EEVBlog 121GW Discussion thread

[ChrisG]

Another set.

[ChrisG]

3rd set. Comments to follow.

[The Soulman]

ChrisG, can you post pictures of the wiper contacts as well? Just curious.

[ChrisG]

They are spotless.  And I just put the whole thing together again... :-)

[ChrisG]

Appears to have less contamination but we have no idea how many cycles were on the two meters posted so far.   Can you tell with your eyes if any of that is metallic?   

I wonder how the people with the intermittent switches would compare with the ones the work as expected.  It would seem that the intermittent meters make have less contact force and would be less prone to wear but we really don't know.   

Does not seem to be metallic but I'm not an expert. On purpose I did not clean the contacts though. Btw I just noticed that on the wiper side there also one track very close to the hole of the selector pivot. As already said the wiper dots where pretty clean and shiny. Let's keep this conversation going please. Perhaps other people dare to open up their 121GW and see, check or take pictures. BTW Dave and Dave could also open up their 121GW's right?

[Candid]

Appears to have less contamination but we have no idea how many cycles were on the two meters posted so far.   Can you tell with your eyes if any of that is metallic?

I posted the first pictures and checked today. I think it's not metallic at least not magnetic. I checked with a strong neodyn magnet and it doesn't feel metallic.

I cleaned it and will check in some weeks.

[EEVblog]

Practical world sounds like a marketing term.  I really have no idea idea about the 87Vs reputation, if it has one or how it came to be.

After almost 30 years in the industry I can assure you it has that reputation. But I guess you have to be in the industry and closely follow this stuff to know.
How did it get that rep? - With a great deal of time giving dependable and trustworthy service.
But this is a 121GW thread so I'll stop talking about this now.

[1anX]

I also would like to get more updates on the meter. Any rumors, estimations, etc. So, we could see it's not abandoned. If possible. Or, better, a UEI representative here in the thread. Like other manufacturers do on this forum (e.g., MicSig, R&S, Siglent (tautech), Rigol, etc).

Dave, I think it is time that you found your voice and provide an update of the current situation!

What progress, (if any) has been made by you and UEi in rectifying the faults?

I am becoming concerned by the lack of communication and wondering if the root cause of the "bugs" is hardware related and only you can provide details on the meter to make clear the situation.

Considerable time has passed without you providing meaningful info on the rectification of the meter's abundant faults.

The meter comes with a warranty and is covered under ACCC laws, but I'm hoping you will be proposing a meter replacement/fix program for the shipped meters and a thorough revision for the meters awaiting manufacture/shipping.

With great respect, what is the current state of play Dave?

[amspire]

I also would like to get more updates on the meter. Any rumors, estimations, etc. So, we could see it's not abandoned. If possible. Or, better, a UEI representative here in the thread. Like other manufacturers do on this forum (e.g., MicSig, R&S, Siglent (tautech), Rigol, etc).

Dave, I think it is time that you found your voice and provide an update of the current situation!

What progress, (if any) has been made by you and UEi in rectifying the faults?

I am becoming concerned by the lack of communication and wondering if the root cause of the "bugs" is hardware related and only you can provide details on the meter to make clear the situation.

Considerable time has passed without you providing meaningful info on the rectification of the meter's abundant faults.

The meter comes with a warranty and is covered under ACCC laws, but I'm hoping you will be proposing a meter replacement/fix program for the shipped meters and a thorough revision for the meters awaiting manufacture/shipping.

With great respect, what is the current state of play Dave?

IanX,

If you have a meter that is so faulty that you are talking about invoking ACCC laws, then I hope you have contacted Dave directly - as the ACCC would expect you to do first.

Dave has actually given a lot of communication - far more then any other manufacturer I have seen and he does not even design or make the meters. Just be patient.

I am puzzled by your mention of the "abundant faults". My meter is pretty functional. The biggest issue that people seem to raise is the slow autoranging that is not a fault, but it is being improved.

I am prepared to wait for issues and improvements to be fully tested before release. I have a 121GW meter that seems to be safe and measures volts, amps, resistance, temperature, capacitance, frequency and duty factor all within specifications. I am very happy that Dave has already seen updates that will improve the meter.

What sort of meter did you get?

[EEVblog]

I also would like to get more updates on the meter. Any rumors, estimations, etc. So, we could see it's not abandoned. If possible. Or, better, a UEI representative here in the thread. Like other manufacturers do on this forum (e.g., MicSig, R&S, Siglent (tautech), Rigol, etc).

Dave, I think it is time that you found your voice and provide an update of the current situation!
What progress, (if any) has been made by you and UEi in rectifying the faults?

The issues are being investigated, stuff just takes time, especially when you have more than one issue. There is no point in me speculating daily.
There is essentially very little I can do personally to speed this up or really even help them out from a technical perspective.
They are still waiting on the range switch parts, due in a few days I believe, and then there is testing etc.
On top of that of course are several software issues, some like slow autoranging for example has been solved (to match that of the U1282A), others are still being worked on (presumably one by one).
Please be patient, when I know for certain, you'll know.

[Brumby]

1anX - you seem to be running hot and cold....

I cant see us hearing from Dave anytime soon as he is keeping a very low profile and with good reason. He cant comment until he has a defined course of action to rectify the bugs and switch issue. Its up to UEi at this stage to pull their finger out and and turn this first "production" run meter into what it should be.

Dave needs to be extremely diplomatic with both his backers/buyers and with UEi. Nothing to be gained by pounding a fist on a table! As much as I dislike the situation with this meter, I am following Dave's lead with patience and high expectations for speedy fix.

and now this...

Dave, I think it is time that you found your voice and provide an update of the current situation!

What progress, (if any) has been made by you and UEi in rectifying the faults?

I am becoming concerned by the lack of communication and wondering if the root cause of the "bugs" is hardware related and only you can provide details on the meter to make clear the situation.

Considerable time has passed without you providing meaningful info on the rectification of the meter's abundant faults.

The meter comes with a warranty and is covered under ACCC laws, but I'm hoping you will be proposing a meter replacement/fix program for the shipped meters and a thorough revision for the meters awaiting manufacture/shipping.

With great respect, what is the current state of play Dave?

My definition of "patience" in this scenario is a little more than a couple of days.  It's taken over 2 years of development to get the 121GW to where it is today - and the first run of units were bound to encounter some issues when released to the sort of people who frequent here.

In my limited experience, Dave has been pretty candid with his comments - and usually only says something when there is something worth saying.  If you want comforting on a daily basis, then that's not going to happen here.

This stuff takes time.

Oh, and even mentioning the ACCC at this point is really unfair.  Not only that - but it comes across as a threat.

Besides, the ACCC won't give a flying fig about getting the meter fixed to your satisfaction.  They would look at these threads and immediately understand what the situation is and they will tell you to talk to Dave.  They will simply ensure that you haven't been swindled.  An issue which would be immediately resolved by returning the meter and getting a full refund.

[Brumby]

My post above was written as Dave made his reply - but I posted it on principle.

I also would like to get more updates on the meter. Any rumors, estimations, etc. So, we could see it's not abandoned. If possible. Or, better, a UEI representative here in the thread. Like other manufacturers do on this forum (e.g., MicSig, R&S, Siglent (tautech), Rigol, etc).

Dave, I think it is time that you found your voice and provide an update of the current situation!
What progress, (if any) has been made by you and UEi in rectifying the faults?

The issues are being investigated, stuff just takes time, especially when you have more than one issue. There is no point in me speculating daily.
There is essentially very little I can do personally to speed this up or really even help them out from a technical perspective.
They are still waiting on the range switch parts, due in a few days I believe, and then there is testing etc.
On top of that of course are several software issues, some like slow autoranging for example has been solved (to match that of the U1282A), others are still being worked on (presumably one by one).
Please be patient, when I know for certain, you'll know.

This shows exactly what I expected.

... and you can't say he isn't responsive!!

[rstofer]

This certainly won't be my first or only meter.  Whether I get it today, next month or even a few months from now really makes no difference.  Sure, I'm looking forward to receiving it because, among other things, it is already paid for.  But I would rather have the upgraded switch and firmware rather than try to figure out what to do next. It takes what it takes and getting these issues resolved before it ships to me makes all the sense in the world (to me).

[lowimpedance]

About a week ago I reported the mechanical issue I had on my 121 with the switch and input terminals here ;
https://www.eevblog.com/forum/testgear/eevblog-121gw-multimeter-issues/msg1403241/#msg1403241

So after returning from holiday I thought I would have a look inside at what may be happening with my meter.
What I can see happening with mine to cause my issue was the small flexing of the PCB by the input plugs movement
was causing referred movement at the switch area and since my switch was rather 'sloppy' in its location hole this had caused
either contact or contact location issues on their relevant PCB tracks and thus the mode failures.

Further investigation of the switch rotor showed quite a lot of lateral movement in the PCB locating hole.
So next step was what to do to reduce this lateral 'slop'!.
After removing the PCB from the switch knob I could test fit various rod diameters, (rear ends of twist drills in 0.1mm inc.), to lightly expand
the switch plate plastic fingers that go through the main PCB to remove the slop but still have easy rotational movement.
The best size for my meter was 6.1mm OD. To help alignment further I made the ID be a good sliding fit to the hex shaft of the knob at 5.6mm.
 The resulting sleeve I machined up can be seen in the photo below in its final position (I left it longer so its easily removed.). The material used was PEEK which is
reasonably stiff and has good insulation properties, (I also had it anyway  ). Did not want to use PTFE which would have been too soft at the cross section used.

As an additional exercise I also cut up some 0.3mm PTFE sheet as spacers under the knob circlip, see photo below. Two pieces provided a good take up
of the space between the housing and the circlip. This reduced the knob free play up and down but allowed some compliance in the well in the top housing and the circlip too
due to the soft flexible nature of the PTFE.

So the end result of this exercise was a considerably less 'wobbly' feel to the knob and more importantly all the issues I had reported have now gone
completely and any movement of the input plugs have no effect on the switch nor does wiggling the knob itself set off any failures of the mode its in.

[1anX]

This certainly won't be my first or only meter.  Whether I get it today, next month or even a few months from now really makes no difference.  Sure, I'm looking forward to receiving it because, among other things, it is already paid for.  But I would rather have the upgraded switch and firmware rather than try to figure out what to do next. It takes what it takes and getting these issues resolved before it ships to me makes all the sense in the world (to me).

Agreed!

I have a collection of meters and was going to by a second Fluke, but then Dave came along with his integration of a few good ideas to make a "unique" meter. I put the Fluke on hold as I watched his ideas develop into a best bang for buck meter. Hackable and with BLE and the low burden voltage won me over, as I was seriously considering a uCurrentGold.

So Dave has my money and I'm looking forward to a best bang for buck meter turning up in March or whenever. I'm concerned that UEi have stumbled at the first production run and the switch is an issue. Look forward to seeing if Dave's good ideas get the build quality they deserve!

[1anX]

I'm hoping to use this meter for measuring ucurrent on my ESP32 projects. Has anyone used the meter for ucurrent measurement by placing a zero ohm shunt in place of the fuses? If you have tried it what burden voltage was present and how does it compare to the uCurrentGold?

[EEVblog]

I'm hoping to use this meter for measuring ucurrent on my ESP32 projects. Has anyone used the meter for ucurrent measurement by placing a zero ohm shunt in place of the fuses? If you have tried it what burden voltage was present and how does it compare to the uCurrentGold?

You can calculate that from the schematic value for the shunt resistors.
http://www.eevblog.com/files/121GW%20EEVBlog%20Circuit%20diagram.pdf
10mohm, 1.01ohm, and 100.1ohms.
x1 or x10 amp depending on range using that shunt.

[JonM]

For anyone interested in Dave candidly discussing the 121GW issues:

 https://theamphour.com/377-debugger-vs-printeffer/

Spoilers: Dave chuckles over the irony of the slow resistance settling time and what he has said about other meters.  He also noted that the slowness was introduced in a late firmware update and he noted that they should have had a formal list of tests to be done for each release.

I'm standing by for the US deliveries....

[lowimpedance]

About a week ago I reported the mechanical issue I had on my 121 with the switch and input terminals here ;
https://www.eevblog.com/forum/testgear/eevblog-121gw-multimeter-issues/msg1403241/#msg1403241
....
The best size for my meter was 6.1mm OD. To help alignment further I made the ID be a good sliding fit to the hex shaft of the knob at 5.6mm.
 The resulting sleeve I machined up can be seen in the photo below in its final position (I left it longer so its easily removed.). The material used was PEEK which is
reasonably stiff and has good insulation properties, (I also had it anyway  ). Did not want to use PTFE which would have been too soft at the cross section used.
....
So the end result of this exercise was a considerably less 'wobbly' feel to the knob and more importantly all the issues I had reported have now gone
completely and any movement of the input plugs have no effect on the switch nor does wiggling the knob itself set off any failures of the mode its in.

By adding the sleeve, I wonder if the PCB will wear away at shaft where it rotates in the hole.  Time will tell.

I did apply a small amount of contact type grease in that area but I also made sure the pressure from the sleeve was just enough to mitigate the lateral movement without grossly effecting the rotary movement of the switch.

[JonM]

I forgot to mention that a couple of days ago I dumped the batteries out of a set top box remote control I need to return and was surprised to see that they were made by UEI. I wonder if the 121GW comes with a UEI battery?

[lowimpedance]

I forgot to mention that a couple of days ago I dumped the batteries out of a set top box remote control I need to return and was surprised to see that they were made by UEI. I wonder if the 121GW comes with a UEI battery?

For us in Oz it was duracell.

[amspire]

So after returning from holiday I thought I would have a look inside at what may be happening with my meter.
What I can see happening with mine to cause my issue was the small flexing of the PCB by the input plugs movement
was causing referred movement at the switch area and since my switch was rather 'sloppy' in its location hole this had caused
either contact or contact location issues on their relevant PCB tracks and thus the mode failures.

Further investigation of the switch rotor showed quite a lot of lateral movement in the PCB locating hole.
So next step was what to do to reduce this lateral 'slop'!.
After removing the PCB from the switch knob I could test fit various rod diameters, (rear ends of twist drills in 0.1mm inc.), to lightly expand
the switch plate plastic fingers that go through the main PCB to remove the slop but still have easy rotational movement.
The best size for my meter was 6.1mm OD. To help alignment further I made the ID be a good sliding fit to the hex shaft of the knob at 5.6mm.
 The resulting sleeve I machined up can be seen in the photo below in its final position (I left it longer so its easily removed.). The material used was PEEK which is
reasonably stiff and has good insulation properties, (I also had it anyway  ). Did not want to use PTFE which would have been too soft at the cross section used.

As an additional exercise I also cut up some 0.3mm PTFE sheet as spacers under the knob circlip, see photo below. Two pieces provided a good take up
of the space between the housing and the circlip. This reduced the knob free play up and down but allowed some compliance in the well in the top housing and the circlip too
due to the soft flexible nature of the PTFE.

In the following, when I refer to just the "switch", I meant all the parts attached to the circuit board.

I decided to fix my 121GW wobbly switch up a while ago, but after looking at the mechanism, I took a very different approach. I did look at your method but I think it was making the situation even worse rather then better mechanically.

First, instead of putting a spacer under the circlip, I put a spacer between the knob and the the nylon detent spring so that the circlip was in contact with the case plastic. This lifts the knob upwards. I didn't make it tight. I just wanted to stop any signifigant vertical movement. The reason for this method was so that the circlip does not press on the switch at all. If you put spacers under the circlip, the circlip is pressing on the switch which I think is wrong. The spacer was actually just many layers of duct tape. I had a punch to make a nice hole in the middle.

But now the knob was even looser. So I added a few strips of Kapton tape around the knob sides to widen it till it stopped wobbling much. You want to change a cone shape that is narrowest at the top to one widest at the top, so start off with a thin strip of tape just near the top side of the knob . Then a thicker one that goes to 2/3 down the knob  sides. Then some that goes around the whole sides. The idea is to increase the knob diameter to the point it can hardly wobble, but still have a tiny clearance to the case. Correctly done and the knob is no stiffer then before.

A bit of heat (I just used some hot water) to set the Kapton adhesive and it stays in place really well.

The way you have fixed the switch is you are trying to use the switch to stop the knob from moving which is putting more stress on the switch. It should be the knob that correctly locates the switch. The switch should be free to float a bit.

After this fix, I don't think anyone would notice an issue. I didn't take the tape right to the top and you cannot actually see it at all.

Richard

[ChrisG]

Appears to have less contamination but we have no idea how many cycles were on the two meters posted so far.   Can you tell with your eyes if any of that is metallic?

I posted the first pictures and checked today. I think it's not metallic at least not magnetic. I checked with a strong neodyn magnet and it doesn't feel metallic.

I cleaned it and will check in some weeks.

I wouldn't have expected it to be magnetic but it's good to know.

Copper, aluminum, gold etc are non-ferro and this test will not work IMHO.

[Towger]

I have reattached one of ChrisG's photos with a couple of questions:



1) What is the point of the two notches in the outside track?
    Are these to give slightly more resistance when moving the switch from the Off position, however there appears to be an slightly worn segment with no track, so this maybe the 'Off' position.

2) A number of the vias too close to the rubbing line/area of the track for my liking.   Joe's testing has clearly shown the wear caused by vias in the track is a major contributor to early failure of the switch.  There appears to be plenty of space in the other side so this only costs initial design time to be done right.

[lowimpedance]

So after returning from holiday I thought I would have a look inside at what may be happening with my meter.
What I can see happening with mine to cause my issue was the small flexing of the PCB by the input plugs movement
was causing referred movement at the switch area and since my switch was rather 'sloppy' in its location hole this had caused
either contact or contact location issues on their relevant PCB tracks and thus the mode failures.

Further investigation of the switch rotor showed quite a lot of lateral movement in the PCB locating hole.
So next step was what to do to reduce this lateral 'slop'!.
After removing the PCB from the switch knob I could test fit various rod diameters, (rear ends of twist drills in 0.1mm inc.), to lightly expand
the switch plate plastic fingers that go through the main PCB to remove the slop but still have easy rotational movement.
The best size for my meter was 6.1mm OD. To help alignment further I made the ID be a good sliding fit to the hex shaft of the knob at 5.6mm.
 The resulting sleeve I machined up can be seen in the photo below in its final position (I left it longer so its easily removed.). The material used was PEEK which is
reasonably stiff and has good insulation properties, (I also had it anyway  ). Did not want to use PTFE which would have been too soft at the cross section used.

As an additional exercise I also cut up some 0.3mm PTFE sheet as spacers under the knob circlip, see photo below. Two pieces provided a good take up
of the space between the housing and the circlip. This reduced the knob free play up and down but allowed some compliance in the well in the top housing and the circlip too
due to the soft flexible nature of the PTFE.

In the following, when I refer to just the "switch", I meant all the parts attached to the circuit board. The "switch rotor" is the knob part attached to the case.

I decided to fix my 121GW wobbly switch up a while ago, but after looking at the mechanism, I took a very different approach. I did look at your method but I think it was making the situation even worse rather then better mechanically.

First, instead of putting a spacer under the circlip, I put a spacer between the switch rotor and the the nylon detent spring so that the circlip was in contact with the case plastic. This lifts the rotor upwards. I didn't make it tight. I just wanted to stop any signifigant vertical movement. The reason for this method was so that the circlip does not press on the switch at all. If you put spacers under the circlip, the circlip is pressing on the switch which I think is wrong. The spacer was actually just many layers of duct tape. I had a punch to make a nice hole in the middle.

But now the switch rotor was even looser. So I added a few strips of Kapton tape around the switch rotor sides to widen it till it stopped wobbling much. You want to change a cone shape that is narrowest at the top to one widest at the top, so start off with a thin strip of tape just near the top side of the switch rotor. Then a thicker one that goes to 2/3 down the rotor sides. Then some that goes around the whole sides. The idea is to increase the switch rotor diameter to the point it can hardly wobble, but still have a tiny clearance to the case. Correctly done and the switch rotor is no stiffer then before.

A bit of heat (I just used some hot water) to set the Kapton adhesive and it stays in place really well.

The way you have fixed the switch is you are trying to use the switch to stop the rotor from moving which is putting more stress on the switch. It should be the switch rotor that correctly locates the switch. The switch should be free to float a bit.

After this fix, I don't think anyone would notice an issue. I didn't take the tape right to the top and you cannot actually see it at all.

Richard

For reference when referring to the rotor, that is the plastic carrier for the contact fingers, the PCB is the Stator and the knob would be the 'clicker plate' just like any rotary wafer switch assembly.
 Okay when making it worse are you referring to the small sleeve or the PTFE under the circlip ?.
For the sleeve i disagree that any additional stress is being applied to the contacts or the PCB. The dimensions of it are carefully chosen by measurement to only
apply enough pressure to the plastic fingers of the rotor body to the PCB hole reducing only lateral free play ,(which in my opinion is excessive ), to a minimum without
adversely effecting the effort required to rotate. This mechanical change guarantees alignment of the switch contacts on the rotor with the PCB stator and how will this
introduce ant further stress on the contacts.
 Now for the PTFE under the circlip, has anyone measured the distance from the top of the rotor body to the underside of the circlip to determine how much distance
there actually is ?. I will admit it has crossed my mind and yet I did not actually do it myself , so as soon as I can I will take the meter apart again I will attempt to get some
measurement done there and post back, as well as try the knob without the PTFE spacers to see if only the sleeve cured my issue or it is indeed needed to apply light pressure to the rotor.
If even that is actually happening with the spacers installed.