Author Topic: Has anyone else seen this paper that claims to have impoved the uCurent?  (Read 414 times)

0 Members and 1 Guest are viewing this topic.

Offline AE7OO

  • Regular Contributor
  • *
  • Posts: 59
  • Country: us
Greetings,

While I was gathering information on methods of  current measurement, I ran across this paper.  It was published in TELKOMNIKA Indonesian Journal of Electrical Engineering in January 2014.  A link to where I got it from is at the bottom.

* uCurrentPaper.pdf (422.87 kB - downloaded 59 times.)

It looks like they tore bits and pieces of the uCurrent, slapped it together and sent it in.   
It looks like they even call it the uCurrent.  :wtf:

A brief glance has me  scratching my head over things.  Some of the descriptions just don't make sense. 
As an example of the hardware, it looks like they are using 1/2 of a MCP6002 as a comparator(for display of battery status),  open loop, with a Si2302 MOSFet on the output (along with a LED).  I don't know, but it really looks like that thing will at a minimum ring like a bell.   Another thing, if your going to use 4 alkaline batteries,  why bother with the virtual ground?  Going with a +/-3 system will cut more than a couple of items from the BOM.  And in the place of all of that, go with a monitor chip that let's you know when your batteries are low.

The first citation is to what I guess is the user manual for Dave's uCurrent from back in 2006.  So it's obvious they looked at it.

GB - AE7OO

BTW, I searched via both Google and the forum to see if anyone else had seen this.... 8)

https://www.semanticscholar.org/paper/Accurate-Low-Current-Measurement-Circuit-for-and-Geng-Liu/185abc0a03d14931b5f288374e333b91335763cc
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 5621
  • Country: gb
Re: Has anyone else seen this paper that claims to have impoved the uCurent?
« Reply #1 on: January 27, 2020, 12:55:30 pm »

It looks like they tore bits and pieces of the uCurrent, slapped it together and sent it in.   
It looks like they even call it the uCurrent.  :wtf:



It's far from being a unique circuit configuration, it's just a shunt and gain stage with split rail battery supply.

He also credited Dave, among others, in the References section. Nothing wrong with that.


EDIT: It bears little resemblance to Dave's circuit, It's single stage, uses an LDO (better supply rejection) and Opamp for rail splitting rather than a dedicated splitter IC.

That block isn't what I would describe as a true I/V converter though.


P.S. I wish someone would get around to putting some sensible input protection on the Opamp input, 100R (R7) is even worse than Dave's 270R.
« Last Edit: January 27, 2020, 01:53:21 pm by Gyro »
Chris

"Victor Meldrew, the Crimson Avenger!"
 

Online SiliconWizard

  • Super Contributor
  • ***
  • Posts: 5344
  • Country: fr
Re: Has anyone else seen this paper that claims to have impoved the uCurent?
« Reply #2 on: January 27, 2020, 04:18:20 pm »
Couple notes...

As you said, since they are going for a 5V regulated supply, they'll need several batteries in series (or an external PS?). Then they could as well go directly for a symmetric supply instead of a virtual ground. Performance would be better with less to worry about. The whole point of Dave's design was to be able to run on a single Li cell (3V).

Second note, they claim a bandwidth of 1MHz. I dunno how they'd get that as the rated GBW of the MAX4239 is 6.5MHz and they have implemented a fixed gain of 50. You won't get 1MHz out of that configuration. Quoting: "The I/V converter measures the electric current with  uA and nA selectable ranges in the frequency range from direct current to 1MHz." Really?

Third note: they claim a typical use for "Accurately measuring the sleep and operating current of a microcontroller". Problem with this is that in such a case, you're frequently dealing with currents in the µA range (or less) in sleep mode, and sometimes tens of mA in run mode. The smallest shunt they have is 20 ohm. That's 1V burden voltage @50mA. Like, yeah, sure, that will work fine. Could be a problem. It's a very common issue that they have not addressed here IMO. (Additional note: from table 1 "uA  1mV/uA  ±2200uA" the max range is ±2200uA anyway. How are you going to use this to measure the "sleep and operating current of a microcontroller" in real world? Come on.)

So yes, it's largely inspired by Dave's work, but they don't even hide that, it's the first reference they list. It's not much of an improvement though IMO (I'd say the opposite actually) and I wonder how this could warrant writing a scientific paper about it.
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 5621
  • Country: gb
Re: Has anyone else seen this paper that claims to have impoved the uCurent?
« Reply #3 on: January 27, 2020, 04:23:06 pm »
Yes, I have to agree, it is a poor quality paper, especially for 5 people. It would be interesting to know what else they publish in that journal.
« Last Edit: January 27, 2020, 04:24:40 pm by Gyro »
Chris

"Victor Meldrew, the Crimson Avenger!"
 
The following users thanked this post: tooki

Offline AE7OO

  • Regular Contributor
  • *
  • Posts: 59
  • Country: us
Re: Has anyone else seen this paper that claims to have impoved the uCurent?
« Reply #4 on: January 27, 2020, 09:02:54 pm »

I was not trying to claim they copied it, but rather that they were inspired by it, sort of, kind of, maybish.
However, they do refer to their device as the uCurrent in more than place.

At what level would people talking throw off  a current measurement?   I quote:
"An  additional  feature  is  the  nA  current  range,  which is not found in most multimeters. Measuring the current with probes is not convenient for oscillographs.  Such  as  speaking  voice  and  human  body  movement  caused  by  vibration  will  affect the measurement instrument." 

I've measured nA with no problem and even higher pA(Using a circuit from NS using a LMC660), and at no time did I notice talking/people moving as being a problem.  Having the AC kick on was a disaster, though... :palm:

To be truthful, that paper looks like something thrown together to justify the money they were given.  What they produced was from what I can tell lower quality in all aspects compared to Dave's uCurrent.
And they had his design and reasoning right in front of them.

 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf