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
Search
8566B Performance Verification with 8620C
|
I have been puttsing through the 8566B Performance Tests and Adjustments manual for fun and am wondering if I can use an 8620C (with associated plugin) for the frequency response check instead of the 8340 series. To me, it seems the main thing you need is a nice flat sweep at -10dBm. However, I am an RF neophyte and have never used a dedicated sweeper.
?
8566B flatness performance requirements (at -10dBm):
0.06 to 2.5 GHz is 0.6dB 2 to 12.5 GHz is 1.7dB? 12.5 to 20 GHz is 2.2dB 20 to 22 GHz is 3dB ?
8620C flatness with 86222A ?(0.01 to 2.4 GHz) is 0.25dB internally leveled and 0.1dB externally leveled.
8620C flatness with 86290B (2 to 18.6 GHz) is 0.7dB to 0.9dB internally leveled and 0.15dB externally leveled. ?
Seems to be within the spec up to 18.6 GHz. 86222A hits it on the lower frequencies even with internal leveling. However, I am not sure about getting these specs at an output of -10dBm. Do I need an attenuator? Will this compromise the output flatness?
?
Seemed worthwhile to compare the 8620 to the quoted 8340A. However, from a quick glance at the 8340A spec sheet, it is not as straightforward for externally leveled info. For the internally leveled flatness at output power of -10dBm it quotes 0.9dB (0.01 to 2.3 GHZ), 1.6dB (2.3 to 20 GHz), 2.1dB (20 to 26.5 GHz). So, this makes sense why you need to have the 8340 externally leveled in order to be within the flatness spec of the 8566B.
?
As a side note, it is interesting to view the progression from "Sweep Oscillator" to "Synthesized Sweeper" in the HP Catalogs and how it really seemed to be a game changer.
?
Thanks for any insights!
?
-Michael Bierlein |
|
Just a follow up on this. I got a hold of the Performance Tests section of an old 8566A manual (Part 4 of the "Operating and Service Manual" which is contained in Volume 2, such a confusing manual history for this line). They DO use an 8620C with 86222A for up to 2.4GHz and the 86290B Opt H08 (option gets it up to 22GHz) for the higher frequencies. Interestingly for the 2.4GHz sweep, you dont use the external ALC, but rather you take power meter readings at select frequencies before the sweep, then do the sweep, then correct the displayed results with your previous power meter readings. All of this is done with a 20dB attenuator at the input of either the 8566 or the power sensor. Power meter is used to set output of 86222A to -15dB (after the attenuator). The higher frequency setup has you use external ALC but with diode detector (33330C) feedback instead of power meter feedback like in the later manuals. You still set the 86290B output power with a power sensor/meter beforehand. So the 86222A seems nice for the lower frequency check- I will go that route and hold out for an 8340A/B later. The exotic H08 option for the 86290B seems harder to come by than any 8340A/B. Of course could probably check on the 8350 series, too....? -Michael Bierlein On Wed, Sep 25, 2024 at 18:36 Michael Bierlein via <bierl008=[email protected]> wrote:
|
|
The original frequency response alignment procedure for the 8566A was done with first and 8620C and an appropriate plug-in: 86222C for 10 MHz to 2.4 GHz; 86290C for 2GHz to 18.4 GHz; 86260C for 17.0 GHz to 22GHz, all normalized in output against a 436A using 8482A and 8485A sensors. The align procedure is in the 8566A Test and Adjust manual, later adapted to the 8350B and 83592A/B/C and 83595A/B/C plug-ins. And finally adapted for the 8340B/C sweeper. You can also use the 836xx family of sweepers for the Test and Align/ Performance Test procedures.
Don Bitters |
|
Don, thanks for the reply. That is exactly what I was looking for. 8620C + plugins seem to show up much cheaper than 8340A/B's (for good reason), so I feel this is a good option for those on a budget. I couldn't find a digital copy of any relevant manual specifying that this was acceptable but now I have a paper copy of the old 8566A Test and Adjust manual. WISH I had an 836xx series! Those who can't afford have to go back in time, but it's more fun that way anyway. -Michael Bierlein On Fri, Sep 27, 2024 at 00:17 Don Bitters via <donbitters=[email protected]> wrote: The original frequency response alignment procedure for the 8566A was done with first and 8620C and an appropriate plug-in: 86222C for 10 MHz to 2.4 GHz; 86290C for 2GHz to 18.4 GHz; 86260C for 17.0 GHz to 22GHz, all normalized in output against? a 436A using 8482A and 8485A sensors. The align procedure is in the 8566A Test and Adjust manual, later adapted to the 8350B and 83592A/B/C and 83595A/B/C plug-ins. And finally adapted for the 8340B/C sweeper. You can also use the 836xx family of sweepers for the Test and Align/ Performance Test procedures. |
|
Regarding older 8566 performance verification procedures: for those interested I have uploaded my scan of the 8566A performance verification. Fun to compare equipment requirements to the well known 8566B performance verification manual floating around.
?
It looks like the 8566A service material was released in four volumes, none of which appear to have separate part numbers, but rather share the same part number (either 08566-90001 or 08566-90006 ?). I have uploaded Volume 2 of this series which includes the Performance Tests as HP-8566A-Operating-and-Service-Manual-Volume-Two-Performance-Tests.pdf in the "?8566A, 8566B Spectrum Analyzer " directory.
?
I have also uploaded my scan of the 8566B performance verification (08566-90168) as the freebie one floating around has numerous gross text replacement errors. The one with errors is currently in the 8566 folder, so if people feel my new upload is unnecessary I will remove it, but you can still view it on my archive.org account ( ).?
?
I also have a handful of other HP manuals I've scanned and processed that I would like to upload. Should I upload here or maybe on the HPAK wiki or both?
?
-Michael Bierlein |
|
You might as well upload to both unless someone complains :)
?
Another option for a sweeper is the synthesised HP 83752A from the 1990s. These only cover up to 20 GHz but they can be found for sensible money sometimes. I've got one here and the internal levelling is remarkably good up to 20 GHz. Much better than the datasheet.
?
Many years ago I used it to check my HP 8566B analyser using external levelling up to about 10 GHz using a resistive splitter and I did the external levelling with a power meter. I found that the HP 8566B was much flatter than the spec. It was remarkably good from about 40 MHz to 2 GHz on the first range. Above 10 GHz I just connected the 83752A to the input of the 8566B using a really short Gore cable (after checking it with a power meter) and this gave good results to about 15 GHz. Again, much better than the spec limits. I didn't bother checking it above 15 GHz. This was quite a few years ago though. I stopped using the 8566B about 6/7 years ago after upgrading to something a bit more modern. Otherwise I would have checked it more recently.?
?
It's easy to get in a muddle with the Yig preselector tuning feature on the 8566B and this can mess up the flatness. I found I got the best results with the factory default settings. Whoever used the analyser before me (I purchased it as an ex-rental from Microlease in about 2010) had really messed up the Yig tuning using the front panel buttons and this had put some deep dips in the frequency response. Fortunately, the factory default settings proved to be very good once I reset it. |
to navigate to use esc to dismiss