[volt-nuts] HP 3458A Mem test 1 hight. How do you read 32K memory chips?
J. L. Trantham
jltran at att.net
Wed Aug 20 22:08:07 EDT 2014
I don't think I would worry too much about messing up the 'CALRAM'. If you
'mess up' the 'CALRAM', and your meter is, otherwise, OK, you will spend
about $500 with Keysight for a 'Keysight' calibration, which I would
If they get your meter and it is not 'suitable' for calibration (meaning
that, in some way, it's 'broken'), you will be notified and offered the
option of having them 'repair' the meter (for about $2700, which includes
the 'Keysight' calibration) or returning the meter to you for your repair at
a cost of half the calibration charge, IIRC.
The key point about reading these NVRAM's is to, first and foremost, let the
chips cool after removal before trying to read them. I learned about that
the hard way.
Second, make sure your programmer specifically supports the chip you are
trying to read and program. If so, reading then archiving the data should
be straight forward and programming a new chip should be equally straight
I am not familiar with the 'Mem test 1 high' issue but I suspect the
Assembly Level Repair Manual or the Component Level Repair Manual should
address the issue. I agree with others that this likely means that, in some
way, your meter is 'broken'. If you can chase that down to a simple problem
and repair it, getting the calibration with Keysight would be the next order
of business. Once it passes calibration, you would be able to purchase the
'Repair Agreement' for a couple hundred dollars per year, with a discount
for multiple year purchases up to a maximum of 5, IIRC, which gives you a
meter with a 'factory warranty' for that time period.
Personally, I would look for a 'professional' programmer rather than one of
the 'cheaper' ones, especially if you are anticipating getting into vintage
equipment and programming EPROM's, etc. BP Micro, Advin, DATA I/O, Elnec,
etc., would ones I would look for. I'm sure others on the list would have
their own recommendations. The prices on theBay right now are particularly
ridiculous but good deals can be had. The majority of these units use
parallel port connections and need WinXP as the OS. The later versions,
starting around $500 will connect via USB and be able to utilize Win7 or
From: volt-nuts-bounces at febo.com [mailto:volt-nuts-bounces at febo.com] On
Behalf Of John Phillips
Sent: Wednesday, August 20, 2014 8:41 PM
To: Discussion of precise voltage measurement
Subject: Re: [volt-nuts] HP 3458A Mem test 1 hight. How do you read 32K
I do have a ni gpib-usb-hs
I was hoping to remove the chip, read the data with a usb programer and
wright the data back into a new chip before installing it.
Then do a full calibration.
If the 16k chip has the cal vars what do the other 32k chips have?
Thanks for the input.
I should be able to read to read the data from a good meter with the MREAD
and wright it with a programer.
I have been using excl to read program 3458As I will see if I can get MREAD
to work there.
On Wed, Aug 20, 2014 at 5:33 PM, Mark Sims <holrum at hotmail.com> wrote:
> You REALLY don't want to mess with that 24 pin NVRAM chip... it
> contains the calibration memory. If you bugger the data in it, you
> get to spend a couple of grand getting the meter back into working
> order. I know a couple of people that did just that.
> The solution is to first make a backup copy of the memory contents
> using the (undocumented) MREAD command over the GPIB bus.
> Poul-Henning Kamp figured out how to do it... buy the man a beer... even
better a case of
> beer... Search the archives for details.
> I just backed up all the memory in my 3 HP3458A's for when the inevitable
> BIG BAD DAY comes and the backup batteries go bye-bye. If you happen to
> have a NI GPIB-232CV-A (or possibly a Prologix) RS-232 to GPIB
> converter and can run a DOS or WIN98 program, I have a program that
> dumps the memory chips. It has code for the Prologix in it, but I've
> only used it with my a Prologix compatible unit I built... I don't
> have a real Prologix to test it with.
> volt-nuts mailing list -- volt-nuts at febo.com To unsubscribe, go to
> and follow the instructions there.
volt-nuts mailing list -- volt-nuts at febo.com
To unsubscribe, go to
and follow the instructions there.
More information about the volt-nuts