| Products > Test Equipment |
| HP 856X Battery Backup and Various Memory Questions |
| << < (4/4) |
| kipp:
Hi, I know this is old, but reading through the responses I don't think the original questions were clearly answered, and the discussion drifted onto the 859x series, which evidently are sufficiently different that the information doesn't apply. --- Quote from: Electro Fan on May 13, 2015, 03:50:30 am ---Here are a bunch of questions, please/thanks: 1. Is it correct that there is a battery for: a. the Spectrum Analyzer itself --- End quote --- YES. It can be accessed without opening the unit, by opening a removable plate hiding under the mass memory module on the back of the unit. It's a 3 V lithium battery in a standard format. --- Quote ---b. the Mass Memory Module --- End quote --- YES. If I recall correctly, it's format is identical to the one in the analyzer itself. Accessing it requires the module to be opened by removing four torx screws, which might be in there good and tight. There is a label on the mass memory module that should say when the battery was last replaced. Mine says "1997", it is now 2023, and when I got my module open after a fight with the screws the battery was totally fine, still well over 3 V. --- Quote ---c. the Memory Card that inserts into the Mass Memory Module --- End quote --- YES. It's a CR2016 watch battery sitting in a tray inserted into the end opposite the connectors. The card might be marked "battery" where the tray is located. It is removed by getting a fingernail into a tiny little slot and pulling until it hurts. --- Quote ---I'm pretty sure the answer to a. and c. is yes, but I'm not so sure about b. (the Mass Memory Module). 2. Regarding the battery replacement process for a. the Spectrum Analyzer itself, below is a photo of a SA without the Mass Memory Module in place, and also an image of the battery replacement process from the manual. The manual says there are two screws to be removed - are those the screws that are highlighted with the red circles? --- End quote --- That is the correct panel. I actually have only ever replaced the battery once, when I had the unit opened up, so I did it from inside. I've not removed that plate. I would have guessed, however, that you should only remove the one screw that's different from the others next to the label that says "to remove ...", but you can't break anything by experimenting. There is nothing delicate in that part of the instrument. Start taking out screws until the plate is loose, then put the ones back that you didn't need to remove. I've attached a photo of the inside of the unit to show what's there. The battery holder is on the back, just behind the rear end of the CRT assembly (I seem to have taken this photo when I had my battery removed for some reason; at the time I was debugging my CRT driver board). --- Quote ---3. What type of battery is needed for the SA itself? (The manual says it's a 1420-0341; is there some other more commonly found equivalent?) Anyone know where this battery can be purchased in the U.S.? --- End quote --- It's a 3 V lithium battery in a standard format. I don't remember what format, exactly, but it says on the battery. CR2 maybe? It's not a kind with wires, it's a cylindrical battery that is inserted into a battery holder. The format is widely available, though probably not at your local convenience store. Try camera shops or drug stores. --- Quote ---4. Is the battery in the SA itself backing up calibration data? (If this battery dies and the SA is turned off will the calibration be lost? Or are there other scenarios under which the cal data can be lost?) If this battery (in the SA itself) isn't backing up cal data, what is it doing? Where is the cal data being stored? --- End quote --- No, it is not backing up calibration data. Calibration data is stored in a flash memory chip, and does not require any power to be retained. That chip is physically write-protected with a jumper on the CPU board, it is next to impossible to alter its contents inadvertently. The battery can be safely removed, or die on its own, without negatively impacting the instrument. The battery is used to store front panel settings in RAM so you can press "LAST STATE" when you turn it on and get it back to how you had it set last time you used it. The instrument knows when the RAM has been wiped due to a battery failure (presumably it stores some magic cookie in memory and looks for it to detect corruption), so there's no risk of putting the instrument into a harmful impossible state by attempting to restore its configuration from corrupted RAM. I have removed and re-installed the battery many times during a repair of my instrument without consequences, other than to get a "battery failure" message on boot. --- Quote ---5. If the battery in the SA itself is backing up cal data what data is being stored in the Memory Module (if any)? Or is all other data (outside the SA itself) being stored resident in the Memory Card? (Is the Mass Memory Module just a housing for the Memory Card or does the Mass Memory Module store some data without a Memory Card?) --- End quote --- The mass memory module includes its own memory, and a clock. With the module installed, the spectrum analyzer can optionally display the date and time on the screen. The module can be disconnected from the instrument, and it retains the time and its memory contents. --- Quote ---6. Can someone please explain the typical/primary intended uses of the Mass Memory Module and the Memory Card? Are there other memory (beyond the cal memory) functions commonly used with the SA which use memory resident within the SA itself (ie, not in the Mass Memory Module and/or Memory Card)? --- End quote --- In practice, today, for a hobbyist, the answer is "not much". The mass memory module provides the date and time. You can store trace data in it, for example to do comparisons to compute ratios. I don't know what else besides the mass memory module itself is capable of reading and writing the memory cards, so I don't know what you'd do with one after storing data on it. Nothing else I own can read and write them, I got two cheaply, mostly out of curiosity, but they're useless. The mass memory module is a large, expensive, heavy (it's extremely solidly build) add-on just to get the time displayed on the screen. The memory cards can store program data. HP sold a phase noise measurement routine in the form of one of these cards, part number 85671A. Also a spurious response utility, part number 85672A. If you need or get your hands on one of those, you would need the mass memory module to read the card contents to run the routines. There are other modules that can be plugged into that port on the back of the instrument. That I/O port provides access to the instrument's CPU bus, and you can install what are effectively firmware replacements in that port. HP sold a calibration and diagnostics module that would replace the instrument's normal user interface with a system to automate the calibration sequence for you. It would prompt you to take measurements with bench tools and enter the results. The calibration guide is over a 1000 pages, so automating it would probably have been a god send if you were running a cal shop, but that module is optional, it isn't required if you ever need to do a calibration. --- Quote ---7. As you can see, I'm unclear (actually very confused) on the overall architecture for memory functions (and backup batteries) on the 856X, and the Mass Memory Module, and the Memory Card; any overview on this would be very welcome and appreciated. Thanks! --- End quote --- |
| kipp:
--- Quote from: dfnr2 on May 18, 2015, 11:10:20 pm --- --- Quote from: Electro Fan on May 15, 2015, 10:22:52 pm ---If the data is in EEPROM - which is typically (always?) non-volatile - why does it need to be backed-up with a battery? Plus, how does a user write new cal data to EEPROM? EEPROM would seem like a nicely secure place to put cal data but maybe the cal data winds up in SRAM since SRAM is designed to be used with battery backup? I don't know, just conjecturing... --- End quote --- The Cal data is altered only during service, and is not meant to be modified or backed up by the user. It also should not be lost when the removable battery is taken out to clear the user data (programs, traces, etc.) Looking in the manual (page 7) confirms that the Cal/config data is in EEPROM. Of course, if the EEPROM is removable (or even if it's just accessible), I think it would be prudent to back it up anyway. This is different from the 859xE SA's, which have no EEPROM, and have a battery permanently mounted on the PCB. Since it can't be removed casually, the risk of casually erasing cal data is small, although these SA's do have a procedure for backing up the cal data in case of battery failure. Such a procedure is not necessary (and not described) for the 856XE series. --- End quote --- Hi, That is correct. The calibration data is stored in an EEPROM. The battery is not used to retain the calibration data, only user data. BTW, these instruments are now older than the EEPROM's datasheet promises those chips can retain data. The chip is socketed, so, yes, if you can, it is a very good idea to open the unit up, pull out the chip, and dump its contents. But to answer an earlier question, other than by doing that, for the 856x series there is no "backup" procedure for the calibration data. Also, on a related note, there are numerous PAL and/or GAL programmable chips in the digital parts of these units that are *also* now beyond what their datasheets promise their program retention time is. They are not socketed. The calibration EEPROM's contents can be recovered by following the calibration procedure, if you have the right test equipment, but replacing the ADC's internal state machine programming is maybe impossible if that gets lost (unless someone has a friend at HP/Agilent/Whatever who can dig that up out of a filing cabinet). |
| Jamie_A:
The information in the thread is contradictory to the manual for the 8562A. See attached photo. I can't find any more information as to what is stored in battery backed ram, and how I could back it up over GPIB. Anyone know where I could dig for this? Is there a separate calibration manual? edit: I looked more carefully through the manual, and confirmed that the important calibration (DAC values for YTF/mixer/YTO biases) is stored in write protected EEPROM. I haven't read all of the adjustment sections, so still unsure what "calibration" data the manual refers to for the battery backed RAM. |
| Navigation |
| Message Index |
| Previous page |