Possible problem with Prema 6047 or 6048 meters (new EPROM received)

[perdrix]

Hi all,

I've noticed something about the behaviour of my Prema 6048 that leads me to suspect a firmware problem

I'd like to determine whether this is a problem unique to me or a common one.

If you own either Prema 6047 or 6048, I'd be most grateful if you could undertake the following test sequence with the meter nicely warmed up:

• Power cycle the meter to ensure we are starting with a clean sheet.
• Select the 20V DC range and 4 second integration time
• Connect a stable 10V (or so) source
• Record the meter reading after allowing it to stabilise for a few readings
• Swap the +ve and -ve leads
• Record the meter reading after allowing it to stabilise for a few readings
• Remove the leads and install a shorting link
• Record the meter reading after allowing it to stabilise for a few readings
• Press "Zero" and allow it to complete the null
• Re-connect the 10V source
• Record the meter reading after allowing it to stabilise for a few readings
• Swap the +ve and -ve leads
• Record the meter reading after allowing it to stabilise for a few readings

Please post your readings to the thread.

Thanks
David Partridge

[carl_lab]

What exact problem do you expect?

[Andreas]

I think the description is here:
https://www.febo.com/pipermail/volt-nuts/2018-January/005593.html

with best regards

Andreas

[hankee]

I have seen this effect , although much smaller, with all  Prema's I have (had) , a 5000, 2 times 6001 and a 6030.
It looks to me this is something in the Prema design.
The effect was 10 to 20 ppm with my dmm's, this is probably within the specs but this difference between positive and negative voltage measurement it is quite annoying.

[Kleinstein]

Comparing positive and negative reading of the same stable voltage is one way for a spot check on linearity. Depending on the meter and speed of the readings there can be settling effects too.  Of cause the measurement only makes sense with the zero offset well set.

[Mickle T.]

This is the Prema's design flaw due to unipolar ADC and Inp-Amp output polarity switching. AFAIR it described in the Service Manual.

[perdrix]

The issue is to do with HOW the meter handles the zero offset when you use the Zero command.  Please if you do have one of these meters, I'd appreciate you doing the test.  If (as I suspect) there is a firmware problem, I'd like to persuade Pream to issue updated EPROM images.   I do however need the data to determine if this is just a problem with this meter or a general one.

Dave

[perdrix]

And yes the description of Volt-nuts does describe what happened on MY meter.

[perdrix]

It would appear that Prema will send me (for a small fee) a replacement EPROM with the latest firmware, but without the factory calibration constants specific to my meter.

I've asked If I can post the ROM contents ...

Dave

[voltampere]

Very quick check.



4.  + 7,118760
6.  -  7,117125

8.     0,00000005

11. + 7,118760
13.  - 7,117800


unfortunately i have a lot of noise when measuring the negative voltage, almost 100 microvolt.
Positive reading is only about 5 microvolt noise...

My Meter was at the PREMA factory 6 months ago.
So it should have the latest firmware?

What are the readings on your meter ?

Regards

Joerg

[perdrix]

Thanks for your readings.

For my meter see: https://www.febo.com/pipermail/volt-nuts/2018-January/005593.html

Prema are sending me a replacement EPROM with the latest firmware, and have given me permission of post the image once I have it.

The noise on the -ve readings doesn't sound right ...

Dave

[Magnificent Bastard]

Thanks for your readings.

For my meter see: https://www.febo.com/pipermail/volt-nuts/2018-January/005593.html

Prema are sending me a replacement EPROM with the latest firmware, and have given me permission of post the image once I have it.

The noise on the -ve readings doesn't sound right ...

Dave

You might ask Prema where in the EPROM the factory default calibration constants (for each individual meter) are kept.  Then you can read the new EPROM into memory, and read the old EPROM into memory, then cut and paste that specific section from the old EPROM into the new EPROM's memory image; then burn a new (modified) EPROM with your meter's default calibration constants.  If I had to guess, these will probably be at the end of the EPROM, and there might be some kind of checksum that needs to be modified (or perhaps they have a separate checksum for the firmware and calibration constants, so that would just copy over without modification).  Some hacking may be needed to make this work, but if you figure it out, please post here how you did it!

[perdrix]

I'm ahead of you

I already determined where the factory calibration data is stored in the EPROM, so just a bit of work with a hex editor (with luck and a following wind) ...

Dave

[perdrix]

A replacement EPROM with the latest firmware for the 6048 meters arrived from Prema on Saturday morning.

I've attached a zip file with the images of the EPROM and details on how to add the calibration data for your own 6048 to the image before you burn to your EPROM.

Make sure you take a backup of the original image before you start (just in case).

I'm also going to upload this to ko4bb.com and Bruce Lane's repository at ftp.bluefeathertech.com

Dave

[voltampere]

Thanks for publishing the firmware.

Did you "upgrade" your meter with the new firmware already?

Did it solve your issues?

regards

Joerg

[perdrix]

Yes Joerg,

I have installed the updated firmware, and it appears to have resolved the problems.

I also did a full calibration.

David

[perdrix]

Please note the updated EPROM is not suitable for the Prema 6047.

David