Author Topic: HP 8510C error message  (Read 1238 times)

0 Members and 1 Guest are viewing this topic.

Offline MRIKenTopic starter

  • Newbie
  • Posts: 9
  • Country: us
HP 8510C error message
« on: January 11, 2025, 09:11:35 pm »
I recently acquired a complete 8510C network analyzer system (8510C, 8515A S-parameter test set, 83631A Source)
I get two error messages, "No IF Found" and "Caution: Source 1 Failure".  The messages flash between these 2.
The other thing that is odd, if the start frequency is above 8GHz the messages go away and everything seems to work fine.

I have done some searching and can not find anything about why this is happening.
 

Offline jjoonathan

  • Frequent Contributor
  • **
  • Posts: 865
  • Country: us
Re: HP 8510C error message
« Reply #1 on: January 12, 2025, 12:06:49 am »
Can you show us log amplitude vs frequency?

There are two ways to measure it, both informative in their own way: internally, using the 8510C's a1 parameter measurement (incident signal, port 1), and externally, using a spectrum analyzer attached to port 1 (S11 mode, correction off, so it doesn't hop between ports).
 

Offline MRIKenTopic starter

  • Newbie
  • Posts: 9
  • Country: us
Re: HP 8510C error message
« Reply #2 on: January 12, 2025, 02:39:33 am »
One thing I did try is S12 and S21 response.  With the cables connected it is basically random noise below about 8 GHz, from 8-26 GHz the response is what one would expect to see.
 

Offline jjoonathan

  • Frequent Contributor
  • **
  • Posts: 865
  • Country: us
Re: HP 8510C error message
« Reply #3 on: January 12, 2025, 05:26:55 am »
S parameters are ratios. While it's true that we can look at the noise and hypothesize that amplitudes are low in the a (incident) and b (reflected) measurements, the 8510C also allows you to plot the a and b measurements directly. Plotting a will likely show that it is low up to 8GHz. It looks more like 13GHz to me, but amplitude will tell.

The next question is why. Is the source failing to produce a suitable signal below 13GHz? Or are the samplers failing to convert it down to IF? The 83631 has a frequency band that ends at 13.5GHz so I suspect the issue lies there, but the nature of the issue will need to be determined by tuning the source to various frequencies (using either the VNA in zero-span mode or the 8360 Front Panel Emulator) and inspecting its output with a spectrum analyzer.
 

Offline MRIKenTopic starter

  • Newbie
  • Posts: 9
  • Country: us
Re: HP 8510C error message
« Reply #4 on: January 12, 2025, 02:36:30 pm »
I do not have a spectrum analyzer yet, but I can use my scope and look at it to a bit over 1 GHz.  But I will need to get an SMA cable/adapter so I can check the output. 

It is odd that the 'noise' ends at different freq when I change the start freq. 
 

Offline jjoonathan

  • Frequent Contributor
  • **
  • Posts: 865
  • Country: us
Re: HP 8510C error message
« Reply #5 on: January 12, 2025, 03:22:27 pm »
Can you get access to the a trace?
 

Offline MRIKenTopic starter

  • Newbie
  • Posts: 9
  • Country: us
Re: HP 8510C error message
« Reply #6 on: January 12, 2025, 04:32:50 pm »
I am unsure of what you are referring to?
 

Offline jjoonathan

  • Frequent Contributor
  • **
  • Posts: 865
  • Country: us
Re: HP 8510C error message
« Reply #7 on: January 12, 2025, 05:58:42 pm »
In order to measure S11, the VNA applies an incident signal a1, observes the reflection b1, and calculates S11=b1/a1. The 8510C has buttons for S11, S12, etc but it is also capable of plotting a1 if you look in the menu near the S buttons. Plotting the a1 trace is far superior to trying to read tea leaves in the noise floor of S11, so I am asking you to plot a1.

I don't want to be a downer, but this is intro-level principle-of-operation stuff. I think your problem is fixable, but the steps are going to get considerably harder before they get easier. I can provide high-level guidance, but I have had to ask you 4 times to take the easiest step on this whole 1000-step journey. I can't keep this up. If I ask for something that requires more self-direction on your part, like "open the 8360, put a scope probe on V/GHz and another on detector level, it's on the A10 card, details in the service manual" do you think you can manage? If so, great, we can continue, if not, it's best to pursue a return and obtain something that works out of the box. I'm about to get busy at work and if we get bogged down I'm going to have to leave you hanging: I can't spend time I don't have on this.
 

Offline MRIKenTopic starter

  • Newbie
  • Posts: 9
  • Country: us
Re: HP 8510C error message
« Reply #8 on: January 12, 2025, 08:12:16 pm »
I would like to state I have many years of using a network analyzer, but in all those years I have never had a reason to look for that measurement.  That being said, I do believe this is the trace you asked about.  The markers are set at ~13.5 GHz and ~20 GHz. When I first powered the system up the trace was good from 13.5-26.5GHz, the 20-26.5 range flipped a few times then settled into the screen shot here.
 

Offline jjoonathan

  • Frequent Contributor
  • **
  • Posts: 865
  • Country: us
Re: HP 8510C error message
« Reply #9 on: January 12, 2025, 08:52:33 pm »
Ok great, let's keep going.

The a1 trace shows us that we probably don't have the problem where max amplitude just gets a bit low due to YTM/YTF tracking drift, which is the most common problem to show up when shipping 8360s.

Our next goal is to decide between debugging the 8515 test set and the 8360 source. The principle of operation here is that the source operates completely independently and the 8515 sampler tries to lock on to it, rather than these having independent PLLs like in modern instruments. This is to support VNA measurements using unsynthesized (!) sources from the 8350 series. You have an 8360, which is synthesized, but we still have to deal with the architectural choice which will slightly complicate the matter of assigning blame: is the 8515 bad or is it only going bad  because the 8360 is giving it a bad signal that it can't lock to? I am leaning towards thinking something is wrong with the 8360 because the bands going good/bad are distinctively 8360 bands rather than 8515 bands. The 8360 frequency bands are:
Code: [Select]
    0    .01  - 2.0  GHz
    1    2.0  - 7.0  GHz
    2    7.0  - 13.5 GHz
    3    13.5 - 20   GHz
    4    20   - 26.5 GHz

and from what I remember the 8515 bands are considerably shorter and more numerous. Fortunately for us, even though you don't have a spectrum analyzer, the problem is present on band 0 that should be within the range of your oscilloscope. Could you park the VNA on S11 at CENTER=50MHz (or whatever is center of your scope BW), SPAN=0Hz, SWEEP TIME=100s and look at the signal coming out of Port 1? Does it look like a clean sinusoid on FFT?
 

Offline veedub565

  • Frequent Contributor
  • **
  • Posts: 444
  • Country: gb
Re: HP 8510C error message
« Reply #10 on: January 13, 2025, 06:24:19 pm »
I'm sure you have already done this, but worth double checking the preflight checkout. I've been caught out before with a cable not connected or connected in the wrong place.

Also, I found this in the field service manual, may be helpful ?

No IF Found

This error is detected by the level detector and the 85102 A21 IF counter board. This is done using an algorithm in the 85101C math CPU to calculate the IF frequency. The A21 counter and the CPU check if the IF is within range and has enough power. If not, the CPU initiates a search mode that attempts to relocate the test set VTO harmonic to re-establish the 20 MHz IF. This message is displayed if the IF is not found, or is not within range.

Probable cause of failure:
• RF source signal may be absent, low in power, or at an incorrect frequency
• 85102 A21 IF counter board
• Test set samplers. Be sure that the test set A3 summing amplifier assembly has not biased the samplers off. Measure the voltage on the sampler control pin to each sampler to verify TTL levels. 0 volts
   turns the sampler on and approximately 3 volts or greater turns it off. You can press the S-parameter front panel keys to toggle the samplers.
• Defective cable in the IF signal path
• Test set component failure: A weak VTO, bad power amplifier, step recovery diode, or 4-way power splitter in the VTO
• 85101C A5 math CPU
• 85102 stop sweep cable not connected
• Occasionally, the 85102 A20 sweep ADC can cause this error, accompanied by a Source Sweep Sync error message.

Troubleshooting:
• Check the RF source to test set cable.
• Run the 85102 service program in the Run All mode.
• Run 85101C service program test 1 (CPU A5), and the 85102 tests 2 (A20) and A (A21).
• Perform the unratioed power test in “Unratioed Power Failures” to see if the samplers are working.
• Check the IF test set interconnect cable.

If the test set is an 8514A or 8515A, compare the S11 and S22 traces. If one looks bad and the other looks good, the 85102 and 85101C are probably working. Therefore, suspect a problem with the test set or the IF test set interconnect cable. If both S11 and S22 look bad, first suspect a problem with the 85102 or the source.

NOTE This check does not work for 8512A and 8513A test sets. However, you can compare the a1 and b1 channels for unratioed power levels. Refer to “Unratioed Power Failures.”
 

Offline MRIKenTopic starter

  • Newbie
  • Posts: 9
  • Country: us
Re: HP 8510C error message
« Reply #11 on: January 14, 2025, 02:55:35 pm »
I have triple checked that cable between the units.  I am waiting on some adapters I ordered to do more testing.  I didn't have any 7mm adapters or SMA adapters, which I thought I had.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf