Tektronix 2445,2465 Cal settings EAROM ER1400

[nicnac117]

I am opening this thread to try and get some information from the the members  on the early 2445,2465  cal settings .
There are plenty of articals on the later 2445a,2465a,2465b etc but very little on the ER1400 EAROM in the 2445,2465.
Also there are very few programmers that cover this chip .
For interest sake I intend to build a ER1400 program interface and attempt to save the settings in the EAROM to file ,in similar fashion to the well published later versions.
I do npot have the test equipoment to do a full calibration  but would like to be able to save the cal settings.
Any input from members is appreciated
M5G1400P will be checked as a possible replacement.

[TurboTom]

Some 15+ years ago I had my 2465 re-calibrated after it suffered from EAROM data loss. Prior to the recalibration, I replaced the ER1400's with new ones. I think there are even today some NOS spares available. Actually, after that I might have used the 'scope ten times in total. It's a real shame how quickly the comfort of the DSOs rendered these analog engineering marvels obsolete. I often thought about making a quick-and-dirty ER1400 reader/writer so I could store the cal information in them but in the end, I'll probably never use my good old 2465 again for any "real" job. So I'm not really sure if it's worth the hassle. I may do it anyway...

Cheers,
Thomas

[med6753]

I am very interested to see how this turns out. I also have an older 2465 DMS with this EAROM. The mainframe EAROM is fine but I am getting a checksum error on the Buffer board EAROM which means it either has bad data or the device itself is going bad. And it appears to be the same type as the mainframe.

Post up what you find out.

[MarkL]

I also have an old 2465 with the EAROM.  It's a fine example of Tektronix engineering at its peak.


There's been some investigation on how data is stored in the EAROM. Take a look starting around here:

  https://www.eevblog.com/forum/repair/tektronix-2445-bad-microprocessor-s68a08/msg685727/#msg685727

For backup of EAROM data, I've verified that EXER 02 displays everything that's in there.  You can take a video of flipping through the addresses (the lazy method) or write them down one at a time.  In the event of failure, reprogramming the values is another matter, but at least you have the backup.

Good luck with building a reader/programmer.  The -35V requirement is going to be a hassle.  You might want to consider building something that takes advantage of the circuitry already in the 2465, perhaps by modifying the processor code in the EEPROM(s) or by using a microcontroller that takes control of the scope's bus.


My EAROM had intermittent checksum errors and I was able to get it working by lowering the operating voltage 2V (replaced VR2003 with a 9.1V zener).  It's been working fine with that modification for at least 20 years now.


Since you're collecting information, I spent some time figuring out the checksum algorithm used in the 2465.  It's mentioned in the above thread, but I later discovered the buffer board used a slightly different algorithm.  Attached below is an updated earom.c which includes the buffer board.

[Orange]

I am opening this thread to try and get some information from the the members  on the early 2445,2465  cal settings .
There are plenty of articals on the later 2445a,2465a,2465b etc but very little on the ER1400 EAROM in the 2445,2465.
Also there are very few programmers that cover this chip .
For interest sake I intend to build a ER1400 program interface and attempt to save the settings in the EAROM to file ,in similar fashion to the well published later versions.
I do npot have the test equipoment to do a full calibration  but would like to be able to save the cal settings.
Any input from members is appreciated
M5G1400P will be checked as a possible replacement.

Many years ago I had the same questions. I found an home brew ER1400 programmer.  The programmer and software works fine on an older PC. (W98 / parallel port)
I never tried it on an XP system due to the parralell port I/O. It could be working.

http://www.ic-prog.com/index1.htm

[nicnac117]

Thats what i am going to build . The modern version 1.06c supposedly works with Win 7 ,but if necessary I will put an older system on a spare pc i have with parallel port .Waiting for the bits to arrive . I gave up repair around 2000 and ditched my workshop around 2006 when i moved to a new house . Just doing this as a hobby , interest in scopes , which i used to repair ,amongst other things.

[nicnac117]

I am opening this thread to try and get some information from the the members  on the early 2445,2465  cal settings .
There are plenty of articals on the later 2445a,2465a,2465b etc but very little on the ER1400 EAROM in the 2445,2465.
Also there are very few programmers that cover this chip .
For interest sake I intend to build a ER1400 program interface and attempt to save the settings in the EAROM to file ,in similar fashion to the well published later versions.
I do npot have the test equipoment to do a full calibration  but would like to be able to save the cal settings.
Any input from members is appreciated
M5G1400P will be checked as a possible replacement.

Many years ago I had the same questions. I found an home brew ER1400 programmer.  The programmer and software works fine on an older PC. (W98 / parallel port)
I never tried it on an XP system due to the parralell port I/O. It could be working.

http://www.ic-prog.com/index1.htm

Thanks Mark,
You certainly did a lot of digging !!! well done and thanks.I hate the fact that the settings can not be backed up easily .At this stage it is not really feasable or cost effective to cal the scopes of this age , unless you have the equipment to hand.
I just want it to work .,without error messages .The cal would be close enough for anything I will be doing with it . I bought this 2445 in 1992 so its like an old dog to me .

[nicnac117]

Just a quick update ,. Got the parts and built the ER1400 programmer to work with ICPROG 1.06c .
 Put a copy of XPSP3 on a Dell 670 Workstation i have spare and did a quick check .
Programmer works ,read write ok  so i am going to experiment further over the next while .
 When ive done this I will write a worktru' for those who are interested.
 Also checked the M5G1400P as a replacement .
Seems to be a direct replacement .....more later!!!!

[nicnac117]

Just a quick update ,. Got the parts and built the ER1400 programmer to work with ICPROG 1.06c .
 Put a copy of XPSP3 on a Dell 670 Workstation i have spare and did a quick check .
Programmer works ,read write ok  so i am going to experiment further over the next while .
 When ive done this I will write a worktru' for those who are interested.
 Also checked the M5G1400P as a replacement .
Seems to be a direct replacement .....more later!!!!

Another quick update:
I have successfully copied the 2445 EAROM ( ER1400 )data to the new M5G1400P.It would be really good to know what parts of the hex does what . I have only a limited basic knowledge of this side of things .
The programmer I made was the ER1400 circuit on the IC prog site. I used 2 power supplies using 8 v and 27v respectively both referenced to board earth with obviously the + of the 27v supply to board earth. Because i had 2 regulated power supplies I didnt need the 8v regulator in the circuit , but all else remained as published.
For someone who has a corrupt ER1400 it would be nice to be able to know exactly what does what so that checksum errors could be changed as necessary . Also if your data is already giving an error its a bit late and a proper recal is probably called for , but these days its just not worth the money to buy the necessary equipment .
 I was looking at the Tek Cal equipment going on ebay  (TG501,PG506A,SG503) .
The last set went for $1025 which is too much for this old equipment , especially when you consider postage and customs to UK.
I wonder are there any EX TEK  guys who have some relevant info??

[med6753]

Yea, that's the issue. If you already have a check sum error how do you know what caused it? And that's exactly what I have on the Buffer Board and I really have no idea what cal data it stores...although I suspect it might be the Counter and the DMM options. I don't think it affects anything in the mainframe. I'm tempted to purchase a M5G1400P, plug it in, and see what happens. If it FUBARS the Counter and DMM calibrations I have the equipment to fix that. 

Where did you purchase the M5G1400P?

[nicnac117]

Theres not much point in putting in a blank one!.If someone had a no error buffer board earom file I will program it in a chip for you .
Ebay for blanks!!

[MarkL]

Regarding EAROM contents...

I wonder are there any EX TEK  guys who have some relevant info??

You might pose the question in the Yahoo TekScopes group.  There are some ex/retired Tek engineers that hang out on the list who know the older equipment very well.

It would be even more interesting to get the original source code for these old scopes (assembler, I'm assuming).  It's hard to imagine it has any trade secrets or industry value after 30+/- years.

[nicnac117]

I have posed the question on the TEK groups site.
What we really need is the memory allocations that each calibration setting occupies.
It could be done manually i suppose on a test scope .
Take a copy of the ER1400 hex.
Adjust one scope calibration value .
Then use Hex difference to see what allocation has changed .
It would take forever to map what the allocations are .
I dont think there is much interest in these scopes any more . Everybody wants an "A" or "B" version .
I ve a 2467B opt 10 , 2465B opt10 , 2 x 2465, and 1 x 2445 .
I have backed up all the settings and if I sell any I will send a spare eeprom/earom with them , so the classic scopes can continue into the future. Nostalgic ...I know .
If I dont get a reply on the TEK scope forum ....that will be it !!!!!!!!!!!!!

[nicnac117]

Well, I tried on the TEK website  ...............no reply ............actually not impressed at all with that group .............much better here.......

[med6753]

Well, I tried on the TEK website  ...............no reply ............actually not impressed at all with that group .............much better here.......

Very interesting. I thought I was the only one that felt that way. When I initially joined the Tek Yahoo group I did get some informative and useful responses to fix issues with the 2465 as well as sources for documentation. But it seemed after a while that as I gained experience and fixed specific issues that as I asked additional questions the responses slowed to a crawl and eventually it was nothing but crickets. And I don't understand why. Granted, that Yahoo group format really sucks and I wish it was in a blog format such as here. Or maybe...perish the thought....some professional snobbery going on here. That crossed my mind and I truly hope that's not the case.

Anyway, back on topic. Been busy with some other activities over the past few weeks but I should be getting back to the remaining issues with this 2465 DMS. I did discover some thing very interesting related to the Buffer Board Checksum error which I will update soon in this thread....

https://www.eevblog.com/forum/testgear/tektronix-2465b-oscilloscope-teardown/

[Satbeginner]

Hi all,

I know it's an older thread, but here I go:

I just love these 2465x scopes, I own two 2465B's, that I restored to almost new. (Dallas, PS recap, cleaned inside out,. topbags, etc)
Still looking for frontcovers....

Anyway, I am now working on two 2465's, and one of them is showing a row of dots at the bottom of the screen.
When I perform a calibration, these dots disappear, but after about 2 minutes they are back again?

I now go with my assumption it's the er1400 memory suffering from memory loss.
I ordered a 'new' one, and will replace it when it gets here.

Since I am not familiar with this type of memory and it's long term behaviour my question is:

Is the chip loosing it's content related to being old AND used, or just age?
If it is the first, a NOS replacement would not really help?

Curious what your experience is,

Un saludo,

Satbeginner (Leo)

[MarkL]

My 2465 had an ER1400 that was intermittent.  The intermittent behavior was related to the power supply voltage.

It would be my first guess that your ER1400 has gone bad, for whatever reason.  You should at least verify power supply and signal levels on the ER1400 pins before going through the trouble of replacing it.

If still not sure, after you get the dots you could try running the EAROM test (TEST 04) to see if it shows a failure.  Maybe run it repeatedly in loop mode.

You could also see if there's anything interesting in the EAROM examine (EXER 02), such as parity errors in any of the CAL locations, or if the CAL constants change by themselves.  The values at 0x01 to 0x4C should all have odd parity (see service manual pg. 6-16).

[mikeselectricstuff]

If it helps anyone, I have a standalone ER1400 programmer which can copy devices, and one good ER1400.
No easy way to get data in/out of it though.
It's a unit I built 20+years ago for reading & setting unlock codes on old ford car stereos.

ISTR the reason the ER1400 is not widely supported is it needed something like 36V for programming.

[tggzzz]

I have a 2465, and I've thought about this problem but not done anything about it (yet).

My strategy would be...

Use the test routines to step through all the memory contents, videoing the screen to record the values for posterity. If and when I need to, I would replace the EAROM with a cheap small processor such as an Arduino Trinket.

Since that is 5V only, I would simply bypass the 42V level shifters, and write a small program that used crude bit-banging to respond to the EAROM's serial address/control/data lines from the 2465's processor.

Initially I would simply hard-code the EAROM's contents into Arduino's memory, and ignore any new data coming from the 2465. That would preserve the calibration, but would prevent the 2465 restarting with the settings it had when it was turned off.

Later I might, or might not, monitor the EAROM's datain line, and store any new values in the Arduino's EAROM.

No, I haven't looked at the timings and use-cases in detail, but at first glance there don't seem to be any killers.

[MarkL]

...
Use the test routines to step through all the memory contents, videoing the screen to record the values for posterity.
...

Or if you happen to have the GPIB interface (option 10), you can also dump the EAROM contents to a computer.  See here:

  https://www.eevblog.com/forum/testgear/tektronix-2465b-oscilloscope-teardown/msg1167050/#msg1167050

A bit of simple programming/scripting is needed.