Nixie Tube Fluke 8100B DMM with Video

[FlyingHacker]

Hi,

I have acquired a Fluke 8100B DMM. This is a Burroughs based Nixie Tube meter. Sadly, this unit will not power up.

FIXED: thanks to Tautech and Paul Moir for their incredible assistance. 

Here is a follow up video of it working... Lots of Nixie Tube Porn:

https://youtu.be/bOKkrFakrys

Manual and schematic here: http://meanmutha.com/fluke_8100b_digital_multimeter.pdf

I will post some pictures later. (Pics here now... see below)

I followed the procedure for checking voltages on page 37 of the PDF. On mains power I get ~35V out of the transformer (on either terminal of F2). I have no battery pack, but the unit DOES have the battery option installed. This 35V does seem a little high since it is 24V nominal.

On the -18V rail (case of Q1) I get around -1.7V. I followed the manual and disconnected the 15V side of R5 and hooked it up to a bench power supply. I still get the crazy low reading on the -18V rail.

The power supply schematic is page 89 of the PDF. (and posted below)

I disconnected from the mains and hooked up a bench supply to the battery terminals (as mentioned above this unit has the battery option, but I have no battery pack). I tried voltages from 24V (nominal) to 34V (what the transformer was putting out after rectification). I still get around -0.7 to -1.7V in this configuration.

I swapped out Q2 and Q3, which had no effect.

I measured R1, R2, R3, R4, R5, and R195. All are in spec.

I replaced C58 with a MLCC at 0.1uF (original measure OK too).

I replaced C3 and C5 (electrolytics, whose ESR was a tad higher than normal, but not terrible).

I have a replacement for Q1 on order. I took it out, and it tests OK on the hFE and "GM328 Transistor Tester," but could still be bad I suppose.

Note that testing with either the mains or the power supply attached to the battery terminals results in the meter (m1 on the schem) showing a "charge" up at the high level.

Is there anyone out there with a working 8100 who might be able to take some readings for comparison?

Thoughts?

Thanks.

[FlyingHacker]

See the attached image of the power supply.

Would it be safe to remove F2 and connect up a (current limited) -18V supply such that the negative terminal goes to the highlighted wire and the positive terminal goes to ground? I would disconnect R5 as shown by the blue "Cut." I already have this disconnected anyway in order to isolate the -18v supply (per the manual).

I would NOT have the power supply connected to the battery terminals, and obviously it would NOT be plugged in.

Thoughts? I am trying to see if I can get this thing to work at all.

Thanks.

EDIT: That ground marking would actually go to TP1, which the manual describes as circuit ground. So the + side of the power supply would go to TP1, and the negative side would go to the wire highlighted in yellow.

[FlyingHacker]

For those who do not want to dig through the PDF... Here is the power supply schematic.

[tautech]

Just a thought.....
IF the switching oscilator Q4 & Q5 was not working would the load of the primary of T102 drag the -18V to levels you are measuring? Check if Q4 & 5 are warm. 

You've drawn a Gnd symbol.... it doesn't match others on the schematic. 
Where are you referencing the -18V to?

[FlyingHacker]

Interesting. I will check Q4 and Q5.

I was referencing ground to TP1 in the manual (marked as circuit ground). I guess I drew and earth ground symbol.

I seems Q1 has an hFE of 33. I spoke with another owner and EEVBlog user Paul Moir in my other thread, and he said his Q1 had an hFE of 61.8  So possibly Q1 is bad after all.

I have one on order, but what do you think of hooking up a power supply as shown (with the "ground," which in this case is the + side of the 18V power supply, going to TP1, and the negative side of the 18V power supply going to the highlighted connection).

Thanks.

[tautech]

I have one on order, but what do you think of hooking up a power supply as shown (with the "ground," which in this case is the + side of the 18V power supply, going to TP1, and the negative side of the 18V power supply going to the highlighted connection).

Take note the -18V is referenced to +15V but the -18V rail has to power on first. Any excessive load on the -18V or faulty Q1 will prevent it firing up.

Any external power supply should be current limited to prevent additional damage. Even a 24V bulb in series will limit current draw. If it glows real bright, you still have a problem.

[FlyingHacker]

I have one on order, but what do you think of hooking up a power supply as shown (with the "ground," which in this case is the + side of the 18V power supply, going to TP1, and the negative side of the 18V power supply going to the highlighted connection).

Take note the -18V is referenced to +15V but the -18V rail has to power on first. Any excessive load on the -18V or faulty Q1 will prevent it firing up.

Any external power supply should be current limited to prevent additional damage. Even a 24V bulb in series will limit current draw. If it glows real bright, you still have a problem.

All my supplies are current limited. Learned a long time ago the value of those!

Does my wiring for the -18 supply look correct, though? Positive side to TP1 and negative side to the yellow highlighted wire (there is actually a jumper wire on the board). I would remove the fuse as well. Q1 is already out of the board (though R1 is still there, but I don't think this will matter as indicated). R5 is already disconnected as shown by the blue CUT line.

Am I correct in thinking that feeding in -18V should make everything work, assuming there are no other problems?

I would set the current limit around 1/4 amp or less and see if Q4 and Q5 get hot or what.

Thanks.

[tautech]

Am I correct in thinking that feeding in -18V should make everything work, assuming there are no other problems?

I would set the current limit around 1/4 amp or less and see if Q4 and Q5 get hot or what.

Check Q4 & 5 for heating when mains powered.
Check components associated with T102, FB windings, caps and flyback components for +200V.
The snubber, C2 and R155 if failed might have damaged other components too, check all this simple stuff.

Then it might be safe to proceed with an external supply, you don't want more damage.

[FlyingHacker]

Check Q4 & 5 for heating when mains powered.
Check components associated with T102, FB windings, caps and flyback components for +200V.
The snubber, C2 and R155 if failed might have damaged other components too, check all this simple stuff.

Forgive my ignorance... What is a "FB winding?" You mean check the resistance of the coils in the transformer? Use small words. I am a hobbyist ;-)

I will check the other components. I think I will hold off on checking the Q4 and Q5 until I get the new Q1, or try an external supply. I don't want to put Q1 back in again just to remove it. I already messed up some pads on Q2 and Q3. The initially came out easily, but then removing my replacements did not go well. Thanks.

[tautech]

Check Q4 & 5 for heating when mains powered.
Check components associated with T102, FB windings, caps and flyback components for +200V.
The snubber, C2 and R155 if failed might have damaged other components too, check all this simple stuff.

Forgive my ignorance... What is a "FB winding?" You mean check the resistance of the coils in the transformer? Use small words. I am a hobbyist ;-)

I will check the other components. I think I will hold off on checking the Q4 and Q5 until I get the new Q1, or try an external supply. I don't want to put Q1 back in again just to remove it. I already messed up some pads on Q2 and Q3. The initially came out easily, but then removing my replacements did not go well. Thanks.

Feedback.
T102 pins 6 & 7 is the feedback winding.
C68 and C4 couple the feedback to Q4 & 5, and set the oscillator frequency.
T102 is a switching transformer.

Component removal requires practice and certain methodology to minimise the risk of damage to PCB's.
Good sucker and a range of solders can help dilute the existing solder for easy removal of components, don't use too much heat and too cold will make things difficult too. Bigger tips can also be of assistance.
Use old PCB's and practice, practice and practice. 

If you've got any of that lead free muck solder, biff it as far as you can. 
Quality low temp solders are invaluable for rework.

[FlyingHacker]

Ok. Thanks. Pins 6 and 7 have 8.7 Ohms in circuit. Seems reasonable for the winding, no?

C2 and R55 seem ok. The resistor measure correctly, but I did not desolder C2, though. I don't have a replacement and it has rather short legs.

R6 and R7, and the diodes CR3,5,6, and CR45 look OK on the tester.

Dare I apply the -18V via external supply?


I only use lead solder. Specifically 60/40. I am open to suggestions on what works best for rework.

What temperature would you recommend for both the desoldering station, and the iron for this type of repair? The desoldering iron sucked most of it out, but then I ended up chewing up the pads trying to get the legs of Q2 and Q3 free with the iron. You know that last bit of solder that keeps this pins stuf when they were bent upon installation? I have the desoldering set for 625F and the iron was at 750F. I think the iron was too hot and that damaged the board.

Thanks for all the help.

[tautech]

Your call on external power......
Before you do, re-read this thread, study the manual, have a think, make a coffee, sleep on it. What's the hurry? 

For soldering temp, I only use adjustment settings as a guide, the rest is by feel. 
Fresh solder and then re-suck can do wonders, I often use this method if solder is hard/tricky to remove.

Bent component pins can be a curse, if I have to bend any to retain components it is done very slightly, just enough to retain components.

[FlyingHacker]

Some pictures...

[FlyingHacker]

Note in the second picture I am aiming a flashlight (torch) to light up the Nixies. They are not actually on... yet.

[FlyingHacker]

The insides... Wires are where I am contemplating applying -18V. Q1 is removed, as is F2.

[FlyingHacker]

Your call on external power......
Before you do, re-read this thread, study the manual, have a think, make a coffee, sleep on it. What's the hurry? 

I will check things over again. I do think it would be safer to apply power with a current limited supply then to turn the thing on again. I just want to make sure I am applying the power correctly. My diagram, assuming TP1 for the "ground" (positive side of the power supply) looks right, no? 

Do you think it is odd that I was getting 37V from the rectified mains? That seems awfully high. But I know line voltage used to be consistently closer to 115V here and my lab is 120V.

For soldering temp, I only use adjustment settings as a guide, the rest is by feel. 
Fresh solder and then re-suck can do wonders, I often use this method if solder is hard/tricky to remove.

Bent component pins can be a curse, if I have to bend any to retain components it is done very slightly, just enough to retain components.

Yeah, perhaps... Suck, Resolder with new solder, and suck again? I have done this and it does help. But those bent pins can be a real pain. Makes you think twice about bending them when you put stuff together. It is just the lazy way to get the stuff to stay in place while you solder it

[tautech]

I just want to make sure I am applying the power correctly. My diagram, assuming TP1 for the "ground" (positive side of the power supply) looks right, no? 

Do you think it is odd that I was getting 37V from the rectified mains? That seems awfully high. But I know line voltage used to be consistently closer to 115V here and my lab is 120V.

TP1 correct.

Unloaded I wouldn't worry, it's likely to be very different when operating correctly

Yeah, perhaps... Suck, Resolder with new solder, and suck again?

Exactly, even repeat at times for difficult cases.

[FlyingHacker]

I applied the -18V as described... Q1 still removed.

The current limiter kicked in, and I let it go up to a little over 200mA (do you think that is high enough? F2 is spec'd as 1/4 amp).

Q5 was starting to warm up. Not hot, but warm to the touch. Note that I did not leave it on very long. Q4 did not seem to be warming.

Whatever the cause of the load is it has pulled the voltage down to the -1.7V again. So I bet it is not Q1, but, as Tautech suggested, the load applied.

Q4 and Q5 are listed as 2N4946 transistors.  I found some on eBay, though in a slightly different package.

I guess I will pull Q4 and Q5 and compare their hFE measurements.

Any other thoughts? I am thinking going more current is probably not a good plan.

Thanks.

[FlyingHacker]

The pictures say it all (as Dave might say, "Oh wait! Here we go... Bingo!")

Q4 reads fine.... Q5 reads as a pair of diodes.

[Paul Moir]

Just pulled Q5 now.  Getting ~140 hFE.

Ic on the hFE tester is about 0.75ma.

So I guess the square window on the front is for the battery charge meter?  I always wondered why that was there!

[tautech]

Just pulled Q5 now.  Getting ~140 hFE.

Ic on the hFE tester is about 0.75ma.

So I guess the square window on the front is for the battery charge meter?  I always wondered why that was there!

You got one to fix too Paul?

[Paul Moir]

No I've got a junker and coincidentally the same multimeter with hFE meter as FlyingHacker.  The power of the internet eh?

Further coincident is that I just bought a 8120A to replace my old 8100B, which died about 25 years ago in the ADC end.  I'm waiting for it to arrive.

[FlyingHacker]

Just pulled Q5 now.  Getting ~140 hFE.

Ic on the hFE tester is about 0.75ma.

So I guess the square window on the front is for the battery charge meter?  I always wondered why that was there!

Odd... My "good" Q4 reads an HFE on the Micronta of 11.98 

The bad transistor will not even read, which I guess is a good sign that it is bad.

Yep, there is a little meter in that window. I kinda wish I had a battery pack for it. If I ever get it working I am tempted to make one, but at 21.6V that is 18 cells! Those NiCd batteries are kind of pricey at Mouser... $2.50USD each... So $45USD for a battery pack I would never use.. But it would be cool to carry around this "portable" like the guy in the picture in the manual. (Attached)

[FlyingHacker]

No I've got a junker and coincidentally the same multimeter with hFE meter as FlyingHacker.  The power of the internet eh?

Further coincident is that I just bought a 8120A to replace my old 8100B, which died about 25 years ago in the ADC end.  I'm waiting for it to arrive.

Just shows you have good taste 

At least the 8120A measures current. I mainly use my 8600A for current measurements. It may not be Nixies, but red LEDs, and no batteries to change makes it nice. I use my 8810A for voltage and ohm, or one of my Digital Precision 245s, or the Simpson if I want to watch it. Something nice about numbers that glow!

That Micronta was pretty slick when I bought it 25 years ago! It was the most expensive one at Radio Shack. Of course, even then Radio Shack was already going downhill mainly selling rechargeable NiCds for exorbitant prices.

[FlyingHacker]

So, given my bad transistor. Do you more experienced folks think that that the bad transistor is the disease, or a symptom of some other problem?