Author Topic: Firmware update for 859xE spectrum analyzers?  (Read 1404 times)

Miti, MarkL and 2 Guests are viewing this topic.

Offline hp3310aTopic starter

  • Frequent Contributor
  • **
  • Posts: 307
  • Country: 00
  • 50Hz Territory
Re: Firmware update for 859xE spectrum analyzers?
« Reply #25 on: December 06, 2025, 07:50:28 am »
If you have the EMC personality card it is much easier.

OK, great, what do I have to do?

Which scenario are you referring to?  If you have the EMC card?  Or don't have the EMC card?

I have option 103, I'm not sure if that means I have the EMC card.
 

Online Miti

  • Super Contributor
  • ***
  • Posts: 1666
  • Country: ca
Re: Firmware update for 859xE spectrum analyzers?
« Reply #26 on: December 06, 2025, 11:49:58 am »
I have option 103, I'm not sure if that means I have the EMC card.

There's a slot under the display where you insert a memory card, if you have it, part number 85712, which contains the software, limit lines, etc. It looks something like, or maybe it is, a PCMCIA card. Then from the SA menus you can load the EMC personality in the NVRAM. Manual attached.
Fear does not stop death, it stops life.
 

Offline hp3310aTopic starter

  • Frequent Contributor
  • **
  • Posts: 307
  • Country: 00
  • 50Hz Territory
Re: Firmware update for 859xE spectrum analyzers?
« Reply #27 on: December 06, 2025, 02:20:59 pm »
I have option 103, I'm not sure if that means I have the EMC card.

There's a slot under the display where you insert a memory card, if you have it, part number 85712, which contains the software, limit lines, etc. It looks something like, or maybe it is, a PCMCIA card. Then from the SA menus you can load the EMC personality in the NVRAM. Manual attached.

Ah, thanks for clearing that up for me (I thought it was a PCB/card inside the unit). No, I don't have that memory card. I do have the reader.
 

Offline hp3310aTopic starter

  • Frequent Contributor
  • **
  • Posts: 307
  • Country: 00
  • 50Hz Territory
Re: Firmware update for 859xE spectrum analyzers?
« Reply #28 on: December 06, 2025, 04:55:08 pm »
I think I downloaded them from the same place long time ago when it was working fine. I wonder if it is ok to upload them here.

Edit: Attached.

Thanks a lot, I didn't notice those the attachment (Mark pointed me to it).
 

Offline hp3310aTopic starter

  • Frequent Contributor
  • **
  • Posts: 307
  • Country: 00
  • 50Hz Territory
Re: Firmware update for 859xE spectrum analyzers?
« Reply #29 on: December 07, 2025, 11:18:37 am »
After looking at the ROM files and disassembly for a day I realized: My board doesn't have U6 and U23. How come?

I mean, I'm not complaining. That is great news for me because the haystack just halved. But what do those ROMs contain and why is mine not populated? Is there information available somewhere on what the individual EPROMS contain?

A little background:

I'd like to make a modification to how the screen is updated to make it more compatible with my new LCD display. In particular how the previous sweep is cleared prior to repainting the current sweep. As it is, there's a 10 pixel (about) wide space that is cleared ahead of the trace. However, the new trace is only filled up to 5 pixels (about) before more space to the right is cleared. See image below for screen grabs of how the display blanking progresses during a sweep.

To observe the screen blanking in action on your own 859xE, increase the sweep time to 5 seconds or longer (also on a CRT).

This screen blanking causes no issues on the CRT. Because the phosphor's natural afterglow it just isn't visible with short sweep times. However the LCD display is much faster and the blanking causes random glitches on the trace.

I thought if I could get rid of the big unnecessary black rectangle, then the flickering should be reduced a lot. Ideally there would be only one pixel column cleared ahead of the current column that is being redrawn, but I suspect this was not fast enough and they decided to clear a 10 pixel wide channel in 5 pixel increments.

So much for the motivation to do this. Now I need to make sense of the ROMs. If anyone who already identified some parts in the code... that would be wonderful. I'll take any input I can get.
 

Online Miti

  • Super Contributor
  • ***
  • Posts: 1666
  • Country: ca
Re: Firmware update for 859xE spectrum analyzers?
« Reply #30 on: December 07, 2025, 11:51:34 am »
After looking at the ROM files and disassembly for a day I realized: My board doesn't have U6 and U23. How come?

In one of Mark's reply there is a link to a document that shows how many ROMs each FW revision needs. Attached.
You have Rev. H 94.08.22. If you want to up-rev it to Rev. L 98.06 15 you need four AM27C020.

I'd like to make a modification to how the screen is updated to make it more compatible with my new LCD display. In particular how the previous sweep is cleared prior to repainting the current sweep. As it is, there's a 10 pixel (about) wide space that is cleared ahead of the trace. However, the new trace is only filled up to 5 pixels (about) before more space to the right is cleared. See image below for screen grabs of how the display blanking progresses during a sweep.

To observe the screen blanking in action on your own 859xE, increase the sweep time to 5 seconds or longer (also on a CRT).

This screen blanking causes no issues on the CRT. Because the phosphor's natural afterglow it just isn't visible with short sweep times. However the LCD display is much faster and the blanking causes random glitches on the trace.

I thought if I could get rid of the big unnecessary black rectangle, then the flickering should be reduced a lot. Ideally there would be only one pixel column cleared ahead of the current column that is being redrawn, but I suspect this was not fast enough and they decided to clear a 10 pixel wide channel in 5 pixel increments.

So much for the motivation to do this. Now I need to make sense of the ROMs. If anyone who already identified some parts in the code... that would be wonderful. I'll take any input I can get.

The display only paints, a number of times per second, what's in the video memory so the clearing and the flickering happens in the memory, not in the CRT or LCD. What am I missing here? I changed my CRT with LCD, my design, and it works perfect. See this:

https://www.eevblog.com/forum/projects/hp-8594e-replacing-the-green-crt-with-lcd/
Fear does not stop death, it stops life.
 

Offline hp3310aTopic starter

  • Frequent Contributor
  • **
  • Posts: 307
  • Country: 00
  • 50Hz Territory
Re: Firmware update for 859xE spectrum analyzers?
« Reply #31 on: December 07, 2025, 01:15:30 pm »
After looking at the ROM files and disassembly for a day I realized: My board doesn't have U6 and U23. How come?

In one of Mark's reply there is a link to a document that shows how many ROMs each FW revision needs. Attached.
You have Rev. H 94.08.22. If you want to up-rev it to Rev. L 98.06 15 you need four AM27C020.

Got it, and thanks for pointing me to something that I overlooked. Again.

I'd like to make a modification to how the screen is updated to make it more compatible with my new LCD display. In particular how the previous sweep is cleared prior to repainting the current sweep. As it is, there's a 10 pixel (about) wide space that is cleared ahead of the trace. However, the new trace is only filled up to 5 pixels (about) before more space to the right is cleared. See image below for screen grabs of how the display blanking progresses during a sweep.

To observe the screen blanking in action on your own 859xE, increase the sweep time to 5 seconds or longer (also on a CRT).

This screen blanking causes no issues on the CRT. Because the phosphor's natural afterglow it just isn't visible with short sweep times. However the LCD display is much faster and the blanking causes random glitches on the trace.

I thought if I could get rid of the big unnecessary black rectangle, then the flickering should be reduced a lot. Ideally there would be only one pixel column cleared ahead of the current column that is being redrawn, but I suspect this was not fast enough and they decided to clear a 10 pixel wide channel in 5 pixel increments.

So much for the motivation to do this. Now I need to make sense of the ROMs. If anyone who already identified some parts in the code... that would be wonderful. I'll take any input I can get.

The display only paints, a number of times per second, what's in the video memory so the clearing and the flickering happens in the memory, not in the CRT or LCD. What am I missing here? I changed my CRT with LCD, my design, and it works perfect. See this:

https://www.eevblog.com/forum/projects/hp-8594e-replacing-the-green-crt-with-lcd/

Maybe your design works fundamentally different than that from Newscope or maybe the firmware you're running doesn't expose this quite as strongly. I'm attaching an animated GIF with how it looks, the glitches are clearly visible.
 

Offline bostonman

  • Super Contributor
  • ***
  • Posts: 2690
  • Country: us
Re: Firmware update for 859xE spectrum analyzers?
« Reply #32 on: December 07, 2025, 02:45:39 pm »
Quote
Edit: Please be aware that you may lose everything in the user memory. I’ve lost the EMC personality, but I uploaded it again.

I downloaded the ZIP but wanted to ask a few questions if/when I do this.

What is the EMC and what happens if it's lost? Which part number ROM should be purchased to burn (I need to check whether my burner handles these).
 

Offline hp3310aTopic starter

  • Frequent Contributor
  • **
  • Posts: 307
  • Country: 00
  • 50Hz Territory
Re: Firmware update for 859xE spectrum analyzers?
« Reply #33 on: December 07, 2025, 03:06:50 pm »
Quote
Edit: Please be aware that you may lose everything in the user memory. I’ve lost the EMC personality, but I uploaded it again.

