Fluke 45 - EEROM question

[dramoto]

Hi....
I´m new writing hier.
First of all, thanks for all people of this comunity for share your knowlege and experience. Really usefull.

The matter.
I adquired a Fluke 45 for repair. I fixed one problem and now, I´m trying to fix a corrupted calibration data without pay the 180€ that cost hier the calibration.

The fluke 45 have a litlle eerom with 64 registers where strore the calibration data, the multimeter configuration data and the CRS data check.
I´m trying to go directly to this eerom without pass over the calibration procedure, but....
knows anyone how this registers are organizated?
or what data use the fluke perform the CRC check?
Maybe some one that have a fluke 45, can share his registers data to compare with mine.

It´s probably that fluke have similar eerom for the fluke 45 and other fluke.

Thanks....

[Uup]

I doubt that you will be able to calibrate it that way.

I would recommend downloading the Fluke 45 service manual and doing the calibration yourself, assuming you have or can borrow the equipment to do so.

[dramoto]

Thanks Uup for your opinion.
It´s a pity, but I know noone that can share with me the equipment hier in Spain. I don´t have posibility to perform the calibration myself, just to pay 180€ and I prefer to save this money for a future keihtley 2000.
I tried calibrating the ohms with cal resistor, but no evolution with the cal error.
I think that just one of the register data is corrupted or the CRC data stored. I would like to compare the register datas with others fluke 45 in the attempt to find the corrupted and next step would be to try to calibrate only this, or to try to store a new CRC data assuming and knowing that one step cal is complety out of tolerance.

I know that the normal procedure is just to calibrate the multimeter and that my roundabaout way is strange. But I consider too expensive the calibration cost that fluke spain ask for this multimeter. at least expensive for my hobby intentions.
I think to have read that in USA, to calibrate a Fluke 45 cost about 60$. It´s a great difference. 

Ok......, with my fight       .... are there some fluke 45 owner with a little of time to help???
 

[Uup]

I have a working Fluke 45 in a box stored somewhere. I'll find it and dump the data from its' EEPROM for you. But give me a couple days...

[Uup]

I had some time tonight, so I found my Fluke 45 and checked it out. It still works fine and the calibration is surprising accurate too.

I removed the EEPROM and attempted to read the data from it but encountered a problem. Selected the device in my programmer (EETools TopMax2) but I neglected to see that the device selected was a SOP8. It didn't read the chip properly and that's when I realised the different package sizes. I don't think the EEPROM was damaged.

The part in the Fluke 45 is a Nat Semi 93C46WM in a very uncommon SOP14. Did a search but was unable to find a datasheet for the SOP14 variant. The latest software for the TopMax2 doesn't appear to support a 1K EEPROM in a SOP14 either, oddly enough. So I'll need to make an adaptor.

Do you have a pin-out of the SOP14 93C46?

Otherwise, I can figure it out by checking the connections back to the main processor (6303). Only 6 of the 14 pins would be used, two for power, data in, data out, clock and select. I assume that you've already worked it out as you read the data from your one.

[free_electron]

I have that data.

Calibration is not enough as the eerom stores other stuff as well. If the eerom is corrupted you need a gpib card to reset the serial etc ... Otherwise cal will fail.

I will post a link to the file here later today. First fleamarket !

[amyk]

Do you have a pin-out of the SOP14 93C46?

http://www.datasheets360.com/pdf/589509130356737480

[Uup]

Do you have a pin-out of the SOP14 93C46?

http://www.datasheets360.com/pdf/589509130356737480

Looks good, thanks! 

I tried 'datasheetarchive.com' and 'alldatasheet.com' with no luck, only had the SOP8 variant. I'll have to bookmark 'datasheets360.com'.

[dramoto]

Thanks for your good intentions!!!

I have work on the fluke by GPIB for extract the register information (at the moment no succes with rs-232) . It´s easy and not need to remove the IC from the pcb. Thanks Uup for your time.

I have order for some 93c46 to play with them after put in game the fluke integrity. And I´m now investigating for use a Arduino to read/write the 93c46

That know at the moment is that the register have a total of 64 address of 16 bit.
32 address used for calibration (the multimeter have 32 step cal) and 32 adresss used for configuration status, CRC, etc.
I´m going to play changing configuration datas and recalibrating the ohms looking for changes in the memory. In the attempt to assign functions to memory adrress.

 

[dramoto]

Thanks Orange for your information. 

Step by step they are progress. I found interesting things. With a little of work for change the hex data from the BIN file to the decimal data that report my GPIB, and comparing both fluke datas... I can confirm that the calibration data is stored from the 21 to 53 eerom address. (Counting from 0 to 63). And it appear to be ordered exactly like using the "CALCONST?" GPIB command (from 1 to 33 like explain the table 4-9 from fluke 45 service manual)

I´m continue investigating....

"Uup", please, and of course without any type of hurry (I have time to "this game").... I would like to see  your fluke eerom data. Becouse they are some great differences in the first 16 address of the Orange´s eerom and my eerom.
If you have GPIB adapter (maybe too with RS232), you can interrogate the fluke eerom values just using the "EEREG? xx" command.

Or maybe I can see the free_electron´s eerom...if don´t have sell it in the fleamarket 

[Uup]

"Uup", please, and of course without any type of hurry (I have time to "this game").... I would like to see  your fluke eerom data. Becouse they are some great differences in the first 16 address of the Orange´s eerom and my eerom.

Attached is the data from the EEPROM in my Fluke 45. I also included the data from the V1.7 program ROM, just in case the EEPROM data arrangement is particular to that version.

[dramoto]

[/quote]

Attached is the data from the EEPROM in my Fluke 45. I also included the data from the V1.7 program ROM, just in case the EEPROM data arrangement is particular to that version.
[/quote]

Thanks Uup....
I hope not to need the EEPROM program. But I´ll save it carefully for the posibilty of a future not desired.
I have located the CRC´s address. In the address 20 are stored de CRC check for the "Start Up User Configuration". And in the  address 60 are stored the CRC data for chek the Calibration Data. At the moment I´m working in to analyze the parity check procedure (every calibration constant have his parity check). I hope that my problem is only the Parity Check of some constants. If not... I´ll not be sure if I´ll can discover how to calculate the CRC data...... or spend part of my life reprograming the value from 0 to 65535 until take the right one   
After this I´ll can go to the bank at night to open the strongbox   

[dramoto]

"Uup", please, and of course without any type of hurry (I have time to "this game").... I would like to see  your fluke eerom data. Becouse they are some great differences in the first 16 address of the Orange´s eerom and my eerom.

Attached is the data from the EEPROM in my Fluke 45. I also included the data from the V1.7 program ROM, just in case the EEPROM data arrangement is particular to that version.

Fix it!!!
Thanks for all people colaboration, especially to Uup ant to Orange for share yours eeroms data.
After a lot of time experimenting with my fluke, I have been able to isolate the corrupted data and correct It.

It was just a partity error in the Vcd cal data values. And it appear to be in tolerances, without the "error 7"

Next step...... are there spanish eevblog people? when and where for one "beer drinking + hobby calibration party"?? 
 

[tom_iphi]

Fix it!!!

Can you please give some hints how you fixed it? I am having the same issue.
Did you reprogram the EEPROM externally or in built-in state?
How did you determine how to fill the parity and checksum info?

Many thanks and best regards,
Tom

[dramoto]

Hi Tom...
It was a really hard work to resolve the problem. I have forgotten near all the steps, but I have some notes of my study. I can send to you more info, but I need time to check my notes and remmember the manuals and steps that I did.

For give you first steps...
More or less, you will need:
- The fluke 45 with GPIB port
- A GPIB adapter (or a original by USB, o make a Arduino GPIB adapter)
- A spare 93c46 eerom (not needed, but better if you don´t want to put in risk the original)
- Study the service manual and user manual to be a little bit get in use with RS232 and GPIP commands
- Capacity to remove and re-solder the eerom IC from/to the pcd (SMD solder station)
- A eeprom programmer
- A IC-holder to do easy to move the eerom from the fluke to the programmer during the steps
- Some resistors to run a calibration. Better if you have cal resistors. Not needed if you have other capacity to calibration like Amps, or Volts
- And some/much time to waste....

Just think...
You can solve it just sendind the fluke to calibratión.
In Spain the cost was 180€. The same that you can pay for a serviciable unit.

