Hey friends
sorry for my late reply. But the last few days it has been such nice weather that I spent most of my time outside and was not hanging out in my home lab :-) and also didn't check the emails too frequently. So my apologies for that, but I didn't forget this thread.
I can share the Python source code as soon as I have a more readable version of it. The current one is quite ugly I must say. But I sent messages off-list to some of you.
|
Re: Opens and shorts for 4285A LCR meter owners
Hello,
and it's off:
With best regards
Tam HANNA
Enjoy electronics? Join 15k7 other followers by visiting the Crazy Electronics Lab at
On 2020. 04. 13. 2:41, Dr. David
Kirkby, Kirkby Microwave Ltd wrote:
toggle quoted message
Show quoted text
If you own a 30 MHz 4285A LCR meter, you may have
discovered that you should perform a cable compensation if using
1 m or 2 m cables. You need
* Supplied 100 ohm resistor?
* 42090A open
* 42091A short?
The latter are over $1200 each. The 4285A LCR
manual tells you how cobble together the open and short using
BNC adapters if you don¡¯t have the 42090A and 42091A, but I
decided to make them up. Here¡¯s an internal view of the short.
There are two suboptimal things here, but I decided to ignore
them.
1) One should use metal boxes That would short the
braids with a lower inductance, but I had the plastic boxes and
believe that the tags on the BNC connectors which are about 3 mm
wide, are not going to present much inductance.
2) I shorted the inner conductors with a scrap of
FR4 PCB. A bit of wide copper would have been better, but again
I don¡¯t think it will matter. At the maximum frequency (30 MHz),
the wavelength is 10 m.?
The only difference between the open and short is
that the open joins the two pairs of outer BNCs, but doesn¡¯t
short all 4.?
Tomorrow I will make up some proper labels for
these.?
I also a photograph that I made some time ago to
use with a 1 MHz 4284A. It has red and black connectors for the
DUT - there¡¯s no provision for adding DC bias. That seems to
work okay with the 30 MHz 4285A. I intend making another of
these, with 100 mm spacing between the connectors, to handle
larger inductors.?
The LCR meter in the photographs in a 4284A, but
the 4285A looks identical.?
Note, that you don¡¯t need the open and short for
the 4284 - only the 4285A. I don¡¯t know what other LCR meters
these are used with. Keysight discontinued the 4285A, and do not
have a replacement.?
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: GPSDO connected to external ref inputs
That's 4 Volts peak to peak unterminated. 2 Volts peak to peak or +10 dBm into 50 ohms, which I assume the 5343A and 8566A have.
As to radiation, not as far as I know. I've not tried to work the HF band that I hear gets obliterated by stray 10 MHz from distribution systems. Unless one takes care with double shielded coax and terminators on all unused outputs. Not there with the ham radio gear yet (will take recommendations on a 70 cm capable rig, though).
Thanks.
Jim
toggle quoted message
Show quoted text
------ Original Message ------ From: "Bruce" <bruce@...> To: [email protected]Sent: 4/13/2020 9:25:03 AM Subject: Re: [HP-Agilent-Keysight-equipment] GPSDO connected to external ref inputs No problem - I do think 4v is a bit high though - any radiation problems?
Cheers!
Quoting Jim Ford <james.ford@...>:
Thanks, Bruce and Lou.I figured as long as the units were powered up
everything should be OK. Just paranoid I guess.JimSent from my
Verizon, Samsung Galaxy smartphone
-------- Original message --------From: Lou Blasco
<vk3alb@...> Date: 4/12/20 9:02 PM (GMT-08:00) To:
[email protected] Subject: Re:
[HP-Agilent-Keysight-equipment] GPSDO connected to external ref
inputs Agree with Bruce,Been running my 10 MHz 24x7 into a HP-5342A,
HP-8901B and Fluke 6060A/AN for years with no ill
effects.RegardsLouVK3ALB
|
Re: GPSDO connected to external ref inputs
The output of my x10 dist amp was running at about +12 dBm driving two radios and seven TE.? I put a 10dB pad on the input side because my Icom IC-9700 was complaining about the high level.? e.g.,?? Your +10 dBm is about 5 dB higher than normal practice but definitely won't hurt anything even if not mitigated.
best regards....
Frank Mackowick / WA3NHK
|
Re: GPSDO connected to external ref inputs
Jim,
?
I have a HP 5087A distribution amp feeding output from an older HP Z3816 GPSDO to several pieces of equipment.? These include:
?
Agilent E4425B signal generator
Agilent 8714ET network analyzer
EIP 548A microwave frequency counter
HP 3575A gain/phase meter
HP 5335A counter
HP 8920A RF communication test set
HP 53310A modulation domain analyzer
HP K34-9991A phase comparator
PTS 250 frequency synthesizer
Tracor 527E frequency difference meter
Tracor 895A linear phase/time comparator
Wayne Kerr PSG2400L signal generator
?
Except for the Z3816/5087 units most of the driven equipment is in a power-down state unless needed.? The port on the 527E is switchable to an external connector on the bench for other test applications.? In all, I feed one volt p-p or less to all of the equipment and it stays happy.? The advantage with the 5087A is that each output is individually adjustable to suit the needs of the attached hardware.
?
If any of your units cannot tolerate a 4V amplitude I suggest that you pick up an in-line 50 ohm attenuator either from your favorite auction site or from Mini-Circuits (no affiliation however I rely on them for their low-cost high-quality products).? 4 volts does seem a little excessive for most external reference inputs since I am assuming you are referring to RMS amplitude and not peak-to-peak value.
?
Greg
|
opt 001 upgrade of a 8511A, 8512A or 8515A
Can this be done as user upgrade? Has anyone done this or know of documentation for doing it?
I just bought an 08510-60102 for $65, so buying one or two more to allow multiplexing at that price would suit me. The IF mulitpexer board is available for $100 so I figure $200 or less to add a cable and multiplexer if I am patient.
Thanks,
Reg
|
Re: GPSDO connected to external ref inputs
Thanks, Clif.? One more data point on the curve.
Jim
Sent from my Verizon, Samsung Galaxy smartphone
toggle quoted message
Show quoted text
-------- Original message -------- From: Clif Holland <clif@...> Date: 4/13/20 5:08 AM (GMT-08:00) Subject: Re: [HP-Agilent-Keysight-equipment] GPSDO connected to external ref inputs
I have a GPSDO going to a 4 port distribution amp which is connected to a HP 8920A, HP 8594E, 8656B, and HP 53131A. The Ref signal is never turned off but the equipment is turned off daily. Never had a problem
|
hp8350b/hp83595a front panel mod
Hi all,
I'm trying to get my RF output on the front panel instead of at the rear on an hp83595a (26.5 GHz), while leaving the rear-panel option intact. I already have the front-panel connector from ebay and need to make a semi-rigid with 3.5mm connectors to (I guess, need to verify) the 55 dB attenuator.
Does anyone know what type of semi-rigid is used by HP in the hp83595a for the RF output and/or has suggestions for sourcing the connectors? (Sorry, complete semi-rigid noob here...) BTW, the front-panel connector I have is an APC3.5 feed through.
Thanks,
Jan
|
Re: Bill West keytop puller
On 4/13/20 11:43 AM, Chuck Harris wrote: As to deoxit being bunkum: I disagree. It improves the contact,
and does so for a long, long time. I am not saying that something
else wouldn't work... show me something better. I was wondering what "bunkum" meant. I stopped wondering whether DeOxit was any good when I had a tour of NIST in Maryland many years ago. There were cans of it everywhere. That told me all I needed to know; I went home and placed an order, and have been using it ever since. The makers do get needlessly huckstery about their product. That they do. I think that harms their credibility in the eyes of people who don't know their products' reputation, or those who have never heard of Cramolin. Fader lube is wonderful for resurrecting pots of all kinds. Yes. And their gold wipes are fantastic for gold-plated card-edge connectors. They leave a very slight bit of lubricant behind that makes boards much, much easier to insert and remove. We go through them like crazy at the museum, for Unibus and Qbus boards in particular. (you know how tough they can be to insert and remove) -Dave -- Dave McGuire, AK4HZ New Kensington, PA
|
Re: GPSDO connected to external ref inputs
No problem - I do think 4v is a bit high though - any radiation problems?
Cheers!
Quoting Jim Ford <james.ford@...>:
toggle quoted message
Show quoted text
Thanks, Bruce and Lou.I figured as long as the units were powered up
everything should be OK.? Just paranoid I guess.JimSent from my
Verizon, Samsung Galaxy smartphone
-------- Original message --------From: Lou Blasco
<vk3alb@...> Date: 4/12/20 9:02 PM (GMT-08:00) To:
[email protected] Subject: Re:
[HP-Agilent-Keysight-equipment] GPSDO connected to external ref
inputs Agree with Bruce,Been running my 10 MHz 24x7 into a HP-5342A,
HP-8901B and Fluke 6060A/AN for years with no ill
effects.RegardsLouVK3ALB
|
Understood. It turned out that disassembling the pot wasn't all that difficult. Maybe I can remember to report back next year... ;)
Thanks,
Barry - N4BUQ
toggle quoted message
Show quoted text
----- Original Message ----- From: "Dave Wise" <david_wise@...>
To: [email protected]
Sent: Monday, April 13, 2020 11:01:03 AM
Subject: Re: [HP-Agilent-Keysight-equipment] HP 334A Balance Pots
That's good news, Barry. I didn't mean to advocate drilling except as a last
resort. In my case the track was bad, with high end-to-end resistance.
Please report in a year; I'd like to know how well the cleaning persists.
Dave Wise
________________________________________
From: [email protected]
<[email protected]> on behalf of n4buq via groups.io
<n4buq@...>
Sent: Sunday, April 12, 2020 3:03 PM
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] HP 334A Balance Pots
I'm happy to report that I was able to disassemble the noisy pot, clean out
the carbon dust, put a couple of drops of DeoxiT on the moving parts (wish I
had FaderLub but...) and reassembled it. It works so well now! Glad I
decided to do that.
Thanks,
Barry - N4BUQ
----- Original Message -----
From: "Barry" <n4buq@...>
To: [email protected]
Sent: Friday, April 10, 2020 2:22:53 PM
Subject: Re: [HP-Agilent-Keysight-equipment] HP 334A Balance Pots
I may try prying them open as well so as to be a bit more careful about how
I'm cleaning things in there.
Thanks,
Barry - N4BUQ
----- Original Message -----
From: "Richard Knoppow" <dickburk@...>
To: [email protected]
Sent: Friday, April 10, 2020 1:45:25 PM
Subject: Re: [HP-Agilent-Keysight-equipment] HP 334A Balance Pots
I have had some luck by prying the covers off for cleaning
and then crimping the tabs back. Can't be done where the pots are
stacked. AB were the best of the bunch are a pain to clean.
Ohmite sold these same pots under their own name. Same for fixed
composition resistors, Ohmite and AB are actually all AB. Ohmite
actually says this in some of its early advertising. I have no
idea what the story behind this was.
On 4/10/2020 11:30 AM, ebrucehunter via groups.io wrote:
Barry,
I have one of these instruments, but it is too buried to easily
allow inspection to determine what type of pots were used.
If the values are above 1k, they are probably not wire-wound.
If they happen to be Allen-Bradley (AB) pots, an age old remedy
is to take a 1/16-inch drill bit and drill a hole in the cover
- in the outer crevice - at the bottom of the pot away from the
terminals. The snout of a can of contact cleaner can then be
inserted in the drilled hole to blast the interior of the pot.
This cleaning will typically last for two or three years and
then you might have to do it again. The access hole remains
available. In drilling the hole, metal chips seem to be pulled
outward and are unlikely to enter the pot. But be careful so
that when the drill bit breaks through, that it does not enter
and damage the pot.
Often these pots can be cleaned by pulling outward on the shaft
-- it will move about 1/16-inch -- while quickly turning the
knob from end to end.
Bruce, KG6OJI
|
That's good news, Barry. I didn't mean to advocate drilling except as a last resort. In my case the track was bad, with high end-to-end resistance. Please report in a year; I'd like to know how well the cleaning persists. Dave Wise ________________________________________ From: [email protected] < [email protected]> on behalf of n4buq via groups.io <n4buq@...> Sent: Sunday, April 12, 2020 3:03 PM To: [email protected]Subject: Re: [HP-Agilent-Keysight-equipment] HP 334A Balance Pots I'm happy to report that I was able to disassemble the noisy pot, clean out the carbon dust, put a couple of drops of DeoxiT on the moving parts (wish I had FaderLub but...) and reassembled it. It works so well now! Glad I decided to do that. Thanks, Barry - N4BUQ ----- Original Message ----- From: "Barry" <n4buq@...>
To: [email protected]
Sent: Friday, April 10, 2020 2:22:53 PM
Subject: Re: [HP-Agilent-Keysight-equipment] HP 334A Balance Pots
I may try prying them open as well so as to be a bit more careful about how
I'm cleaning things in there.
Thanks,
Barry - N4BUQ
----- Original Message -----
From: "Richard Knoppow" <dickburk@...>
To: [email protected]
Sent: Friday, April 10, 2020 1:45:25 PM
Subject: Re: [HP-Agilent-Keysight-equipment] HP 334A Balance Pots
I have had some luck by prying the covers off for cleaning
and then crimping the tabs back. Can't be done where the pots are
stacked. AB were the best of the bunch are a pain to clean.
Ohmite sold these same pots under their own name. Same for fixed
composition resistors, Ohmite and AB are actually all AB. Ohmite
actually says this in some of its early advertising. I have no
idea what the story behind this was.
On 4/10/2020 11:30 AM, ebrucehunter via groups.io wrote:
Barry,
I have one of these instruments, but it is too buried to easily
allow inspection to determine what type of pots were used.
If the values are above 1k, they are probably not wire-wound.
If they happen to be Allen-Bradley (AB) pots, an age old remedy
is to take a 1/16-inch drill bit and drill a hole in the cover
- in the outer crevice - at the bottom of the pot away from the
terminals. The snout of a can of contact cleaner can then be
inserted in the drilled hole to blast the interior of the pot.
This cleaning will typically last for two or three years and
then you might have to do it again. The access hole remains
available. In drilling the hole, metal chips seem to be pulled
outward and are unlikely to enter the pot. But be careful so
that when the drill bit breaks through, that it does not enter
and damage the pot.
Often these pots can be cleaned by pulling outward on the shaft
-- it will move about 1/16-inch -- while quickly turning the
knob from end to end.
Bruce, KG6OJI
|
Re: Bill West keytop puller
On this, I have to completely agree: There is no cause to
remove the key caps from these keyboards to clean any part.
The keycaps become essentially welded on after a quarter
century. Where they may have pulled off easily when they
were new (if you used the manufacturer sanctioned puller),
now, things break.
The best way I have found to service these keyboards is to
wash them liberally with IPA. Liberally, meaning soaking,
dripping wet over a pan.
I use a pump sprayer filled with IPA over a pan to collect
the wash. I soak every part of the key mechanism, while
pressing the key multiple times to help wash the grease that
HP so foolishly used, out from under the springs.
After a thorough wash, I shake, and then blow off the remaining
IPA.
No lubrication is necessary, or even desirable. None. Don't
do it.
If a spring has taken a set, just bend it back to straight, and
put it back.
As to deoxit being bunkum: I disagree. It improves the contact,
and does so for a long, long time. I am not saying that something
else wouldn't work... show me something better.
The makers do get needlessly huckstery about their product.
Fader lube is wonderful for resurrecting pots of all kinds.
-Chuck Harris
Roy Thistle wrote:
toggle quoted message
Show quoted text
On Sun, Apr 12, 2020 at 02:05 PM, Maurice Smulders wrote:
What is the safest way to pull the keytops of a Bill West keyboard? I have
a 3325A and 8131A with sticky keys I want to fix...
AT
From 3:10 to 5:08... Curious Mark pulls the caps off a 3325A Bill West keyboard... and shows cleaning and lubing (with DeoxiT... which I think, DeoxiT is mostly bunkum). There is a shot of the tool he is using (looks like a small nail puller) at exactly 4.03/15:18. Seems a bit brutal... He tore the plunger out of the 7 key's frame! (...you can see that if you watch closely.)
If you look in the comments you'll see? Raymond Domp Frank's? recommendations (...same as he has posted in this thread.)
Best regards and wishes.
Roy
|
Thanks for all the replies, ? further poking hasn't revealed much yet, though I haven't completely ruled out the front end connector yet (seems solid though). Power supplies seem ok. ? I'm slightly suspicious of the sampler driver at this point. It occurs to me that if somehow that were damaged in a way that suppressed even harmonics (idk how that could occur yet though) this seems like the symptoms that I'd expect to see. ? Further troubleshooting of that is on hold while I fix my 8555A, which apparently decided now was the perfect time to lose its +10V and +20V power supply:( But once that's fixed I'll come back to this. ? Best, Daniel ? ? Check the power lines for ripple. I have a HP 5351B dumpster find, that did not work to well, needed higher input level, and did not work on all frequencies... It was missing one of the capacitors in the power supply, so lots of ripple. After finding the missing cap, and putting in a new one, it behaved great. Might not be your issue, but its an easy check.. ?
toggle quoted message
Show quoted text
On Fri, Apr 10, 2020 at 7:49 PM Daniel Sheen < danielsheen5@...> wrote: Hi all, ? I have a rather bizarre problem with a 5343A counter. Observed symptoms are that the high frequency input works reliably up to about 250MHz, reads garbage from around 250MHz-800MHz, and then seems to work fine from 800MHz upward through 1GHz (I only have an 8640B handy atm so can't really test higher than that). Cranking up the input power a bunch seems to reduce the dead band to closer to 400MHz-600MHz. ? The counter initially failed troubleshooting step 4 in the service manual. That was resolved by replacing the A10 board (from another counter with a blown sampler). It now passes step 5 and is marginal on step 6 (sensitivity is a little lower than it ought to be but it seems to work). ? The one other observation I've made is the IF amplitude into and out of A11 drops off rapidly if an input is tuned into the range of frequencies that it's insensitive to, but I'm struggling to find a plausible explanation for why that could occur given my understanding of the counter design (that wouldn't also make higher frequencies not be detected). ? If anyone has any thoughts or has run into this before, suggestions would be appreciated. ? Thanks, Daniel, KC1EPN
|
Re: HP 8920A spectrum analyser amplitude calibration
Ah! Got it. I shall remember if I should need it.
Rgds
toggle quoted message
Show quoted text
On 13 Apr 2020, at 13:13, george edmonds via groups.io <G6HIG@...> wrote:
?
Hi Colin
CLIP stands for Component Layout Information Package, HP speak for circuit diagrams and board layouts. When you get a service manual for later HP kit it is the text only, the interesting stuff is in the CLIP.
Kind regards
George G6HIG
On Monday, 13 April 2020, 11:28:34 BST, Colin Smithers <colin@...> wrote:
Hi George,
?
Now were motoring! I was unaware of Amtronix, thanks. Clearly they must know and perhaps might help.
?
You will have to inlighten me as to what the CLIP is, please?
?
Kind regards
?
Colin
?
From: [email protected] [mailto:[email protected]]On Behalf Of
george edmonds via groups.io
Sent: 13 April 2020 10:32
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] HP 8920A spectrum analyser amplitude calibration
?
Hi Colin
I do not understand the posting about making manual adjustments, I can see that using the suggested approach for fault finding is valid, but not for adjusting the calibration given that the 8920A is software calibrated only. The calibration constants are held
in none volatile memory in each module which are read into main ram on power up.
As far as I remember the Cal switch just enables and disables the ability to carry out a calibration via the GPIB port.
Prior to 1999 all 8920A's had a mechanical attenuator, which your appears to have. It then became option 055 with the PIN diode attenuator as option 009.
The only remaining commercial support for the 8920A is provided by Amtronix in the US, who do provide full repair and calibration facilities for them.
Are you aware of the information at Amtronix.com/diagnostic/htm?
Do you have the CLIP for the 8920A, I do if you need it.
Regards
George G6HIG
On Monday, 13 April 2020, 07:25:14 BST, Colin Smithers <colin@...> wrote:
Hi George,
?
Just measured, the unit is substantially flat (+/-0.2dB) across the range from 1MHz -500MHz, but with the Main input around -2.2dB and Ant input -2.6dB. But these back my point that for a subsidiary
function this Spec An is very good indeed.
?
The Main input shows >-25dB return loss, I am yet to measure the Ant input.
?
Looking at the Spec An block diagram two things stick out:-
?
-?????????A signal switch near the front with a Cal setting
-?????????A 4-in-1 DAC which seems to control variable gain amp as well as IF filters
?
So I remain thinking that ?I¡¯m on the hunt for a deeply buried Cal routine. Fiddling with loss internally is clearly possible (from another rreply), but only finding the routine would be future
proof.
?
Thanks for info about that HP IEEE 488 program. My Win95 machine does still boot up¡??J
Gonna have to find an IEEE488 card though. Or info on what the code drives.
?
Rgds
?
Colin
?
From: [email protected] [mailto:[email protected]]On Behalf Of
george edmonds via groups.io
Sent: 13 April 2020 00:37
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] HP 8920A spectrum analyser amplitude calibration
?
Hi Colin
I have also seen your posting on the HP8924 group which has additional information in it.
First it is very important to say at what frequencies you are experiencing the errors, the main reason for this is that the front panel on a HP8920A gives the operating range as 400KHz to 1GHz. This is ONLY true for the original electromechanical attenuator
version, all the ones that I have seen in the UK have had the PIN diode attenuator fitted which in my experience cannot be trusted below 50MHz. HP claimed 10MHz in their dreams.
Most of these test sets were purchased in the UK for analogue cellular use so the use of a PIN diode attenuator was not an issue.
I would first check to see if the input load measures 50 ohms at the N port, if it does then check and re torque the internal hard line SMA's using a torque spanner, this is very important to avoid damage to the SMA's. There were a number of optimal power handling
input loads available, it is not unknown for excess RF power to damage them.
I have not had to work on a 8920A for some years, but as I recall the calibration/adjustment was carried out via the GPIB port using dedicated HP software that would not run on anything later than Windows 95 or possibly 98.
Hope that this helps
George G6HIG
On Sunday, 12 April 2020, 22:18:40 BST, Colin Smithers <colin@...> wrote:
Hi all,
I have an 8920A and the spectrum analyser amplitude level is 3dB out. I have read enough to understand that calibrating this is a bit of a secret. Does anyone know it please?
Similarly for calibrating the power meter.
I have asked the same question on the 8924-series group.
Kind regards
Colin Smithers G4CWH
|
Re: HP 8920A spectrum analyser amplitude calibration
Hi Colin
CLIP stands for Component Layout Information Package, HP speak for circuit diagrams and board layouts. When you get a service manual for later HP kit it is the text only, the interesting stuff is in the CLIP.
Kind regards
George G6HIG On Monday, 13 April 2020, 11:28:34 BST, Colin Smithers <colin@...> wrote:
Hi George, ? Now were motoring! I was unaware of Amtronix, thanks. Clearly they must know and perhaps might help. ? You will have to inlighten me as to what the CLIP is, please? ? Kind regards ? Colin ?
From: [email protected] [mailto:[email protected]]On Behalf Of george edmonds via groups.io
Sent: 13 April 2020 10:32
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] HP 8920A spectrum analyser amplitude calibration ? Hi Colin
I do not understand the posting about making manual adjustments, I can see that using the suggested approach for fault finding is valid, but not for adjusting the calibration given that the 8920A is software calibrated only. The calibration constants are held in none volatile memory in each module which are read into main ram on power up.
As far as I remember the Cal switch just enables and disables the ability to carry out a calibration via the GPIB port.
Prior to 1999 all 8920A's had a mechanical attenuator, which your appears to have. It then became option 055 with the PIN diode attenuator as option 009.
The only remaining commercial support for the 8920A is provided by Amtronix in the US, who do provide full repair and calibration facilities for them.
Are you aware of the information at Amtronix.com/diagnostic/htm?
Do you have the CLIP for the 8920A, I do if you need it.
Regards
George G6HIG On Monday, 13 April 2020, 07:25:14 BST, Colin Smithers <colin@...> wrote: Hi George, ? Just measured, the unit is substantially flat (+/-0.2dB) across the range from 1MHz -500MHz, but with the Main input around -2.2dB and Ant input -2.6dB. But these back my point that for a subsidiary function this Spec An is very good indeed. ? The Main input shows >-25dB return loss, I am yet to measure the Ant input. ? Looking at the Spec An block diagram two things stick out:- ? -?????????A signal switch near the front with a Cal setting -?????????A 4-in-1 DAC which seems to control variable gain amp as well as IF filters ? So I remain thinking that ?I¡¯m on the hunt for a deeply buried Cal routine. Fiddling with loss internally is clearly possible (from another rreply), but only finding the routine would be future proof. ? Thanks for info about that HP IEEE 488 program. My Win95 machine does still boot up¡??J Gonna have to find an IEEE488 card though. Or info on what the code drives. ? Rgds ? Colin ? From: [email protected] [mailto:[email protected]]On Behalf Of george edmonds via groups.io
Sent: 13 April 2020 00:37
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] HP 8920A spectrum analyser amplitude calibration ? Hi Colin
I have also seen your posting on the HP8924 group which has additional information in it.
First it is very important to say at what frequencies you are experiencing the errors, the main reason for this is that the front panel on a HP8920A gives the operating range as 400KHz to 1GHz. This is ONLY true for the original electromechanical attenuator version, all the ones that I have seen in the UK have had the PIN diode attenuator fitted which in my experience cannot be trusted below 50MHz. HP claimed 10MHz in their dreams.
Most of these test sets were purchased in the UK for analogue cellular use so the use of a PIN diode attenuator was not an issue.
I would first check to see if the input load measures 50 ohms at the N port, if it does then check and re torque the internal hard line SMA's using a torque spanner, this is very important to avoid damage to the SMA's. There were a number of optimal power handling input loads available, it is not unknown for excess RF power to damage them.
I have not had to work on a 8920A for some years, but as I recall the calibration/adjustment was carried out via the GPIB port using dedicated HP software that would not run on anything later than Windows 95 or possibly 98.
Hope that this helps
George G6HIG On Sunday, 12 April 2020, 22:18:40 BST, Colin Smithers <colin@...> wrote: Hi all,
I have an 8920A and the spectrum analyser amplitude level is 3dB out. I have read enough to understand that calibrating this is a bit of a secret. Does anyone know it please?
Similarly for calibrating the power meter.
I have asked the same question on the 8924-series group.
Kind regards
Colin Smithers G4CWH
|
Re: GPSDO connected to external ref inputs
I have a GPSDO going to a 4 port distribution amp which is connected to a HP 8920A, HP 8594E, 8656B, and HP 53131A. The Ref signal is never turned off but the equipment is turned off daily. Never had a problem
|
Re: HP 8920A spectrum analyser amplitude calibration
CLIP == Component Level Information Pack == Schematics and Parts Lists (no sorry no Theory of Operation sections) ? David ?
toggle quoted message
Show quoted text
From: [email protected] [mailto:[email protected]] On Behalf Of Colin Smithers
Sent: 13 April 2020 11:28
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] HP 8920A spectrum analyser amplitude calibration? Hi George, ? Now were motoring! I was unaware of Amtronix, thanks. Clearly they must know and perhaps might help. ? You will have to inlighten me as to what the CLIP is, please? ? Kind regards ? Colin ? From: [email protected] [mailto:[email protected]] On Behalf Of george edmonds via groups.io
Sent: 13 April 2020 10:32
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] HP 8920A spectrum analyser amplitude calibration ? Hi Colin
I do not understand the posting about making manual adjustments, I can see that using the suggested approach for fault finding is valid, but not for adjusting the calibration given that the 8920A is software calibrated only. The calibration constants are held in none volatile memory in each module which are read into main ram on power up.
As far as I remember the Cal switch just enables and disables the ability to carry out a calibration via the GPIB port.
Prior to 1999 all 8920A's had a mechanical attenuator, which your appears to have. It then became option 055 with the PIN diode attenuator as option 009.
The only remaining commercial support for the 8920A is provided by Amtronix in the US, who do provide full repair and calibration facilities for them.
Are you aware of the information at Amtronix.com/diagnostic/htm?
Do you have the CLIP for the 8920A, I do if you need it.
Regards
George G6HIG On Monday, 13 April 2020, 07:25:14 BST, Colin Smithers <colin@...> wrote: Hi George, ? Just measured, the unit is substantially flat (+/-0.2dB) across the range from 1MHz -500MHz, but with the Main input around -2.2dB and Ant input -2.6dB. But these back my point that for a subsidiary function this Spec An is very good indeed. ? The Main input shows >-25dB return loss, I am yet to measure the Ant input. ? Looking at the Spec An block diagram two things stick out:- ? -?????????A signal switch near the front with a Cal setting -?????????A 4-in-1 DAC which seems to control variable gain amp as well as IF filters ? So I remain thinking that ?I¡¯m on the hunt for a deeply buried Cal routine. Fiddling with loss internally is clearly possible (from another rreply), but only finding the routine would be future proof. ? Thanks for info about that HP IEEE 488 program. My Win95 machine does still boot up¡??J Gonna have to find an IEEE488 card though. Or info on what the code drives. ? Rgds ? Colin ? From: [email protected] [mailto:[email protected]]On Behalf Of george edmonds via groups.io
Sent: 13 April 2020 00:37
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] HP 8920A spectrum analyser amplitude calibration ? Hi Colin
I have also seen your posting on the HP8924 group which has additional information in it.
First it is very important to say at what frequencies you are experiencing the errors, the main reason for this is that the front panel on a HP8920A gives the operating range as 400KHz to 1GHz. This is ONLY true for the original electromechanical attenuator version, all the ones that I have seen in the UK have had the PIN diode attenuator fitted which in my experience cannot be trusted below 50MHz. HP claimed 10MHz in their dreams.
Most of these test sets were purchased in the UK for analogue cellular use so the use of a PIN diode attenuator was not an issue.
I would first check to see if the input load measures 50 ohms at the N port, if it does then check and re torque the internal hard line SMA's using a torque spanner, this is very important to avoid damage to the SMA's. There were a number of optimal power handling input loads available, it is not unknown for excess RF power to damage them.
I have not had to work on a 8920A for some years, but as I recall the calibration/adjustment was carried out via the GPIB port using dedicated HP software that would not run on anything later than Windows 95 or possibly 98.
Hope that this helps
George G6HIG On Sunday, 12 April 2020, 22:18:40 BST, Colin Smithers <colin@...> wrote: Hi all,
I have an 8920A and the spectrum analyser amplitude level is 3dB out. I have read enough to understand that calibrating this is a bit of a secret. Does anyone know it please?
Similarly for calibrating the power meter.
I have asked the same question on the 8924-series group.
Kind regards
Colin Smithers G4CWH
|
Re: HP 8920A spectrum analyser amplitude calibration
Hi George,
?
Now were motoring! I was unaware of Amtronix, thanks. Clearly they must know and perhaps might help.
?
You will have to inlighten me as to what the CLIP is, please?
?
Kind regards
?
Colin
?
toggle quoted message
Show quoted text
From: [email protected] [mailto:[email protected]]
On Behalf Of george edmonds via groups.io
Sent: 13 April 2020 10:32
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] HP 8920A spectrum analyser amplitude calibration
?
Hi Colin
I do not understand the posting about making manual adjustments, I can see that using the suggested approach for fault finding is valid, but not for adjusting the calibration given that the 8920A is software calibrated only. The calibration constants are held
in none volatile memory in each module which are read into main ram on power up.
As far as I remember the Cal switch just enables and disables the ability to carry out a calibration via the GPIB port.
Prior to 1999 all 8920A's had a mechanical attenuator, which your appears to have. It then became option 055 with the PIN diode attenuator as option 009.
The only remaining commercial support for the 8920A is provided by Amtronix in the US, who do provide full repair and calibration facilities for them.
Are you aware of the information at Amtronix.com/diagnostic/htm?
Do you have the CLIP for the 8920A, I do if you need it.
Regards
George G6HIG
On Monday, 13 April 2020, 07:25:14 BST, Colin Smithers <colin@...> wrote:
Hi George,
?
Just measured, the unit is substantially flat (+/-0.2dB) across the range from 1MHz -500MHz, but with the Main input around -2.2dB and Ant input
-2.6dB. But these back my point that for a subsidiary function this Spec An is very good indeed.
?
The Main input shows >-25dB return loss, I am yet to measure the Ant input.
?
Looking at the Spec An block diagram two things stick out:-
?
-?????????A
signal switch near the front with a Cal setting
-?????????A
4-in-1 DAC which seems to control variable gain amp as well as IF filters
?
So I remain thinking that ?I¡¯m on the hunt for a deeply buried Cal routine. Fiddling with loss internally is clearly possible (from another rreply),
but only finding the routine would be future proof.
?
Thanks for info about that HP IEEE 488 program. My Win95 machine does still boot up¡??J
Gonna have to find an IEEE488 card though. Or info on what the code drives.
?
Rgds
?
Colin
?
From: [email protected]
[mailto:[email protected]]On Behalf Of george edmonds via groups.io
Sent: 13 April 2020 00:37
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] HP 8920A spectrum analyser amplitude calibration
?
Hi Colin
I have also seen your posting on the HP8924 group which has additional information in it.
First it is very important to say at what frequencies you are experiencing the errors, the main reason for this is that the front panel on a HP8920A gives the operating range as 400KHz to 1GHz. This is ONLY true for the original electromechanical attenuator
version, all the ones that I have seen in the UK have had the PIN diode attenuator fitted which in my experience cannot be trusted below 50MHz. HP claimed 10MHz in their dreams.
Most of these test sets were purchased in the UK for analogue cellular use so the use of a PIN diode attenuator was not an issue.
I would first check to see if the input load measures 50 ohms at the N port, if it does then check and re torque the internal hard line SMA's using a torque spanner, this is very important to avoid damage to the SMA's. There were a number of optimal power handling
input loads available, it is not unknown for excess RF power to damage them.
I have not had to work on a 8920A for some years, but as I recall the calibration/adjustment was carried out via the GPIB port using dedicated HP software that would not run on anything later than Windows 95 or possibly 98.
Hope that this helps
George G6HIG
On Sunday, 12 April 2020, 22:18:40 BST, Colin Smithers <colin@...> wrote:
Hi all,
I have an 8920A and the spectrum analyser amplitude level is 3dB out. I have read enough to understand that calibrating this is a bit of a secret. Does anyone know it please?
Similarly for calibrating the power meter.
I have asked the same question on the 8924-series group.
Kind regards
Colin Smithers G4CWH
|
Re: HP 8920A spectrum analyser amplitude calibration
Hi Colin
I do not understand the posting about making manual adjustments, I can see that using the suggested approach for fault finding is valid, but not for adjusting the calibration given that the 8920A is software calibrated only. The calibration constants are held in none volatile memory in each module which are read into main ram on power up.
As far as I remember the Cal switch just enables and disables the ability to carry out a calibration via the GPIB port.
Prior to 1999 all 8920A's had a mechanical attenuator, which your appears to have. It then became option 055 with the PIN diode attenuator as option 009.
The only remaining commercial support for the 8920A is provided by Amtronix in the US, who do provide full repair and calibration facilities for them.
Are you aware of the information at Amtronix.com/diagnostic/htm?
Do you have the CLIP for the 8920A, I do if you need it.
Regards
George G6HIG
On Monday, 13 April 2020, 07:25:14 BST, Colin Smithers <colin@...> wrote:
Hi George, ? Just measured, the unit is substantially flat (+/-0.2dB) across the range from 1MHz -500MHz, but with the Main input around -2.2dB and Ant input -2.6dB. But these back my point that for a subsidiary function this Spec An is very good indeed. ? The Main input shows >-25dB return loss, I am yet to measure the Ant input. ? Looking at the Spec An block diagram two things stick out:- ? -?????????A signal switch near the front with a Cal setting -?????????A 4-in-1 DAC which seems to control variable gain amp as well as IF filters ? So I remain thinking that ?I¡¯m on the hunt for a deeply buried Cal routine. Fiddling with loss internally is clearly possible (from another rreply), but only finding the routine would be future proof. ? Thanks for info about that HP IEEE 488 program. My Win95 machine does still boot up¡??J Gonna have to find an IEEE488 card though. Or info on what the code drives. ? Rgds ? Colin ?
From: [email protected] [mailto:[email protected]]On Behalf Of george edmonds via groups.io
Sent: 13 April 2020 00:37
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] HP 8920A spectrum analyser amplitude calibration ? Hi Colin
I have also seen your posting on the HP8924 group which has additional information in it.
First it is very important to say at what frequencies you are experiencing the errors, the main reason for this is that the front panel on a HP8920A gives the operating range as 400KHz to 1GHz. This is ONLY true for the original electromechanical attenuator version, all the ones that I have seen in the UK have had the PIN diode attenuator fitted which in my experience cannot be trusted below 50MHz. HP claimed 10MHz in their dreams.
Most of these test sets were purchased in the UK for analogue cellular use so the use of a PIN diode attenuator was not an issue.
I would first check to see if the input load measures 50 ohms at the N port, if it does then check and re torque the internal hard line SMA's using a torque spanner, this is very important to avoid damage to the SMA's. There were a number of optimal power handling input loads available, it is not unknown for excess RF power to damage them.
I have not had to work on a 8920A for some years, but as I recall the calibration/adjustment was carried out via the GPIB port using dedicated HP software that would not run on anything later than Windows 95 or possibly 98.
Hope that this helps
George G6HIG On Sunday, 12 April 2020, 22:18:40 BST, Colin Smithers <colin@...> wrote: Hi all,
I have an 8920A and the spectrum analyser amplitude level is 3dB out. I have read enough to understand that calibrating this is a bit of a secret. Does anyone know it please?
Similarly for calibrating the power meter.
I have asked the same question on the 8924-series group.
Kind regards
Colin Smithers G4CWH
|
Dave McGuire
David C. Partridge