Handheld meter robustness testing

[joeqsmith]

Thanks for the offer but I really don't have a use for it.   I would guess there are people on this site that would really appreciate having some of the parts off of it or even the whole meter. 

[mzacharias]

OK.

[joeqsmith]

I am trying to determine why the older Fluke 87V failed at such a low voltage.  With most meters it's pretty easy to see why they fail.  In many cases where meters were damaged at lower levels, I have shown how some like the 121GW, UT181A, UT61E and even the cheap ZT102 could be improved.  We even tackled a cheap analog meter for the fun of it.    In the case of the older Fluke 87V, it has not been so simple.  I did record all of the testing and plan to post it. 

[joeqsmith]

Running tests on the older Fluke 87V. 

https://youtu.be/GhslaLKOpKE

[kcbrown]

Very interesting outcome.  Makes it clear that it’s not a problem with the fundamental design. I think the previous failure was just a fluke.

I’ll be here all night.   Try the veal.



Sent from my iPhone using Tapatalk

[joeqsmith]

Very interesting outcome.  Makes it clear that it’s not a problem with the fundamental design. I think the previous failure was just a fluke.

I no longer believe this was a design problem or a component problem.  I also don't believe it was a quality problem on Fluke's part.  This meter was fairly old and I wonder if something conductive had come off at some point.  Even if this were the case, I don't see where it would have been able to arc and cause this sort of damage.   I monitor pretty much each transient during the tests to see if something starts to breakdown.  Even if the transient generator has some major fault and I missed seeing it with the scope,  it's not capable of putting out more than what that last test I ran.   

[ChrisLX200]

The fact it was an old meter with no dependable record of its history leaves open the possibility those components had already been stressed to the point of failing prior to the test being run. I suspect the result using the repaired meter (with new diodes) is more typical of what you may expect.

The fascinating thing to me is the apparent ease with which you manage to repair a blown up meter

[joeqsmith]

The fact it was an old meter with no dependable record of its history leaves open the possibility those components had already been stressed to the point of failing prior to the test being run. I suspect the result using the repaired meter (with new diodes) is more typical of what you may expect.

The fascinating thing to me is the apparent ease with which you manage to repair a blown up meter

You make a very good point and while I have had several people offer to provide me with older meters or request me to run them, I typically decline mainly for that reason.  In this case, I personally know the person who bought the meter brand new and know they do not work on line voltages, normally.    It was a gamble to run it and it seems in this case, it corrupted the data.  Like you, I do not believe a new 87V would ever fail like this after seeing how the meter is designed and with all of the tests I ran.   

So yes, I am fully willing to eat crow over this particular 87V now.  As far as electrical robustness, I now have absolutely nothing bad to say about any of the Fluke meters I have looked at.  From a mechanical standpoint, I don't like the 87V switch wear but it was making good contact and still better than most of the meters I have looked at.   

They are old designs for the most part and they don't really make a meter that I like or would use.  Maybe one day they will bring back a modern version of the 189.  That IMO, was just a nice clean design.   Not a lot of fluff and has all the basic features.     

[joeqsmith]

Today, after almost three years, we set a new milestone with this thread.  The admins were called in to censor a few posts.  To try and avoid future problems, I have updated the first post to include some basic rules and guidelines to follow if you plan to post on my YT channel.

[ChrisLX200]

...
 From a mechanical standpoint, I don't like the 87V switch wear but it was making good contact and still better than most of the meters I have looked at.   
...

As an 87V owner I was also surprised and concerned by the awful rough movement developing in the rotary switch, although it's difficult to make sense of what is happening (how it 'feels') based on the sound as reproduced on the video. I gave it some thought and suspect it is the result of two things: the spring contacts being slightly 'loose' in their housing (by 'loose' I mean the tension holding it in place may be lower than optimal), and secondly the fact you use a constant speed when turning the switch on your jig. I would suggest the latter, in combination with the less than secure spring retention, allowed a sympathetic chattering to develop.