I downloaded the ZIP but wanted to ask a few questions if/when I do this.

What is the EMC and what happens if it's lost? Which part number ROM should be purchased to burn (I need to check whether my burner handles these).

As for the EPROMs MarkL said:

You will need 4x AM27C020-150 or equivalent if you want to do this, plus access to a programmer and (probably) UV eraser.

I ordered those from a seller on Ebay.
 

Offline bostonman

  • Super Contributor
  • ***
  • Posts: 2690
  • Country: us
Re: Firmware update for 859xE spectrum analyzers?
« Reply #34 on: December 07, 2025, 03:13:30 pm »
Ah yes, I didn't re-read back far enough.
 

Online Miti

  • Super Contributor
  • ***
  • Posts: 1666
  • Country: ca
Re: Firmware update for 859xE spectrum analyzers?
« Reply #35 on: December 07, 2025, 08:54:39 pm »
Maybe your design works fundamentally different than that from Newscope or maybe the firmware you're running doesn't expose this quite as strongly. I'm attaching an animated GIF with how it looks, the glitches are clearly visible.

Is the .gif supposed to be animated because it is not for me. I don't see any glitch.
Fear does not stop death, it stops life.
 

Online MarkL

  • Supporter
  • ****
  • Posts: 2419
  • Country: us
Re: Firmware update for 859xE spectrum analyzers?
« Reply #36 on: December 07, 2025, 10:11:39 pm »
To be clear about this EMC thing, the EMC option is two things: A "personality card" that plugs into the front of the analyzer (85712D) AND a card that plugs into the card cage inside the analyzer.  The personality card is software only, and the internal card provides the quasi-peak detector (QPD) measurement function (aka Option 103) that is specified by regulatory agencies for EMC measurements.

The personality card contains the driver for the QPD card, and various radiated and conducted EMC limit profiles for DUTs as defined by regulatory agencies.  It will plot the received emissions vs. the limit lines and indicate pass/fail for each test.

There's a lot of set up to fully utilize this feature (RF-quiet rooms, LISN's, special antennas, etc.).  The user manual that Miti posted provides the details on all of this.

There is also a personality card (11946-10004) that contains only the driver for the QPD card, and omits the various automated EMC limit testing provided by the EMC card.

If you have the QPD card and want to use it, you will need either of the above two personality cards.  If you already have one of the above two cards loaded, you will lose this functionality if you wipe out the NVRAM and don't have the personality card to reload it.  It is possible to reload the QPD driver and most of the EMC personality card without the original personality card, but it is a painful process as Miti can probably attest.

If you don't intend to do any EMC measurements, I actually would not worry about it.

There is one more thing:  If the 859xE SA came from the factory with the EMC option, the front glass over the CRT is special.  It has a metallic coating on the inside to help reduce emissions from the analyzer itself.  If you ever move the QPD card to a different analyzer, move the bezel and special glass too.  The bezel has to move too because the special glass is thicker.

Hope this clears it up.  If still not clear, please look through the EMC user manual before diving in any further.
 
The following users thanked this post: Miti

Online MarkL

  • Supporter
  • ****
  • Posts: 2419
  • Country: us
Re: Firmware update for 859xE spectrum analyzers?
« Reply #37 on: Yesterday at 04:23:08 am »
I should mention that the QPD card (Option 103) also has built-in AM/FM demodulation.  The option includes a speaker installed inside the analyzer cabinet which allows you to listen to the demodulated signal.  The audio is also available on an earphone jack on the back.

If you are interested in using the demod feature, you do need the QPD driver on one of the previously mentioned personality cards to enable it.


EDIT: Turns out this is not correct.  Read ahead a couple of posts.
« Last Edit: Yesterday at 09:42:48 pm by MarkL »
 

Offline hp3310aTopic starter

  • Frequent Contributor
  • **
  • Posts: 307
  • Country: 00
  • 50Hz Territory
Re: Firmware update for 859xE spectrum analyzers?
« Reply #38 on: Yesterday at 08:11:07 am »
To be clear about this EMC thing, the EMC option is two things: A "personality card" that plugs into the front of the analyzer (85712D) AND a card that plugs into the card cage inside the analyzer.  The personality card is software only, and the internal card provides the quasi-peak detector (QPD) measurement function (aka Option 103) that is specified by regulatory agencies for EMC measurements.

The personality card contains the driver for the QPD card, and various radiated and conducted EMC limit profiles for DUTs as defined by regulatory agencies.  It will plot the received emissions vs. the limit lines and indicate pass/fail for each test.

There's a lot of set up to fully utilize this feature (RF-quiet rooms, LISN's, special antennas, etc.).  The user manual that Miti posted provides the details on all of this.

There is also a personality card (11946-10004) that contains only the driver for the QPD card, and omits the various automated EMC limit testing provided by the EMC card.

If you have the QPD card and want to use it, you will need either of the above two personality cards.  If you already have one of the above two cards loaded, you will lose this functionality if you wipe out the NVRAM and don't have the personality card to reload it.  It is possible to reload the QPD driver and most of the EMC personality card without the original personality card, but it is a painful process as Miti can probably attest.

If you don't intend to do any EMC measurements, I actually would not worry about it.

There is one more thing:  If the 859xE SA came from the factory with the EMC option, the front glass over the CRT is special.  It has a metallic coating on the inside to help reduce emissions from the analyzer itself.  If you ever move the QPD card to a different analyzer, move the bezel and special glass too.  The bezel has to move too because the special glass is thicker.

Hope this clears it up.  If still not clear, please look through the EMC user manual before diving in any further.

Thanks Mark, this clears up everything.

I did swap out the CRT for an LCD a couple of days ago and noticed the thick glass you mentioned. It is 3mm thick and apparently has a metal coating on the inside and a conductive strip around it (picture attached). The dark gray front screen bezel also is slightly different. The fingers that hold the glass in place are slightly taller (about 1mm).

As for the NVRAM and EMC software, I did get confused and I thought the firmware upgrade will wipe that out. Apparently my mind made that up, too. When I go and exchange the NVRAM for an FRAM, I will make sure the data it contains is transferred, so the unit shouldn't notice the change.

I should mention that the QPD card (Option 103) also has built-in AM/FM demodulation.  The option includes a speaker installed inside the analyzer cabinet which allows you to listen to the demodulated signal.  The audio is also available on an earphone jack on the back.

If you are interested in using the demod feature, you do need the QPD driver on one of the previously mentioned personality cards to enable it.

When I played around with a square wave signal generator the other day, there were some specific settings of signal frequency and RESBW and span where the unit would start emitting strange noises, sounded like  a capacitor whine. It turned out to come from the speaker.

I think all this means that the original NVRAM is still intact 30 years after it was built. I better hurry with a replacement so I don't loose all the software that is on it.
 

Online MarkL

  • Supporter
  • ****
  • Posts: 2419
  • Country: us
Re: Firmware update for 859xE spectrum analyzers?
« Reply #39 on: Yesterday at 09:41:30 pm »
I said:
...
If you are interested in using the demod feature, you do need the QPD driver on one of the previously mentioned personality cards to enable it.
This is what the QPD user manual implies, but out of curiosity I tried it without the QPD driver loaded (ERASE MEM ALL).  Surprisingly, the AM/FM demod feature continued to work with no problem.  And a further look finds that the demod commands are in the EPROM firmware.

So, if all you're looking to get out of the QPD (Option 103) card is the demod feature, you *don't*  actually need the QPD driver.  However, if you want to do quasi-peak measurements, you definitely need the driver from the EMC or QPD personality cards.

One other way to preserve a personality card loaded on an SA is to copy it to an SRAM card.  There are four sizes of SRAM cards listed in the catalog: 32k (85700A), 128k (85702A), 256k (85704A), and 512k (85705A).  I've only ever seen the 32k and 128k on the used market.  They plug into the front slot where the personality cards also go. 

The 32k card can be used to copy the QPD personality loaded in an SA, but the EMC personality needs the 128k card to fit what gets loaded (everything except the limit lines).  I've verified both of these scenarios work.  However, take note that the SRAM cards are powered by a coin cell, so are subject to losing data due to a depleted battery.
 

Online Miti

  • Super Contributor
  • ***
  • Posts: 1666
  • Country: ca
Re: Firmware update for 859xE spectrum analyzers?
« Reply #40 on: Today at 01:27:19 pm »
The 32k card can be used to copy the QPD personality loaded in an SA, but the EMC personality needs the 128k card to fit what gets loaded (everything except the limit lines).  I've verified both of these scenarios work.  However, take note that the SRAM cards are powered by a coin cell, so are subject to losing data due to a depleted battery.

Should I design a card with FRAM?  :-DD
« Last Edit: Today at 02:41:09 pm by Miti »
Fear does not stop death, it stops life.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf