1. Home
  2. Electronics Design
  3. HP 167x Logic Analyzer Issue.. Running out of options

HP 167x Logic Analyzer Issue.. Running out of options

Guys.. Hopefully, someone can help me or knows a contact that I might get a hold of that could help.

I own an early (1996) HP1670A model logic analyzer that has a problem where the NVRAM which held the LAN Ether (MAC) address became corrupted. When the analyzer's OS detects this corruption, it decides to 'zero out' the MAC address, rendering the LAN capability useless. The owner is expected to send the LA (or CPU card) to HP for them to reprogram the MAC address. There is no hook from user interface (neither RS232/GPIB) that will allow you to make the change on your own. There were 4 revisions of this analyzer, A,D,E,G. The A & D models used a CRT, while the E,G used a flat panel color display. HP also decided in the E & G models to allow the operator to repair the MAC should it become corrupted.

There must be some sw utility that would allow HP to setup the MAC address, but I've contacted them directly and they 'claim' it doesn't exist anymore. I've also contacted the members of the HP/Agilent Yahoo Group to see if they have any answers. No luck there. The plan is to remove the NVRAM and attempt to reverse engineer the setup and reprogram it accordingly. Of course the part is soldered in but that doesn't scare me.

Anyway, I was just curious if anyone in the community may have some first hand knowledge as to how other's may have resolved this problem. It would be nice to find someone that just so happens to have this special utility but I'm sure that is not likely to happen. So, if someone has some additional information to offer, I would appreciate it.

thanks Jim

Reply to
Jim Flanagan
Loading thread data ...

If you don't care whether the MAC is unique, maybe someone has read and posted the contents of another unit's NVRAM?

(I just bought an old Advin Pilot U40 so that I could do that with all my old gold HPs and Teks.)

Cheers

Phil

--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
hobbs at electrooptical dot net
http://electrooptical.net
Reply to
Phil Hobbs

It had to be set up with a MAC address at the factory, though... is there a header connector or a few test points nearby that could engage a bed-of-nails connection? It only takes two or three wires to program a NVRAM of the serial-flash type, and it might have been easier to bypass the serial inputs for the MAC setting.

This reminds me of the old Apple Laserwriter problem: those machines had security controls, would take a password to enable if you set it up that way, and units in public use (at copy centers and libraries) occasionally got a patron-applied password. Apple's only solution was to buy a new logic board for $1200 or so.

Reply to
whit3rd

Yes, that crossed my mind. I've searched for that as well with no luck. Another stumbling block is that the NVRAM part is actually a Dallas RTC that happens to have 4K of NVRAM. That's bad enough except the reading/writing of the 4K memory is NOT standard. I have access to three device programmers (two being high $ DataIO) and none of them support this part. Therefore I've built a quicky reader/programmer on vector board to support this effort. Guess I'll remove the part from the CPU and have a go at it. take care - Jim

Reply to
Jim Flanagan

Maybe worth a post to the Agilent forum

formatting link
There seem to be some helpful HP types there

Reply to
Mike Harrison

Long shot but if the card uses a common ethernet chipset, there may be guidance in the source of linux-kernel for that particular chip.

This is no the first time I hear of chips losing there MAC, but I've not had to deal with it in your situation. NVRAM corruption can be prevented with addition of CPU supply monitor chip to hold CPU reset during 'stupid' zone when power is going down (usually) and there's a region where the CPU issues nonsense signals on output lines. There's a random tendency to scribble NVRAM that is easily fixed by adding a CPU power supervisor to hold CPU reset while power below spec.

Your two issues is programming the NVRAM (if lucky you may find what numbers go where from kernel source) and add the supervisor chip to prevent this happening to you again.

Grant.

Reply to
Grant

Hello Jim,

Did you find solution to your LAN corruption problem?

I've got the same problem on a HP1662cs. I can change the LAN parameters but they always turn up the same (192.0.2.3) and the Ethernet statistics (with IP 0.0.0.0) are all 0. I have a DALLAS RTC that is working fine but I don't whether it is also used for LAN data.

Greetings

Claude

Reply to
clwelsch

Welcome to Usenet.

Jim's phone is very likely to have been disconnected after a decade. If the rest of the config memory still works, and you can stand the slow speed, you could put a GPIB/Ethernet dongle on it and use that.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC / Hobbs ElectroOptics 
Optics, Electro-optics, Photonics, Analog Electronics 
Briarcliff Manor NY 10510 

http://electrooptical.net 
http://hobbs-eo.com
Reply to
Phil Hobbs

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.