Fluke 87V adversely affected by cool temperatures

[CB350F]

Hello All,

I have a Fluke 87V that is misbehaving.  If the temperature is above 70 Fahrenheit, the meter starts in a second or less.  By start, I mean the "F87" is displayed and the meter transitions to the proper mode.  If the temperature is around 65 Fahrenheit, it takes ~5 seconds for the meter to start.  If the temperature is below 60 Fahrenheit, the meter will usually not start at all.  In all circumstances, after the meter has successfully started, I can turn it off for a few seconds and back on and it will run fine.  If I leave the meter off for 30 seconds, it will be slow to start again (if the temperature is low).   When it is running, the meter is stable and all functions appear to work fine. 

I'd appreciate any thoughts on what the issue might be, or how to fix.

Thanks!

[mzacharias]

The battery is in good shape?

[CB350F]

yes, brand new and tested battery. 

[coromonadalix]

some schematics for the 83 and 87 :
http://www.schematicsunlimited.com/f/fluke/fluke-83-85-87-500-service-manual

from fluke

http://assets.fluke.com/manuals/83_85_87smeng0500.pdf

[retiredcaps]

If the temperature is below 60 Fahrenheit, the meter will usually not start at all.

60F is 15.5C.  While my house isn't that cold during winter, I do keep the thermostat around 17C and never had a problem with my 87V starting.

[retiredcaps]

yes, brand new and tested battery.

What is the 9V battery voltage (unloaded) as tested by another multimeter?  What brand?

Take apart the 87V and post some clear focused pictures of the pcb around (both sides) and around the 9V battery snap.

[retiredcaps]

some schematics for the 83 and 87 :

The manuals above are for the Fluke 80 series I, not the V.  While it may be helpful, there are significant changes between the 2 series.

[Electro Detective]

Sounds like a dud battery, loose battery connections, crusty green crap on the battery terminals and trailing along the cable to the board.. or all three!   



or it's a MIC TooHungLow 87C    

[CB350F]

It is an Amazon Basics alkaline battery.  It measures 9.55 Volts.  I have also attached the 87v to a power power supply at 9V and the problem persisted. 

I’ll post some pics of the pcb. 

[CB350F]

Here are pics of the PCB.  I couldn't see any obvious issues like burnt traces/components, or bad solder joints.

[CB350F]

Few more pics.

[CB350F]

Here are three more pics. 
 - The first shows the battery voltage. 
 - The second show the current draw of the 87v. 
 - The third is a shot of the battery connector.

[CB350F]

some schematics for the 83 and 87 :
http://www.schematicsunlimited.com/f/fluke/fluke-83-85-87-500-service-manual

from fluke

http://assets.fluke.com/manuals/83_85_87smeng0500.pdf

Unfortunately my meter is a newer model.  The PCB does not match what is shown in those manuals.

I appreciate the reply!

[CB350F]

yes, brand new and tested battery.

What is the 9V battery voltage (unloaded) as tested by another multimeter?  What brand?

Take apart the 87V and post some clear focused pictures of the pcb around (both sides) and around the 9V battery snap.

Thanks for the reply.  I added some pics on another post.

[floobydust]

I've seen a similar problem in other gear, the cause was the crystal oscillator circuit.
PCB corrosion/flux residue, bad crystal/meter dropped, cracked loading caps etc.

I would pull the DMM out of the fridge and warm up the crystal and see if it starts. Or a few taps.

Y1 131.072kHz crystal 3x8mm, Fluke 831453. Might be an Epson tuning fork crystal, C-2 131.0720K-P but size is 2x6mm.
Any lettering on it might help.

[CB350F]

If the temperature is below 60 Fahrenheit, the meter will usually not start at all.

60F is 15.5C.  While my house isn't that cold during winter, I do keep the thermostat around 17C and never had a problem with my 87V starting.

I have other Flukes (not an 87v, only have one of those) and they all start fine in cooler/cold temperatures.

[CB350F]

I've seen a similar problem in other gear, the cause was the crystal oscillator circuit.
PCB corrosion/flux residue, bad crystal/meter dropped, cracked loading caps etc.

I would pull the DMM out of the fridge and warm up the crystal and see if it starts. Or a few taps.

Y1 131.072kHz crystal 3x8mm, Fluke 831453. Might be an Epson tuning fork crystal, C-2 131.0720K-P but size is 2x6mm.
Any lettering on it might help.

This sounds like a possibility.  I think the crystal is under the display, not sure if I can get to it without taking the display off.  Maybe I'll have to apply some localized heat on the bottom of the PCB directly below the crystal.

Really appreciate the assistance!

[retiredcaps]

I have also attached the 87v to a power power supply at 9V and the problem persisted. 

How?  Did try connecting a power supply directly to the pcb?  Thus, in the process of bypassing the 9V battery snap cable?

PS. I have an original 87 series 1 and I know it works in -20C weather when I was troubleshooting my car battery a few years ago.

[Electro Detective]

Have you tried some different brand batteries ? 

Nowadays you can't trust any brand at any price to do something simple..like just work as advertised at most temperatures,
from normal room temp to an extreme that a human with warm clothing, sheepskin boots and gloves can tolerate!   

[CB350F]

I've seen a similar problem in other gear, the cause was the crystal oscillator circuit.
PCB corrosion/flux residue, bad crystal/meter dropped, cracked loading caps etc.

I would pull the DMM out of the fridge and warm up the crystal and see if it starts. Or a few taps.

Y1 131.072kHz crystal 3x8mm, Fluke 831453. Might be an Epson tuning fork crystal, C-2 131.0720K-P but size is 2x6mm.
Any lettering on it might help.

This sounds like a possibility.  I think the crystal is under the display, not sure if I can get to it without taking the display off.  Maybe I'll have to apply some localized heat on the bottom of the PCB directly below the crystal.

Really appreciate the assistance!

I did some experiments to try to isolate the component that is causing the problem.  Cooling the crystal seem to have no affect.  However, cooling the microprocessor consistently caused the issue.  I used an ice cube in a zip lock bag and carefully held it to the chip without touching any other components.  Ever heard of a chip or microprocessor exhibiting this sort of behavior?

[CB350F]

Have you tried some different brand batteries ? 

Nowadays you can't trust any brand at any price to do something simple..like just work as advertised at most temperatures,
from normal room temp to an extreme that a human with warm clothing, sheepskin boots and gloves can tolerate!   

It had a RayoVac in it that tested OK, but I replaced it with the new Amazon battery. 

[vtp]

However, cooling the microprocessor consistently caused the issue.  I used an ice cube in a zip lock bag and carefully held it to the chip without touching any other components.  Ever heard of a chip or microprocessor exhibiting this sort of behavior?

Yep, TI MSP430 and its LFXTAL oscillator. Seen that happen lately in other MSP430 based products too. Very slow to start at low temperatures, slow enough to cause problems.

Even as normal it is not that fast to start, see datasheet for maximum startup time, for one version they stated max 5 seconds.

[CB350F]

However, cooling the microprocessor consistently caused the issue.  I used an ice cube in a zip lock bag and carefully held it to the chip without touching any other components.  Ever heard of a chip or microprocessor exhibiting this sort of behavior?

Yep, TI MSP430 and its LFXTAL oscillator. Seen that happen lately in other MSP430 based products too. Very slow to start at low temperatures, slow enough to cause problems.

Even as normal it is not that fast to start, see datasheet for maximum startup time, for one version they stated max 5 seconds.

Thanks for the info!  Guess I need to decide if I want to try to replace the chip.

[Rabid Badger]

Did you buy the meter new? An issue like that should be covered under warranty.

[floobydust]

TI says the start-up time could be several seconds; a firmware fix of toggling the high speed XT osc. setting on and off:
MSP430 crystal osc app note
It could even be old Fluke firmware?

I'd try a different loading cap first, easiest to try.
MSP430F448 datasheets