Author Topic: EEVBlog 121GW Discussion thread  (Read 229453 times)

Terry01 and 1 Guest are viewing this topic.

Online CDaniel

  • Regular Contributor
  • *
  • Posts: 249
  • Country: ro
Re: EEVBlog 121GW Discussion thread
« Reply #1275 on: December 10, 2018, 08:08:04 am »
The shim thichness is not relevant because normally there is a empty space between the switch and the case   ... so actually could be a very little push from the shim . My multimeter with the board out of the case is working well if I turn the switch , is not that loose not to make contact without the shim .
So the PCB needs to be just a little thicker . Thats the good solution , because with the shim , all the case tolerances , how the pcb sits on the case  screw mounts play a role ... and you could have in the end too little pressure or to much . If something is not straight the wear will be more in one side .
« Last Edit: December 10, 2018, 08:26:23 am by CDaniel »
 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 5803
  • Country: us
Re: EEVBlog 121GW Discussion thread
« Reply #1276 on: December 10, 2018, 06:25:32 pm »
I don't disagree with what you have posted but from that last picture I posted, we can see the wear on the locking tab side.  I have mentioned my concern about the hole not being plated.  Its a rough surface.   If the locking tabs are what will be used to hold the tolerances on the spings, I don't see a problem as long as the tabs do not wear.  Sadly, they appear to be. 

I think we can both agree that once they get a handle on the switch and remove the shim, it should be retested.  Hopefully Dave will revamp that cycle testing jig and maybe this time ditch all the sticky notes.   
How electrically robust is your meter?? https://www.youtube.com/channel/UCsK99WXk9VhcghnAauTBsbg
 

Offline 1anX

  • Regular Contributor
  • *
  • Posts: 184
  • Country: au
Re: EEVBlog 121GW Discussion thread
« Reply #1277 on: December 11, 2018, 09:50:01 pm »
The shim thichness is not relevant because normally there is a empty space between the switch and the case   ... so actually could be a very little push from the shim . My multimeter with the board out of the case is working well if I turn the switch , is not that loose not to make contact without the shim .
So the PCB needs to be just a little thicker . Thats the good solution , because with the shim , all the case tolerances , how the pcb sits on the case  screw mounts play a role ... and you could have in the end too little pressure or to much . If something is not straight the wear will be more in one side .

UEI stated early on, when the first batch of meters were released and the faulty switch reports started coming back, that it was the thin (out of spec) PCB causing the problem. The shim was to correct for the thin board providing insufficient pressure for the switch contacts. At least that was my understanding of what EEVBlog was stating as causing the problem at the time.

Soon enough we will all know if that is the case, as the 4000 out of spec PCBs must now have been fitted into the final batch of meters? That being the case then, if UEI decides to continue the manufacture of the 121GW meter they will be on a correct thickness PCB and the switch will no longer require the shim.

EEVBlog commented earlier that they were unsure of when the next batch of meters produced on the new PCBs and incorporating the updated hardware mods would be produced. It cant be too far away as it seems sales have been strong up to this point. Soon we will have our answers and hopefully photos from the new meter's owners.

« Last Edit: December 11, 2018, 09:59:12 pm by 1anX »
 

Offline beanflying

  • Super Contributor
  • ***
  • Posts: 3875
  • Country: au
  • Toys so very many Toys.
Re: EEVBlog 121GW Discussion thread
« Reply #1278 on: December 18, 2018, 02:03:22 am »
Just by way of contrast to Joe's pictures above I knocked mine down for the first time since I installed the shim today firstly to upgrade the inferior 1.15 firmware that I have been using since it was first released to 1.57, yes I am that lazy when something works.

I got a first production kickstarter meter stickered #137 and the follow up shim kit when first released was installed straight away.

These photos are taken with no clean up of any sort to the board or contacts and no colour or contrast corrections. Just here for comparison not to start a pissing contest. A bit of wear on one side but most of that was there if I recall pre shim.

Coffee, Food, R/C and electronics nerd in no particular order :)
 
The following users thanked this post: joeqsmith

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 5803
  • Country: us
Re: EEVBlog 121GW Discussion thread
« Reply #1279 on: December 18, 2018, 04:00:18 am »
How did it look on the back side where the locking tabs ride?   
How electrically robust is your meter?? https://www.youtube.com/channel/UCsK99WXk9VhcghnAauTBsbg
 

Offline beanflying

  • Super Contributor
  • ***
  • Posts: 3875
  • Country: au
  • Toys so very many Toys.
Re: EEVBlog 121GW Discussion thread
« Reply #1280 on: December 18, 2018, 04:15:52 am »
Looked fine to me. From a strictly engineering point of view I would have liked to see an additional washer, longer barrel and maybe a circlip used instead of the clip but so far the clips are doing their job.

Just a potted theory yours spent more time than most without a shim? (not sure if you got a second meter from Dave?) So is yours indicative of what those of us with shims installed from scratch or fairly early into the life of the meter can expect or is it just what is possible without the shim for period X and then fitting one later?

EDIT: Just trawled my photos from Shim arrival time and none of the meter or PCB
« Last Edit: December 18, 2018, 04:31:03 am by beanflying »
Coffee, Food, R/C and electronics nerd in no particular order :)
 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 5803
  • Country: us
Re: EEVBlog 121GW Discussion thread
« Reply #1281 on: December 18, 2018, 12:20:22 pm »
Looked fine to me. From a strictly engineering point of view I would have liked to see an additional washer, longer barrel and maybe a circlip used instead of the clip but so far the clips are doing their job.

Just a potted theory yours spent more time than most without a shim? (not sure if you got a second meter from Dave?) So is yours indicative of what those of us with shims installed from scratch or fairly early into the life of the meter can expect or is it just what is possible without the shim for period X and then fitting one later?

EDIT: Just trawled my photos from Shim arrival time and none of the meter or PCB

Most of the pictures I posted were taken from what others uploaded.  I am not sure why anyone would attempt to extrapolate anything about switch wear from the prototype.  The pictures I have shown of the prototype are presented only to give some insight into how the design has evolved.   If it is still causing confusion, I can go ahead and scrub this area as well. 

***

I spent a few minutes and cleaned up this area as well.  Hopefully this will avoid further confusion with readers.   
« Last Edit: December 18, 2018, 12:51:57 pm by joeqsmith »
How electrically robust is your meter?? https://www.youtube.com/channel/UCsK99WXk9VhcghnAauTBsbg
 

Offline beanflying

  • Super Contributor
  • ***
  • Posts: 3875
  • Country: au
  • Toys so very many Toys.
Re: EEVBlog 121GW Discussion thread
« Reply #1282 on: December 18, 2018, 01:00:57 pm »
Only because I decided to go AWOL from these threads due to the speculators and armchair gurus before most even had meters so I am not in touch with what has gone on and been said or not said was why I asked about your photos. Your much punished prototype deserves a spot on a shelf and retirement as I did watch the torture done to it :)

As the upgrade to the firmware went as it should have and providing the meter test in spec I still will be a happy user. I have some more gear now than when I first got it so a few hours playing across the ranges will be in order but no torture outside of spec ;)
Coffee, Food, R/C and electronics nerd in no particular order :)
 
The following users thanked this post: GeoffreyF

Offline lowimpedance

  • Super Contributor
  • ***
  • Posts: 1061
  • Country: au
  • Watts in an ohm?
Re: EEVBlog 121GW Discussion thread
« Reply #1283 on: December 18, 2018, 11:37:55 pm »
Just by way of contrast to Joe's pictures above I knocked mine down for the first time since I installed the shim today firstly to upgrade the inferior 1.15 firmware that I have been using since it was first released to 1.57, yes I am that lazy when something works.

I got a first production kickstarter meter stickered #137 and the follow up shim kit when first released was installed straight away.

These photos are taken with no clean up of any sort to the board or contacts and no colour or contrast corrections. Just here for comparison not to start a pissing contest. A bit of wear on one side but most of that was there if I recall pre shim.
Have you had a close up look at the switch spring contact dimples to see if there is any notable wear either even or uneven ...or not !.
My meter which is from the first shipment showed a bit more PCB contamination than yours but after cleaning the PCB was still in excellent condition with no obvious track 'grooving' from the dimples.
The area of wear does appear to be the 'dimples'. Some wear is to be expected, just how much and how fast.... well  :-//.
Photo's from mine are already posted here and in the Firmware thread for comparison.
The odd multimeter or 2 or 3 or 4...or........can't remember !.
 

Offline beanflying

  • Super Contributor
  • ***
  • Posts: 3875
  • Country: au
  • Toys so very many Toys.
Re: EEVBlog 121GW Discussion thread
« Reply #1284 on: December 19, 2018, 12:03:56 am »
Nothing obvious under the microscope even holding them off axis to drop the glare like in the photo. All contacts on all switches wear including the ones in my high end decade boxes have wear. The aim should be to engineer and design within a budget (fact of modern design) the best solution. Maybe the budget in hindsight should have been stretched a bit more.

My thoughts above would be minor in production cost but would require retooling costs to then be recovered across the production. So while the washer, extra plastic and circlip would be in the order of cents the tool for the brush holder would be in the order of four figures even using a composite RIM tool and a lot more in metal for injection molding. The bonus of this sort of solution is it is could be retrofitted to the existing meters. If available I would most likely buy one.
« Last Edit: December 19, 2018, 01:37:10 am by beanflying »
Coffee, Food, R/C and electronics nerd in no particular order :)
 

Offline beanflying

  • Super Contributor
  • ***
  • Posts: 3875
  • Country: au
  • Toys so very many Toys.
Re: EEVBlog 121GW Discussion thread
« Reply #1285 on: December 19, 2018, 03:11:47 am »
Capacitance Linearity Tests.

Used Kickstarter V1 Meter running 1.57 firmware. Approximate Temp 23C
HP/Yokagawa Decade Capacitance box. Basic Accuracy 0.25%+3pF (order of magnitude better than the 121GW) Silvered Mica Caps. 0 - 1.2 uF

I used the stock leads with the screw on 4mm adapters as it is a real world device for most of us. Data below for anyone interested.

Small amount of drift generally up on the small capacitance's in the order of generally 10pF or so.

For those not wanting to look at the data it shows well inside spec across a range sliding toward spec at either end which is to be expected.  :-+
Coffee, Food, R/C and electronics nerd in no particular order :)
 
The following users thanked this post: Marco1971

Offline beanflying

  • Super Contributor
  • ***
  • Posts: 3875
  • Country: au
  • Toys so very many Toys.
Re: EEVBlog 121GW Discussion thread
« Reply #1286 on: December 19, 2018, 06:01:22 am »
Resistance Linearity Tests.

I did some very quick tests across a limited range of values way back on page 1 of this thread. The Data is included in the spreadsheets below in red. It cross references the Russian decade box to the 121GW to my Agilent 34401A. The Russian Box is at least an order of magnitude better than the 121GW at 0.03 and 0.02% across it's ranges.

Above the 119.99k Limit of the Russian Box I have used my own Vishay 0.1% decade box from 0.1 to 1Meg cross referenced to my 34401A (mainly because I haven't tested it fully before) above that I currently have a limited range. Homebrew box is better than the resistors spec it seems :-+

I have chosen not to test below 0.1 Ohm as it is well outside what I would use for 2W testing generally and not a fair test. The figures for below 1 ohm I would consider fairly good on 2W

Not sure if my meter is better than average but the Resistance readings are way better than the Spec above a few ohms to 50K they are freaky good. I finished up running a check of the same values of the Russian Box on 2W on the 34401a as I also have another few meters to test.  Looks like one of the calibration factors could be tweaked a fraction above 50K but still within spec.

Overall Linearity against a better meter and decent Resistance boxes :-+
« Last Edit: December 19, 2018, 06:05:31 am by beanflying »
Coffee, Food, R/C and electronics nerd in no particular order :)
 
The following users thanked this post: Marco1971

Offline nudge

  • Newbie
  • Posts: 4
  • Country: nl
    • Pareidolic Labs
Re: EEVBlog 121GW Discussion thread
« Reply #1287 on: December 19, 2018, 06:49:53 pm »
Hey all, I see the 121GW can measure temperatures >1000C in the manual! Just wondering if the thermocouple is safe to put into a conventional oven to test the temperature (multimeter of course outside).

Cheers!
An Aussie living in Amsterdam | My GitHub
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 9584
  • Country: us
Re: EEVBlog 121GW Discussion thread
« Reply #1288 on: December 19, 2018, 07:27:04 pm »
Hey all, I see the 121GW can measure temperatures >1000C in the manual! Just wondering if the thermocouple is safe to put into a conventional oven to test the temperature (multimeter of course outside).

For most ordinary thermocouples supplied with meters the insulation material around the probe is the limiting factor. It will melt or burn long before reaching the maximum temperature of the thermocouple junction.
I'm not an EE--what am I doing here?
 

Offline GeoffreyF

  • Regular Contributor
  • *
  • Posts: 233
  • Country: us
Re: EEVBlog 121GW Discussion thread
« Reply #1289 on: December 19, 2018, 07:28:53 pm »
Hey all, I see the 121GW can measure temperatures >1000C in the manual! Just wondering if the thermocouple is safe to put into a conventional oven to test the temperature (multimeter of course outside).

Cheers!

Yes, I have done exactly that myself.  I don't know if it is safe to put in food itself but probably is. I would check the door just to make sure there are no sharp edges where you run it in.  I assume that by "Conventional" you mean "Gas".  If electrical, just position the probe so it doesn't touch the element directly.  Microwaves would be a bad idea.  I just put the probe end in the oven and closed the door.

I am not sure if the probe itself goes to 1000c but it's within the idea of a baking oven.
US Amateur Extra W1GCF.
 

Offline beanflying

  • Super Contributor
  • ***
  • Posts: 3875
  • Country: au
  • Toys so very many Toys.
Re: EEVBlog 121GW Discussion thread
« Reply #1290 on: December 19, 2018, 11:16:52 pm »
The limits on the 121GW for Temperature are generally the limits of the 'K type' thermocouple junction used. In practice it goes a little lower than this so there may be hardware or software things going on but if you are south of -200C you won't be using this meter for your readings.

In the photo below is a K type bead termocouple I use for characterising and tuning commercial ovens as it can be poked in through a door seal and looped around a shelf. I haven't run it above 250C but after many years of use the insulation is fine. I would be fairly happy running the 121GW in a similar use but I haven't needed to yet.

A few of the others shown are silicone and teflon coated wires and should be fine in the same region and up toward 400C. The screw in types will easily handle more as only the junction and silica filled barrel are exposed to much heat the metal shield contains just heat resistant braid like the bead ones and is for mechanical strength not heat. Moving into the 1000C+ range they tend to be ceramic carriers for things like kiln etc.

The ones on my baby Coffee Roaster as shown are 6 or 7 years old and working fine at 200-250C.

Just one thing with the Meter is keep it away from as much heat as possible if the internal sensor or board gets hot the readings you will get are likely to be out. If it all goes to poo and you melt it or knick a cable they are cheap and no harm will come to the meter but let us see the carnage  ;)
« Last Edit: December 19, 2018, 11:18:23 pm by beanflying »
Coffee, Food, R/C and electronics nerd in no particular order :)
 

Offline nudge

  • Newbie
  • Posts: 4
  • Country: nl
    • Pareidolic Labs
Re: EEVBlog 121GW Discussion thread
« Reply #1291 on: December 20, 2018, 08:18:11 am »
Very cool info, thanks guys!
An Aussie living in Amsterdam | My GitHub
 

Offline jazper

  • Contributor
  • Posts: 39
Re: EEVBlog 121GW Discussion thread
« Reply #1292 on: December 20, 2018, 11:23:24 pm »
Hey I have seen a lot of negative stuff about the 121GW Multimeter. Lots of people recommending Brymen etc.

I took a chance, I bought the 121GW, and I have to say it's one of the easiest to use and fully featured multimeters I've ever laid my hands on. I'm very happy with the purchase.

I have two minor qualms, one is that the rubber probe holder on one side fits a little too snug, and the other is that the case needs to be purchased separately (and I didn't know that when I bought the meter). That said, it's a lot of bang for not many bucks.

If Dave is reading this - first of all, thankyou! - Second it'd be a really good idea to put an option to add the case on the purchase page of the store, or have a 121GW + Case item on the page.

One very happy customer.
 

Offline GeoffreyF

  • Regular Contributor
  • *
  • Posts: 233
  • Country: us
Re: EEVBlog 121GW Discussion thread
« Reply #1293 on: December 21, 2018, 12:31:54 am »
There are meters which cost 50% more than the 121 GW and still lack features which it has.   Yes, those meters have some other features and I might even buy one of them too *IF* I need those features.  However, I needed those of the 121GW and that's why I bought it.

Not one person with "Issues" actually described a real world task they could not complete and how they might have completed it with something else ... not one.
US Amateur Extra W1GCF.
 

Offline beanflying

  • Super Contributor
  • ***
  • Posts: 3875
  • Country: au
  • Toys so very many Toys.
Re: EEVBlog 121GW Discussion thread
« Reply #1294 on: December 22, 2018, 05:26:05 am »
Thermal Stability test - sort of but really rough.

Reference is a Fluke 515A portable Calibrator. Switched on for several days now as have the 34401a and Advantest. Overall drift on this gear is way less than the 121GW's ability to resolve it and has been sitting at +- 3LSD regardless of any Temperature variations in the shack over the last day. The Advantest has been out of use for some years and has crept up a fraction.

Not a complete test by any means but just an idea of DC stability in a walk up portable multimeter type test (which is what it is). The GW121 was flicked on and readings taken no more than a minute or two after that. There was 1-2 LSD of drift leaving it on for longer periods of time after that. If I was being serious I would run a graph against Temperature and stuff the whole lot in my chamber but it has other stuff in it held at a stable long term temperature. Any Northerners want to go play in the snow for us to learn more  ;)

The 10M input impedance loads the Fluke a miniscule amount (0.14mV) when connected with the other meters compared to theirs in the GigOhms range. Something to bear in mind when comparing References which may suffer even more with loading than the Fluke. But it is a comparison of stability not a test of 'absolute accuracy'.

There was a post I made way back in January at 42C  :phew: https://www.eevblog.com/forum/testgear/eevblog-121gw-discussion-thread/msg1403277/#msg1403277 Same cheap and cheerful Chinese reference today at 27C is 5.0003V compared to the 4.9998V at 42C. Either could have drifted so draw nothing really from this other than both are fairly decent for general use.

Fluke Comparison Results in the Pics and temperatures on the 121GW.
« Last Edit: December 22, 2018, 05:33:06 am by beanflying »
Coffee, Food, R/C and electronics nerd in no particular order :)
 
The following users thanked this post: Kean, Marco1971

Offline beanflying

  • Super Contributor
  • ***
  • Posts: 3875
  • Country: au
  • Toys so very many Toys.
Re: EEVBlog 121GW Discussion thread
« Reply #1295 on: December 22, 2018, 06:04:11 am »
How did it look on the back side where the locking tabs ride?

Dropped the back off it again and took a few happy snaps by rotating the knob rather than a tear down. I haven't done any cleanup at all on this area and I don't think I did when I installed the shim either?

Looks fine for another 10 years of use for me at least.  :)
Coffee, Food, R/C and electronics nerd in no particular order :)
 

Offline beanflying

  • Super Contributor
  • ***
  • Posts: 3875
  • Country: au
  • Toys so very many Toys.
Re: EEVBlog 121GW Discussion thread
« Reply #1296 on: December 23, 2018, 06:02:30 am »
Mains Frequencies and Voltages

While I have included the 121GW and the data I really wanted to have a proper look at the relative accuracy of a Valhalla 2703 AC Calibrator I have picked up during the year. I have run tests over 50 and 60 hz in the ranges of 100-130V and 220-250V today so this covers the majority of mains used in the world.

Other than spot checks on meters I haven't run a proper series through the Valhalla so time to start and figure out if it is in or out of spec. and look at if it needs a tweak. I will run some higher frequency tests and bigger voltage ranges against the 34401a but unless someone wants anything specifically on the 121GW I won't most likely get any additional data. The 121GW works within spec on limited testing to above 5kHz as claimed as I have done some other testing prior.

Doing this for me was to improve confidence in mains logging I have done from time to time on the 121GW and to get a better idea of the accuracy of the data. Not metrology grade results by any means but to give a reasonable level of confidence.

Valhalla 2703 stated accuracy +- 0.03% reading +0.03 range
Agilent 34401a accuracy +- 0.06 reading +0.03 range (two ranges used <120 and <750)
121GW accuracy +- 0.3 + 10 counts (0.1V in these tests I have manually added it to the errors as parsing data is a PITA)

I have included columns with calculated errors for the readings from the Specifications and while they sit well inside that there appears to be a minor issue with the Valhalla outputting consistently low by circa 0.1V/100V in this sort of frequency range. Voltage Steps of 2V are very consistent against both the meters. If a correction of that 0.1/100V was done the absolute error drops dramatically. Man with two clocks problem but I only have one good clock  :palm:

Call it the 121GW on mains around the world way better than spec regardless :-+
« Last Edit: December 23, 2018, 06:06:47 am by beanflying »
Coffee, Food, R/C and electronics nerd in no particular order :)
 

Offline MiroS

  • Regular Contributor
  • *
  • Posts: 54
  • Country: pl
Re: EEVBlog 121GW Discussion thread
« Reply #1297 on: December 25, 2018, 01:46:58 pm »
One very happy customer.
After watching review of other multimeters on YT of UEI, I was quite sure how it will go . Just look at old review of Dave for UEI multimeter - I think it was 100$ multimers review -  and reflect the same criteria to 121GW.
 

Offline GeoffreyF

  • Regular Contributor
  • *
  • Posts: 233
  • Country: us
Re: EEVBlog 121GW Discussion thread
« Reply #1298 on: December 27, 2018, 03:59:33 pm »
Here is another use for the 121GW.  In keeping with the traditions of obscure complaints, I have a quibble with the migraine inducing high frequency of the "continuity test".  Also, it seems to not "unlatch" quite to my satisfaction as I send dots and dashes at 60wpm.  Finally, it was rather thoughtless that the meter can't receive Morse code to change the configuration.  Wouldn't that be so handy if one is swinging from the rungs of a transmission tower, making measurements, I should not deign to actually touch the meter?  If I paid all this money for Code Practice Oscillator, I think it should be better than $10.00 one that is sold for that purpose.   

KC1KNS
US Amateur Extra W1GCF.
 
The following users thanked this post: hammy, beanflying

Offline MiroS

  • Regular Contributor
  • *
  • Posts: 54
  • Country: pl
Re: EEVBlog 121GW Discussion thread
« Reply #1299 on: December 27, 2018, 05:12:16 pm »
Here is another use for the 121GW.  In keeping with the traditions of obscure complaints, I have a quibble with the migraine inducing high frequency of the "continuity test".  Also, it seems to not "unlatch" quite to my satisfaction as I send dots and dashes at 60wpm.  Finally, it was rather thoughtless that the meter can't receive Morse code to change the configuration.  Wouldn't that be so handy if one is swinging from the rungs of a transmission tower, making measurements, I should not deign to actually touch the meter?  If I paid all this money for Code Practice Oscillator, I think it should be better than $10.00 one that is sold for that purpose.   

KC1KNS

Is left side of LCD starting to crack by Morse vibration? 73.
 
The following users thanked this post: GeoffreyF


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf