Author Topic: EEVblog 121GW Multimeter Issues  (Read 668787 times)

0 Members and 4 Guests are viewing this topic.

Offline Iagash

  • Regular Contributor
  • *
  • Posts: 69
  • Country: de
Re: EEVblog 121GW Multimeter Issues
« Reply #50 on: January 11, 2018, 07:48:09 pm »
Hi firewalker,

yes, I took it appart after I noticed the issue.

Contacts look fine there are 4 of them, two on each side of the axis, each with two contact points.
They look "goldish" the pcb also looks like ENIG finish. All looks good, so I have no idea why it happens.
 

Offline exe

  • Supporter
  • ****
  • Posts: 2557
  • Country: nl
  • self-educated hobbyist
Re: EEVblog 121GW Multimeter Issues
« Reply #51 on: January 11, 2018, 07:51:31 pm »
They look "goldish" the pcb also looks like ENIG finish. All looks good, so I have no idea why it happens.

May be they don't reach well the pcb?
 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 11601
  • Country: us
Re: EEVblog 121GW Multimeter Issues
« Reply #52 on: January 11, 2018, 07:54:00 pm »
Hi joeqsmith,

yes, I can handle line voltages and do so regularly.

Do you think it's related to spikes on the mains? I usually use a 0-300V 6A insulating transformer for tests like this, but if you think it makes sense I can test directly on mains but have only 230V available then. So your request would be >50VA for >6h logging to SD-card? Resistive load would be fine (they are more quiet)?

I will run it over night and hope it doesn't burn down the house. :-)

Just tell me if with or without the transformer.

I was checking the power using an arb, amplifier and transformer coupled.  It seems like it was limited to 50V when I started to turn it up.  You don't need a lot of current.  Obviously the meter has no problems reading just voltage.  It was tied to the power measurement.   

The data logging problem I don't think was mode specific.  I was wanting to run a stability test and had the input pins shorted or some voltage reference attached to it.   I tried recording  24 hours and when I checked it, the card only had a few points.  I tried different sample rates and such but it seemed like there was maybe a delay and timeout when recording to the card and it would just stop recording.   I want to say I tried to narrow down how long it took to fault out and it was maybe two hours.   

So just a quick test at for the AC VA mode and a short or something for the long term so we don't burn down the house. 

Thanks again for looking into this.

Offline Iagash

  • Regular Contributor
  • *
  • Posts: 69
  • Country: de
Re: EEVblog 121GW Multimeter Issues
« Reply #53 on: January 11, 2018, 08:30:29 pm »

I was checking the power using an arb, amplifier and transformer coupled.  It seems like it was limited to 50V when I started to turn it up.  You don't need a lot of current.  Obviously the meter has no problems reading just voltage.  It was tied to the power measurement.

You are completely right. The display switches to OFL when I exceed a voltage of 55V AC.

I can't test with DC as my power supply has max 32V.

Can someone else please check VA mode with more than 55V DC?

I will run the logging over night.
 

Offline Iagash

  • Regular Contributor
  • *
  • Posts: 69
  • Country: de
Re: EEVblog 121GW Multimeter Issues
« Reply #54 on: January 11, 2018, 08:49:59 pm »
They look "goldish" the pcb also looks like ENIG finish. All looks good, so I have no idea why it happens.

May be they don't reach well the pcb?

I don't think so. In that case the meter would also show odd behavior during operation when you touch the switch which is not the case.
I'd rather guess it might be an debouncing issue in the firmware.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: EEVblog 121GW Multimeter Issues
« Reply #55 on: January 11, 2018, 08:58:11 pm »
If you change to mVA/VA to manual ranges, you realize there are only 4 combinations available:

mVA mode:
5.5V + 5.5mA
55V + 5.5mA
5.5V + 55mA
55V + 55mA


VA mode:
5.5V + 550mA
55V + 550mA
5.5V + 10A
55V + 10A

Definitely the whole VA feature is underveloped. If I was logging VA, I would definitely want to use a manual range to avoid any autoranging during logging.
« Last Edit: January 11, 2018, 09:00:37 pm by amspire »
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: EEVblog 121GW Multimeter Issues
« Reply #56 on: January 11, 2018, 09:20:06 pm »
In lowZ mode the meter only starts to display a value >0 when the voltage rises to around 11.5V. When the voltage is lowered from 12V it works down to around 3V. If this is expected behaviour, it should be documented in the manual.
I haven't had a meter with a Low Z mode so I am not familiar with how it is meant to work. It seems to be trying to autodetect whether the input is AC or DC and it needs at least 12V to do this. It has no ranges so I assume the 550V range is fixed. The Low Z input resistance looks to be about 2.7K ohms which would dissipate 112W at 550V!

To limit dissipation to, say, 5W, then you are limited to around 110V maximum. At 240V, the dissipation would be 21W which is a lot of heat to be contained inside the meter. You could do a quick measurement, but a long term measurement would be really cooking the insides of the meter.

I think I will be replacing the 2.7K resistors with something much larger. 27K or higher. This seems like a dangerous feature as it is right now. Like the VA mode, very underveloped as you can see from the lack of documentation and specifications in the manual.
« Last Edit: January 11, 2018, 09:25:48 pm by amspire »
 

Offline idpromnut

  • Supporter
  • ****
  • Posts: 613
  • Country: ca
Re: EEVblog 121GW Multimeter Issues
« Reply #57 on: January 11, 2018, 09:32:01 pm »
In lowZ mode the meter only starts to display a value >0 when the voltage rises to around 11.5V. When the voltage is lowered from 12V it works down to around 3V. If this is expected behaviour, it should be documented in the manual.
I haven't had a meter with a Low Z mode so I am not familiar with how it is meant to work. It seems to be trying to autodetect whether the input is AC or DC and it needs at least 12V to do this. It has no ranges so I assume the 550V range is fixed. The Low Z input resistance looks to be about 2.7K ohms which would dissipate 112W at 550V!

To limit dissipation to, say, 5W, then you are limited to around 110V maximum. At 240V, the dissipation would be 21W which is a lot of heat to be contained inside the meter. You could do a quick measurement, but a long term measurement would be really cooking the insides of the meter.

I think I will be replacing the 2.7K resistors with something much larger. 27K or higher. This seems like a dangerous feature as it is right now. Like the VA mode, very underveloped as you can see from the lack of documentation and specifications in the manual.

My Aglient U1232A has an impedance of 3.6kOhm on its LowZ mode. I think 2.7kOhm on the 121GW is fine.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: EEVblog 121GW Multimeter Issues
« Reply #58 on: January 11, 2018, 09:43:03 pm »
In lowZ mode the meter only starts to display a value >0 when the voltage rises to around 11.5V. When the voltage is lowered from 12V it works down to around 3V. If this is expected behaviour, it should be documented in the manual.
I haven't had a meter with a Low Z mode so I am not familiar with how it is meant to work. It seems to be trying to autodetect whether the input is AC or DC and it needs at least 12V to do this. It has no ranges so I assume the 550V range is fixed. The Low Z input resistance looks to be about 2.7K ohms which would dissipate 112W at 550V!

To limit dissipation to, say, 5W, then you are limited to around 110V maximum. At 240V, the dissipation would be 21W which is a lot of heat to be contained inside the meter. You could do a quick measurement, but a long term measurement would be really cooking the insides of the meter.

I think I will be replacing the 2.7K resistors with something much larger. 27K or higher. This seems like a dangerous feature as it is right now. Like the VA mode, very underveloped as you can see from the lack of documentation and specifications in the manual.

My Aglient U1232A has an impedance of 3.6kOhm on its LowZ mode. I think 2.7kOhm on the 121GW is fine.
But how does it work? Is it a PTC thermistor and not a 2k7 resistor? I was just looking at the Fluke 114 manual and its LowZ goes to 600V and has about a 3K impedance. It better be a PTC thermistor, because 120W of heat inside a multimeter is a massive problem.
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 37626
  • Country: au
    • EEVblog
Re: EEVblog 121GW Multimeter Issues
« Reply #59 on: January 11, 2018, 09:45:43 pm »
In lowZ mode the meter only starts to display a value >0 when the voltage rises to around 11.5V. When the voltage is lowered from 12V it works down to around 3V. If this is expected behaviour, it should be documented in the manual.
I haven't had a meter with a Low Z mode so I am not familiar with how it is meant to work. It seems to be trying to autodetect whether the input is AC or DC and it needs at least 12V to do this. It has no ranges so I assume the 550V range is fixed. The Low Z input resistance looks to be about 2.7K ohms which would dissipate 112W at 550V!

No! That's not how this meter or any other Low-Z meter works.
The "resistor" is actually a PTC themristor, it's resistance increases when it heats up.
 

Offline Iagash

  • Regular Contributor
  • *
  • Posts: 69
  • Country: de
Re: EEVblog 121GW Multimeter Issues
« Reply #60 on: January 11, 2018, 09:48:48 pm »
I think I will be replacing the 2.7K resistors with something much larger. 27K or higher. This seems like a dangerous feature as it is right now. Like the VA mode, very underveloped as you can see from the lack of documentation and specifications in the manual.

No, the lowZ mode seems fine apart from the low sensitivity. It consumes around 5mA at 12V AC input voltage.
If I ramp up the voltage very quickly to 310V AC it consumes a maximum of 245mA for a very short time and drops immediately to around 3mA which is a power consumption of about 1W.
So I guess everything is save.
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: EEVblog 121GW Multimeter Issues
« Reply #61 on: January 11, 2018, 09:53:03 pm »
In lowZ mode the meter only starts to display a value >0 when the voltage rises to around 11.5V. When the voltage is lowered from 12V it works down to around 3V. If this is expected behaviour, it should be documented in the manual.
I haven't had a meter with a Low Z mode so I am not familiar with how it is meant to work. It seems to be trying to autodetect whether the input is AC or DC and it needs at least 12V to do this. It has no ranges so I assume the 550V range is fixed. The Low Z input resistance looks to be about 2.7K ohms which would dissipate 112W at 550V!

No! That's not how this meter or any other Low-Z meter works.
The "resistor" is actually a PTC themristor, it's resistance increases when it heats up.
Thanks. Even Fluke didn't bother to explain that in their manual. I guess it is one of those things that you are just meant to know.
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 37626
  • Country: au
    • EEVblog
Re: EEVblog 121GW Multimeter Issues
« Reply #62 on: January 11, 2018, 10:01:34 pm »
NOTE: Can we please limit this thread to reporting and confirmation of bugs and issues, any discussions should be on the discussion thread.
Thanks.
 
The following users thanked this post: beanflying

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: EEVblog 121GW Multimeter Issues
« Reply #63 on: January 11, 2018, 10:09:28 pm »
NOTE: Can we please limit this thread to reporting and confirmation of bugs and issues, any discussions should be on the discussion thread.
Thanks.
I think we have been doing that. If I crossed a line by not understanding the undocumented low Z mode, I apologize.
 

Offline massivephoton

  • Contributor
  • Posts: 25
  • Country: br
Re: EEVblog 121GW Multimeter Issues
« Reply #64 on: January 11, 2018, 10:19:06 pm »
I haven't had a meter with a Low Z mode so I am not familiar with how it is meant to work. It seems to be trying to autodetect whether the input is AC or DC and it needs at least 12V to do this. It has no ranges so I assume the 550V range is fixed. The Low Z input resistance looks to be about 2.7K ohms which would dissipate 112W at 550V!

To limit dissipation to, say, 5W, then you are limited to around 110V maximum. At 240V, the dissipation would be 21W which is a lot of heat to be contained inside the meter. You could do a quick measurement, but a long term measurement would be really cooking the insides of the meter.

I think I will be replacing the 2.7K resistors with something much larger. 27K or higher. This seems like a dangerous feature as it is right now. Like the VA mode, very underveloped as you can see from the lack of documentation and specifications in the manual.

The 2k7 impedance seems to be pretty standard. Fluke 115, 116 and 117 LoZ specs indicates 3k with 500pF, as the datasheet states:
http://media.fluke.com/documents/2793260_6116_ENG_A_W.PDF

As already suggested at least here...
Probably some thermistor or similar.

Instruction Sheet for Fluke SV225 says:

Specifications
Voltages up to 1000 volts continuous can be safely applied to
the Adapter without damage.
Operation Temperature
-20 °C to +55 °C (-40 °F to 131 °F)
Altitude
2,000 Meters Operating
Humidity
90 % at 0 to 35 °C (32 °F to 95 °F), 70 % at 35 to 55 °C
(95 °F to 131 °F)
Nominal Resistance
3,000 ? @ 25 °C (77 °F)

... and here ...

I would think it uses a PTC thermistor, so it draws a semi constant power over the voltage range. You find a similar one in the cheap voltstick devices.

... and way hotter here ...

https://www.google.com/patents/US8480301

thermistors are used to limit the power.


- J'ai vu une maison de cent mille francs.
- Comme c'est joli!
 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 11601
  • Country: us
Re: EEVblog 121GW Multimeter Issues
« Reply #65 on: January 11, 2018, 10:38:50 pm »
NOTE: Can we please limit this thread to reporting and confirmation of bugs and issues, any discussions should be on the discussion thread.
Thanks.
I think we have been doing that. If I crossed a line by not understanding the undocumented low Z mode, I apologize.
I think part of the problem as well is that it's not always cut and dry if the perceived problems were not that way by design.   To me, the slow auto range was by design.  It's known and not really a bug.  For the most part, the feedback appears to be going in one of two places.   

Strange on the VA mode.   I guess I just assumed it would allow for a wider voltage range.  Keep us posted on the data logging.     
 
The following users thanked this post: amspire

Offline Iagash

  • Regular Contributor
  • *
  • Posts: 69
  • Country: de
Re: EEVblog 121GW Multimeter Issues
« Reply #66 on: January 12, 2018, 10:21:23 am »
Keep us posted on the data logging.   

I did run VA logging for about 9h over night.
It produced a logfile with  130662 entries, which is about 4 lines per second.
So this seems fine.

It has a small display glitch during logging. The "k" symbol behind the record counter in the upper display is flickering.

As I stated in my previous post the bargraph is also flickering some times when two values are measured, so I guess there are some race conditions in the display update code of the firmware or they didn't use double buffering for the display.

Quote
mm@ovid:~/121gw$ head -10 18011200.CSV ; echo "[...]" ; tail -10 18011200.CSV
START,2018/01/12,01:48:47,
ID,170800000,
INTERVAL,000,sec,
,MAIN,,,SUB-1,,,SUB-2,,,Remark,
No. ,Func. ,Value,Unit,Func. ,Value,Unit,Func. ,Value,Unit,
1,DCVA,00000.0,mVA,DCV,00.0001,V,DCA,-0001.41,mA,,
2,DCVA,00000.0,mVA,DCV,00.0001,V,DCA,-0001.41,mA,,
3,DCVA,00000.0,mVA,DCV,00.0001,V,DCA,-0001.41,mA,,
4,DCVA,00000.0,mVA,DCV,00.0001,V,DCA,-0001.41,mA,,
5,DCVA,00000.0,mVA,DCV,00.0001,V,DCA,-0001.41,mA,,
[...]
130656,DCVA,01292.9,mVA,DCV,03.2999,V,DCA,0391.83,mA,,
130657,DCVA,01292.9,mVA,DCV,03.2999,V,DCA,0391.83,mA,,
130658,DCVA,01292.9,mVA,DCV,03.2999,V,DCA,0391.81,mA,,
130659,DCVA,01292.9,mVA,DCV,03.2999,V,DCA,0391.81,mA,,
130660,DCVA,01292.9,mVA,DCV,03.2999,V,DCA,0391.81,mA,,
130661,DCVA,01292.9,mVA,DCV,03.2999,V,DCA,0391.81,mA,,
130662,DCVA,01292.9,mVA,DCV,03.2999,V,DCA,0391.81,mA,,
MAX,49,DCVA,01294.1,mVA,
MIN,1,DCVA,00000.0,mVA,

Should I try something else?
« Last Edit: January 12, 2018, 10:34:24 am by Iagash »
 

Offline Iagash

  • Regular Contributor
  • *
  • Posts: 69
  • Country: de
Re: EEVblog 121GW Multimeter Issues
« Reply #67 on: January 12, 2018, 10:30:37 am »
I found something else with logging.

If you long press [MEM] in Ohm mode (no matter if AUTO or MANUAL range) the meter crashes immediately.
Only turning it OFF and ON again brings it back.


Forget  it. It was my stupidity. I had the SD-card still in my desktop.
So the bug now is:

The meter crashes if you enable logging without a SD-card present.

And another one:

Autoranging doesn't work when logging to SD-card and it stays disabled even after you switch off logging but it is displayed the whole time on the display as enabled. You have to long press the RANGE button to get it working again.

I added both issues to post https://www.eevblog.com/forum/testgear/eevblog-121gw-multimeter-issues/msg1397729/#msg1397729
« Last Edit: January 12, 2018, 10:51:54 am by Iagash »
 

Offline fanOfeeDIY

  • Supporter
  • ****
  • Posts: 412
  • Country: jp
    • YouTube Channel
EEVblog 121GW Multimeter Issues
« Reply #68 on: January 12, 2018, 12:14:58 pm »
I am not sure this is the how it is designed or a bug.

When I was reviewing the 121GW meter, almost every time I turn the knob from OFF to Low-Z, the back-light flickers.

The back-light going on for about 0.1 sec seems to consume battery every time start using the meter.

The firmware version is 1.01.

I uploaded the video of back-light flashing at the bellow link.
https://youtu.be/7Plq4HR34cI
« Last Edit: January 12, 2018, 12:35:06 pm by fanOfeeDIY »
 

Offline Iagash

  • Regular Contributor
  • *
  • Posts: 69
  • Country: de
Re: EEVblog 121GW Multimeter Issues
« Reply #69 on: January 12, 2018, 12:25:51 pm »
When I was reviewing the 121GW meter, almost every time I turn the knob from OFF to Low-Z, the back-light flickers.

The back-light going on for about 0.1 sec seems to consume battery consumtion every time start using the meter.

Confirmed with Firmware U-1.02.
It blinks always when it is switched on from any of the "OFF" positions, but it doesn't blink when turning on again from auto power off.

I guess it's related to the micro controller initialization on a cold boot.

The power consumption for this should be negligible, so I'd say it'a purely cosmetic issue, if at all.
« Last Edit: January 12, 2018, 12:54:47 pm by Iagash »
 

Offline fanOfeeDIY

  • Supporter
  • ****
  • Posts: 412
  • Country: jp
    • YouTube Channel
EEVblog 121GW Multimeter Issues
« Reply #70 on: January 12, 2018, 12:27:52 pm »
Another point I found was when I was doing the overnight SD card data logging.
one of the file had garbled data in the file, when I was testing for the thread in the forum.

https://www.eevblog.com/forum/testgear/new-eevblog-branded-multimeter-coming/msg1398530/#msg1398530

The data is corrupted as bellow in the middle of the data.
59625,DCV,049.997,V,,,,,,,,
59626,DCV,049.997,V,,,,,,,,
59627,DCV,049.996,V,,,,,,,,
59628,DCV,049.997,V,,,,,,,,
59629,DCV,                                                                                                                                                                                        59641,DCV,050.001,V,,,,,,,,
59642,DCV,049.996,V,,,,,,,,
59643,DCV,049.996,V,,,,,,,,
59644,DCV,049.998,V,,,,,,,,
59645,DCV,049.996,V,,,,,,,,

The garbled line only happened in one line out of four files of data logging.
The other three files had no issue.

The firmware is 0.01 and the SD card is the SunDisk which is included in the meter as a factory default.

I attached the original file with zip compression to work around the file size limitation on the forum.

Hope these report help improving the meter.
« Last Edit: January 12, 2018, 01:10:19 pm by fanOfeeDIY »
 

Offline firewalker

  • Super Contributor
  • ***
  • Posts: 2450
  • Country: gr
Re: EEVblog 121GW Multimeter Issues
« Reply #71 on: January 12, 2018, 12:32:26 pm »
Maybe a self test thing? Like show all segments and test the back light?
Become a realist, stay a dreamer.

 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 11601
  • Country: us
Re: EEVblog 121GW Multimeter Issues
« Reply #72 on: January 12, 2018, 01:02:58 pm »
I did run VA logging for about 9h over night.
It produced a logfile with  130662 entries, which is about 4 lines per second.
So this seems fine.

Should I try something else?


Thanks for running these longer tests!  I know a bit of a pain.   Glad to hear they have it working.   It's been some time since I did anything with the prototype and I can't remember any other real major functional problems with it.  I remember most were simple problems.   If you check my main thread where I post my results, some of the problems were discussed there but my guess is they have been corrected.   

What's the open circuit HV diode test putting out now? 

Offline Iagash

  • Regular Contributor
  • *
  • Posts: 69
  • Country: de
Re: EEVblog 121GW Multimeter Issues
« Reply #73 on: January 12, 2018, 01:16:04 pm »
What's the open circuit HV diode test putting out now?

17.332V
 

Offline joeqsmith

  • Super Contributor
  • ***
  • Posts: 11601
  • Country: us
Re: EEVblog 121GW Multimeter Issues
« Reply #74 on: January 12, 2018, 02:43:50 pm »
What's the open circuit HV diode test putting out now?
17.332V
That's interesting.  I think the prototype was under 17.   I was a bit concerned with the mux they were using had an absolute max rating.  It looked like they stayed with the same part in Dave's close up video showing the board.   Maybe they came up with a better way to protect it.   From what I saw, the mux was the weak point as far as the transient tests I run.    That mux is an HEFxxxx  located near the edge of the PCB if you wanted to pull the datasheets to have a look.   I think its an absolute max of 18 on that one.  You can drive it a little above the rail.   Easy enough to trace the path to the input pin to see how it is being protected.   Again, hard to say if you would call it a bug or issue anyway.  Maybe just something to be aware of.   


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf