Products > Test Equipment
Agilent PNA E835xA with 85092 ECal, Problems
<< < (4/5) > >>
joeqsmith:
It appears that if you run through a calibration using the mechanical standards, then use one of the User calibrations, you can change the frequency range.   If you set the range outside of the ECal and then attempt to run though the User cal, it will not let you.  So order is important.   It does appear to work correctly so no custom software or hacking is needed.    :phew:

Next step is to borrow a set of mechanical standards again.   
joeqsmith:
Another thing to be aware of, the documentation states how there is NO limit to the number of data points allowed in a User Characterization stored to disk memory. When stored in the ECal module, the number of data points is limited by the PNA firmware.   This is not true.  The Rev B ECal is limited to 7519 points.   

If you run a mechanical set of standards using say 16001 points, then attempt to create a user cal with that set, it will only use 201 data points.   Using 6401 data points for the mechanical standards, then transferring to the Rev B ECal gives the expected number of data points. 

***
It turns out that the 7519 data points is the total allowed, divided across all of the user characterizations.  So you could have one user characterizations using 6401 and another at 801, with 300 points left.   
joeqsmith:
I've found another Rev C that supports 30kHz to 9GHz, so not all Rev C's were the same.  This one is about 2000 serial numbers lower than the newest one I have.  It will need a new connector before it could be used but the basic internals appear functional so may be worth saving it.  I've been looking for a spanner wrench (PN# 08513-20014) that went with these but no luck.  This may have to be a custom made part.   

The 85033E was used as a reference to setup the other two.  While doing transfer, discovered the Rev C also does not have unlimited storage.  I had attempted to setup 4 user configurations, each with 16001 points, and ran out of memory.  Something else to keep in mind if you are planning to save your setups with the ECal.   

***
While searching for replacement connectors, I came across some cloned parts out of China.  The photos are poor and there are no details about them.   
https://www.ebay.com/itm/355906797299

The OEM connectors are works of art.   Here are a few damaged ones:
https://www.microwaves101.com/encyclopedias/how-to-not-trash-a-calibration-kit
joeqsmith:
I pulled the RevB & C hardware apart to see what it would take to swap some of the parts.  I was amazed how different these two designs were.  Of course, the PCBs are different as expected.  But the RF calibration hardware is also completely different as well.  The pin out, the number of screws, screw locations, all different.  Rev B even has some blue wires (factory patches) and IMO looks more like a prototype than something in production.   

I sketched up a design for a spanner to pull the connectors but upon further inspection, I realized the mating connector has no place for a wrench.  The connectors are also loctited into place.  I would be very hesitant to apply any torque to the case.   

In the following video, they pull down a damaged Rev C E-Cal.  Sadly, they do not attempt to pull the connectors so not much help in this regard.   

My advice if you wanted to buy one, first realize there are no parts available, Keysight will no longer cal or service them and the risk is high that someone has damaged the precision connectors.  I suspect some of the newer E-Cals will work with the old PNAs but I have not looked into it.   

 
joeqsmith:
Upon further inspection, it appears that what I thought was Loctite was actually where the bad connector had been marked with paint.  I made up a tool and it came out very easily.  Once removed, it had been marked with bad on the inside as well.  There was no damage to the inside connectors.   So it was a matter of removing the connector from the Rev B and installing it into the Rev C.  The connectors on the Rev B were both in very good condition and I have no plans to use that ECal.   

These connectors are are crazy complex.  I wonder what their cost was new.  Because Agilent never meant these to be serviced by the end user, I could not find any information on torque.  It would need to be higher than a standard 3.5mm as we don't want them to move when connecting to them.   They use a large thread with a seat so they are designed for much higher torque.  I used 20"/lbs and called it good.  Better to error on the light side.   

The last step is to run though the user calibration. 
Navigation
Message Index
Next page
Previous page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod