Keyboard Shortcuts
ctrl + shift + ? :
Show all keyboard shortcuts
ctrl + g :
Navigate to a group
ctrl + shift + f :
Find
ctrl + / :
Quick actions
esc to dismiss
Likes
- HP-Agilent-Keysight-Equipment
- Messages
Search
|
Re: HP 3325A Phase noise (and reference clock) as discussed in CuriousMarc's video of Soyuz Clock, Part 6
Several years ago I did phase noise measurements on a 3325B. The
toggle quoted message
Show quoted text
results are here: The unit had the standard (not high-stability) oscillator and my measurements showed a dramatic PN improvement using a high-quality external reference. John On 4/14/20 8:50 AM, Raymond Domp Frank wrote:
Hi Szabolcs, |
|
Re: HP 3325A Phase noise (and reference clock) as discussed in CuriousMarc's video of Soyuz Clock, Part 6
Hi Szabolcs,
Only now it occurs to me that Option 001 is only exactly like the Options 001 in so many HP instruments: High Stability Oscillator (OCXO)! Until now I thought that it (also) contained some special (PLL-) circuitry, tuning or performing whatever trick to be able to spec. the phase noise. I hadn't even looked it up in the catalog... I'd think the lower phase noise from an OCXO source only applies at higher frequencies, not at frequencies like a few Hz, where apparently, PLL influence drowns out the OCXO's good phase noise specs. At jitter amounts in the tens of ppm's at LF, the only good the OCXO could do is less phase jitter into the PLL (control) circuitry. Is the incoming (reference) phase jitter the main/only component? IOW, is the frequency reference source the main/only cause of the huge phase noise at LF? Raymond |
|
Re: HP 334A Balance Pots
Just as a slight follow-up, it occurred to me that if this pot does go bad, I can replace the plastic resistive component as well as the wiper with parts from a NEW or NOS pot from the same series (they're Clarostat pots). Those parts are interchangeable between standard and concentric pots.
toggle quoted message
Show quoted text
Thanks, Barry - N4BUQ ----- Original Message -----
From: "Barry" <n4buq@...> |
|
5335A keyboard LED interesting thing
Hi,
Sorry if it is know, but it is new to me. While fixing the LED in one of the keys of my 5335A, I have stumbled upon some interesting thing. All keys have LEDs in them, even those where the keycap does not have the transparent insert. So far this is not unusual, as manufacturing could probably be simpler if one has to deal with only one type of keys, although I found no particular reason why the actual LED had to be populated as it is not part of the key mechanism. I was able to replace it with some fiddling and without removing the switch. However, what I found is that the Gate Cycle Norm key is supposed to have no light, but the LED there is actually controlled by the CPU and it lights up if pressed. I saw it light up when the keyboard was removed from the instrument front panel for testing and then tried it with the cap removed to verify. So likely at the late stages of development sometime, someone decided that this button should not light, and it was accomplished by replacing the keycap but not modifying the firmware. I find this quite amusing. Szabolcs |
|
Re: Application Note 122: EMI Measurement Procedure
toggle quoted message
Show quoted text
On Apr 14, 2020, at 09:33, Richard Katezansky <rk@...> wrote:
|
|
Re: HP 3325A Phase noise (and reference clock) as discussed in CuriousMarc's video of Soyuz Clock, Part 6
Hi,
toggle quoted message
Show quoted text
Yes, I somehow expected that without the OCXO option there isn't much point in specifying phase noise. At least it wouldn't be any stellar value anyway. The good news is that you can feed it any good quality 10MHz reference and I would assume one could still rely on the phase noise specification. Since option 001 is nothing than a separate OCXO which you loop back into the ext ref input on the back panel of the instrument, I would find it hard to believe that with any other equal or better reference the 3325A/B would not meet the specification. Btw if you are still planning to upgrade your instrument, opt 001 is the same for A and B model and usually you can find a -broken- 3325A with opt 001 cheaper that the quite rare opt 001 oscillator by itself. Go figure! But yeah, I know the feeling when you have an instrument without some option and the test equipment OCD kicks is to desperately obtain that missing option, even if there is no other rational reason for it :-) Szabolcs Raymond Domp Frank <hewpatek@...> ezt ¨ªrta (id?pont: 2020. ¨¢pr. 14., K, 14:50):
|
|
Re: HP 8920A spectrum analyser amplitude calibration
Hi Colin Having checked the Agilent specifications for a 8920A RF power meter accuracy is plus or minus 10% of reading. Spectrum analyser is plus or minus 2.5dB. Looks like your 8920A is almost in calibration. Best wishes George G6HIG On Tuesday, 14 April 2020, 13:50:34 BST, Colin Smithers <colin@...> wrote: Gem! Thank you. Colin On 14 Apr 2020, at 13:08, jyelmgren via groups.io <jyelmgren@...> wrote:
|
|
Re: hp8350b/hp83595a front panel mod
¿ªÔÆÌåÓýHi Jan, The semi-rigid is standard .141 with precision stainless steel SMA connectors. It is L-shaped with a bit of an offset in the leg that goes to the front panel connector: Please note that the front panel connector is not just a
feedthrough, it has a DC block built into it. If you substitute a
straight bulkhead connector you lose that DC protection. I would
recommend removing the rear panel connector and using it instead.
You can always revert to rear panel output if you need to. There's
precious little room in the plug in to add a relay to allow
switching back and forth. If you use your 3.5mm bulkhead connector
I'd recommend a connector saver and an external DC block. Better
yet go to an SMA bulkhead connector instead of the 3.5mm as I'd
bet you'll be using SMA for everything else as 3.5mm connectors
are expensive and fragile. Since you will be mating 3.5mm to SMA do be aware that those
connections should be torqued to 5 in/lb, .57 N/m, not the usual
SMA to SMA or 3.5 to 3.5 value of 8 in/lb, .9 N/m If amplitude accuracy needs to be within specs you'll need to do an RF amplitude calibration once everything is in place as you've made changes to the RF output path. Good luck with the project and let us know how it goes. I've switched an 8340B from rear to front panel output but my 83595A's came with front panel output so I've not needed to modify them.
Cheers, Steve K. WB?DBS
On 4/13/2020 11:27 AM, Jan de Jongh
wrote:
Hi all, |
|
Re: HP 8920A spectrum analyser amplitude calibration
toggle quoted message
Show quoted text
On 14 Apr 2020, at 13:08, jyelmgren via groups.io <jyelmgren@...> wrote:
|
|
Re: HP 3325A Phase noise (and reference clock) as discussed in CuriousMarc's video of Soyuz Clock, Part 6
Hi Szabolcs,
Thanks for your reaction and pointer. I have been able to download the 3325B service manual you refer to, from the Agilent website, no problem... At least now I've got some numbers re. the phase noise. The qualitative part is well known of course but as Lord Kelvin said... Apparently, one needed to buy the option 1 version to get phase noise specified. Re. Marc: It seems so much out of character (as far as we outsiders know). It's annoying and spoils my fun watching his videos. Just banning and no reaction whatsoever to two polite requests for explanation c.q. restoration to his known email addresses is rude but apparently in-character. It's possible that he saw (/youtube shows?) each iteration of the main message (I edited it maybe five times in quick succession) as a separate reaction, much like the individual mail messages that we sometimes get in a digest with each editing change by a poster but resulting in one changed online post. Online, I only saw just the most recent version. OTOH, I'm sure he would have recognized it and could have admonished me to better prepare and not edit my reactions if he wanted. I've been thinking about contacting Ken but I'd dislike going behind Marc's back and Ken does not decide what Marc says in his videos. I can't imagine Ken not seeing (upfront) what I saw, it's not rocket science (pun intended). In that respect, it's enough for me that I had the thought and was able to verify it. Anyway, I've got some quantitative info on the phase noise subject for the 3325(B, option 1)... Raymond |
|
Re: HP 8920A spectrum analyser amplitude calibration
¿ªÔÆÌåÓýTemporary fix ? All, ? Based on some thoughts delivered by this thread I have found a temporary fix, but it is really only that: ? -????????? Set SA working -????????? Inject known power into Ant port -????????? Set to 1dB/div and establish the error (needs the precision of being on 1dB/div) -????????? Go to RF Analysis screen -????????? Inject known power into main port and establish that error -????????? Go into CONFIG screen -????????? Set port offsets ¡®ON¡¯ -????????? Set those error offsets in the Ant port and Main port windows, to 2 decimal places -????????? Save these settings to internal non-volatile memory as a pre-set by a useful name. I use ¡®CALFIX¡¯ -????????? As the CONFIG screen port offsets do not survive power reset, on first turning the instrument on, recall ¡®CALFIX¡¯ ? The instrument is now reading far more accurately on both Ant input dBm and the Power Meter, and these can all be done from the front panel. ? There remains another offset on SA accuracy from the Main RF port that can¡¯t be removed without also affecting the Power Meter. ? But as I said, it¡¯s not a particularly desirable or elegant solution, though I hope this is helpful to some ? Rgds ? Colin G4CWH/ZF2CA/M3E |
|
Re: HP 3325A Phase noise (and reference clock) as discussed in CuriousMarc's video of Soyuz Clock, Part 6
Hi,
toggle quoted message
Show quoted text
There is a version of the 3325B service manual, which has some service notes and other documents inside. It used to be available at Keysight's web page, I can't find it now there, but I have it saved if you need it. It is manual part no. 03325-90003. On page 186 it discusses the jitter and concludes that you have to measure it right (with very long gate times) to be in spec. It also says that phase noise is only specified for opt 001 instruments, and it is -60dBc for 30kHz at center of 20MHz. It is interesting of what you say about Marc. I had the opposite experience, I pointed out to him in the previous video, where he fixed the 3325A that he should actually read the whole adjustment section in the SM, as having the PSU voltages in spec is the first step, they may need to be adjusted out of spec in later steps when adjusting the AD levels. He even responded to my post and i was not banned. However, you may want to contact Ken, he seems to be quite responsive on his - absolutely fascinating - blog, where among other awesome stuff, he discusses the same Soyuz clock. Szabolcs Raymond Domp Frank <hewpatek@...> ezt ¨ªrta (id?pont: 2020. ¨¢pr. 12., V, 18:42):
|
|
Re: WTB: 3467A Printer module.
You may consider signing up with??They will prov ide you with a USA address. From there they will ship to you. Just a happy user, no affiliation. Regards, Harke
On Tuesday, April 14, 2020, 12:27:26 AM GMT+2, Jared Cabot via groups.io <jaredcabot@...> wrote:
Hmm, seems that ebay seller doesn't want to ship internationally.... Does anyone else have a printer module they don't need? |
|
Re: HP 8920A spectrum analyser amplitude calibration
¿ªÔÆÌåÓýHi George,I have assumed that all I am fixing are long term calibration drifts as I use this unit at home with its last formal cal a long time ago.
Given the wide formal spec and the third party SA build, I now suspect that cal error could have been in place for literally decades. But again, the performance is very stable which is why it would be great to get it right. Clearly the hardware has the
capability for it.
Kind regards
Colin On 14 Apr 2020, at 01:04, george edmonds via groups.io <G6HIG@...> wrote:
|
|
Re: GPSDO connected to external ref inputs
With a disciplined oscillator, the crystal oscillator will determine the short term characteristics of the output; the GPS will determine the "long term stability".
If you use a poor crystal oscillator, you will have jitter and most likely poor phase noise performance. The same applies with the Rb standards. The short term performance is that of the crystal oscillator. The Rb device is not a primary standard with so so short term stability; That is why they contain to good quality crystal oscillator. In addition you still have to calibrate it once in a while. With GPS that isn't necessary. A disciplined oscillator, using GPS should result in about 100 times better stability than that of the crystal oscillator alone. Again, if you use a poor crystal oscillator, the performance will not be too great. Mine use SC crystal oscillators in ovens. I use two different disciplined oscillators in two location at home. I have compared them using a vector voltmeter and can see some phase changes now and again, but I don't think that for my purposes that is important. Stuart K6YAZ Los Angeles, USA |
|
Re: GPSDO connected to external ref inputs
¿ªÔÆÌåÓýThat's sorta my attitude, too.? For less than $250, I have a frequency reference that is better than I will ever need, and I don't have the time to fool with anything better.? Now, if they were to give me that rubidium frequency standard that crapped out at work, I would try to revive it with a hair dryer or heat gun.? Just to have holdover.? It's really embedded in a rack, though, and not easy to get out, so I haven't volunteered to remove it.? ?Yes, I live on the Time Nut Precipice, the TNP for short! Jim Ford? Sent from my Verizon, Samsung Galaxy smartphone -------- Original message -------- From: "saipan59 (Pete)" <[email protected]> Date: 4/13/20 8:02 PM (GMT-08:00) Subject: Re: [HP-Agilent-Keysight-equipment] GPSDO connected to external ref inputs You are standing on the precipice of becoming a Time Nut. FairThanks Dave! I'm sorta already there, although my focus is "obsolete gear that I'm unlikely to ever use". For example, the super-cool HP115AR Clock, with its mechanical display, and really awkward startup procedure... Here is a very short video of it running, working with the audio from WWV on a Hammarlund SP-600 rcvr (not shown), signal displayed by an HP122AR scope, and comparing the WWV timing to the oscillator in an HP5245L (not shown). But I'm thinking I'll set up this GPSDO in a little box, so that I can point to it and say "the output of that thing is really accurate". Pete |
|
Re: GPSDO connected to external ref inputs
On Mon, Apr 13, 2020 at 08:08 PM, Dave McGuire wrote:
You are standing on the precipice of becoming a Time Nut. FairThanks Dave! I'm sorta already there, although my focus is "obsolete gear that I'm unlikely to ever use". For example, the super-cool HP115AR Clock, with its mechanical display, and really awkward startup procedure... Here is a very short video of it running, working with the audio from WWV on a Hammarlund SP-600 rcvr (not shown), signal displayed by an HP122AR scope, and comparing the WWV timing to the oscillator in an HP5245L (not shown). But I'm thinking I'll set up this GPSDO in a little box, so that I can point to it and say "the output of that thing is really accurate". Pete |
Jos Raven PA0AMX
saipan59 (Pete)
to navigate to use esc to dismiss