121GW Range Switch Cycle Testing
Вставка
- Опубліковано 8 жов 2024
- Full testing of the 121GW multimeter using the new shim solution.
Inspections after 1000, 11000, 25000, and 50,000 cycles.
Sorry about ad-hoc nature of the edit, this was pieced together from lots of random footage.
UPDATE: We completely screwed the pooch! We installed the shim upside down! There was supposed to be contact friction between the shim and the plastic case (which created all the dust). Apparently we were too dumb to read UEI's instructions precisely. Doh.
UPDATE: We completely screwed the pooch! We installed the shim upside down! There was supposed to be contact friction between the shim and the plastic case (which created all the dust). Apparently we were too dumb to read UEI's instructions precisely. Doh.
EEVblog2 haha nice :D
// Start again after RTFM
1
2
3
4
5
6
7
Sometimes you just have to RTFM! I suppose they should have included *pictures for you Dave!* ;-)
ForViewingOnly - but you can see how a simple picture would have helped! If it isn't obvious then the manufacture should have shown a picture or two. Not just for Dave but this fix is likely going to be sent out to all the backers that have already been sent their meters so including pictures will *help* prevent others from doing the same mistake.
Hey Dave! Perhaps add a small magnet to the rotating bit so it collects the presumably ferrous dust and keeps it from accumulating near any of the high voltage stuff
Dave, I've have to know. Did you name this meter after the 1.21Gigawatts scene from Back to the Future? I'm sure you've said it in one of your many videos about it, but I admittedly haven't watched them all.
Why the solder mask on the second inner ring at 20:08 that connects the two contact patches? If there is supposed to be a disconnect at that point they really should have connected the patches with a trace that is out of the contact path.
I agree- and all of the plate-throughs should be out of the wiper path as well.... I can see how *I* would fix that layout!
I hope that in the long run you will either offer a replacement plastic contact retainer (due to the plastic wear) or actually plate the pivot hole in the PCB. I understand that because of normal wear and tear, the replacement plastic would not normally be covered under warranty, but still having the part available is a good idea.
The case is probably glass bead reinforced, so yes the dust would be glass, I suggest a nylon (PA66) shim
A bit of silicone grease should help reduce the amount of abrasive dust that spews out onto the contact area and accelerates the wear
Yeah, grease reduces friction massively and that will reduce micro fretting and the amount of accumulated crap.
Will it not just turn into an abrasive goop?
If it never grinds there won't be any abrasive, but either way it will be squeegeed to the side. I rather have pristine contacts and change grease every 10 years than let it grind with no maintenance and possible faults. The contacts in my ignition are greased, and I know that because my car got recalled after 27 years because it catches fire when it dries out.
10 years?!?! I throw away one of these per year... Not because they are poorly made or designed, but because they get damaged in so many ways...
Automate also inserting-pulling out probe plugs
Having tracks with solder mask under the contacts is just a dumb design! I have had multi-meters fail because of the solder mask failing and the contacts shorting between wrong tracks!
You also need grease in some particular places to prevent extra dust and such.
Hmm... moving the holes by the contacts so they are not as close to the wipers on the pcb might reduce wear over time a bit.
im feeling that strange sense of deja vu!...
how about a peice of felt instead of the fibreglass shim? would stop that dust
David thank you for your Meter Switch testing update.
I would be very interested if you opt to use grease on the shim and if so the type of grease.
I dont like, how this knob is mounted on the PCB... using a sharp hole to pivot a soft plastic knob, without any axis or somewhat a support, it's gonna wear (but not within 50000 cycles, obviously :) - and the mess with this plastic particles inside is gonna be worse
The plastic part with the contacts is supported by the brass shaft on the knob which is in turn is supported by the hole in the front half of the casing. IIRC most flukes do it the exact same way, only difference is that the hole in the PCB is copper through-plated on the Flukes rather than raw FR4 - Dave makes mention that it could be copper plated.
19:50 - I am really curious, what is causing the resistance spikes? Seems strange and is about 5,000 cycles between which makes me wonder even more.
@EEVBLOG2 why some of the contacts have 1 nipple and some of them have 2?
What about using XX phenolic in lieu of G10 plate? Would be much easier on the items in question and the wear dust would be paper. CE phenolic would be a second guess but then again, you are getting into a cloth. Paper phenolic has an added advantage of being self cleaning.
Very nice. I was around for the livestream to see the 10K test in action.
hi Dave,
watching this and previous videos with that range switching tests i am wondering if it would be an issue, when there are vias very close to the range switching traces?
i mean, when the plastic of the range switching nobe is worn, the contact may be not exactly centered anymore. that could cause that the contact points "fall" to the via and the contact is weak.
Mechanical bodge. Plastic washer would be nice, but producing the mold for it would probably cost the same as the new mold for the switch itself. So there is a fiber glass part that is easy to machine but way too abrasive.
Love those keysight meters!
You're going to want to use a more lubricious material for that shim. Fiberglass and some phenolics have little bits of glass in them, which will be in any dust or shavings from them.
What's the point of the fiberglass shim? Just for the testing, or will also be present on production?
To prevent a problem with intermittent contacts caused by a thinner production main PCB.
Can you do a video about how you designed this switch and got it manufactured?
Edit: I assume that's also custom plastic around the banana plugs.
Does the stepper-motor apply any force? When people turn a knob they don't just lightly grip it and apply force at 90º.
Hi Dave,
use Delrin and a delrin bushing
Hi Dave! I was making my own circuit with a knob. For the contact which kind of PCB finishing is it? ENIG?
Hey Dave, is the model number "121GW" a refference to 1.21 Gigawatt ? :-D
12:12 Use vias and route that bit under in the next revision?
Is it possible to build range switch, that instead of actual contacts, you have smd reed switches on board, and range switch has magnets instead of wiper contacts? Wouldn't that be virtually no wear out contacts?
can't the big hole in for the switch be thruplated? That way the plastic would rub on smooth gold and not actual fiberglass. I think that would extend the plastic lifespan quite a bit!
did the diodetester work properly after the last cleanup @51k ?
when this will be available on the market? thanks
that is 1 hell of a switch
I guess this means that every 3 years you have to take the thing apart and clean off the switch contacts to avoid the thing from shorting out.
Can you use any special washers in the center shaft since it seems that the PCB material is eating away at the plastic and the plastic clips?
Is the 121GW for sale anywhere ? I can't seem to find it.
Well you def proved your design, I have an older DVOM that is at least 10 years old and it only has roughly 3k ish cycles. That would be a guess, and I used it a lot lol. And anyone down brlow in the comments not likeing some part of it just is not thinking clearly, 50k+ cycles, thats easily 15 years of service with no lube.......
Thanks for sharing, a bit like Joe Smith :-)
Even if the life of that rotary switch is 51000 times, then if we use this meter 10 times a day and everyday then it should last about 14 years!!!!
Would software de-bouncing help with that wiggling switch issue?
maybe. Problem is that the range switch interfaces directly with the HY3131 multimeter chip, not the STM32 micro. Any 'software fix' implemented by the STM32 would likely just be a kludge that ignores fast/erratic range changes, also slowing down the responsiveness of perfectly functioning meters. The range switch also provides some basic analogue signal switching functionality on the frontend, so if the contacts are going high impedance because they are worn, it will also degrade some of the analogue functionality of the meter which is invisible to software.
can you use a delrin shim?i think it would hold up
Conclusions: circular contacts are not the best, it makes sense to use contactless sensors with the combination of relays, or at least sealed magnet switches, also does it have L measuring circuit ?
i dont think it does mate, along with the usual functions it has C and i think temp and hz, but for a hobby/semi professional meter i think L is more useful than temp, IMHO.
but then i wind a lot of coils and inductors.
Well, for its price it is good, but there is a possibility to pack much more in that case, with the respective price - and L is actually very useful (as you can't use resistors for high efficient circuits), temperature can be measured by implementing a powerful MCU with loads of memory and sensor profiles.
Thats some pretty brutal testing, I'm amazed how well it held up! Are these available to purchase yet, Dave?
Not yet, no. Need to fulfill all the Kickstarter orders first.
No worries, I'm definitely having one when you've got stock :D All the best from the UK ;)
Keep in mind that dust might get nocked out with use since it's moved during use.
How about a UHMW shim --
UHMW? what's that?
Ultra High Molecular Weight (plastic). It's the standard slippery plastic used for sliders.
Machining a thin shim like that is going to be a pain, it would cost 10x+ as much as the FR4 disk. If you could find a way to eliminate the inner features it would be possible though, eg a circle of UHMW tape would work.
you just punch the shim from a sheet, super easy
UHMW... PE.
Very long linear polyethylene. Like Nylon but has even better propertys here and there.
dave, how can i get one of this? ty
wow, I would have expected the contacts to be worn out after that
Bit of dry lube should last ages too
I don't like the fibreglass shim.
If you're going ahead with that the testing should be done with the meter in the bench position like on its feet.
Could be that more glass dust made it in then to ruin your plating.
Ok, so not alot of dust is from the fibreglass, you still only need a stray hair of glass from a edge to wreak havoc.
How's the hp all in one doing?
do the 51k cycles, without opening it up once, let all that dirt and debris get really gunked in there, and effect it as much as possible...
If it was possible to make a multimeter with reed switches and magnet in the knob there was selecting them, it would live very long. I guess magnet distance would be the problem and maybe outside magnets could jam it!. :-)
Also much more expensive.
Since it's all fancy and Bluetooth, there wont be an app in 5 years time. Sure, the protocol is open source, but I don't think the switches lifetime will have any effect whatsoever. The thing will become obsolete way before the switch finally breaks down for good.
PTFE?
1 plus 5 plus 5 is 9! dave math!
Dear Lord: Please don't let Dave take on the Flat-Earthers with hours of squeaky diatribes.
Go until it breaks, 1-10M.
Dave, unrelated but ... news just hit me on Intel's fubs. Any thoughts on that one? Here is link ... www.zdnet.com/article/why-intel-x86-must-die-our-cloud-centric-future-depends-on-open-source-chips-meltdown/
Hey Dave! Perhaps add a small magnet to the rotating bit so it collects the presumably ferrous dust and keeps it from accumulating near any of the high voltage stuff.
I am losing confidence in buying your meter. I am not trying to be mean, just wanted to let you know.
mini hop Based on other meter selector switch cycles I assume. Testing a meter by the designer and manufacturer online is extremely brave, and proof of confidence in their products. Including spacer fitting in error.
OK, I will be like these kids who type *FISRT* hahaha