want a really clean signal for
some specific application, in which case a little bandpass filter
could clean things up further.

Cheers,
Tom

If I remember correctly it uses a double balanced mixer as the frequency
doubler, and only works at high output levels from the 8640 (+10dBm), so you
would require an external attenuator to get lower levels.

The HP part No is 08754-60057.

73
Jeff
G8HUL


"Scott" wrote in message
.. .
I recently picked up an HP 8640B Opt. 323 (8640B in a can) and wonder if
anyone has information on the external frequency doubler so that it can use
it's upper frequency position (512-1024 MHz). I want to get 902 MHz out of
it to test the feed for my parabolic dish's return loss (SWR).

I can build a diode doubler to generate harmonics and then put a 1/4
wavelength of shorted-end coax across its output to select the 902 MHz
harmonic, but I'd like to be able to tune all freqs between the 512 and
1024 MHz without building a lot of shorted stubs. How does HP do it with
their doubler? Bandpass filtering? Could one bandpass filter cover that
much range?? Maybe something as simple as a low pass filter with an Fco
of 1050 MHz or so, but with this, how would I keep the fundamental freq
(256-512 MHz) out of the output??

Thanks for any leads!

--
Scott N0EDV
http://corbenflyer.tripod.com/
Gotta Fly or Gonna Die
Building RV-4 (Super Slow Build Version)

I think I might have found a modern alternative...a very small surface
mount device that looks like it will do exactly what the HP doubler does
AND only $5.95 in SINGLE quantities (The AMK-2-13+)!!

http://www.minicircuits.com/products...pliers_sm.html

Doesn't get much easier than this!!

Scott
N0EDV




Jeff wrote:
If I remember correctly it uses a double balanced mixer as the frequency
doubler, and only works at high output levels from the 8640 (+10dBm), so you
would require an external attenuator to get lower levels.

The HP part No is 08754-60057.

73
Jeff
G8HUL


"Scott" wrote in message
.. .

I recently picked up an HP 8640B Opt. 323 (8640B in a can) and wonder if
anyone has information on the external frequency doubler so that it can use
it's upper frequency position (512-1024 MHz). I want to get 902 MHz out of
it to test the feed for my parabolic dish's return loss (SWR).

I can build a diode doubler to generate harmonics and then put a 1/4
wavelength of shorted-end coax across its output to select the 902 MHz
harmonic, but I'd like to be able to tune all freqs between the 512 and
1024 MHz without building a lot of shorted stubs. How does HP do it with
their doubler? Bandpass filtering? Could one bandpass filter cover that
much range?? Maybe something as simple as a low pass filter with an Fco
of 1050 MHz or so, but with this, how would I keep the fundamental freq
(256-512 MHz) out of the output??

Thanks for any leads!

--
Scott N0EDV
http://corbenflyer.tripod.com/
Gotta Fly or Gonna Die
Building RV-4 (Super Slow Build Version)




--
Scott
http://corbenflyer.tripod.com/
Gotta Fly or Gonna Die
Building RV-4 (Super Slow Build Version)

Scott wrote:
I think I might have found a modern alternative...a very small surface
mount device that looks like it will do exactly what the HP doubler does
AND only $5.95 in SINGLE quantities (The AMK-2-13+)!!

Not "exactly".

Guaranteed F1 in the output is only -20
dBc. Not nearly as good as a DBM.

Chuck

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----

In article ,
Chuck wrote:

Scott wrote:
I think I might have found a modern alternative...a very small surface
mount device that looks like it will do exactly what the HP doubler does
AND only $5.95 in SINGLE quantities (The AMK-2-13+)!!

Not "exactly".

Guaranteed F1 in the output is only -20
dBc. Not nearly as good as a DBM.

Scott-

I've read past postings that discussed Minicircuits.com and the 8640B
doubler. Why not contact them and ask if they have a better
alternative? They may be familiar with the 8640B, and have addressed
the question before.

Regarding Jeff's comment, I was of the impression that the signal was
generated at a high level prior to the attenuator, and attenuator
settings were correct for the 512-1024 band as well as the others.
Since your "8640B in a can" has the 512-1024 switch position, it should
be possible to install a home-made doubler in the same space where the
HP doubler would go, prior to the attenuator.

Fred
K4DII

Regarding Jeff's comment, I was of the impression that the signal was
generated at a high level prior to the attenuator, and attenuator
settings were correct for the 512-1024 band as well as the others.
Since your "8640B in a can" has the 512-1024 switch position, it should
be possible to install a home-made doubler in the same space where the
HP doubler would go, prior to the attenuator.

Fred
K4DII

The doubler is an external device that screws onto the front panel output
socket, there is no provision for putting it internally prior to the
attenuator. The 512-1024 switch position merely changes the displayed
frequency.

Jeff