Chattering will be dependant on a number of factors: the amount of friction between spring contact face with plated track, the surface area of the contact face (which will increase as small flats are worn into them as they bed in), and the base spring tension pressing the two components together (oh, and speed!). If the spring retention mechanism allows enough flexibility such that friction overcomes its ability to slide smoothly (and thus sticks for a fraction of a second) then you will get sympathetic chattering, however this will be exacerbated by using a constant speed allowing the pattern to be imprinted into the track itself. Once established there will be positive feedback and things will only get worse as the procedure continues. Note that a lubricated track may have reduced the friction to the point where sticking would not occur - but all the other factors need to be in place before this would become a critical factor.

(There are similarities with using a parting tool in the lathe, the factors that cause chattering are similar, and if nothing can be done about the stiffness of tool mounting or play in headstock bearings then adding lubrication and/or adjusting cutting speed often resolves the problem)

Such perfectly constant speed provided by your test jig is not something that is seen in real life when simply turning the switch manually, so the chances of developing such a pattern of chattering will therefore be reduced. I do understand your idea about reducing the possibility of heat being generated due to  turning the switch too rapidly but I also think you should vary the speed too.

It may be too late for that test example now (given the state of the tracks), but perhaps re-tensioning the spring contacts so they are held in place more firmly,  a few seconds on a buffing wheel to round off any sharp flats formed on the spring contact faces, and perhaps a little lubrication would restore proper function.

[rsjsouza]

Today, after almost three years, we set a new milestone with this thread.  The admins were called in to censor a few posts.  To try and avoid future problems, I have updated the first post to include some basic rules and guidelines to follow if you plan to post on my YT channel.

Wow, that is quite odd. The discussion was quite civil in my opinion. Perhaps there is no desire to criss cross feuds happening in another platform. Oh well... Back to your normal programming I guess.

[joeqsmith]

Today, after almost three years, we set a new milestone with this thread.  The admins were called in to censor a few posts.  To try and avoid future problems, I have updated the first post to include some basic rules and guidelines to follow if you plan to post on my YT channel.

Wow, that is quite odd. The discussion was quite civil in my opinion. Perhaps there is no desire to criss cross feuds happening in another platform. Oh well... Back to your normal programming I guess.

I was surprised myself and did not think I posted anything that would warrant having the posts pulled.  I didn't even consider it a feud of any sort.  I suspect they just did not like being called out for advertising on my channel and posting misinformation.   Read the rules and if you can't follow them, there is no need to tell me you are unsubscribing over them.  I won't care.  Just unsubscribe and move on.

[joeqsmith]

...
 From a mechanical standpoint, I don't like the 87V switch wear but it was making good contact and still better than most of the meters I have looked at.   
...

As an 87V owner I was also surprised and concerned by the awful rough movement developing in the rotary switch, although it's difficult to make sense of what is happening (how it 'feels') based on the sound as reproduced on the video. I gave it some thought and suspect it is the result of two things: the spring contacts being slightly 'loose' in their housing (by 'loose' I mean the tension holding it in place may be lower than optimal), and secondly the fact you use a constant speed when turning the switch on your jig. I would suggest the latter, in combination with the less than secure spring retention, allowed a sympathetic chattering to develop.

Chattering will be dependant on a number of factors: the amount of friction between spring contact face with plated track, the surface area of the contact face (which will increase as small flats are worn into them as they bed in), and the base spring tension pressing the two components together (oh, and speed!). If the spring retention mechanism allows enough flexibility such that friction overcomes its ability to slide smoothly (and thus sticks for a fraction of a second) then you will get sympathetic chattering, however this will be exacerbated by using a constant speed allowing the pattern to be imprinted into the track itself. Once established there will be positive feedback and things will only get worse as the procedure continues. Note that a lubricated track may have reduced the friction to the point where sticking would not occur - but all the other factors need to be in place before this would become a critical factor.

(There are similarities with using a parting tool in the lathe, the factors that cause chattering are similar, and if nothing can be done about the stiffness of tool mounting or play in headstock bearings then adding lubrication and/or adjusting cutting speed often resolves the problem)

Such perfectly constant speed provided by your test jig is not something that is seen in real life when simply turning the switch manually, so the chances of developing such a pattern of chattering will therefore be reduced. I do understand your idea about reducing the possibility of heat being generated due to  turning the switch too rapidly but I also think you should vary the speed too.

It may be too late for that test example now (given the state of the tracks), but perhaps re-tensioning the spring contacts so they are held in place more firmly,  a few seconds on a buffing wheel to round off any sharp flats formed on the spring contact faces, and perhaps a little lubrication would restore proper function.

Fact
 noun
1. something that actually exists; reality; truth:
Your fears have no basis in fact.
2. something known to exist or to have happened:
Space travel is now a fact.
3. a truth known by actual experience or observation; something known to be true:
Scientists gather facts about plant growth.
...

[Fungus]

It might be worth investigating where the dial noise comes from.

eg. Put some very viscous grease in the white switch dial thing to dampen any movement of the contact springs where they touch it.

This would let you know if the noise comes from PCB scraping or rattling of the springs in the housing.

[joeqsmith]

It might be worth investigating where the dial noise comes from.

eg. Put some very viscous grease in the white switch dial thing to dampen any movement of the contact springs where they touch it.

This would let you know if the noise comes from PCB scraping or rattling of the springs in the housing.

If you have not taken the time to watch the video, I added a couple of pictures that may interest you.  It's really up to Fluke to determine if they consider it an issue and how they want to address it.

[joeqsmith]

After doing some maintenance on the generators,  played with a few light bulbs for fun.  Funny to see how the voltage levels that damaged so many cheap meters effects a simple light bulb.   

https://youtu.be/aaZxWMgOz70

[joeqsmith]

Viewer posted:

chrimony
4 hours ago
@6:10 Surprised to see the plasma outside the bulb.?

I responded that I suspect this was from the arc that occurred at the base of the bulb and those gases escaping out of the socket.  With a 1ms frame rate, the camera is just not fast enough to tell the whole picture.

Here are the 5ms of footage.

[joeqsmith]

Getting ready to look at another meter.  I've been busy repairing an old scope that a friend of mine gave me that happens to be just like one I had given away a few years ago.   I have finished it up but have some cleanup to do along with some other spring time work that needs to get done but I hope to get started this weekend.

Again, I want to than those of you who have offered to pitch in and buy me this meter to run, or out right offered to pay for it.  I appreciate the gesture.  Maybe one day I will open some sort of account to allow people to donate.   For now, consider the small bit of data I supply as my way of giving back to the technical community.

Yes, the meter is here..  It looks scared.. 

[tautech]

Getting ready to look at another meter.  I've been busy repairing an old scope that a friend of mine gave me that happens to be just like one I had given away a few years ago.   I have finished it up but have some cleanup to do along with some other spring time work that needs to get done but I hope to get started this weekend.

Again, I want to than those of you who have offered to pitch in and buy me this meter to run, or out right offered to pay for it.  I appreciate the gesture.  Maybe one day I will open some sort of account to allow people to donate.   For now, consider the small bit of data I supply as my way of giving back to the technical community.

Yes, the meter is here..  It looks scared..

That's the same one my sparky mate uses for pretty much everything domestic and industrial.
Will be very interested to see how it fares. 
TIA Joe.

[joeqsmith]

Poor Flukes.  Like any other meter, they always have to earn my respect.   

[MacMeter]

Looks like it tried to escape you, LOCKED in the Panavise now!

[joeqsmith]

There may be a part 2 to talk about the sensor's design.   Enjoy.

https://youtu.be/k8hhtTtWfVc

[MacMeter]

Break down & repair, very detailed as usual. Great job Joe!

[IanB]

There may be a part 2 to talk about the sensor's design.   Enjoy.

That was remarkably accurate. It mostly got the voltage correct to ±1 volt with no direct contact. I'm impressed.

[joeqsmith]

Even with me being fairly insulated I was amazed how well it works.   

I soldered some test points the to the board and started to poke around it.  They seem to use a 2.4KHz sinewave for their reference.  I have not done a lot of tests but it appears to always be running (while in the non-contact mode).   This make some sense as they need to actively measure the capacitance.   The capacitance from a wire to the sensor is in the 10s of pf.  Looks like maybe two plates, one with a lower capacitance.   

They must have to work with uninsulated wires as the area near the jaw is very thick.  Moving a wire inside the saw area with the cover removed, it is really sensitive.   

With the low freq. reference and working in the few megs of coupling, it may explain why the unit has troubles with high frequency signals.  I wonder if they would be better off making the final measurement with the reference disabled.  Talking out of my ass.  I really have no idea how it even works yet....