Author Topic: EEVBlog 121GW Discussion thread  (Read 769734 times)

0 Members and 9 Guests are viewing this topic.

Offline Iagash

  • Regular Contributor
  • *
  • Posts: 69
  • Country: de
Re: EEVBlog 121GW Discussion thread
« Reply #375 on: March 07, 2018, 02:27:48 pm »
@Dave: at the risk of sounding like an ungrateful asshole, while the beeper on/off option is fantastic (nice and quite now!  :-+ ), it renders the continuity beeper also silent, which I would argue either should have it's own configuration option, or at least be on all the time, even if the beeper for functions/buttons is turned off.

That being said, I don't know that I would classify this as an "issue".  A big thanks to UEi for getting issues in the FW addressed quickly!

On first thought I'd also say that the continuity beep should be always on. But on second thought it's maybe a matter of taste.

A bonus would be a blinking backlight if the continuity beep is off.

Anyways thanks for the beeper option.
 

Offline Hydron

  • Frequent Contributor
  • **
  • Posts: 978
  • Country: gb
Re: EEVBlog 121GW Discussion thread
« Reply #376 on: March 07, 2018, 03:29:44 pm »
 I personally think the beeper is necessary in the following circumstances:
- Continuity (which even uses the sound symbol!)
- Diode test
- When you've been an idiot and left a probe in the A jack when the meter is in voltage mode (assuming it's using sensing jacks - can't remember if this is the case)
- Over-load warning on ranges where an overload could cause damage

Other than those, I'd normally want the beeper off, so I'd agree with everyone asking for a middle option (preferably as a third setting in addition to the current two)
 

Online EEVblog

  • Administrator
  • *****
  • Posts: 37661
  • Country: au
    • EEVblog
Re: EEVBlog 121GW Discussion thread
« Reply #377 on: March 08, 2018, 12:24:22 am »
Bluetooth and VA mode is still completely broken, even if you filter the invalid records.
It's also broken in the android app. It displays the same wrong readings.

The 391.2 mVA reading is the correct one.

I can confirm this, I see infrequent large scale data value excursion in VA mode. Reported.
 

Offline mtdoc

  • Super Contributor
  • ***
  • Posts: 3575
  • Country: us
Re: EEVBlog 121GW Discussion thread
« Reply #378 on: March 08, 2018, 01:13:06 am »
I personally think the beeper is necessary in the following circumstances:
- Continuity (which even uses the sound symbol!)
- Diode test
- When you've been an idiot and left a probe in the A jack when the meter is in voltage mode (assuming it's using sensing jacks - can't remember if this is the case)
- Over-load warning on ranges where an overload could cause damage

Other than those, I'd normally want the beeper off, so I'd agree with everyone asking for a middle option (preferably as a third setting in addition to the current two)

I agree. That seems like a no brainer.
 

Offline Seppy

  • Supporter
  • ****
  • Posts: 189
  • Country: au
  • Curious
Re: EEVBlog 121GW Discussion thread
« Reply #379 on: March 09, 2018, 03:27:22 am »
a) Nothing was "removed in editing". The entire original 23 hour video is still being processed by youtube and will eventually be available online. It seems that only the final 2 hours is currently available.

b) By the "edited video" I presume you mean the one "How to do Lifecycle Testing". I simply extracted some content from the stream to create a new video were I talked about general range switch testing as I thought it might be interesting to people as separate video.

c) There is a uCurrent x100 amplifier used in the jig across the switch contacts, that is not shown in the equations.

The 24 hour video was no longer listed when I checked, so yes I was referring to your how-to video.   

Quote
.. r=v/2300 and R=2740/((1/r)-1) . The second note stated the first column was V.    So for v=128,  R=2740/((2300/128)-1) = 152.5. 

I assume it's not as simple as it being off by 100X or 0.152 ohms,  seems way too low but maybe.     Looking at V=1272, R = 1.516 in this case and I think it was much higher.  Strange that the basics are not clear or just coded.



There is a 100x amplifier, that formula didn't show the uCurrent gain.
 

Offline Seppy

  • Supporter
  • ****
  • Posts: 189
  • Country: au
  • Curious
Re: EEVBlog 121GW Discussion thread
« Reply #380 on: March 09, 2018, 03:43:12 am »
Maybe someone can explain to me what the problem is:
2 different function generators tested, JDS6600 and FeelTech FY6600. Identical settings, identical results.

For example: 4kHz SQRT with 2Vpp. Measured ACV with the 121GW and as comparison a Fluke 28II (highres mode).

Range AUTO: 121GW selects 5V range, Fluke selects 6V range
121GW reads 1,37xxVac, Fluke reads 0,99xxVac

Manual range 5V 121GW, 6V Fluke 28II:
121GW reads 1,37xxVac, Fluke reads 0,99xxVac

Manual range 50V 121GW, 60V Fluke 28II:
121GW reads 0,95xVac, Fluke reads 0,94xxVac

Did you mean 2V peak to peak square wave, what did you mean by square root (SQRT)?
I reproduced the test and do not get your results. I get 1.0022 Vrms on a production meter for the above setup.

Did you signal generator have a DC offset, if so make sure that the 121GW isn't in DC + AC mode.
DC + AC mode for a 0 - 2V square wave results in approximately 1.4V, which is close to what you measured.

« Last Edit: March 09, 2018, 03:50:23 am by Seppy »
 

Offline Candid

  • Regular Contributor
  • *
  • Posts: 156
  • Country: de
Re: EEVBlog 121GW Discussion thread
« Reply #381 on: March 09, 2018, 12:24:58 pm »
Sorry, my mistake. square wave for sure.

Can you reproduce the following measurements?
2.0V peak to peak sine wave 4kHz, No DC offset, duty cycle 50%. Both meters in AC mode. Firmware 1.07 on my 121GW with serial number EEVblog 000084

Auto mode on both meters, Fluke in highres mode:
121GW selects 5V range, reads 1.1133Vac
Fluke 28II selects 6V range, reads 0.7031Vac

manual mode on both meters:
121GW 5V range, reads 1.1133Vac
Fluke 28II 6V range, 0.7032Vac

121GW 50V range, reads 0.659Vac (and it looses the frequency measurement and displays 0.00kHz)
Fluke 28II 60V range, reads 0.663Vac

See the attached pictures.
« Last Edit: March 09, 2018, 12:27:19 pm by Candid »
 

Offline fanOfeeDIY

  • Supporter
  • ****
  • Posts: 412
  • Country: jp
    • YouTube Channel
Re: EEVBlog 121GW Discussion thread
« Reply #382 on: March 11, 2018, 09:52:36 pm »
I uploaded video regarding roughly comparing Burden Voltage.

https://youtu.be/sSyZtqiWmvg
 

Online EEVblog

  • Administrator
  • *****
  • Posts: 37661
  • Country: au
    • EEVblog
Re: EEVBlog 121GW Discussion thread
« Reply #383 on: March 11, 2018, 11:26:42 pm »
I uploaded video regarding roughly comparing Burden Voltage.

Quote
On your 121GW, you do have a fairly high resolution. In the Amps low range, I think its 10uA resolution which is what you have with the other meters. However, your burden voltage is to be VERY low. Maybe 200ish uv!! Try that and see what you think.? 

I posted the above comment on YT.   Looking in the manual, maybe it no longer works this but I'm pretty sure that the prototype did.

The 500mA range uses the Amps jack (and a x10 amplifier) unlike most meters. This gives a very low maximum burden voltage of 30mV/A, or 15mV for 500mA full scale.
Very few meters will be this low on the 500mA range.
With 50,000 count the resolution is 10uA.
 

Offline fanOfeeDIY

  • Supporter
  • ****
  • Posts: 412
  • Country: jp
    • YouTube Channel
Re: EEVBlog 121GW Discussion thread
« Reply #384 on: March 12, 2018, 12:15:47 am »
I uploaded video regarding roughly comparing Burden Voltage.

Quote
On your 121GW, you do have a fairly high resolution. In the Amps low range, I think its 10uA resolution which is what you have with the other meters. However, your burden voltage is to be VERY low. Maybe 200ish uv!! Try that and see what you think.? 

I posted the above comment on YT.   Looking in the manual, maybe it no longer works this but I'm pretty sure that the prototype did.

The 500mA range uses the Amps jack (and a x10 amplifier) unlike most meters. This gives a very low maximum burden voltage of 30mV/A, or 15mV for 500mA full scale.
Very few meters will be this low on the 500mA range.
With 50,000 count the resolution is 10uA.
Thanks.   

I had DL'ed the manual when you made it available.  The one I have is 25th Nov. 17.   I see a mention of the 500mA mode on page 11. 

Quote
A and 500mA    11A/1000V DC/AC, IR 20kA HRC FAST + DIODE
and again on page 55
Quote
A/500mA current input fuse: 11A/1000V DC/AC, IR 20kA HRC FAST

The table on page 14 does not mention it.   
Quote
The following modes will use the x10 amplifier and may have additional off-set error that can be nulled out before measurement.

Maybe add something to page 32 where the current measurement is shown.   When I looked at the burden on the pre-production, is was pretty impressive.  From fanOfeeDIY's video, the manual may need a few small additions. 

Maybe you (Dave) would consider making a short video showing the meter's burden compared to a few others and how to use it properly.   

I am not quite following the discussion.  :'(
Especially "200ish uv!!" part.
Should I have used different range in my video?

 

Offline fanOfeeDIY

  • Supporter
  • ****
  • Posts: 412
  • Country: jp
    • YouTube Channel
EEVBlog 121GW Discussion thread
« Reply #385 on: March 12, 2018, 12:44:13 am »
I am not quite following the discussion.  [emoji22]
Especially "200ish uv!!" part.
Should I have used different range in my video?

Do the same test with your 121GW except attach the leads from the A 500mA connector and the COM.   The 121GW will display the current with a resolution of 10uA just like the others you show.  However, now if you look at the burden voltage for the meter, I suspect you will see something around 200uV  rather then several mV as your video shows.   I use "ish" as close enough.   

I had modified a UNI-T UT61E which has a very high burden voltage, in an attempt to improve it.   There was only so much I could do in the area available and the way the board was laid out.  I think are real low current, the burden on this meter was lower than the 121GW but once I changed ranged, it was very clear just how good the 121GW was.   It's impressive.   Give it a try and see what happens.  Post your updates, text is good enough.

Ah, I completely understood now. [emoji4]

Let me take and upload a short video tonight as your advice with A/500mA connector.
I still have the setup on my desk.
« Last Edit: March 12, 2018, 12:57:57 am by fanOfeeDIY »
 

Offline fanOfeeDIY

  • Supporter
  • ****
  • Posts: 412
  • Country: jp
    • YouTube Channel
Re: EEVBlog 121GW Discussion thread
« Reply #386 on: March 12, 2018, 10:31:45 am »
I am not quite following the discussion.  :'(
Especially "200ish uv!!" part.
Should I have used different range in my video?

Do the same test with your 121GW except attach the leads from the A 500mA connector and the COM.   The 121GW will display the current with a resolution of 10uA just like the others you show.  However, now if you look at the burden voltage for the meter, I suspect you will see something around 200uV  rather then several mV as your video shows.   I use "ish" as close enough.   

I had modified a UNI-T UT61E which has a very high burden voltage, in an attempt to improve it.   There was only so much I could do in the area available and the way the board was laid out.  I think are real low current, the burden on this meter was lower than the 121GW but once I changed ranged, it was very clear just how good the 121GW was.   It's impressive.   Give it a try and see what happens.  Post your updates, text is good enough.

Thank you for your valuable comment on both YouTube and here.
I uploaded another video.
https://youtu.be/ffP1Y0zp9KE
 
The following users thanked this post: rodcastler, Cliff Matthews, 1anX, genghisnico13

Online EEVblog

  • Administrator
  • *****
  • Posts: 37661
  • Country: au
    • EEVblog
Re: EEVBlog 121GW Discussion thread
« Reply #387 on: March 12, 2018, 10:50:59 am »
Maybe add something to page 32 where the current measurement is shown.   When I looked at the burden on the pre-production, is was pretty impressive.  From fanOfeeDIY's video, the manual may need a few small additions. 

Yep, the manual needs some updating.
The burden voltage may not be much better than, or on par with some other meters depending upon the range selected. It's a combination of what range uses the x10 amplifier and what shunt is used.
You could get the same excellent burden voltage on every range, but you'd need a effectively manual range switch positions with an optimised shunt for every range.
Very few meters will use the A jack for the 500mA range, so if your needs are on that range, it's going to be hard to beat.
 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 11630
  • Country: us
Re: EEVBlog 121GW Discussion thread
« Reply #388 on: March 12, 2018, 11:07:20 am »
Thank you for your valuable comment on both YouTube and here.
I uploaded another video.
Thank you for taking the time to run the test a second time. 

Online EEVblog

  • Administrator
  • *****
  • Posts: 37661
  • Country: au
    • EEVblog
Re: EEVBlog 121GW Discussion thread
« Reply #389 on: March 13, 2018, 10:50:39 pm »
V1.09 firmware https://www.eevblog.com/wp-content/uploads/2017/11/EEVBlog1_09.zip
It fixes the issue with the VA mode Bluetooth connection sending garbage data.

 
The following users thanked this post: VK5RC, haoleboy, Candid, ChrisG, 1anX

Offline dcac

  • Frequent Contributor
  • **
  • Posts: 336
Re: EEVBlog 121GW Discussion thread
« Reply #390 on: March 14, 2018, 12:31:06 pm »
I am not quite following the discussion.  :'(
Especially "200ish uv!!" part.
Should I have used different range in my video?

Do the same test with your 121GW except attach the leads from the A 500mA connector and the COM.   The 121GW will display the current with a resolution of 10uA just like the others you show.  However, now if you look at the burden voltage for the meter, I suspect you will see something around 200uV  rather then several mV as your video shows.   I use "ish" as close enough.   

I had modified a UNI-T UT61E which has a very high burden voltage, in an attempt to improve it.   There was only so much I could do in the area available and the way the board was laid out.  I think are real low current, the burden on this meter was lower than the 121GW but once I changed ranged, it was very clear just how good the 121GW was.   It's impressive.   Give it a try and see what happens.  Post your updates, text is good enough.

Thank you for your valuable comment on both YouTube and here.
I uploaded another video.
https://youtu.be/ffP1Y0zp9KE

Is the 121GW current measurement really within spec. here, it seems to show something like 4.6000mA when using the mA/uA jack but using the 10A jack it shows about 4.23mA - seems quite a bit of to me.

 

Offline ChrisG

  • Supporter
  • ****
  • Posts: 125
  • Country: nl
Re: EEVBlog 121GW Discussion thread
« Reply #391 on: March 14, 2018, 08:04:03 pm »
The beeping beeper is a bit funny indeed now. When switched off it does not beep at all, whilst the beeper icon is showing still. So agree and concur with the others. Suggestion for the Manual: make clear that the bin file needs to be named EEVblog.bin exactly. When I used the EEVblog1.09.bin file name it did not work, waited for 5 minutes and then decided to abort.
 

Offline Iagash

  • Regular Contributor
  • *
  • Posts: 69
  • Country: de
Re: EEVBlog 121GW Discussion thread
« Reply #392 on: March 15, 2018, 04:54:40 pm »
V1.09 firmware https://www.eevblog.com/wp-content/uploads/2017/11/EEVBlog1_09.zip
It fixes the issue with the VA mode Bluetooth connection sending garbage data.

Not really, the main display is correct now, but the upper display lost any connection to reality.

The corrupted data records are still there but less frequent.

 :palm:
 

Offline Iagash

  • Regular Contributor
  • *
  • Posts: 69
  • Country: de
Re: EEVBlog 121GW Discussion thread
« Reply #393 on: March 15, 2018, 06:05:55 pm »
I'm sure they run some sort of regression testing before they release updates and that is partly why UEI's response is slow.

Well, then you are a lot more optimistic than me.

Then again, now that Dave has stepped into the role of running life cycle testing maybe he needs to expand into some sort of firmware quality role as well.

Well I'll attach my console tool for the 121gw.
You can use it via:
Code: [Select]
gatttool -b 88:6B:XX:XX:XX:XX --char-write-req -a 0x0009 -n 0300 --listen | ./121gw.pl

To log into a file for later examination use:
Code: [Select]
gatttool -b 88:6B:XX:XX:XX:XX --char-write-req -a 0x0009 -n 0300 --listen | ./121gw.pl 2>&1 | tee logfile.txt

To find the BT address use:
Code: [Select]
hcitool lescan
and press ctrl-C when it is found.

You just need a linux system with bluetooth installed. Anything like debian jessie or newer should do. The gatttool command is in the bluez package. Nothing else is required.


Maybe Dave should make a video about what a ringbuffer is, how to find out how much space is left and skip writing a record now and then if bluetooth is too slow instead of overwriting not yet sent data.

I guess UEI is clearly in need of one.
« Last Edit: March 15, 2018, 06:19:09 pm by Iagash »
 

Offline 1anX

  • Regular Contributor
  • *
  • Posts: 195
  • Country: au
Re: EEVBlog 121GW Discussion thread
« Reply #394 on: March 15, 2018, 09:32:22 pm »
V1.09 firmware https://www.eevblog.com/wp-content/uploads/2017/11/EEVBlog1_09.zip
It fixes the issue with the VA mode Bluetooth connection sending garbage data.

Not really, the main display is correct now, but the upper display lost any connection to reality.

The corrupted data records are still there but less frequent.

 :palm:

Just when I think headway is being made in correcting and fixing this meter's faults, the reoccurring theme, which is a lack of testing yet again raises its head!

Like Joe said, its Dave's name/brand on the meter so he must assume ultimate responsibility as the marketer and seller of this device to ensure it actually works as specified!
 

Offline 1anX

  • Regular Contributor
  • *
  • Posts: 195
  • Country: au
Re: EEVBlog 121GW Discussion thread
« Reply #395 on: March 16, 2018, 12:45:02 am »
It seems 1.09 was a regression of other functionality in the VA mode.
The link has been removed.

Not sure what you mean by this? Can you please explain in a little detail just what testing is done and how you are releasing the updates!

I looked up regression testing as opposed to retesting fixed faults and found this defined as below...

"Retesting is done to make sure that the tests cases which failed in last execution are passing after the defects against those failures are fixed. Regression testing is not carried out on specific defect fixes. ... In Regression testing, you can include the test cases which passed earlier."


What process is UEi and Dave following for firmware releases?

Thanks in advance for any info you may wish to post, I find this educational and am learning how the development for this meter is evolving.
 

Offline Seppy

  • Supporter
  • ****
  • Posts: 189
  • Country: au
  • Curious
Re: EEVBlog 121GW Discussion thread
« Reply #396 on: March 16, 2018, 01:06:28 am »
It appears that the firmware may not be a regression, I believe I just found a new error in AC mode and thought it was a regression.

V1.09 firmware https://www.eevblog.com/wp-content/uploads/2017/11/EEVBlog1_09.zip
It fixes the issue with the VA mode Bluetooth connection sending garbage data.

Not really, the main display is correct now, but the upper display lost any connection to reality.

The corrupted data records are still there but less frequent.

 :palm:


Can you confirm your setup was in DC mode? What were the conditions which resulted in the incorrect values of the secondary display on the 121GW?

Old post about this was removed as it seems it was not a regression and instead a different error.
« Last Edit: March 16, 2018, 01:38:48 am by Seppy »
 

Offline Iagash

  • Regular Contributor
  • *
  • Posts: 69
  • Country: de
Re: EEVBlog 121GW Discussion thread
« Reply #397 on: March 16, 2018, 01:54:48 am »

Can you confirm your setup was in DC mode? What were the conditions which resulted in the incorrect values of the secondary display on the 121GW?

Old post about this was removed as it seems it was not a regression and instead a different error.

I used DC mode, 400mA, 3.7V, 1.42VA. Both displays on the meter itself are perfect.

The display via bluetooth on the secondary display is 2.9V and 255.86A, the main display is fine.
 

Offline Seppy

  • Supporter
  • ****
  • Posts: 189
  • Country: au
  • Curious
Re: EEVBlog 121GW Discussion thread
« Reply #398 on: March 16, 2018, 02:29:08 am »

Can you confirm your setup was in DC mode? What were the conditions which resulted in the incorrect values of the secondary display on the 121GW?

Old post about this was removed as it seems it was not a regression and instead a different error.

I used DC mode, 400mA, 3.7V, 1.42VA. Both displays on the meter itself are perfect.

The display via bluetooth on the secondary display is 2.9V and 255.86A, the main display is fine.

Are you on windows or Android, which version of the app?
 

Offline Iagash

  • Regular Contributor
  • *
  • Posts: 69
  • Country: de
Re: EEVBlog 121GW Discussion thread
« Reply #399 on: March 16, 2018, 02:34:45 am »
Are you on windows or Android, which version of the app?

I tried Android Version 0 eevblog.x121gw, but my perl script displays the same values.

It doesn't matter, fix the BT transmission with a correct buffering implementation.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf