¿ªÔÆÌåÓý

  1. HP-Agilent-Keysight-Equipment
  2. Messages
Expanded Topics Messages
Date
Re: Frequency range of HP 5352B frequency counter

 

On Wed, 25 May 2022 at 16:49, Lothar baier <Lothar@...> wrote:

As to my knowledge option 005 was never covered by the regular service manual ,? back in my tucker days I remember seeing a addendum that was published to be added to the service guide that covered the part# for the sub panel , dress panel , connector and cable assy as well as the specs.

Also note that the 46GHz is not a hard spec , it merely means that the unit meets a published spec for sensitivity at this point , it more than likely will work at 47 or higher but the sensitivity will be crappy


The unit I bought arrived at my friends, but the higher frequency input was non-functional. Luckily the seller had another, which functioned okay. But there seems to be a hard limit at 43 GHz. I think it worked at 43, but not at 43.01, suggesting there's some software stopping it from going higher in frequency, and it's not limited to the connector.?

The unit was working down to a power of -24 dBm at 43 GHz, so well within the specification of -15 dBm?@ 40 GHz.? It worked with an input of 12 mV RMS on the lower frequency range, which is much better than the 25 mV RMS specification. That's not bad for a ?200 (around $250) frequency counter.

But I don't think without some changes outside the connector, will this count above 43 GHz - there seems to be some software limit there.

Dave
Re: HP 3465A offset when input shorted

Chris Smith
 

Got it. That's confirmed here as well. Doing too many tasks at the same time here to do any of them properly. I shall sit in the corner for a few mins with the dunce hat on!

By elimination of ideas, the JFET looks like the most sane solution now so it looks like I'll have to obtain one. I'll stick this DMM in the cupboard though until I've filled a Mouser basket.

Thanks for everyone's input on this; it was a valuable learning exercise for me :)
Re: HP 3465A offset when input shorted

 

On Wed, May 25, 2022 at 10:31 AM Ozan <ozan_g@...> wrote:
On Wed, May 25, 2022 at 09:22 AM, Chris Smith wrote:
I did a simulation of the clamp and it clamps at 5v, both polarities which is under the 6v reverse breakdown so the thing shouldn't get zenered. It's a bit tight though!

----
When the node is clamped to let's say +5V, CR9 has (+5V-(-3.92V))=8.92V reverse bias. In the sim you should see CR8 and CR9 seeing ~ 2*3.92V+0.7V reverse bias in the extremes of the over voltage case.?


Agreed.? The anode of CR9 is held at approximately -3.92 V unless something?takes current away from CR11 and the only thing that can do that is reverse current through CR9.

Orin.
Re: HP 3465A offset when input shorted

 

On Wed, May 25, 2022 at 09:22 AM, Chris Smith wrote:
I did a simulation of the clamp and it clamps at 5v, both polarities which is under the 6v reverse breakdown so the thing shouldn't get zenered. It's a bit tight though!

----
When the node is clamped to let's say +5V, CR9 has (+5V-(-3.92V))=8.92V reverse bias. In the sim you should see CR8 and CR9 seeing ~ 2*3.92V+0.7V reverse bias in the extremes of the over voltage case.?

Ozan
HP 3465A offset when input shorted

 

Chris Smith:

>I did a simulation of the clamp and it clamps at 5v, both polarities which is?
>under the 6v reverse breakdown so the thing shouldn't get zenered.

Looking at the schematic there is indeed polarity clamps for + & - excursions using?
CR9 & CR 8 going to 3.9V zenerdiodes, CR11 & CR10, that reverse bias the actual?
clamping diodes?CR 8 & CR9 when not in use.

So CR 9 and CR 8 is effectively in anti parallel so zenering the B-E diode of the 2N3904?
cannot occur.


------


Best regards,

Goran Finnberg
The Mastering Room AB
Goteborg
Sweden

E-mail: mastering@...

Learn from the mistakes of others, you can never live long enough to
make them all yourself.??? -?? John Luther

(\__/)
(='.'=)
(")_(") Ranglet, Aron, VovVov, Nero & Smurfen:RIP
Re: HP 8662A Power up - overvoltage shutdown

 

Agreed. That would be extremely useful.

The -40V rail's operating spec max is 0.5A - so I can increase the current dummy load I have in place by quite a way yet - I'll try that next.

thanks

Tony
Re: HP 3465A offset when input shorted

Chris Smith
 

I did a simulation of the clamp and it clamps at 5v, both polarities which is under the 6v reverse breakdown so the thing shouldn't get zenered. It's a bit tight though!

As for the JFET pricing, it's reasonable from Mouser but it's ?15 delivery (what I paid for the meter) or min spend. I have a shopping list a mile long so it might be used to bulk an order up. I'll do some thinking. I will most likely throw a 3904 in there soon, at least from the perspective of the experimental side of this and then replace it with a JFET later.
Re: HP 3465A offset when input shorted

 

FWIW, Mouser have over 8000 PN4117A. at $3.17 each.? I don't know how easy/expensive to get them to ship in the UK though.

You don't need the metal can version - HP switched to plastic packages for many of the front end parts in later 3456A meters for example.

On Wed, May 25, 2022 at 8:53 AM Orin Eman via <orin.eman=[email protected]> wrote:


On Wed, May 25, 2022 at 8:12 AM Chris Smith <me@...> wrote:
Thanks all for the replies. I looked into this as well and came across similar FETs. The only problem is obtaining them cost more than I paid for the meter. It turns out the B-E junction of a 2n3904 is similarly specified i.e. roughly the same order of magnitude of leakage, and a lot cheaper. I will try that and see if it works first, reverting to more expensive solutions later if it doesn't.

?Yes, a 2N3904 B-E is next best after the FETs for leakage, but doesn't have a very good reverse breakdown spec - only 6V.
Re: HP 3465A offset when input shorted

 

Orin Eman:

?>Yes, a 2N3904 B-E is next best after the FETs for leakage,?
>but doesn't have a very good reverse breakdown spec - only 6V.

The B-E diode is zenered by back reversing it.

It increases 100s times or more increases the B-E diode leakage if zenered once.

There is a need for back to back 2N3904 B-E diodes to avoid that.

-----------

Best regards,

Goran Finnberg
The Mastering Room AB
Goteborg
Sweden

E-mail: mastering@...

Learn from the mistakes of others, you can never live long enough to
make them all yourself.??? -?? John Luther

(\__/)
(='.'=)
(")_(") Ranglet, Aron, VovVov, Nero & Smurfen:RIP
Re: HP 3465A offset when input shorted

 

The 1901-0586 is also used in the 34702A, in which it is listed as "DIODE: SI 30 WV 10PA LEAKAGE", so I would say that a low-leakage JFET would be a perfect choice as a replacement. A 2N4117 or its plastic version the PN4117, with drain and source connected together, would work great.

IHTH !

Joel Setton
Re: HP 3465A offset when input shorted

 



On Wed, May 25, 2022 at 8:12 AM Chris Smith <me@...> wrote:
Thanks all for the replies. I looked into this as well and came across similar FETs. The only problem is obtaining them cost more than I paid for the meter. It turns out the B-E junction of a 2n3904 is similarly specified i.e. roughly the same order of magnitude of leakage, and a lot cheaper. I will try that and see if it works first, reverting to more expensive solutions later if it doesn't.

?Yes, a 2N3904 B-E is next best after the FETs for leakage, but doesn't have a very good reverse breakdown spec - only 6V.
Inert Gas

 

Has anyone used canned air (R134a, er=1.0113, plus a bittering agent) to backfill sealed HP modules? I don't have any dry nitrogen on hand, canned air doesn't seem to kill the performance, and it seems likely that it's better than nothing. Since the failure modes are slow (condensation, oxidation) I thought I'd ask around to see if anyone has any hard won lessons they'd like to share.
Re: Frequency range of HP 5352B frequency counter

Lothar baier
 

¿ªÔÆÌåÓý

As to my knowledge option 005 was never covered by the regular service manual ,? back in my tucker days I remember seeing a addendum that was published to be added to the service guide that covered the part# for the sub panel , dress panel , connector and cable assy as well as the specs.

Also note that the 46GHz is not a hard spec , it merely means that the unit meets a published spec for sensitivity at this point , it more than likely will work at 47 or higher but the sensitivity will be crappy

?

From: [email protected] <[email protected]> On Behalf Of Lothar baier via groups.io
Sent: Wednesday, May 25, 2022 10:18 AM
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] Frequency range of HP 5352B frequency counter

?

Dave ,

Incorrect, the 46GHz used a 2.4mm , it was a connector that is similar to the ones used on VNA (NMD type) , google pictures of the 52B with option 05 and you see the difference?



On May 25, 2022, at 05:01, Dr. David Kirkby, Kirkby Microwave Ltd via groups.io <drkirkby@...> wrote:

?

On Wed, 25 May 2022 at 06:34, Lothar baier <Lothar@...> wrote:

It depends ,? the 46GHz option changed the input connector , the regular 5352 just as the 5351 used a 3.5mm male connector which will work to 40GHz with degraded specs which doesn¡¯t matter for a counter ,? the 46GHz option used a 2.4mm male NMD connector which is good to 50GHz ,

?

As far as I can see from the manual, the input connector on the 46 GHz option is 3.5 mm, not 2.4 mm, which does surprise me.?

?

I bought the counter from eBay and had it delivered to a friend who had a 50 GHz signal generator. He found that the high frequency input is kaput, but luckily the seller has a couple more counters, so hopefully one of the 3 will work.?

I reckon the also changed the cable assy as the 0.141 cable used already has a bunch of loss at 40GHz so using a 0.085 cable would have been a sensible choice .

?

?

I also know that initially HP used factory selected samplers for the 52 and 52 with 46GHz option so the 52 samplers had a different part#? , later on they scrapped the concept of having different part numbers and then just sold one part !

?

Do you know at what S/N that was stopped? It would be nice if the S/N of the unit I may get was new enough that it would be suitable.

?

Do you know if a 46 GHz one is likely to work at 47 GHz? There¡¯s nothing of interest for me between 40 and 46 GHz, but there¡¯s an amateur band in the UK covering 47-47.2 GHz. Although I have no immediate plans to work at 47 GHz, if the counter is likely to work there, it would provide an incentive to upgrade.?

?

Dave

--

Dr. David Kirkby,
Kirkby Microwave Ltd,
drkirkby@...

Telephone 01621-680100./ +44 1621 680100

Registered in England & Wales, company number 08914892.
Registered office:
Stokes Hall Lodge, Burnham Rd, Althorne, Chelmsford, Essex, CM3 6DT, United Kingdom
Re: HP 3465A offset when input shorted

Chris Smith
 

There is separate protection for the AC, ohms and current ranges from what I can see. I haven't traced the whole thing out here or checked the whole DMM out yet, only the DC path.?
HP 3465A offset when input shorted

 

Chris Smith:

>Low capacitance is not an issue as it's after the AC converter so low voltage slew.?
>It can be considered a DC amplifier with very low bandwidth at that point.?
>There is actually 0.1uF to ground anyway at the junction point so that would be?
>several orders of magnitude larger than any parasitic capacitance anyway.

Good to know.

I will look at the schematic later this evening to see what they have done.

Strange to clamp later on leaving the input buffer amp and AC converter open to burn out.

But maybe there is overload protection there too ?


------------



Best regards,

Goran Finnberg
The Mastering Room AB
Goteborg
Sweden

E-mail: mastering@...

Learn from the mistakes of others, you can never live long enough to
make them all yourself.??? -?? John Luther

(\__/)
(='.'=)
(")_(") Ranglet, Aron, VovVov, Nero & Smurfen:RIP
Re: HP 3465A offset when input shorted

Chris Smith
 

Low capacitance is not an issue as it's after the AC converter so low voltage slew. It can be considered a DC amplifier with very low bandwidth at that point. There is actually 0.1uF to ground anyway at the junction point so that would be several orders of magnitude larger than any parasitic capacitance anyway.
HP 3465A offset when input shorted

 

Chris Smith:

>It turns out the B-E junction of a 2n3904 is similarly specified?
>i.e. roughly the same order of magnitude of leakage, and a lot cheaper.

What about capacitive loading at the clamp point ?

Looking at the service manual for 5 minutes I see no spec for the?
accuracy when fed 100 kHz compared to 50/60 Hz.

I am thinking that it is not only low leakage but also low capacitance?
that you need in that point.

----------

Best regards,

Goran Finnberg
The Mastering Room AB
Goteborg
Sweden

E-mail: mastering@...

Learn from the mistakes of others, you can never live long enough to
make them all yourself.??? -?? John Luther

(\__/)
(='.'=)
(")_(") Ranglet, Aron, VovVov, Nero & Smurfen:RIP
HP/Agilent/Keysight VEE for a reasonable price?

 

Hi all,

I'm looking to get my hands on a copy of the HP/Keysight VEE Pro software development package for tinkering around, but I'm not able to fork out the full cost of the current version for my hobbyist use...
Is there anywhere I can get maybe an earlier copy for a (much) cheaper price, or does anyone here have a license they are willing to sell?

My use case is to create runtime versions of software front-ends for automated testing with such equipment as 8903B Audio Analyzer, 4145B semiconductor Parameter Analyzer, 4276A LCZ Meter, etc etc.

Thanks!
Re: Frequency range of HP 5352B frequency counter

Lothar baier
 

¿ªÔÆÌåÓý

Dave ,
Incorrect, the 46GHz used a 2.4mm , it was a connector that is similar to the ones used on VNA (NMD type) , google pictures of the 52B with option 05 and you see the difference?


On May 25, 2022, at 05:01, Dr. David Kirkby, Kirkby Microwave Ltd via groups.io <drkirkby@...> wrote:

?
On Wed, 25 May 2022 at 06:34, Lothar baier <Lothar@...> wrote:

It depends ,? the 46GHz option changed the input connector , the regular 5352 just as the 5351 used a 3.5mm male connector which will work to 40GHz with degraded specs which doesn¡¯t matter for a counter ,? the 46GHz option used a 2.4mm male NMD connector which is good to 50GHz ,


As far as I can see from the manual, the input connector on the 46 GHz option is 3.5 mm, not 2.4 mm, which does surprise me.?

I bought the counter from eBay and had it delivered to a friend who had a 50 GHz signal generator. He found that the high frequency input is kaput, but luckily the seller has a couple more counters, so hopefully one of the 3 will work.?

I reckon the also changed the cable assy as the 0.141 cable used already has a bunch of loss at 40GHz so using a 0.085 cable would have been a sensible choice .



I also know that initially HP used factory selected samplers for the 52 and 52 with 46GHz option so the 52 samplers had a different part#? , later on they scrapped the concept of having different part numbers and then just sold one part !


Do you know at what S/N that was stopped? It would be nice if the S/N of the unit I may get was new enough that it would be suitable.

Do you know if a 46 GHz one is likely to work at 47 GHz? There¡¯s nothing of interest for me between 40 and 46 GHz, but there¡¯s an amateur band in the UK covering 47-47.2 GHz. Although I have no immediate plans to work at 47 GHz, if the counter is likely to work there, it would provide an incentive to upgrade.?

Dave
--
Dr. David Kirkby,
Kirkby Microwave Ltd,
drkirkby@...

Telephone 01621-680100./ +44 1621 680100

Registered in England & Wales, company number 08914892.
Registered office:
Stokes Hall Lodge, Burnham Rd, Althorne, Chelmsford, Essex, CM3 6DT, United Kingdom
Re: HP 3465A offset when input shorted

Chris Smith
 

Thanks all for the replies. I looked into this as well and came across similar FETs. The only problem is obtaining them cost more than I paid for the meter. It turns out the B-E junction of a 2n3904 is similarly specified i.e. roughly the same order of magnitude of leakage, and a lot cheaper. I will try that and see if it works first, reverting to more expensive solutions later if it doesn't.

I have checked the voltages with the diode clipped out and it matches my LTspice simulation of the clamp which is good.
25101 - 25120 of 151469