EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: DerKeksinator on April 18, 2022, 01:29:40 pm
-
Hi, I recently acquired said spectrum analyzer. The seller couldn't tell me if or when parts got replaced, however it looks like someone might have taken out or replaced something at some point in time. It is not known if this happened before or after the last calibration, but judging by the markings on the cables and the converter, I assume this was not done by qualified service personnel.
The good news:
It's working fine and produces reasonable measurements across the whole band.
The not so good news:
1:
I was trying to judge how good the shielding of a 900MHz VCO is with an improvised near field probe, but I noticed a second tone at 900MHz, which was not from the VCO nor any other Equipment running in the background. After switching off everything else and terminating both, the RF-input and the Cal-output, it was still present. It's around 900MHz at -122dBm (see Image below).
[attach=1]
There are no other tones visible (300MHz, 450MHz, 600MHz and 1.8GHz).
It's amplitude is not affected by the attenuator setting.
It disappears with the case removed.
After removing the case I noticed that there's an unterminated connection on the second converter (A3A4 J3). The service manual notes that it's not connected in my case, however it does not state if it should be terminated or not. It's meant as the input for the High band of the first converter in other models according to the service manual.
[attach=2]
If I bring my flat hand above this connector while the unit is running it appears again.
Should I terminate it? Would that change other characteristics? Didi I miss something in the manual along the lines of, "terminate if unused"?
2:
When I openend the Case of the analyzer, a yellow jumper fell out, I asked the seller and he mentioned that it might belong somewhere on the Log Amplifier board should the NBW option not be installed. I'm not sold on this yet. Any Ideas, where it might belong, or if this statement is correct? I'm trying to find more information on that, but so far I couldn't find a satisfying answer.
3:
Because I wasn't sure when the battery was last replaced, I wrote down the calibration data and while doing so I noticed that some of the DAC values are at 255. This was after finding the yellow jumper, so I'm not sure if those would be different, should the jumper be missing somewhere.
[attach=3]
Maybe some of you can tell if there's something wrong with the data as well.
A big Thank You in advance!
Edit: It appears that my formatting with the images isn't really working out. This is my first post here, so I'm not entirely sure what went wrong.
-
On 1) The 8594E residual responses spec from 150kHz to 2.9GHz is <-90dBm. The signal you're seeing is way below that, so I would claim it's in spec. As far as A3A4-J3 is concerned, I found a couple of photos of other 8594E units here:
https://www.eevblog.com/forum/testgear/hp-8594e-ebay-score/msg713994/#msg713994 (https://www.eevblog.com/forum/testgear/hp-8594e-ebay-score/msg713994/#msg713994)
And here:
https://www.eevblog.com/forum/repair/hp-8594e-spectrum-analyzer-repair-(i-hope)/msg2197383/#msg2197383 (https://www.eevblog.com/forum/repair/hp-8594e-spectrum-analyzer-repair-(i-hope)/msg2197383/#msg2197383)
And neither has anything on that connector. You could try putting an SMA open on it (not a terminator) to see if that gets rid of the spurious signal, if it still bothers you.
FWIW, using your setup I also have a ~900MHz spurious signal on my 8595E, but it's smaller at -134dBm. The 8595E has a different front-end configuration than the 8594E, so it's not a great comparison.
When using an E or M field probe with these units, you usually want to be using an external LNA so you're not operating at such low input levels.
On 2) What kind of jumper? A 0.1" jumper or something else? Can you post a photo?
On 3) It's normal to have 0 and 255 DAC values. I have them on my 8595E in the same rows and they also appear in Assembly Level Repair manual in Figure 13-6: "Typical Calibration Data: Page 1".
Do your CAL FREQ and CAL AMPTD routines complete without error? If so and you're not seeing 0 or 255 values in other locations other than those shown in the "typical" table, you should be fine.
-
Thank you!
1)
Yes, it's fine, but now that I know it's there, it bothers me. I tuned the vco lower, so I don't have to see it.
Ok, putting an open on J3 further reduced the tone by 3dB. I'm now relly curious if termination would reduce it by ~10dB, as in your case. I'll see if there's any DC or HF signal on there and then decide if that's a smart idea. Unfortunately I don't have any hf millivoltmeter probes, designing and characterising some was one of the reasons to get a spectrumanalyzer after all...
2)
RM2.54mm 2x1 in yellow with gold/brass contacts.
[attach=1]
3)
RTFM would have helped, and yes it appears to be fine.
It passes Freq and Ampl cal as well, as the confidence test.
ADD)
The battery seems to be from 1996 though, no solder residue or any vidsibly different solderjoints. So I ordered a new one, after replacing one of the decoupling caps, because I broke it while removing the board.
[attach=2]
I noticed that the Analyzer didn't retain Date and Time after removing the RAM board, though. But maybe that's normal when unplugging it.
Thanks again!
-
On the RTC, yes, it's expected to lose the time/date. The RTC is on the processor board, so when you unplug the SRAM card to replace the battery, the RTC loses power. The DS1210 photo you posted is an NVRAM controller.
When I get some time in the next day or two I will pop the cover on my 8495E to look for any obvious, lone yellow jumper. But that's a long shot because 1) they may have used yellow jumpers everywhere or nowhere, and 2) the 8495E is different in many ways.
A better long shot is if someone with a 8594E could take a look and compare with photos you could post. That sounds like a bunch of busy-work, but I'm not sure how else one could go about this except by comparison.
A really hard way might be to try to find the HP part number for a 0.1" jumper by looking in the CLIP, then find out the places where that part number is referenced, and then figure out if it should be installed or not.
It's also possible the jumper is not important, and fell off because it was being stored in the "off" position on a single pin.