Fluke 1587 insulation multimeter battery fuss

[shakalnokturn]

This meter was only acquired recently, I can't tell all its history other than it had a little corrosion in the battery compartment.
It has been thoroughly cleaned, PCB closely checked for any signs of leakage.

Issue mostly happens on a "cold start", meter has been off for several minutes:

If I turn it on straight to the insulation test position, most of the time it will complain about batteries being low, sometimes enough to not allow the insulation test.
Switching from insulation to any other position makes no change on the low battery indication when returning to insulation test, no battery complaints on other functions. Switching off and back on once or twice will allow normal operation on insulation test.

Switching on to any other function then (after 2-3 seconds) to insulation test won't show low battery indicator/message.

I have tried with two new sets of different branded alkaline batteries, both giving the same behaviour.


Same problem is described here:
http://www.labwrench.com/?community.posts/threadNo/109354/subject/fluke-1587-megger-question/

I have tried the "high current battery test" as described, it reads 5.3V but won't keep the meter happy on the next cold start.

Is this a well known problem, with a known solution or abnormal behaviour?

My firmware version is 1.04

[dacman]

I've seen it a couple of times, once with a unit that had battery leakage residue (white powder substance) and another that looked pretty dingy in the battery compartment (which I suspect was due to battery leakage).

Clean the battery contacts with a switch lubricant and cotton swab, and if the contacts do not (then) look pristine, order replacements from Fluke.

According to the Fluke response in the posted link, a major (US) brand of batteries is known to cause this problem due to high internal resistance.  (My suggested fix is intended to minimize contact resistance.)

[Black Phoenix]

I've seen it a couple of times, once with a unit that had battery leakage residue (white powder substance) and another that looked pretty dingy in the battery compartment (which I suspect was due to battery leakage).

Clean the battery contacts with a switch lubricant and cotton swab, and if the contacts do not (then) look pristine, order replacements from Fluke.

According to the Fluke response in the posted link, a major (US) brand of batteries is known to cause this problem due to high internal resistance. 

My sugested fix is simply stay away from that major US brand of batteries and use Eneloops or normal Panasonic Alkaline. Still waiting for one to fail on me.

[TERRA Operative]

I use the AA lithium primary cells in my unit to ensure a good voltage.
I haven't had any leakage yet (I should check actually....) so I haven't had the battery indicator show unless the batteries were actually flat, usually from zapping the apprentice too many times.

[shakalnokturn]

@ dacman: The contacts are nice and clean in battery compartment and between compartment and PCB.

@ Black Phoenix: I'm using neither Duracell nor Energizer. The batteries currently in the meter are Philips "Power Alkaline"

@ TERRA Operative: That's one solution, quite an expensive one unless you get the apprentice to buy the batteries once he's finished electrolysing. Do primary lithiums ever leak?


I have tried supplying the 1587 from a lab supply, at normal voltages I couldn't get the isolation test to grumble about the batteries being low.

The funny thing being that for any supply voltage between 3.5V and 7V the high current battery test will show 4V... The test current must be quite high but very short, I wasn't able to catch it on the lab supply or DMM.
The next step would be to scope the current during start on high current battery test and when switching to isolation test to check what actually goes on in both cases. (If the meter does the same battery check when it is switched to isolation test that may be the reason for displaying low battery, still I don't think it's normal behaviour.)

How about an actual long term fault with the power switching in the 1587?

[TERRA Operative]

I'm not sure about lithium leakage, they do last a long time though.

Most of the time work paid for them, but I haven't had as much luck finding them in Japan as I did in Australia where they are everywhere on the shelves next to the alkalines.

[shakalnokturn]

Just out of curiosity, I didn't find anything that would explain my 1587's behaviour with this...

The test setup was a lab supply followed by 8.5mF decoupling (still too little, the high current test captures seem to show the decoupling capacitors discharge) close to the 1587 followed by a 0.1ohm resistor for current measurement before supplying the meter.

Scope captures read as 1mV=1mA.