Fluke 845AR rejuvenation...neons, contacts, etc...

[valley001]

Hi all,

I picked up one of these fairly cheap as a defective unit.  Opening the unit revealed two things: 1) Old, leaking electrolytic caps, and 2) flickering neon bulbs in the chopper circuit. 

I managed to replace the neons with NE2U tubes, no fancy aging or testing for ignition voltage etc (though I may need to revisit this).  I also replaced the worst offending caps (10uf, originals were long gone and leaking electrolyte), I am waiting on the remainder from mouser. 

My unit was stable for a while and then began to peg in the 3uv and 1uv range in operate mode, in zero mode it would still zero.  I opened the unit to discover flickering neons (the new ones  ).  But, the reason for my pegging turned out to be the operate/zero switch.  Cleaning the operate contacts allowed the unit once again achieve zero in operate mode in the 10, 3 and 1uv ranges.   I imagine the range switch contacts are another source of potential problems...

My question now is, do these neon based choppers tolerate some flicker as part of the design?  Has anyone opened their functional unit to discover one of the neons flickering?   

[valley001]

I am experimenting with the neons I purchased, NE2U advertised as "Memotronics" by Amazon.  My setup is originally for fixing cathode poisoning on nixie tubes but adapted for firing neons by installing a 30K current limiting resistor in series (I used 10K for nixies). 

These neons seem to fire around 90v or so, fairly constantly.  The problem I see with them is only the top half of the cathode is illuminated.  I can drive them all the way to rated 1.8ma and still only the top half is lit.  I imaging this may be problem for the 845AR...I will need to see if they behave the same way inside the unit.

[Dr. Frank]

These neons are obviously defective, and the half lit bar is definitely a problem for the light guiding rod.
Have you checked the Dark Effect? There's nothing fancy about that, only to check if these are real NE2U with reduced Dark Effect, otherwise that's another severe problem for the chopper.
Flickering neons already indicate a big problem with your purchase, but if you'd have proven good NE2U, then this can be caused by other components in the 84Hz oscillator circuit.
Frank 

[valley001]

Thank you for your input.  I did check and adjust the 84hz oscillator with my oscilloscope.  I could not get it spot on 84hz, it would drift to either side a bit.  I was going to check this again after replacing the rest of the electrolytic caps.  Should this be a rock solid 84hz?

I tested the original lamps in my setup and the entire bar would illuminate on one side, the other (I assume the "used" side) would not illuminate evenly.  They would not flicker unless I really backed off the voltage and seemed happy to draw 2.5ma or, so firing at 110v give or take.  I re installed these in the 845 rotated 180 deg and they still flicker.  They are not darkened like yours in the other large 845AR thread, but they do flicker. 

Perhaps I will buy many different types of neons and see if I can find some that are more suitable.   

These neons are obviously defective, and the half lit bar is definitely a problem for the light guiding rod.
Have you checked the Dark Effect? There's nothing fancy about that, only to check if these are real NE2U with reduced Dark Effect, otherwise that's another severe problem for the chopper.
Flickering neons already indicate a big problem with your purchase, but if you'd have proven good NE2U, then this can be caused by other components in the 84Hz oscillator circuit.
Frank

[Dr. Frank]

The oscillator frequency is not that important, and will never be that stable.
It's trimmed only to 84 Hz to suppress 50 and 60 Hz, including any mixing frequencies, probably.

If one bar behaves differently than the other, that may indicate that you received used bulbs, or a defect... maybe NE2U fakes, I don't believe it..

Anyhow, the proper hit for good neons is the culprit to get the 845 working again, I only had a bit of chance and branadic in front to find cheap unknown neons w/o Dark Effect.

Frank

[valley001]

It was the original old wore out bulbs that behaved differently from one bar to the other. 

I see mouser has a fairly large selection of neon lamps, if I can not get the thing running on these NE-2U lamps after I replace the rest of the suspect capacitors I may order a selection of lamps and experiment.  When I researched the different designations of these small neon lamps I noticed others that are specified for reduced dark effect, like the NE-2H for example. 

Edit- Looking closely at these bulbs, I see a portion of the cathode that is brighter/shinier than the rest of the bar.  This is the portion that is illuminated in my picture above.  Looking at branadic's bulbs from post #167 in the other 845 thread, my new NE-2U look the same as the two on the far left.  I note that the originals (supposedly also NE-2U) the bars are bright/shiny the whole length, as are the  Barthelme 82305.  I wonder if there is something to this.  (That thread below for reference)   

https://www.eevblog.com/forum/metrology/teardown-fluke-845aabar-tweaks-and-mods-(and-repairs)/150/

The oscillator frequency is not that important, and will never be that stable.
It's trimmed only to 84 Hz to suppress 50 and 60 Hz, including any mixing frequencies, probably.

If one bar behaves differently than the other, that may indicate that you received used bulbs, or a defect... maybe NE2U fakes, I don't believe it..

Anyhow, the proper hit for good neons is the culprit to get the 845 working again, I only had a bit of chance and branadic in front to find cheap unknown neons w/o Dark Effect.

Frank

[valley001]

OK-

1) I have replaced all of the electrolytic caps with quality name brand caps.  I was surprised to find that many of the originals tested up to double their rated capacitance, and others that were leaking tested fine?  This from one of those component testes so perhaps not reliable measurements. 

2) I now have more neon lamps here than I care to admit.  I managed to find some that with no dark effect (tested ignition voltage same between light and dark), fairly low and constant ignition voltage, and they illuminate the full cathode.
 

Results are encouraging.  The needle has become more "quiet" though I do see it drifting within a 120-150nv range give or take (1uv range setting on zero function).  This data taken by watching the needle over a short period of time.  I do not have data logging capabilities yet. 

[Dr. Frank]

Well done, I think, as these old electrolytic capacitors also create a lot of problems.
This drift you see may arise from thermo voltages, so you have to let the instrument stabilize for several hours before making any stability measurements.
Even then, the 845A is known to drift in the ballpark of about 100nV/hr, as everybody else also has found, see the other thread, please.
Frank