Bruce,
Thank you for your input. I will definitely look at your recommendation for
a path forward.

(to clarify - I didn't mean at all I'd be onboarding on a recapping project
before I get this working properly. Rather, in case I'm able to identify a
module most likely at fault for it, I'd pull it out and first examine if a
cap failure is triggering this particular symptom. A rather "wide net"
approach, but it does factor in the age of the unit.)

Thank you again, Bruce.

On Sun, Sep 18, 2022 at 11:20 AM Bruce <bruce@...> wrote:

Radu -

Most likely a problem with one of the synthesizers. Before you wast
time checking for caps, start looking at the synthesizer output. The
frequency range is clearly marked on the motherboard of the TF
section. Not that there is a problem with examining for defective
caps, but this is more likely an IC or cable problem. Fix the problem
first, then look at "restoring" the piece. The 3335 is an interesting
piece and you have an interesting problem.

My recommendation:
1) Read over (and understand) the frequency generation section
2) Look at the outputs of the various synthesizer sections - check the
VCOs first, then the divider outputs
3) Analyze the unit that is not working correctly

Best place to look at the synthesizer outputs is at the input of the
mixers (some are sma connectors on top of boards), others are
available on the motherboard.

Some guesses:
1) Defective divider or oscillator in one of the synthesizers
2) Incorrect data to the synthesizers from the front panel
3) The filter switching network problem

Cheers!

Bruce




Quoting Radu Bogdan Dicher <vondicher@...>:

I got some really weird data to report.

I've been playing with adjusting the levels and frequency.

1. It seems I am able to get an output between exactly 14MHz and

19MHz.

One half a MHz up or down from there, output drops to zero.
2. As long as there's output (as above), the level adjustment seems to
be working OK. I haven't looked closer at whether the value is
exactly what
it should be, but it raises and lowers as I command it from the

AMPLITUDE

controls.
3. The frequency doesn't change as I adjust per #1 above. It's always
about 21.5MHz (says my scope).

Not sure what to make of it yet, but I am going to seek to determine
whether this can be triggered by a single module, pull it out and look

for

damaged caps (first line of defense).

Thank you all.
Radu.

On Sun, Sep 18, 2022 at 5:24 AM Patricio A. Greco via groups.io
<patricio_greco@...> wrote:

You are Wellcome. Is a great satisfaction to put these jewels back to

work

again.


Regards, Patricio.


Ing. Patricio A. Greco
Taller Aeronáutico de Reparación 1B-349
Organización de Mantenimiento Aeronáutico de la Defensa OMAD-001
Laboratorio de Calibración ISO 17025 AREA: RF/MW
Gral. Martín Rodríguez 2159
San Miguel (1663)
Buenos Aires
T: +5411-4455-2557
F: +5411-4032-0072
www.servicios-electronicos.com.ar



On 18 Sep 2022, at 01:46, Radu Bogdan Dicher <vondicher@...>

wrote:

Thank you, Patricio, and thank you very much, Tom. Hopefully, my PS

repair

gets reflected soon in a fully functional unit.
I'm a bit disheartened, but I do very much appreciate your

encouragement!

Radu.

On Sat, Sep 17, 2022 at 9:43 PM Radu Dicher <vondicher@...>

wrote:

Bruce,
Thank you for pitching in.

Yes, upon restarting the unit, it goes to 1MHz and -86.98dBm
amplitude, which I think is default. Everything display-wise seems

perfect.

Except maybe the issue of no buffer time to allow the LO to reach
temperature, which I find odd. Maybe the LO is offline? Which may

explain

no sinewave at output.

I reconnected by hand the rigid line. Not sure if properly, as I don't
have a brass-torqued SMD wrench. I assume the worst that can happen

with

that is a non-leveled output, but not missing altogether. Not knowing

more

on the history of the unit, I assume maybe my handling of the shielded
modules may have severed the signal path somehow. Though I generally

have a

very soft and overly planned touch, if anything, so maybe the history

of

the unit is the issue. I dunno.

I will be looking into examining the signal path for any potential
interruptions. Given the history of the unit I tend to think it should

come

back online within spec, but that's not what we're seeing here.
Radu.

On Sat, Sep 17, 2022 at 9:25 PM Bruce <bruce@...> wrote:

Do you know at power up the output is ~ -80dBm - try setting the
amplatude to 0dBm and be sure you reconnected the hardline from the RF
unit to the attenuator. Any front panel diagnostics lit ???

Cheers!

Bruce

Quoting Tom Lee <tomlee@...>:

Great news, indeed. Keep going, Radu!

-- Cheers,
Tom

--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070


On 9/17/2022 20:27, Radu Bogdan Dicher wrote:

Great news, folks.

After finding out both A15CR1 and CR2 were shorted, I replaced them
with 1N4003s, and all rails are now coming online in spec (while
having the board out, I also replaced all 10uF caps). Now the +15V
is just a hair off at 14.987V (should be within 0.01V of 15V), but
I can't imagine that's a huge deal.

However, I don't seem to see any signal coming out of it. Learning
how to use it, so take that with a grain of salt.
Radu.

On Sat, Sep 17, 2022 at 6:54 PM Radu Bogdan Dicher via groups.io
<> <vondicher@...> wrote:

Well, after finally taking this apart (I'd post pictures, but I

am

a bit ptsd'ed...) I have two shorted diodes on A15 - CR1 and

CR2.

I am thinking vanilla 1N4148 would be OK in this position -

thoughts?

Thank you,
Radu.

On Sat, Sep 17, 2022 at 5:30 PM Radu Bogdan Dicher via

groups.io

<> <vondicher@...> wrote:

Raymond, Clay,
You both sort of got beat by a hair with the 4 youtube

videos

Clay is talking about. Must be this:
.

And yes, I take back my statement on the overall assembly
design of this unit. I didn't know there's just one SMA
connecting the shielded modules (there's actually several in
my unit as to, I think, the youtube sample, but they're not
rigid lines), but overall, it's a breeze to take apart (once
you know what the correct path is).

Thank you both! On to the work for me.
Radu.

On Sat, Sep 17, 2022 at 4:42 PM Raymond Domp Frank
<hewpatek@...> wrote:

On Sun, Sep 18, 2022 at 01:22 AM, freshndaire wrote:

I would not call this unit hard to service or a poor
mechanical design. Pull the big aluminum hunk and

you

have lots of access! The synthesizer module has 3
long bolts holding it in. Remove the SMA Semi-rigid
coax cable from the module's backplane then pull

the 3

screws in the middle and you can lift the synth

module

and set it aside with the other cables plugged in

IIRC.

So my memory served me right!

Raymond

On Sep 18, 2022, at 15:00, F1EKU <rfconsulting.fr@...> wrote:

?If I remember correctly, the introduction of Torx screws was made by HP around 1990, at the same time as the patent for Torx screws became public domain.

Back in the '60's, while In the USAF, we knew these as "Reed and
Prince" screw heads.? I've since noted the more commonly used term of
Pozidriv's.? .

I recently picked up a pair a 8640B's for a good price (I think...)

Any advice on how to fix the gears on at least the one working unit? I am mechanically adept. It would be really cool if I could get both units working so that I could test 2/20 KHz selectivity on receivers. I read somewhere that someone (in Italy?) had/does offer brass replacement gears - but I couldn't find supplier or cost, or a review of them or how long ago this was.

I got some really weird data to report.

I've been playing with adjusting the levels and frequency.

1. It seems I am able to get an output between exactly 14MHz and 19MHz.
One half a MHz up or down from there, output drops to zero.
2. As long as there's output (as above), the level adjustment seems to
be working OK. I haven't looked closer at whether the value is exactly what
it should be, but it raises and lowers as I command it from the AMPLITUDE
controls.
3. The frequency doesn't change as I adjust per #1 above. It's always
about 21.5MHz (says my scope).

Not sure what to make of it yet, but I am going to seek to determine
whether this can be triggered by a single module, pull it out and look for
damaged caps (first line of defense).

Thank you all.
Radu.

On Sun, Sep 18, 2022 at 5:24 AM Patricio A. Greco via groups.io
<patricio_greco@...> wrote:

You are Wellcome. Is a great satisfaction to put these jewels back to work
again.


Regards, Patricio.


Ing. Patricio A. Greco
Taller Aeronáutico de Reparación 1B-349
Organización de Mantenimiento Aeronáutico de la Defensa OMAD-001
Laboratorio de Calibración ISO 17025 AREA: RF/MW
Gral. Martín Rodríguez 2159
San Miguel (1663)
Buenos Aires
T: +5411-4455-2557
F: +5411-4032-0072
www.servicios-electronicos.com.ar



On 18 Sep 2022, at 01:46, Radu Bogdan Dicher <vondicher@...> wrote:

Thank you, Patricio, and thank you very much, Tom. Hopefully, my PS repair
gets reflected soon in a fully functional unit.
I'm a bit disheartened, but I do very much appreciate your encouragement!
Radu.

On Sat, Sep 17, 2022 at 9:43 PM Radu Dicher <vondicher@...> wrote:

Bruce,
Thank you for pitching in.

Yes, upon restarting the unit, it goes to 1MHz and -86.98dBm
amplitude, which I think is default. Everything display-wise seems perfect.
Except maybe the issue of no buffer time to allow the LO to reach
temperature, which I find odd. Maybe the LO is offline? Which may explain
no sinewave at output.

I reconnected by hand the rigid line. Not sure if properly, as I don't
have a brass-torqued SMD wrench. I assume the worst that can happen with
that is a non-leveled output, but not missing altogether. Not knowing more
on the history of the unit, I assume maybe my handling of the shielded
modules may have severed the signal path somehow. Though I generally have a
very soft and overly planned touch, if anything, so maybe the history of
the unit is the issue. I dunno.

I will be looking into examining the signal path for any potential
interruptions. Given the history of the unit I tend to think it should come
back online within spec, but that's not what we're seeing here.
Radu.

On Sat, Sep 17, 2022 at 9:25 PM Bruce <bruce@...> wrote:

Do you know at power up the output is ~ -80dBm - try setting the
amplatude to 0dBm and be sure you reconnected the hardline from the RF
unit to the attenuator. Any front panel diagnostics lit ???

Cheers!

Bruce

Quoting Tom Lee <tomlee@...>:

Great news, indeed. Keep going, Radu!

-- Cheers,
Tom

--
Prof. Thomas H. Lee
Allen Ctr., Rm. 205
350 Jane Stanford Way
Stanford University
Stanford, CA 94305-4070


On 9/17/2022 20:27, Radu Bogdan Dicher wrote:

Great news, folks.

After finding out both A15CR1 and CR2 were shorted, I replaced them
with 1N4003s, and all rails are now coming online in spec (while
having the board out, I also replaced all 10uF caps). Now the +15V
is just a hair off at 14.987V (should be within 0.01V of 15V), but
I can't imagine that's a huge deal.

However, I don't seem to see any signal coming out of it. Learning
how to use it, so take that with a grain of salt.
Radu.

On Sat, Sep 17, 2022 at 6:54 PM Radu Bogdan Dicher via groups.io
<> <vondicher@...> wrote:

Well, after finally taking this apart (I'd post pictures, but I am
a bit ptsd'ed...) I have two shorted diodes on A15 - CR1 and CR2.
I am thinking vanilla 1N4148 would be OK in this position -

thoughts?

Thank you,
Radu.

On Sat, Sep 17, 2022 at 5:30 PM Radu Bogdan Dicher via groups.io
<> <vondicher@...> wrote:

Raymond, Clay,
You both sort of got beat by a hair with the 4 youtube videos
Clay is talking about. Must be this:
.

And yes, I take back my statement on the overall assembly
design of this unit. I didn't know there's just one SMA
connecting the shielded modules (there's actually several in
my unit as to, I think, the youtube sample, but they're not
rigid lines), but overall, it's a breeze to take apart (once
you know what the correct path is).

Thank you both! On to the work for me.
Radu.

On Sat, Sep 17, 2022 at 4:42 PM Raymond Domp Frank
<hewpatek@...> wrote:

On Sun, Sep 18, 2022 at 01:22 AM, freshndaire wrote:

I would not call this unit hard to service or a poor
mechanical design. Pull the big aluminum hunk and you
have lots of access! The synthesizer module has 3
long bolts holding it in. Remove the SMA Semi-rigid
coax cable from the module's backplane then pull the 3
screws in the middle and you can lift the synth module
and set it aside with the other cables plugged in IIRC.

So my memory served me right!

Raymond

I recently picked up a pair a 8640B's for a good price (I think...) The short history is that a ham was selling them for the wife of a recent SK. She knows nothing about the hobby. He knew very little about the sig gens' histories. Supposedly the ham was a member of a club and used them to help members repair / test rigs. One of the sig gens is supposed to work, the other was for parts, condx unknown, but it has an "X" on the bezel along with a small plastic bag taped to the top the has 2 very tiny gold fingers in it.

All of the controls on both units rotate through their full range of motion with no issues (as far as I can tell). I powered up the 1 unit (it had a power cord attached), unit seems to run okay, LEDs, lamps seem to cycle okay with the expected illumination, meter indicates output and it varies when I move the attenuator. Fan runs, heatsink seems to run very warm after a couple of hours. I have not looked at the output on a scope yet. I have not powered up the 2nd unit (with the X on it)

The 1 working unit has option 001, the other unit has no options.

Visual inspection of inisde of "good" unit. 3 nylon gears cracked: 1 range gear (1 plc), 1 fm deviation gear (2 plcs), 1 smaller gear that drives the main tuning osc (2 plcs)
The bad (i.e "X" parts) unit has 1 range gear cracked in 1 place but all parts seem to be intact/present in the unit.

Any advice on how to fix the gears on at least the one working unit? I am mechanically adept. It would be really cool if I could get both units working so that I could test 2/20 KHz selectivity on receivers. I read somewhere that someone (in Italy?) had/does offer brass replacement gears - but I couldn't find supplier or cost, or a review of them or how long ago this was.

I've downloaded and did a cursory read of the docs I could find related to the 8640b.

I bought these because I need a decent generator to replace the 8647a that I recently discovered has hidden damage (attenuator output has one range that is not correct) that I was not aware of. And 8647a's appear to be unservicable. I realize that the specs on the 8647a are not good enough for doing the selectivity tests, but it wasn't originally purchased for that purpose.

Did I waste my money? Should I just use the one until the gears just completely break?? This test equipment is just my hobby, I don't make any money from this but I like my equipment to be in good working order.

Thanks,
Jim

Star drive and Torx are NOT the same.
-----Original Message-----
From: [email protected] <[email protected]> On Behalf Of Dave McGuire
Sent: 18 September 2022 05:44
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] Pozidriv screws on HP covers
On 9/17/22 22:23, saipan59 (Pete) wrote:

Torx and square drive are the kings of readily available
high torque screws

Question: Is Torx and "star drive" the same thing? Perhaps the only
difference is that Torx is trademarked??

Yes. Look at the people who call Torx "star drive" with the same
derision as you would the people who equate Phillips, PoziDriv, and JIS
and call them all "crosspoints".
-Dave

On 18 Sep 2022, at 01:46, Radu Bogdan Dicher <vondicher@...> wrote:

Thank you, Patricio, and thank you very much, Tom. Hopefully, my PS repair gets reflected soon in a fully functional unit.?

I'm a bit disheartened, but I do very much appreciate your encouragement!

Radu.?

On Sat, Sep 17, 2022 at 9:43 PM Radu Dicher <vondicher@...> wrote:

Bruce,

Thank you for pitching in.?

Yes, upon restarting the unit, it goes to 1MHz and -86.98dBm amplitude,?which I think is default. Everything display-wise seems perfect. Except maybe the issue of no?buffer time to?allow the LO to reach temperature, which I find odd. Maybe the LO is offline? Which may explain no sinewave at output.?

I reconnected by hand the rigid line. Not sure if properly, as I don't have a brass-torqued SMD wrench. I assume the worst that can happen with that is a non-leveled output, but not missing altogether. Not knowing more on the history of the unit, I assume maybe my handling of the shielded modules may have severed the signal path somehow. Though I generally?have a very soft and?overly planned touch, if anything, so maybe the history of the unit is the issue. I dunno.?

I will be looking into examining the signal path for any potential interruptions. Given the history of the unit I tend to think it should come back online?within spec,?but that's not what we're seeing here.?

Radu.?

On Sat, Sep 17, 2022 at 9:25 PM Bruce <bruce@...> wrote:

Do you know at power up the output is ~ -80dBm - try setting the??
amplatude to 0dBm and be sure you reconnected the hardline from the RF??
unit to the attenuator.? Any front panel diagnostics lit ???

Cheers!

Bruce

Quoting Tom Lee <tomlee@...>:

> Great news, indeed. Keep going, Radu!
>
> -- Cheers,
> Tom
>
> --?
> Prof. Thomas H. Lee
> Allen Ctr., Rm. 205
> 350 Jane Stanford Way
> Stanford University
> Stanford, CA 94305-4070
>?
>
> On 9/17/2022 20:27, Radu Bogdan Dicher wrote:
>> Great news, folks.
>>
>> After finding out both A15CR1 and CR2 were shorted, I replaced them??
>> with 1N4003s, and all rails are now coming online in spec?(while??
>> having the board out, I also replaced all 10uF caps). Now the?+15V??
>> is just a hair off at 14.987V (should be within 0.01V of 15V), but??
>> I can't imagine that's a huge deal.
>>
>> However, I don't seem to see any signal coming out of it. Learning??
>> how to use it, so take that with a grain of salt.
>> Radu.
>>
>> On Sat, Sep 17, 2022 at 6:54 PM Radu Bogdan Dicher via???
>> <> <vondicher=[email protected]> wrote:
>>
>>? ? Well, after finally taking this apart (I'd post pictures, but I am
>>? ? a bit ptsd'ed...) I have two shorted diodes on A15 - CR1 and CR2.
>>? ? I am thinking vanilla 1N4148 would be OK in this position - thoughts?
>>? ? Thank you,
>>? ? Radu.
>>
>>? ? On Sat, Sep 17, 2022 at 5:30 PM Radu Bogdan Dicher via?
>>? ? <> <vondicher=[email protected]> wrote:
>>
>>? ? ? ? Raymond, Clay,
>>? ? ? ? You both sort of got beat by a hair with the 4 youtube videos
>>? ? ? ? Clay is talking about. Must be this:
>>? ? ? ??.
>>
>>? ? ? ? And yes, I take back my statement on the overall assembly
>>? ? ? ? design of this unit. I didn't know there's just one SMA
>>? ? ? ? connecting?the shielded modules (there's actually several in
>>? ? ? ? my unit as to, I think, the youtube sample, but they're not
>>? ? ? ? rigid lines), but overall, it's a breeze to take apart (once
>>? ? ? ? you know what the correct path is).
>>
>>? ? ? ? Thank you both! On to the work for me.
>>? ? ? ? Radu.
>>
>>? ? ? ? On Sat, Sep 17, 2022 at 4:42 PM Raymond Domp Frank
>>? ? ? ? <hewpatek@...> wrote:
>>
>>? ? ? ? ? ? On Sun, Sep 18, 2022 at 01:22 AM, freshndaire wrote:
>>
>>? ? ? ? ? ? ? ? I would not call this unit hard to service or a poor
>>? ? ? ? ? ? ? ? mechanical design.? Pull the big aluminum hunk and you
>>? ? ? ? ? ? ? ? have lots of access!? The synthesizer module has 3
>>? ? ? ? ? ? ? ? long bolts holding it in.? Remove the SMA Semi-rigid
>>? ? ? ? ? ? ? ? coax cable from the module's backplane then pull the 3
>>? ? ? ? ? ? ? ? screws in the middle and you can lift the synth module
>>? ? ? ? ? ? ? ? and set it aside with the other cables plugged in IIRC.
>>
>>? ? ? ? ? ? So my memory served me right!
>>
>>? ? ? ? ? ? Raymond
>
>
>?