[tom_iphi]

Hi dramoto,

many thanks for your help! There is no hurry.
I am familiar with GPIB and equipped for it and I have already read the EEPROM of my Fluke 45 via GPIB.
All but one cal constant seem ok, one offset for AC Volts seems to be out of spec. I guess, this EEPROM cell got corrupted.
I found the instrument on the junk and repaired it. The readings with factory presets seem ok, but I do get that EEPROM error message upon power up.

Did you actually recalibrate the instrument?
Or else with what data did you rewrite the EEPROM?

As far as I understand the manual, I can fix the problem by recalibrating the instrument, but I do not have accurate reference voltage and current sources.
I had hoped I could fix the parity or checksum bits and the one wrong offset manually without going through the whole calbibration process.

Getting the instrument recalibrated professionally is out of the question. This is for hobby and professional recalibration is too expensive. The instrument is even useable as is, but the error on power-up is bugging me ;-).

Best regards from Germany,
Tom

[dramoto]

The Fluke 45 perform two consistency checks on the data of the eerom. One is the parity check, every cal constant must to be parity or odd depends of the rule. And the other check is a Checksum, in fact two checksum, one for cal constants (error 7), and another for system data (error 6).
If the checksum is corrupt. Just need to do a new write by the fluke. For example, go into calibration mode and do the calibration just of resistors. It´s easy with this DMM becouse you can select the value of the resistor to check. You must to check the manual, but for reference you take resistors for the 6 ranges (10ohm, 100ohm, 1kohm, 10K, 100K, 1Mohm) just read with your fluke the value of this resistors (or better with another DMM more accuracy) and you can use this resistors to do a self calibration. The gol is that you can select on the fluke that your reference resistor is a 1004.5 ohms intead a 1000.000 ohms.
This will fix a checksum corrupt (error 7).

I didn´t recalibrate my fluke. Just isolate and fix the corrupted data. Bought some cal resistors. And did a resistor calibration for fix the checksum.
I bought later a voltage reference to check, and it "appear" to be in tolerance. More than really I need.


I played a lot writing and re-writing the fluke eerom. And learned a lot of this system. But 18 months later my volatile memory is with other things....
Please, send me your list of cal constants. I can cross info and check is your offset can be wrong or not.

[tom_iphi]

My error code is indeed 7, i.e. a parity or checksum error.

Here is my instrument ID:

FLUKE, 45, 5675271, 1.7 D1.0

And here are my cal constants as read with calconst?. CalConst 21 seems to be out of spec. It is -169, but should be between -100 and 100.

CalConst#1 = +1.032592
CalConst#2 = +1.032592
CalConst#3 = +1.038925
CalConst#4 = +1.033584
CalConst#5 = +1.032180
CalConst#6 = +1.000534
CalConst#7 = +0.9999085
CalConst#8 = +0.9998169
CalConst#9 = -66
CalConst#10 = +0.9995880
CalConst#11 = -67
CalConst#12 = +0.9998779
CalConst#13 = +0.9992523
CalConst#14 = +1.001663
CalConst#15 = +1.001510
CalConst#16 = +0.9713287
CalConst#17 = +0.9712982
CalConst#18 = +0.9956207
CalConst#19 = +0.9700012
CalConst#20 = +0.9703064
CalConst#21 = -169
CalConst#22 = +0.9947815
CalConst#23 = +1.004806
CalConst#24 = +1.004317
CalConst#25 = +1.008178
CalConst#26 = +1.003326
CalConst#27 = +1.003616
CalConst#28 = +1.003616
CalConst#29 = +0.9965668
CalConst#30 = +0.9963837
CalConst#31 = +1.000213
CalConst#32 = +0.0000000
CalConst#33 = +1.000052

Here is the EEPROM content as read with EEREG?

EEREG? 1 = 0x0567
EEREG? 2 = 0x5271
EEREG? 3 = 0x0080
EEREG? 4 = 0x0000
EEREG? 5 = 0x0008
EEREG? 6 = 0x0100
EEREG? 7 = 0x0000
EEREG? 8 = 0x0000
EEREG? 9 = 0x0000
EEREG? 10 = 0x0000
EEREG? 11 = 0x0000
EEREG? 12 = 0x0000
EEREG? 13 = 0x0000
EEREG? 14 = 0x0000
EEREG? 15 = 0x0000
EEREG? 16 = 0x0000
EEREG? 17 = 0x0F00
EEREG? 18 = 0x0701
EEREG? 19 = 0x0082
EEREG? 20 = 0x0000
EEREG? 21 = 0x9038
EEREG? 22 = 0x90B1
EEREG? 23 = 0x90B1
EEREG? 24 = 0x93EE
EEREG? 25 = 0x9132
EEREG? 26 = 0x907B
EEREG? 27 = 0x8046
EEREG? 28 = 0x7FF4
EEREG? 29 = 0x7FE9
EEREG? 30 = 0xFF7D
EEREG? 31 = 0x7FCB
EEREG? 32 = 0xFF7B
EEREG? 33 = 0x7FF1
EEREG? 34 = 0x7F9E
EEREG? 35 = 0x80DA
EEREG? 36 = 0x80C7
EEREG? 37 = 0x7153
EEREG? 38 = 0x714E
EEREG? 39 = 0x7DC3
EEREG? 40 = 0x70A4
EEREG? 41 = 0x70CD
EEREG? 42 = 0xFEAE
EEREG? 43 = 0x7D55
EEREG? 44 = 0x8277
EEREG? 45 = 0x8236
EEREG? 46 = 0x8430
EEREG? 47 = 0x81B4
EEREG? 48 = 0x81DB
EEREG? 49 = 0x81DB
EEREG? 50 = 0x7E3F
EEREG? 51 = 0x7E27
EEREG? 52 = 0x801C
EEREG? 53 = 0xE874
EEREG? 54 = 0x0145
EEREG? 55 = 0x3F80
EEREG? 56 = 0x01B9
EEREG? 57 = 0x3F80
EEREG? 58 = 0x01A4
EEREG? 59 = 0x0000
EEREG? 60 = 0x0000
EEREG? 61 = 0x142D
EEREG? 62 = 0x0000
EEREG? 63 = 0x0300
EEREG? 64 = 0xFFFF

Many thanks again for your support!
Best regards,
Tom

[Waldo64]

Hallo Tom,

i have two Fluke 45.

No. 1 has no "error 7" and calconst 21 and 31 are out of specs.
I never did any calibration on that instrument, and will never do.
For me as hobbyist all measured values are far good enough.

No.2  has "error 7" and all calconst are in specs.
This instrument showed up "error 7" just after correctly calibrating AC Volts.
Even a new (correct) calibration of AC Volts did not erase the "error 7" messsage.

As far as i can see even a single (correct) calibration of any function (Volts, Amps etc.) can trigger "error 7".
I think it maybe a software problem with the fluke 45.

What i also noticed: The value of register 64 rises from time to time.
Maybe it's a counter for the power on time.
Maybe this register needs a minimum value (warmup time) for a successful calibration without "error 7".

[tom_iphi]

Hi Waldo64,

many thanks for your feedback!
Meanwhile I have borrowed a second Fluke45 without error 7 and I can confirm that Cal Const 21 is out of spec there, too.
So, this is indeed not related to the cal problem.

I intend to do a calibration against that working instrument and I'll see if the error hopefully disappears.
It is a shame that cal constants cannot be rewritten via GPIB.

Best regards,
Tom

[dramoto]

HI..
Sorry for my delay.
Just a fast reply for today.
I discovered how Fluke do the parity check for the "long values". Not too much info about the "short values". I must to check.
Just say to you other EEREG 42 (Calconst 21) values from others (thanks to "Up" and to "Orange"), for example, "FE3F", "FF61", or "006E" or "002C, "0000" will comply with the parity rule.

Your offset in Vac is FF7D and FF7B. This mean that your offset for Aac must to be not so far. For example, the same value FF7D or FF7B or FF61or 0000
You can try to program any of this values on you EEPROM by hand, and then perform any calibration to fix the checksum error.


 

My error code is indeed 7, i.e. a parity or checksum error.

Here is my instrument ID:

FLUKE, 45, 5675271, 1.7 D1.0

And here are my cal constants as read with calconst?. CalConst 21 seems to be out of spec. It is -169, but should be between -100 and 100.