Author Topic: US$150 Class Multimeter  (Read 39019 times)

ogdento and 1 Guest are viewing this topic.

Online EEVblogTopic starter

  • Administrator
  • *****
  • Posts: 37626
  • Country: au
    • EEVblog
Re: US$150 Class Multimeter
« Reply #100 on: May 02, 2020, 01:51:02 am »
Thanks for clarification, on photo it looks to be down on board, at the level of that screw head, touching it...

Even if it was flat on the PCB touching for the entire surface it wouldn't make a difference, the creepage distance is the same length along the PCB anyway.
That's debatable, if you look, at half of resistor creepage distance there is ground underneath the resistor. So looking at that, that is definitely half of what is resistor rated for...

No, it doesn't work like that. There is still creepage from the high voltage end to that ground plane along the PCB surface and then it has to get through the solder mask. Adding the resistor on top makes absolutely no difference to the length of that creepage path. So it makes absolutely no difference if that resistor is vertical or flat touching the PCB. I think I'll have to do a specific video on this.
« Last Edit: May 02, 2020, 01:58:30 am by EEVblog »
 

Offline 2N3055

  • Super Contributor
  • ***
  • Posts: 6407
  • Country: hr
Re: US$150 Class Multimeter
« Reply #101 on: May 02, 2020, 09:55:42 am »
Thanks for clarification, on photo it looks to be down on board, at the level of that screw head, touching it...

Even if it was flat on the PCB touching for the entire surface it wouldn't make a difference, the creepage distance is the same length along the PCB anyway.
That's debatable, if you look, at half of resistor creepage distance there is ground underneath the resistor. So looking at that, that is definitely half of what is resistor rated for...

No, it doesn't work like that. There is still creepage from the high voltage end to that ground plane along the PCB surface and then it has to get through the solder mask. Adding the resistor on top makes absolutely no difference to the length of that creepage path. So it makes absolutely no difference if that resistor is vertical or flat touching the PCB. I think I'll have to do a specific video on this.

You are absolutely correct. It is my English that fails me. Thank you for correcting me.  I went and took a look at verbiage..
Creepage distance is surface path distance (which in this case would be PCB and component surface), and as you say correctly, it would be the same.
I wanted to say clearance distance (distance of over the air discharge path), that is distance between (in this case) two pins of that resistor in divider, and pcb nets with exposed metal connected to it .

Problems are not that simple, when creepage and clearance start to combine (for instance if there is surface contamination or condensation).. In which case over that case you get a voltage gradient over high impedance path.. If I remember correctly, for 8kV overvoltage class min. distance is 8mm. When this SIP divider is mounted vertically, it is clear what is what and how you measure clearance distance . When mounted like this over PCB traces and ground plane, it's not so obvious.  But I'm sure that people that designed this meter know all this and took it into account. It's just unusual to see.

Another problem I encountered in practice was that with any component that is close to PCB (so pretty much any component ), condensation will tend to go under component, between PCB and component.
Especially if board is very clean, and if there is no surface wetting, condensation will tend to form into droplets, and by capillary action get sucked into tight spaces between board and component. 
We used conformal coating and/or TH construction with DIP cases and passives distanced off the board to promote drying. Nowadays, you just go smd, good conformal and you're golden.

But, I doubt anything like that will be a problem with multimeter in normal use.

Regards,
 

Online EEVblogTopic starter

  • Administrator
  • *****
  • Posts: 37626
  • Country: au
    • EEVblog
Re: US$150 Class Multimeter
« Reply #102 on: May 02, 2020, 10:06:38 am »
You are absolutely correct. It is my English that fails me. Thank you for correcting me.  I went and took a look at verbiage..
Creepage distance is surface path distance (which in this case would be PCB and component surface), and as you say correctly, it would be the same.
I wanted to say clearance distance (distance of over the air discharge path), that is distance between (in this case) two pins of that resistor in divider, and pcb nets with exposed metal connected to it .

Problems are not that simple, when creepage and clearance start to combine (for instance if there is surface contamination or condensation).. In which case over that case you get a voltage gradient over high impedance path.. If I remember correctly, for 8kV overvoltage class min. distance is 8mm. When this SIP divider is mounted vertically, it is clear what is what and how you measure clearance distance . When mounted like this over PCB traces and ground plane, it's not so obvious.  But I'm sure that people that designed this meter know all this and took it into account. It's just unusual to see.

You are forgetting the solder mask is an insulator.
You are also forgetting that the input protection  limits the voltage to something in the order of 1kV or less, so the distances here are way more than required, be it creepage or clearance.

Quote
Another problem I encountered in practice was that with any component that is close to PCB (so pretty much any component ), condensation will tend to go under component, between PCB and component. Especially if board is very clean, and if there is no surface wetting, condensation will tend to form into droplets, and by capillary action get sucked into tight spaces between board and component.

If you have condensation in your multimeter then you have more things to worry about than that huge gap on your hybrid resistor divider.

Quote
But, I doubt anything like that will be a problem with multimeter in normal use.

It won't be. Sorry but you are making mountain out of a molehill here, there is absolutely no practical issue here.
 

Offline 2N3055

  • Super Contributor
  • ***
  • Posts: 6407
  • Country: hr
Re: US$150 Class Multimeter
« Reply #103 on: May 02, 2020, 10:09:47 am »
It won't be. Sorry but you are making mountain out of a molehill here, there is absolutely no practical issue here.

I agree, as I repeated several times. I also don't think it will be problem in normal use.
I'm just making comments.
Regards,
 

Online joeqsmith

  • Super Contributor
  • ***
  • Posts: 11601
  • Country: us
Re: US$150 Class Multimeter
« Reply #104 on: May 02, 2020, 06:48:43 pm »
If the various dielectric strengths could not handle the transient, maybe.   If GDTs are used, you can pretty much expect to see the full peak voltage across the network.   If its a 2KV rated part and you put 8KV across it with something near the body, I could see it breaking down.   If MOVs are used, they should be fast enough and have a low enough clamp voltage to prevent any problems but you never know.  It's always fun to try it.   

Designer may not always consider the clearances.  If you watch Dave's fencer video where he compares the Brymen  against the UEI meter, you can clearly hear the 121 breaking down where the Brymen stays silent.  From my testing, the switch contacts that enable the two series PTCs do not have enough clearance and will break down at this level.  I suspect this is what we are hearing.  Then again, I could be totally wrong.  If I had that fencer, I would repeat the test. 
 
https://youtu.be/7j5NtKB0vss?t=1019
 
The following users thanked this post: Marco1971, umbro

Online EEVblogTopic starter

  • Administrator
  • *****
  • Posts: 37626
  • Country: au
    • EEVblog
Re: US$150 Class Multimeter
« Reply #105 on: May 18, 2020, 11:48:41 am »
Update: It's still UL testing which will take until end of June. They think first mass production units in September or October.
Obviously delayed due to the woo-woo virus.
« Last Edit: May 18, 2020, 11:55:44 am by EEVblog »
 
The following users thanked this post: Wytnucls, coromonadalix

Offline rsjsouza

  • Super Contributor
  • ***
  • Posts: 5974
  • Country: us
  • Eternally curious
    • Vbe - vídeo blog eletrônico
Re: US$150 Class Multimeter
« Reply #106 on: May 18, 2020, 02:31:43 pm »
Update: It's still UL testing which will take until end of June. They think first mass production units in September or October.
Obviously delayed due to the woo-woo virus.
So, no pictures before that then?  ::)
Vbe - vídeo blog eletrônico http://videos.vbeletronico.com

Oh, the "whys" of the datasheets... The information is there not to be an axiomatic truth, but instead each speck of data must be slowly inhaled while carefully performing a deep search inside oneself to find the true metaphysical sense...
 

Online EEVblogTopic starter

  • Administrator
  • *****
  • Posts: 37626
  • Country: au
    • EEVblog
Re: US$150 Class Multimeter
« Reply #107 on: May 19, 2020, 12:53:45 am »
Update: It's still UL testing which will take until end of June. They think first mass production units in September or October.
Obviously delayed due to the woo-woo virus.
So, no pictures before that then?  ::)

Sure, here you go. It hasn't been made in EEVblog Blue yet.

 
The following users thanked this post: rsjsouza

Offline rsjsouza

  • Super Contributor
  • ***
  • Posts: 5974
  • Country: us
  • Eternally curious
    • Vbe - vídeo blog eletrônico
Re: US$150 Class Multimeter
« Reply #108 on: May 19, 2020, 12:59:30 am »
Update: It's still UL testing which will take until end of June. They think first mass production units in September or October.
Obviously delayed due to the woo-woo virus.
So, no pictures before that then?  ::)

Sure, here you go. It hasn't been made in EEVblog Blue yet.

(Attachment Link)
Well played, sir. Nice that the rotary switch is switches are lower than the rubber holster.

(edit) so it is a Prova branded job. Interesting.
« Last Edit: May 19, 2020, 01:05:22 am by rsjsouza »
Vbe - vídeo blog eletrônico http://videos.vbeletronico.com

Oh, the "whys" of the datasheets... The information is there not to be an axiomatic truth, but instead each speck of data must be slowly inhaled while carefully performing a deep search inside oneself to find the true metaphysical sense...
 

Offline evava

  • Regular Contributor
  • *
  • Posts: 172
  • Country: cz
Re: US$150 Class Multimeter
« Reply #109 on: May 19, 2020, 08:15:45 am »
 

Offline Mp3

  • Frequent Contributor
  • **
  • Posts: 361
  • Country: us
Re: US$150 Class Multimeter
« Reply #110 on: May 19, 2020, 08:40:19 am »
Would be nice to have another option with usb logging in this price range besides UNI-T  :-+
High school graduate
 

Offline Brumby

  • Supporter
  • ****
  • Posts: 12281
  • Country: au
Re: US$150 Class Multimeter
« Reply #111 on: May 19, 2020, 08:52:06 am »
Update: It's still UL testing which will take until end of June. They think first mass production units in September or October.
Obviously delayed due to the woo-woo virus.
So, no pictures before that then?  ::)

Sure, here you go. It hasn't been made in EEVblog Blue yet.



Cute.
« Last Edit: May 19, 2020, 08:55:01 am by Brumby »
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 16531
  • Country: 00
Re: US$150 Class Multimeter
« Reply #112 on: May 19, 2020, 09:06:11 am »
Would be nice to have another option with usb logging in this price range besides UNI-T  :-+

Brymen has a few models...
 

Offline rsjsouza

  • Super Contributor
  • ***
  • Posts: 5974
  • Country: us
  • Eternally curious
    • Vbe - vídeo blog eletrônico
Re: US$150 Class Multimeter
« Reply #113 on: May 19, 2020, 09:20:04 am »
Something like this Prova 803 ?
http://www.prova.com.tw/product_detail.asp?seq=49
I suspect it is unreleased, as it seems to have two rows of buttons and I don't see a meter on their product line with this. But it meets the 60000 counts and have Prova's style and arrsngement of knobs.

As Dave woukd say: give me some knobs! :-DD
Vbe - vídeo blog eletrônico http://videos.vbeletronico.com

Oh, the "whys" of the datasheets... The information is there not to be an axiomatic truth, but instead each speck of data must be slowly inhaled while carefully performing a deep search inside oneself to find the true metaphysical sense...
 

Online EEVblogTopic starter

  • Administrator
  • *****
  • Posts: 37626
  • Country: au
    • EEVblog
Re: US$150 Class Multimeter
« Reply #114 on: May 19, 2020, 10:31:23 am »
Something like this Prova 803 ?
http://www.prova.com.tw/product_detail.asp?seq=49
I suspect it is unreleased

Yes, I've said several times it's unreleased, still undergoing UL testing. You won't find info on this anywhere.
 
The following users thanked this post: rsjsouza

Online EEVblogTopic starter

  • Administrator
  • *****
  • Posts: 37626
  • Country: au
    • EEVblog
Re: US$150 Class Multimeter
« Reply #115 on: May 19, 2020, 12:45:25 pm »
Would be nice to have another option with usb logging in this price range besides UNI-T  :-+

This one does not have logging. Although I think there is potential bluetooth module footprint inside, I have not checked if it actually outputs anything yet (i..e if the user could potentially retrofit a TTL to bluetooth board)
 
The following users thanked this post: Mp3

Offline Mp3

  • Frequent Contributor
  • **
  • Posts: 361
  • Country: us
Re: US$150 Class Multimeter
« Reply #116 on: May 19, 2020, 11:23:54 pm »
Would be nice to have another option with usb logging in this price range besides UNI-T  :-+

This one does not have logging. Although I think there is potential bluetooth module footprint inside, I have not checked if it actually outputs anything yet (i..e if the user could potentially retrofit a TTL to bluetooth board)

would be more than happy to spend the $ on the BT module separately, for what it's worth. The uni-t's USB-serial interface works, but is far from ideal
High school graduate
 

Online EEVblogTopic starter

  • Administrator
  • *****
  • Posts: 37626
  • Country: au
    • EEVblog
Re: US$150 Class Multimeter
« Reply #117 on: May 20, 2020, 06:53:24 am »
Just tested for serial output data from the chipset on the pin header that looks like it's for a bluetooth module and I got nothing :-(

Fixed the problem with diode mode though, they just forgot to cal it. An internal jumper enables cal mode.
 

Online EEVblogTopic starter

  • Administrator
  • *****
  • Posts: 37626
  • Country: au
    • EEVblog
Re: US$150 Class Multimeter
« Reply #118 on: May 21, 2020, 11:27:17 pm »
Battery current measured at 4mA, and cutoff voltage is 3.55V for 3xAAA's. Has to be 3.55V because there is a 3.5V reg inside. Those playing along at home can do battery life estimations.
 

Offline sequoia

  • Supporter
  • ****
  • Posts: 154
  • Country: us
Re: US$150 Class Multimeter
« Reply #119 on: May 21, 2020, 11:57:22 pm »
Battery current measured at 4mA, and cutoff voltage is 3.55V for 3xAAA's. Has to be 3.55V because there is a 3.5V reg inside. Those playing along at home can do battery life estimations.

So, shouldn't have "over voltage" problem (like BM235) when using Lithium AAA's?

 
 

Offline HKJ

  • Super Contributor
  • ***
  • Posts: 2892
  • Country: dk
    • Tests
Re: US$150 Class Multimeter
« Reply #120 on: May 22, 2020, 04:43:26 am »
Battery current measured at 4mA, and cutoff voltage is 3.55V for 3xAAA's. Has to be 3.55V because there is a 3.5V reg inside. Those playing along at home can do battery life estimations.

That would be about 250 hours with Duracell Alkaline, 300 with Energizer Lithium and about 170 with eneloop.
 

Offline bluey

  • Regular Contributor
  • *
  • Posts: 142
  • Country: au
Re: US$150 Class Multimeter
« Reply #121 on: May 22, 2020, 08:30:06 pm »
Cyrustek???
60,000 count chip is few years old, so a new model not listed? Nearest i can find is this one.
http://www.cyrustek.com.tw/spec/ES51997.pdf
« Last Edit: May 22, 2020, 08:33:41 pm by bluey »
 

Online EEVblogTopic starter

  • Administrator
  • *****
  • Posts: 37626
  • Country: au
    • EEVblog
Re: US$150 Class Multimeter
« Reply #122 on: May 23, 2020, 05:45:57 am »
Cyrustek???
60,000 count chip is few years old, so a new model not listed? Nearest i can find is this one.
http://www.cyrustek.com.tw/spec/ES51997.pdf

Nope, this one does 5 readings per second.

And I'm seriously thinking they took cues from the Fluke 77 IV, it's practically the same size, shape and case design. But it's electronics oriented and better in practically every way.

 

Offline RedSky

  • Contributor
  • Posts: 21
  • Country: au
Re: US$150 Class Multimeter
« Reply #123 on: May 23, 2020, 06:28:18 am »
Okay you have me interested.

I have an old school UNI-T UT70D and a Brymen 867S, 80,000 count & 500,000 count respectively and they still test accurately within their specs according to my cheap voltage reference, it would however be nice to have a third accurate Multimeter at times to compare against the other two if they are showing different readings and be within the birthday budget  ;)

What sort of bandwidth are we talking here 100kHz?
 

Offline Wytnucls

  • Super Contributor
  • ***
  • Posts: 3045
  • Country: be
Re: US$150 Class Multimeter
« Reply #124 on: May 23, 2020, 08:30:39 am »
When it comes to accuracy, the Brymen 867S is a 50,000 count meter (3%+2 LSD) and the UT70D, 8,000 count.
« Last Edit: May 23, 2020, 08:36:21 am by Wytnucls »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf