¿ªÔÆÌåÓý

  1. HP-Agilent-Keysight-Equipment
  2. Messages
Expanded Topics Messages
Date
Re: Off Topic - Looking for assistance with power sensors

 

"Razvan Popescu via groups.io" <yo8ryr@...> writes:

Hi,

I found partially my notes regarding E4413A that I had from a friend.
What I did was the following:

1) Before doing any EEPROM sensor operations run the following command
to stop the auto trigger from the power meter:

INIT:CONT OFF

2) Read the calibration table:
"DIAG:DET:EEPR? MemoryAddress,NumberOfBytesToRead"
Do not read more than 100 bytes, it will crash the power meter. In my
case an E4418B. Table size depends on how many frequency points were
calibrated. You can add additional points if you want for specific
frequencies like ham radio bands. There is enough space in the EEPROM.
Only the calibration table for the 26.5 GHz model was exactly 256
bytes.
H33 model will have it bigger.

Important thing to note is that the calibration table starts at 569
for E441XA sensors.
This is likely just coincidence because they are all calibrated by
keysight - it might be at any location in the EEPROM.

You can only write at specific locations based on the following
formula:
Start byte address in the EEPROM for each conversion entry is x*8+9
First 4 bytes are the frequency, next 2 bytes are the low CF and last
2 bytes are the high CF.

I will give an example below what will be the bytes values for a 50
MHz frequency and CF LOW 100% and CF HIGH 100%.

(0 * (2 ^ 24) + 0 * (2 ^ 16) + 195 * (2 ^ 8) + 80) / 1000 = 50 (MHz)
So if we consider that starting address is 569 for the E441XA sensors
you will need to write the following bytes at the address starting
569.

569 -> 0
570 -> 0
571 -> 195
572 -> 80
---- This will cover the 4 bytes for the frequency at 50 MHz ----

Now the next 2 bytes will be for the CF LOW:

100 * (64 * 256 + 0)/2^14 = 100 (CF LOW 100%)

573 -> 64
574 -> 0

Since the value for the CF HIGH is also 100 the same bytes will be
written to the addresses 575 and 576.
Without updating the checksum of the table the power meter will not
accept the new data... I attached a program i used to rewrite the
calibration data of my sensor. Keysight doesn't do any business with
private persons, so there was no other way i guess.
Re: Off Topic - Looking for assistance with power sensors

 

Hi,

I found partially my notes regarding E4413A that I had from a friend.
What I did was the following:

1) Before doing any EEPROM sensor operations run the following command to stop the auto trigger from the power meter:

INIT:CONT OFF

2) Read the calibration table:
"DIAG:DET:EEPR? MemoryAddress,NumberOfBytesToRead"
Do not read more than 100 bytes, it will crash the power meter. In my case an E4418B. Table size depends on how many frequency points were calibrated. You can add additional points if you want for specific frequencies like ham radio bands. There is enough space in the EEPROM.
Only the calibration table for the 26.5 GHz model was exactly 256 bytes.
H33 model will have it bigger.

Important thing to note is that the calibration table starts at 569 for E441XA sensors.

You can only write at specific locations based on the following formula:
Start byte address in the EEPROM for each conversion entry is x*8+9
First 4 bytes are the frequency, next 2 bytes are the low CF and last 2 bytes are the high CF.

I will give an example below what will be the bytes values for a 50 MHz frequency and CF LOW 100% and CF HIGH 100%.

(0 * (2 ^ 24) + 0 * (2 ^ 16) + 195 * (2 ^ 8) + 80) / 1000 = 50 (MHz)
So if we consider that starting address is 569 for the E441XA sensors you will need to write the following bytes at the address starting 569.

569 -> 0
570 -> 0
571 -> 195
572 -> 80
---- This will cover the 4 bytes for the frequency at 50 MHz ----

Now the next 2 bytes will be for the CF LOW:

100 * (64 * 256 + 0)/2^14 = 100 (CF LOW 100%)

573 -> 64
574 -> 0

Since the value for the CF HIGH is also 100 the same bytes will be written to the addresses 575 and 576.


3) Write the calibration table:
DIAG:DET:EEPR MemoryAddress,ValueOfByte

Based on the values above the commands should be:
INIT:CONT OFF
DIAG:DET:EEPR 569,0
DIAG:DET:EEPR 570,0
DIAG:DET:EEPR 571,195
DIAG:DET:EEPR 572,80
DIAG:DET:EEPR 573,64
DIAG:DET:EEPR 574,0
DIAG:DET:EEPR 575,64
DIAG:DET:EEPR 576,0

This will update the sensor EEPROM for the 50MHz CF value only.
You will need to repeat the process for all frequency points that you need.

For the H33 you will have even more points but there is enough space.
I didn't figure it out where is the EEPROM entry for the Standard vs OPT H33 to convert a normal sensor to the H33 model.

When you finish you can run INIT:CONT ON or just Preset or reboot the power meter. It will activate the trigger again by default.

New sensors like E93xxA also have a temperature compensation table but I never had one to investigate it and even the broken ones on eBay are too expensive. I was told that hardware for 6 GHz and 18 GHz is the same but it is missing the calibration table.

I hope it helps. If someone can build a python script to automate the process it would be great, my programming skills are not very good and now I don't really have any free time... This would be a project for the winter.

Regards,
Razvan

On 22/05/2025 16:38, George/Gyorgy Albert via groups.io wrote:
Sorry, correction: for writing this is the right form: "DIAG:DET:EEPR eepromaddress, eepromvalue" (the ? must be removed).
BR,
G
On 5/22/2025 5:36 PM, George/Gyorgy Albert via groups.io wrote:
EEPROM writing, if my notes are correct it is "DIAG:DET:EEPR? eepromaddress, eepromvalue".
Re: Off Topic - Looking for assistance with power sensors

 

"VeeCal" <Sales@...> writes:

Fantastic!

Okay, so this makes me feel a bit better. I was thinking I was the
only one out here that wanted to try this. Something
that has been made abundantly clear to me is to capture the data from
the eeprom before attempting to write to it, just
in case the file becomes corrupted I have a backup copy to restore
from. With this in mind I will work towards a script
to capture this in all instances.
Be careful. The EEPROM clock line is shared with the chopper pins. I'm
assuming they are using some phantom powering of the power sensors
electronics. Likely there were no pins left and they wanted to stay
compatible with older power sensor.

Downside is that you have to wait after writing each byte to the EEPROM.
I used the following script after bricking my sensor a few times:

------------------------------------------------------------
#!/usr/bin/python3

import pyvisa
from functools import partial
import time
import sys
rm = pyvisa.ResourceManager()
inst = rm.open_resource('TCPIP0::192.168.0.50::gpib0,13::INSTR')

#inst.write("DIAG:TEST 16")
i=0
with open(sys.argv[1], "rb") as file:
for char in file.read():
notmatch=1
while notmatch:
print(f'{i}', end='\r')
value = inst.query(f'DIAG1:DET:EEPR? {i},1')
time.sleep(0.1)
if int(value) != char:
print(f'writing offset {i}, data {char}, old {int(value)}')
inst.write(f'DIAG1:DET:EEPR {i},{char}')
time.sleep(0.1)
else:
i = i + 1
notmatch=0
file.close()
--------------------------------------------------------

This reads each byte, and if it differs reprograms it until it
matches. Never ever abort the program, this will brick your sensor. And
it is all your fault if the script doesn't work for you. ;-)
Re: Off Topic - Looking for assistance with power sensors

 

"F1EKU via groups.io" <rfconsulting.fr@...> writes:

According to HPAK's technical documentation, it's impossible to write
new calibration factors into the sensor because
each time it requires calculating a checksum that the EPMs of the
E4416/17/18/19 series cannot compute. They explain that
for this, we must use the newer EPMs from the N1911/12/13/14
series. Personally, I've been able to read a sensor's entire
block and write it back into another probe, but changing one or more
values within the block always resulted in the
inability to initialize the modified probe on an E44XX series power
meter. Of course, HPAK uses an external program
capable of calculating the checksum when they calibrate a sensor upon
request and write the complete block, even with an
E44XX power meter. Does anyone know the solution for calculating the
checksum?
Yes, i reverse engineered the EEPROM format of the E9XXX power sensors
because i wanted to add option H18 to my E9304A. The EEPROM contains
different tables, and each table header contains a simple 'add all
bytes' style checksum. But one has to be very careful when writing the
EEPROM - if the EEPROM contents are not valid, the power meter shows an
error message (or even locks up completely). Communication with the
power sensor is no longer possible and you need some tricks to
re-establish communication.
Programming codes for option 002 and option 003 for 8663A

 

We are in the middle of upgrading our software to support the RS FSMR measuring receiver and came across a section of the test for pulse and phase mod that we would like to automate. The current test is done manually (which is fine) but we thought it would be fun to snazzy it up by fully automating it. Kicker is, we can find nothing anywhere on the internet for commands to control either 002 or 003. The manuals on the net are SN specific and only seem to reference the base unit. I scoured the Wiki repository here and came up zilch on that as well.

Is it possible there are no commands to control this or would someone be able to point me in the direction of where we should be looking?

With respect, Mike
Re: Off Topic - Looking for assistance with power sensors

 

Fantastic!

Okay, so this makes me feel a bit better. I was thinking I was the only one out here that wanted to try this. Something that has been made abundantly clear to me is to capture the data from the eeprom before attempting to write to it, just in case the file becomes corrupted I have a backup copy to restore from. With this in mind I will work towards a script to capture this in all instances.
Re: Off Topic - Looking for assistance with power sensors

 

Thank you for the response. Yes, we have a lab that is asking for this but again it is not something I have played with.

If you would not mind, I would like to speak to you to get your insights on how to proceed or even if we should proceed. Please send me an email to sales@... I will show you what we have so far and what we would like to try. At the moment we are adding in a RS FSMR measuring receiver as a driver to the 8663A calibration program. THAT is a NICE receiver!?
Re: Off Topic - Looking for assistance with power sensors

 

Sorry, correction: for writing this is the right form: "DIAG:DET:EEPR eepromaddress, eepromvalue" (the ? must be removed).

BR,

G

On 5/22/2025 5:36 PM, George/Gyorgy Albert via groups.io wrote:
EEPROM writing, if my notes are correct it is "DIAG:DET:EEPR? eepromaddress, eepromvalue".
--
Gy?rgy Albert
Mob +40-722-304534
Re: Off Topic - Looking for assistance with power sensors

 

Hi,

A few years ago I also tried to find out the calibration tables of the E4412/4413 power sensors, using my E4418B power meter. The "DIAG:DET:EEPR? eepromaddress, 1" command was definitely working (and returned the EEPROM location content), where I looped eepromaddress from 0 to 2047 (the address seems truncated to 11 bits, so using addresses higher than 2047 interprets the modulo 2048 part). I compared the read value with the dumped EEPROM content, and it matches. The second parameter when the content is read, '1',? is the number of bytes. I used reading only one single byte in a loop, but multiple bytes also can be read.

I tested also the EEPROM writing, if my notes are correct it is "DIAG:DET:EEPR? eepromaddress, eepromvalue". There were (several) checksum bytes for sure, if one single bit was altered in the content, at startup the sensor has been declared defective (corrupt table). But I didn't succeed to decrypt the table locations and checksum calculation.

In one of the EPM power meter manuals (probably this document is the source of inspiration : 9018-01324.pdf) I found some service commands, which I tested. The results are listed in the attached text file.

BR,

George/Gyorgy

On 5/22/2025 12:04 PM, Razvan Popescu via groups.io wrote:
Hi,

I will check my notes. I remember there were some GPIB DIAG and INIT commands that are not documented.

If you look at the PS-Cal and SureCal calibration software you will see they can read/write the new values via the power meter GPIB commands. Maybe for the newest sensors it is a little bit more difficult but for the HP branded E4413A that I had it was OK.

Regards,
Razvan

On May 22, 2025 8:43:32 AM UTC, "F1EKU via groups.io" wrote:
According to HPAK's technical documentation, it's impossible to write new calibration factors into the sensor because each time it requires calculating a checksum that the EPMs of the E4416/17/18/19 series cannot compute. They explain that for this, we must use the newer EPMs from the N1911/12/13/14 series. Personally, I've been able to read a sensor's entire block and write it back into another probe, but changing one or more values within the block always resulted in the inability to initialize the modified probe on an E44XX series power meter. Of course, HPAK uses an external program capable of calculating the checksum when they calibrate a sensor upon request and write the complete block, even with an E44XX power meter. Does anyone know the solution for calculating the checksum?

--
Gy?rgy Albert
Mob +40-722-304534

E4418_service_GPIB_commands.txt
E4418_service_GPIB_commands.txt

Joining group

 

Dear group manager,

Which procedure must be followed to join this group?

73
Raymond ON4DBV
Re: HP 8566B repair

 

Recall reg 7 is the last stored state. You can save instrument states in registers 0-7, 8 is the correction numbers.
Don Bitters
Re: Off Topic - Looking for assistance with power sensors

 

Hi,

I will check my notes. I remember there were some GPIB DIAG and INIT commands that are not documented.

If you look at the PS-Cal and SureCal calibration software you will see they can read/write the new values via the power meter GPIB commands. Maybe for the newest sensors it is a little bit more difficult but for the HP branded E4413A that I had it was OK.

Regards,
Razvan

On May 22, 2025 8:43:32 AM UTC, "F1EKU via groups.io" wrote:
According to HPAK's technical documentation, it's impossible to write new calibration factors into the sensor because each time it requires calculating a checksum that the EPMs of the E4416/17/18/19 series cannot compute. They explain that for this, we must use the newer EPMs from the N1911/12/13/14 series. Personally, I've been able to read a sensor's entire block and write it back into another probe, but changing one or more values within the block always resulted in the inability to initialize the modified probe on an E44XX series power meter. Of course, HPAK uses an external program capable of calculating the checksum when they calibrate a sensor upon request and write the complete block, even with an E44XX power meter. Does anyone know the solution for calculating the checksum?
Re: Off Topic - Looking for assistance with power sensors

 

Of course, I would be very interested in knowing the solution for writing to R&S NRV-ZXX series sensors, as the documentation I've read doesn't mention it. It seems that specific software is required for this.
Re: Off Topic - Looking for assistance with power sensors

 

According to HPAK's technical documentation, it's impossible to write new calibration factors into the sensor because each time it requires calculating a checksum that the EPMs of the E4416/17/18/19 series cannot compute. They explain that for this, we must use the newer EPMs from the N1911/12/13/14 series. Personally, I've been able to read a sensor's entire block and write it back into another probe, but changing one or more values within the block always resulted in the inability to initialize the modified probe on an E44XX series power meter. Of course, HPAK uses an external program capable of calculating the checksum when they calibrate a sensor upon request and write the complete block, even with an E44XX power meter. Does anyone know the solution for calculating the checksum?
Re: Off Topic - Looking for assistance with power sensors

 

Hello Mike,

You need to use some power meter GPIB commands to write/update the sensor EEPROM. It is not only the calibration table coefficients to update but also the temperature compensation table.

I only played with Agilent EPM and Gigatronics power meters to do this but from what I saw in other docs on the internet you have the exact same principle for Boonton and R&S power meters.

I recommend you download the initial/old table and check how it is done and after send new table. I used a text file to read the new cal table and write it to the sensor.

I did this in a semi automated way because I only had 2 sensors to update and it was interesting to try it. For Gigatronics it was not even needed a script because you can update the values on the power meter itself.

I guess you need a fully automated procedure to integrate it in your VeeCal software?

Regards,
Razvan

On 21/05/2025 19:23, VeeCal via groups.io wrote:
Hello again,
I am hoping to find someone here who has done some work with the E44XX series of power sensors - specifically validating and recording the new cal factors to the EEPROM. Likewise for Gigatronics, Anritsu, and Rohde Schwarz. I have done the procedures in the past for the 848X series but this new one is outside of my comfort zone.
With respect, Mike
Re: BNC that does not have the bayonet lugs

 

This version of BNC is commonly used in avionics: Radios, Transponders and the like.?
The RF unit slides into a housing (tray); The (modified BNC Female) coax connector on the RF unit makes contact into the (modified) BNC male connector on the tray??
Installation is probably a lot simpler than it would be if conventional BNCs were used.
All other connections (dc / control/ audio ect) are catered for by multi-pin connectors. (the plug on the RF unit and the socket on the tray.
Similar the that mentioned by Wilko Bulte.

Cliff



On Wed, May 21, 2025 at 6:31?PM Robert G8RPI via <robert8rpi=[email protected]> wrote:
It would help if you told us what it is on.
It does sound like the BWD powerscope connector. They will take a BNC but it won't lock on.

Robert
Re: HP8757A Display Problem (Scalar Network Analyzer)

 

Congratulations!
?
Really glad to hear you located the issue, and it was in the general area we were
discussing.
?
Off Topic - Looking for assistance with power sensors

 

Hello again,
I am hoping to find someone here who has done some work with the E44XX series of power sensors - specifically validating and recording the new cal factors to the EEPROM. Likewise for Gigatronics, Anritsu, and Rohde Schwarz. I have done the procedures in the past for the 848X series but this new one is outside of my comfort zone.
With respect, Mike
Re: BNC that does not have the bayonet lugs

 

It would help if you told us what it is on.
It does sound like the BWD powerscope connector. They will take a BNC but it won't lock on.

Robert
Re: HP 334A distortion analyser problem

 

On Tue, May 20, 2025 at 10:25 AM, Goran Parezanovic wrote:
I applied "supper cold" spray ( -20 deg)to each transistor in the measuring circuit ( Q5 to Q9) but the voltmeter works fine, stays at zero.?
?
I think it's more likely that the problem is one or more electrolytic capacitors which often tend to be "leaky" when they are first charged up. This will go away after a bit of time.? If the required warmup time is short enough for you to tolerate it, just leave it as is. If it's long enough to be annoying, you can try walking thru the circuit, grounding things that won't mind being grounded, until you get to the point where the problem no longer occurs.
?
If you ground a point and find that the meter now stays at zero even when cold, you know that the problem is not after the stage you grounded, but it might be IN the stage you just grounded.
?
If the circuit path is long, the best way to do this is a binary search, where you start in the middle, dividing it in half, then proceed into quarters, etc. Eventually you'll narrow it down to one section of the circuit, where you'll have to decide which element is the problem.
?
--
Jim Adney
Madison, WI USA
21 - 40 of 151493