Author Topic: EEVBlog 121GW Discussion thread  (Read 770216 times)

VEGETA and 2 Guests are viewing this topic.

Offline Pat5

  • Newbie
  • Posts: 4
  • Country: is
Re: EEVBlog 121GW Discussion thread
« Reply #1325 on: January 08, 2019, 07:50:17 am »
I would like to thank everyone for the valuable feedback. I was under the impression that resistance measurement was recently improved to around 4.2-4.5 seconds.

The range switch reliability is something which i haven't thought about and is a valid point: No matter how feature-full and accurate, how are you going to use your device if you can't turn it on ?
The above is an extreme example of course, but the idea of joeqsmith (mechanical stress) is well worth taking into consideration.

@exe Personally, I do not believe in objective, non-polarized opinions of people, which is why i'm interested in everyone's opinion. Polarization is the reason that i've chosen this thread to post in.

Again, Thank you !
« Last Edit: January 08, 2019, 07:55:07 am by Pat5 »
 

Offline CDaniel

  • Frequent Contributor
  • **
  • Posts: 411
  • Country: ro
Re: EEVBlog 121GW Discussion thread
« Reply #1326 on: January 08, 2019, 08:42:24 am »
The big step in my opinion would be to make a firmware capable of "normal" autorange hysteresis in all modes - volts , amps , resistance and so on.
Every multimeter has the advertised number of counts 4000 , 6000 , 50000 all the time and switches up and down above this value , lets say at 4400-4100 , 6600-6300 , 55000-52000 ...

Hysteresis is directly corelated with the autorange speed , if a meter could switch ranges instantly  , then there wouldn't be any hysteresis at all needed .

121GW has the biggest hysteresis I ever saw or heard ... 55000-40000 for resistance , 55000-44000 for volts , probably because it is so slow and an autorange loop could form easily if the voltage , resistance is variable up and down .

« Last Edit: January 08, 2019, 08:46:20 am by CDaniel »
 

Offline exe

  • Supporter
  • ****
  • Posts: 2559
  • Country: nl
  • self-educated hobbyist
Re: EEVBlog 121GW Discussion thread
« Reply #1327 on: January 08, 2019, 09:04:32 am »
Many (most?) meters at this price don't support firmware updates at all.

Because they don't need it?  >:D
 

Offline J-R

  • Frequent Contributor
  • **
  • Posts: 952
  • Country: us
Re: EEVBlog 121GW Discussion thread
« Reply #1328 on: January 08, 2019, 09:09:28 am »
I would like to thank everyone for the valuable feedback. I was under the impression that resistance measurement was recently improved to around 4.2-4.5 seconds.

The range switch reliability is something which i haven't thought about and is a valid point: No matter how feature-full and accurate, how are you going to use your device if you can't turn it on ?
The above is an extreme example of course, but the idea of joeqsmith (mechanical stress) is well worth taking into consideration.

@exe Personally, I do not believe in objective, non-polarized opinions of people, which is why i'm interested in everyone's opinion. Polarization is the reason that i've chosen this thread to post in.

Again, Thank you !

Just to clarify, I mean to test all 4 DMMCheck Plus resistors in a row takes me about about 17 seconds total on the 121GW vs. 5 seconds total on faster meters.  On the BM235, by the time I took my eyes off the probes and looked at the meter, it already had settled on the value.  It's lightning fast.  I have no problems recommending it for quickly checking a pile of random resistors.

« Last Edit: January 08, 2019, 09:19:32 am by J-R »
 

Online beanflying

  • Super Contributor
  • ***
  • Posts: 7358
  • Country: au
  • Toys so very many Toys.
Re: EEVBlog 121GW Discussion thread
« Reply #1329 on: January 08, 2019, 09:17:30 am »
Many (most?) meters at this price don't support firmware updates at all.

Because they don't need it?  >:D


"Just to be clear, I bricked it with own firmware, not with the stock one."



Just to be clear maybe some shouldn't play with it either  ::)
« Last Edit: January 08, 2019, 09:22:26 am by beanflying »
Coffee, Food, R/C and electronics nerd in no particular order. Also CNC wannabe, 3D printer and Laser Cutter Junkie and just don't mention my TEA addiction....
 

Offline CDaniel

  • Frequent Contributor
  • **
  • Posts: 411
  • Country: ro
Re: EEVBlog 121GW Discussion thread
« Reply #1330 on: January 08, 2019, 09:33:35 am »
You can't play with it so much , it is complicated to reverse engineear the code and maybe doesn't worth the pain ... so it is not hackable , just in theory .
 

Online beanflying

  • Super Contributor
  • ***
  • Posts: 7358
  • Country: au
  • Toys so very many Toys.
Re: EEVBlog 121GW Discussion thread
« Reply #1331 on: January 08, 2019, 09:38:13 am »
I would agree not worth the pain. I just upgraded mine February firmware to 1.5x and it has come a long way.

Taking a drive by sling about that firmware upgrade option from someone who broke one with their own was just a bit much. Anyone want to buy a secondhand meter  ::)
Coffee, Food, R/C and electronics nerd in no particular order. Also CNC wannabe, 3D printer and Laser Cutter Junkie and just don't mention my TEA addiction....
 

Offline FrankBuss

  • Supporter
  • ****
  • Posts: 2365
  • Country: de
    • Frank Buss
Re: EEVBlog 121GW Discussion thread
« Reply #1332 on: January 08, 2019, 11:36:02 am »
You can't play with it so much , it is complicated to reverse engineear the code and maybe doesn't worth the pain ... so it is not hackable , just in theory .

Someone did already reverse engineered the firmware, and modified it for something, without even owning a 121GW, and it works. But there are not many such wizards out there.
So Long, and Thanks for All the Fish
Electronics, hiking, retro-computing, electronic music etc.: https://www.youtube.com/c/FrankBussProgrammer
 

Offline CDaniel

  • Frequent Contributor
  • **
  • Posts: 411
  • Country: ro
Re: EEVBlog 121GW Discussion thread
« Reply #1333 on: January 08, 2019, 11:59:50 am »
I'm aware only for that "Hello World" on display , that is not difficult to create if you find in code the text "U-1.57" - the current software version displayed at start-up . You can replace it with everything ... your name maybe .  But to reverse engineer all , without the source code and to understand it , is much harder than to write your own code from scratch . A decompiler will give you something , but is not the original program , much of it must be interpreted and are many ways . That's why it is useless .
« Last Edit: January 08, 2019, 12:33:08 pm by CDaniel »
 

Offline Brumby

  • Supporter
  • ****
  • Posts: 12288
  • Country: au
Re: EEVBlog 121GW Discussion thread
« Reply #1334 on: January 08, 2019, 12:18:56 pm »
It's not useless ... just very challenging.

... but then I started out writing application software on mainframes in assembler.
 

Offline FrankBuss

  • Supporter
  • ****
  • Posts: 2365
  • Country: de
    • Frank Buss
Re: EEVBlog 121GW Discussion thread
« Reply #1335 on: January 08, 2019, 12:34:04 pm »
I'm aware only for that "Hello World" on display , that is not difficult to create if you find in code the text "U-1.57" - the current software version displayed at start-up . You can replace it with everything ... your name maybe .  But to reverse engineer all , without the source code and to understand it , is much harder than to write your own code from scratch . A decompiler will give you something , but is not the original code , much must be interpreted and are many ways . That's why it is useless .

No, he increased the autorange speed, Dave did a video about it, too:



He has published everything in Github:

https://github.com/tpwrules/121gw-re

But you are right, probably not useful as the base for a fully open source version. But the circuit diagram is open source, so someone could develop a cleanroom implementation of an open source firmware for it. Shouldn't be too difficult, just a lot of work for all the functions the original firmware has.
So Long, and Thanks for All the Fish
Electronics, hiking, retro-computing, electronic music etc.: https://www.youtube.com/c/FrankBussProgrammer
 
The following users thanked this post: Marco1971, HarlanLSU1

Online joeqsmith

  • Super Contributor
  • ***
  • Posts: 11631
  • Country: us
Re: EEVBlog 121GW Discussion thread
« Reply #1336 on: January 08, 2019, 01:16:03 pm »
We understand your point of view ... your hobby testing how robust a meter is . As a user you dont care that much if the switch won't last only 20K cycles in the future ... but the firmware is buggy now  :-\  and you can't use it as you would like .
I would be pissed if I paid 300+ for a meter and the switch went bad in the first year or two.  The UNI-T was only $30 and really what I expect from a cheap meter with nominal use.  Same for BK and Mastech.  I consider them disposable.  But at $300 it better be built to stand the test of time.   

Same holds true for basic transients.  Last thing I want is a meter like my first brand new Fluke.  Look at it wrong and a $70+ dollar repair bill, that was in the 80s.  Firmware can be fixed but if the hardware is wrong, there is little you are going to do to correct it. 

My normal use of a handheld meter is pretty basic.  That's why I buy cheap, disposable meters for the garage.  The 121GW would be out of place in that environment.

For potential buyers is important to know the switch longevity , but users can't do anything as you said ( besides beeing pissed  >:()
So we concentrate on what can be changed , the firmware .
And without a mature firmware many switches will last for ever , who would use daily such a multimeter ?

I don't remember the exact count for our testing of the switch but I believe it was 40000. We did our range switch testing live on youtube.

From previous post
Quote
  Yea, I know Dave ran what he called his 50,000 cycle test on it, I watched.   

But again, from pictures people have posted (no my pictures), it appears that the contact has changed.  They had both the single double dimple designs.  I have no idea what is in there today and how it compares with what you tested or why it was changed.   I would like to see you repeat your testing without the yellow sticky notes, with both contact designs and again after the shim has been designed out.  Then again, it's a bit of work.  I've seen switches go bad on enough of these cheap meters that personally, I don't trust it.   Adding the shim last minute didn't build confidence.  That's my feeling. 



No, he increased the autorange speed, Dave did a video about it, too:
...
He has published everything in Github:
...
But you are right, probably not useful as the base for a fully open source version. But the circuit diagram is open source, so someone could develop a cleanroom implementation of an open source firmware for it. Shouldn't be too difficult, just a lot of work for all the functions the original firmware has.
The same person pointed out one of the changes to the code that effected the filter and also found the code that had the logical error for the lead insertion test. 

I suspect that it is more difficult than you think which is why talk surrounding the firmware continues a year after they started shipping. 

Offline Marco1971

  • Contributor
  • Posts: 29
  • Country: it
121GW Range Switch Cycle Testing
« Reply #1337 on: January 08, 2019, 05:20:58 pm »
 

Offline newbrain

  • Super Contributor
  • ***
  • Posts: 1714
  • Country: se
Re: EEVBlog 121GW Discussion thread
« Reply #1338 on: January 10, 2019, 12:28:42 pm »
Every multimeter has the advertised number of counts 4000 , 6000 , 50000 all the time and switches up and down above this value , lets say at 4400-4100 , 6600-6300 , 55000-52000 ...
False, neither Fluke 87-V nor (AFAICR) UNI-T 61D do that.
Nandemo wa shiranai wa yo, shitteru koto dake.
 

Offline CDaniel

  • Frequent Contributor
  • **
  • Posts: 411
  • Country: ro
Re: EEVBlog 121GW Discussion thread
« Reply #1339 on: January 10, 2019, 01:55:23 pm »
Pretty bad , I hope you don't want 121GW to copy the exceptions or slow meters with the hardware that it has ...

If the hysteresis is very big will eat some resolution because for a 50.000 count meter you can't use a giant up level like 65.000 just to have down level above 50.000 count .
I think  I don't have to explain again that if you adjust a voltage up and down you will experience this loss of resolution .
Anyway the switching should be as seamless as possible , not this huge hysteresis and slowness that now the firmware has ... I don't think any user would want different or wouldn't care
« Last Edit: January 10, 2019, 03:21:15 pm by CDaniel »
 

Offline GeoffreyF

  • Regular Contributor
  • *
  • Posts: 234
  • Country: us
Re: EEVBlog 121GW Discussion thread
« Reply #1340 on: January 10, 2019, 06:45:11 pm »
Pretty bad , I hope you don't want 121GW to copy the exceptions or slow meters with the hardware that it has ...

If the hysteresis is very big will eat some resolution because for a 50.000 count meter you can't use a giant up level like 65.000 just to have down level above 50.000 count .
I think  I don't have to explain again that if you adjust a voltage up and down you will experience this loss of resolution .
Anyway the switching should be as seamless as possible , not this huge hysteresis and slowness that now the firmware has ... I don't think any user would want different or wouldn't care

I have to wonder what you use meters for.  When I stick my probe in a circuit, I have a pretty good idea what my expectations are because I know how to read a schematic.   If  it turns out my range choice was wrong (that button which says "Range") then I can select another range.   If I really have no idea at all (which is rare) then auto range is quite handy. 

ALL engineering involves compromises which fit around expected use.  So does equipment purchase selection.  In your posts, that concept rarely if ever appears.  The cost of a meter is not that much of an issue if actually using it makes sense.  The 121GW is a good meter for many purposes. Others are good for other things.   This is why people buy different meters.  The 121GW meets the specifications on its tin and always has.   It's great that there are firmware updates.  THAT IS NOT BAD - IT IS GOOD.
US Amateur Extra W1GCF.
 
The following users thanked this post: newbrain

Offline newbrain

  • Super Contributor
  • ***
  • Posts: 1714
  • Country: se
Re: EEVBlog 121GW Discussion thread
« Reply #1341 on: January 10, 2019, 07:00:10 pm »
[...] to copy the exceptions [...]
I'm honestly interested to know which DMMs have a down-range threshold higher than the nominal target range as you have stated.
Can you demonstrably name some?
It looks like an unnatural design decision.

I have provided counter examples to your statement, to which I can add all the Fortune chip based DMM (just check each link).

I can agree (as I did answering your original post) that 121GW thresholds could be somewhat tighter, but please cease this 4 months prolonged cruelty to a deceased ungulate.
 :horse: :horse: :horse: :horse:
Nandemo wa shiranai wa yo, shitteru koto dake.
 

Offline CDaniel

  • Frequent Contributor
  • **
  • Posts: 411
  • Country: ro
Re: EEVBlog 121GW Discussion thread
« Reply #1342 on: January 10, 2019, 08:59:40 pm »
Hmm , almost all , if not all , multimeters with the microcontroller imbedded with the measurement chipset , old and new , even the cheapest chinese ones . And the fast multimeters with separate microcontroller ... the extreme hysteresis that needs to go below the nominal counts is a clear indication for slow autoranging .
« Last Edit: January 10, 2019, 09:07:26 pm by CDaniel »
 

Online joeqsmith

  • Super Contributor
  • ***
  • Posts: 11631
  • Country: us
Re: EEVBlog 121GW Discussion thread
« Reply #1343 on: January 11, 2019, 01:07:14 am »
The big step in my opinion would be to make a firmware capable of "normal" autorange hysteresis in all modes - volts , amps , resistance and so on.
Every multimeter has the advertised number of counts 4000 , 6000 , 50000 all the time and switches up and down above this value , lets say at 4400-4100 , 6600-6300 , 55000-52000 ...

According to your post, which is presented as an absolute statement of fact, EVERY multimeter does this in EVERY function and EVERY range.  If that is true, it would be easy enough to just run DCV for example and sweep it around the first switch point.  There would be no need to check every switch point and every function, correct?   

If the above is correct, I have a few meters that have survived that I could check easy enough.  Then again, you already KNOW the outcome.  So really, there's no point in checking, right? 

Quote
Hmm , almost all , if not all , multimeters with the microcontroller imbedded with the measurement chipset , old and new , even the cheapest chinese ones .
Sorry missed this one.  So it's not for certain EVERY meter as you first state.  It's all or almost all. 

Not being the king of person that would ever listen to what someone has to post on the WWW about meters, I decided to run a few in the DCV mode at the first switch point.  I ran them a few times.

Gossen, M248B, 300K count,  3.07, 2.89      Not off to a real good start for your Almost all, Every statements...
Fluke, 189, 50K, 5.5, 4.89                             2 for 2
Brymen, BM869s, 50K,  5.25,  4.60               Well my favorite meter is at least in with a good group....
Summit, TPI 194II, 50K, 4.99, 4.03               Didn't even make it to 5.0...
UNI-T, UT181A, 60K, 5.9, 5.6                         Again, it didn't make it to 6.0
Brymen, BM839, 20K, 1.93, 1.84                   I am seeing a trend...
Fluke, 87V HR, 20K, 1.99, 1.69                      Yes, there is a trend..
CEM, DT9939, 40K, 4.04, 3.90                       Full disclosure, I actually ran my Extech EX540.  It's just a rebrand of that meter

Well I put in some effort to try and help your case but it seems in doing so, I discovered like so often on the internet, you havn't a clue.   

*****
Mastech, MS8229, 4K, 4.01, 3.60                This meter has some really interesting features. 
Kasuntest, ZT102, 6K, 6.19, 5.74
Fluke, 107, 6K, 5.99, 5.40
AMPROBE, AM530, 4K, 4.028,  3.60

So are you clueless as I think you are, or am I missing the point of your rants?    Feel free to explain your side.  All I have is data.

*****
OWON, B41T+, 22K, 2.257, 2.02               
UNI-T, UT61E, 22K, 2.25, 1.97

Surely you are going to show what we are doing wrong?
« Last Edit: January 12, 2019, 08:20:21 pm by joeqsmith »
 
The following users thanked this post: Kean, newbrain, beanflying, Marco1971

Offline GeoffreyF

  • Regular Contributor
  • *
  • Posts: 234
  • Country: us
Re: EEVBlog 121GW Discussion thread
« Reply #1344 on: January 11, 2019, 02:22:46 am »
Hmm , almost all , if not all , multimeters with the microcontroller imbedded with the measurement chipset , old and new , even the cheapest chinese ones . And the fast multimeters with separate microcontroller ... the extreme hysteresis that needs to go below the nominal counts is a clear indication for slow autoranging .

Name some meters.  Describe the tests you used to make this benighted determination.  Provide numbers including time and tables of results for these tests you didn't do.  Also, please tell us the accuracy of these meters which you like or think are better in this regard.  - of course you don't even know why I asked you that but you would if you actually read and thought about prior posts.  Name exactly what it is you do with multimeter's that makes your "observations" such a high priority.  Of course you can't do any of this.  You never will.  Nothing you have written is of any actionable use and you obviously have no clue that it isn't.  In all your posts you somehow imagine you will be taken seriously - please consider that a number of people who have never met and don't know each other are simply annoyed with your tedious, redundant, not objective, numerical or measured opinions.  If it's credibility you seek - each post is the opposite of that.   
US Amateur Extra W1GCF.
 
The following users thanked this post: newbrain

Offline 2N3055

  • Super Contributor
  • ***
  • Posts: 6451
  • Country: hr
Re: EEVBlog 121GW Discussion thread
« Reply #1345 on: January 11, 2019, 07:13:10 am »
Metrix 3293, 10.0000 scale :  going up at 9.9999, down at 09.900
 

Offline GeoffreyF

  • Regular Contributor
  • *
  • Posts: 234
  • Country: us
Re: EEVBlog 121GW Discussion thread
« Reply #1346 on: January 11, 2019, 01:24:49 pm »
Metrix 3293, 10.0000 scale :  going up at 9.9999, down at 09.900

And the price? 
US Amateur Extra W1GCF.
 

Offline 2N3055

  • Super Contributor
  • ***
  • Posts: 6451
  • Country: hr
Re: EEVBlog 121GW Discussion thread
« Reply #1347 on: January 11, 2019, 01:54:39 pm »
Metrix 3293, 10.0000 scale :  going up at 9.9999, down at 09.900

And the price?
Less than the Gossen mentioned, we're discussing now where different meters switch scales in auto. Not comparing other things..
 

Offline chronos42

  • Regular Contributor
  • *
  • Posts: 137
  • Country: de
Re: EEVBlog 121GW Discussion thread
« Reply #1348 on: January 11, 2019, 06:54:04 pm »
Metrix 3293, 10.0000 scale :  going up at 9.9999, down at 09.900

Some more data from my handheld meters in the 5V range:

Agilent U1252B, 50000 counts: up 5.2000V, down 4.5000V

Fluke 287, 50000 counts: up 5.5000V, down 4.5000V

Gossen Metrahit Energy, 60000 counts: up 6.1000V, down 5.4000V

« Last Edit: January 11, 2019, 06:56:53 pm by chronos42 »
 
The following users thanked this post: Marco1971

Offline Messtechniker

  • Frequent Contributor
  • **
  • Posts: 774
  • Country: de
  • Old analog audio hand - No voodoo.
Re: EEVBlog 121GW Discussion thread
« Reply #1349 on: January 12, 2019, 08:54:05 am »
VC 940 (UT71E): 4.0 V up 3.8 V down
34465A: 1.2 V up 1.0 V down
« Last Edit: January 12, 2019, 08:55:46 am by Messtechniker »
Agilent 34465A, Siglent SDG 2042X, Hameg HMO1022, R&S HMC 8043, Peaktech 2025A, Voltcraft VC 940, M-Audio Audiophile 192, R&S Psophometer UPGR, 3 Transistor Testers, DL4JAL Transistor Curve Tracer, UT622E LCR meter
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf