Author Topic: 500 ppm LCR Meter at Elektor March 2013  (Read 54974 times)

0 Members and 1 Guest are viewing this topic.

Offline Noize

  • Regular Contributor
  • *
  • Posts: 183
  • Country: gb
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #25 on: March 11, 2013, 02:14:14 pm »
I bought 10 credits this morning for £1 to get the article for that esr meter.


So as I read this, the elektor unit has closed code for the kit, I wouldn't bother and also 4 x $US7.90 for the article series

I paid 10 "elektor credits" for the article, whatever that's worth (I have a bunch from when I got something published).

Dave.
 

Offline DmitryL

  • Regular Contributor
  • *
  • Posts: 242
  • Country: gb
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #26 on: March 11, 2013, 02:26:27 pm »
+1
in the go rlc2 there are "only" 1% resistors in series with the 4052
and as I understood it the software calculates the error
so yes why did elektor use 0.01% resistors in such a configuration ?

The more precise resistors in the differentional amplifiers, the better. I used 0.1% in my RLC2, and it didn't hurt :)
BTW, when you finnish the device, don't froget to make Open, Short and Load calibration with a set of known precise  _smd_ resistors, otherwise measurements won't be good enough.
« Last Edit: March 11, 2013, 02:31:06 pm by DmitryL »
 

Offline BravoVTopic starter

  • Super Contributor
  • ***
  • Posts: 7547
  • Country: 00
  • +++ ATH1
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #27 on: March 11, 2013, 03:00:07 pm »
Thanks to all for the insights of this circuit, glad I didn't buy the credit just to download this issue.

Looking at the test frequency, wonder why they dont make up to 100 KHz as well since this is started from the scratch.  :-//



and my lcr repository is not bad at all too !
My apology, I meant the world most complete diy lcr cicrcuits repository on earth.   :-+

Offline Rufus

  • Super Contributor
  • ***
  • Posts: 2095
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #28 on: March 11, 2013, 03:03:33 pm »
Those chips are in the order of 50-100ohm on-resistance depending upon voltage, and it varies a lot with temp too IIRC.

The channel resistance of the 4052 has no effect on measurements - the design isn't that dumb.
 

Offline Rufus

  • Super Contributor
  • ***
  • Posts: 2095
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #29 on: March 11, 2013, 03:08:04 pm »
It does not use mains frequency as reference, it generates its own square signal and filters it to sine...

It is likely specified measuring at twice the local mains frequency to help reject mains pick up on the component and cables.
 

Offline tinhead

  • Super Contributor
  • ***
  • Posts: 1918
  • Country: 00
    • If you like my hacks, send me a donation
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #30 on: March 11, 2013, 03:20:01 pm »
It uses 0.01% resistors.
Accuracy is only claimed over +/-2 degC range
And I see a 0.01% 100R (and 1K, 10K, 100K) resistor in series with a 74HCT4052 mux  :-//

Dave.

they could use at least matched/selected MAX4052A, but yeah, Elektor.
I don't want to be human! I want to see gamma rays, I want to hear X-rays, and I want to smell dark matter ...
I want to reach out with something other than these prehensile paws and feel the solar wind of a supernova flowing over me.
 

Offline Rufus

  • Super Contributor
  • ***
  • Posts: 2095
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #31 on: March 11, 2013, 04:00:22 pm »
in the go rlc2 there are "only" 1% resistors in series with the 4052
and as I understood it the software calculates the error
so yes why did elektor use 0.01% resistors in such a configuration ?

You can't magic absolute accuracy out of no where. Those resistors are *the* reference in the instrument if they are not precise you would have to calibrate them by measuring external reference parts. You would probably want to use precision parts for them to get better thermal and age stability anyway.
 

Offline kripton2035

  • Super Contributor
  • ***
  • Posts: 2572
  • Country: fr
    • kripton2035 schematics repository
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #32 on: March 11, 2013, 04:38:57 pm »
in the go rlc2 there are "only" 1% resistors in series with the 4052
and as I understood it the software calculates the error
so yes why did elektor use 0.01% resistors in such a configuration ?

You can't magic absolute accuracy out of no where. Those resistors are *the* reference in the instrument if they are not precise you would have to calibrate them by measuring external reference parts. You would probably want to use precision parts for them to get better thermal and age stability anyway.
the point here is that these very precise resistors are in serie with a 74hct4052 that is certainly not 0.01% precise, nor ppm/deg that precise too
so yes you need very precise resistors to make a very precise lcr meter, but why ruin the precision just after putting these resistors ?
 

Offline Rufus

  • Super Contributor
  • ***
  • Posts: 2095
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #33 on: March 11, 2013, 04:51:44 pm »
in the go rlc2 there are "only" 1% resistors in series with the 4052
and as I understood it the software calculates the error
so yes why did elektor use 0.01% resistors in such a configuration ?

You can't magic absolute accuracy out of no where. Those resistors are *the* reference in the instrument if they are not precise you would have to calibrate them by measuring external reference parts. You would probably want to use precision parts for them to get better thermal and age stability anyway.
the point here is that these very precise resistors are in serie with a 74hct4052 that is certainly not 0.01% precise, nor ppm/deg that precise too
so yes you need very precise resistors to make a very precise lcr meter, but why ruin the precision just after putting these resistors ?

That part of the circuit is similar in the Elektor and your RLC-2 meter. The resistance of the 4052 has no effect on measurements and is irrelevant in both.

The whole circuit is similar except the RLC-2 uses a quadrature detector to measure phase and the Elektor seems to do it in software.
« Last Edit: March 11, 2013, 05:05:57 pm by Rufus »
 

Offline enz

  • Regular Contributor
  • *
  • Posts: 134
  • Country: de
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #34 on: March 11, 2013, 04:59:08 pm »
That part of the circuit is similar in the Elektor and your RLC-2 meter. The resistance of the 4052 has no effect on measurements and is irrelevant in both.

Correct, the On-Resistance of the 74HCT4052 is only in series with the input of the buffering OP-Amps. The measurement over the selected range-switch-resistor is done differentially.
It's all written in the article and clearly to see in the schematic.

Martin
 

Offline Amarbir[Lynx-India]

  • Frequent Contributor
  • **
  • Posts: 431
  • Country: in
  • Indian Dealer
    • Lynx-India - Visit Us For Not So Boring Electronic Instruments
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #35 on: March 11, 2013, 05:54:12 pm »
I can't remember any full open source project by Elektor.

That's probably one of the reasons why they turned down my uWatch project  :-//

Dave.

Dave  ,
 When You have silicon chip in your own country why do you need elektor ? .Other magazines beg borrow and steal articles from SC .A Person having a goldmine never understands the importance then a person who does not .I fancy that semtester they had few months back and now epe mag uk has it hee hee  .That was one awesome project .If you come to india stay with me  ;D
Regards

Amarbir Singh Dhillon [ Lynx-India ] , Chandigarh [ India ] - > www.lynxdealerstore.com , www.lynx-india.com
Indian Distributor For  [ Autoelectric , Sofitech , IDEOfy ,Peak Electronic Design [UK ] , Anatek And Creatronica ]
My Electronics Blog - > www.lynxchandigarh.com
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #36 on: March 11, 2013, 06:53:32 pm »
And I see a 0.01% 100R (and 1K, 10K, 100K) resistor in series with a 74HCT4052 mux  :-//
Pardon for being an enthusiast level, are you saying by using that mux defies the purpose of using those expensive resistors ? Again, CMIIW.

Those chips are in the order of 50-100ohm on-resistance depending upon voltage, and it varies a lot with temp too IIRC.

Dave.
Well, it wouldn't be the first time Elektor published a circuit which doesn't work at all.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline BravoVTopic starter

  • Super Contributor
  • ***
  • Posts: 7547
  • Country: 00
  • +++ ATH1
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #37 on: March 11, 2013, 07:29:21 pm »
The channel resistance of the 4052 has no effect on measurements - the design isn't that dumb.
Scratched my head quite hard thinking what you said here, made some rough sketch, something looks or similar like this Rufus ? CMIIW

« Last Edit: March 11, 2013, 07:33:19 pm by BravoV »
 

Offline enz

  • Regular Contributor
  • *
  • Posts: 134
  • Country: de
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #38 on: March 11, 2013, 08:19:47 pm »
The channel resistance of the 4052 has no effect on measurements - the design isn't that dumb.
Scratched my head quite hard thinking what you said here, made some rough sketch, something looks or similar like this Rufus ? CMIIW



I'm not Rufus  ;), but you got it right. That is essentially the concept used in the design.
The signal comming out of Pin13/X is buffered and the signal on the other side of the range resistors (going to the inverting input of the opamp) is buffered as well.
These buffered signals are than fed into an instrumentation amplifier.

Martin
 

Offline BravoVTopic starter

  • Super Contributor
  • ***
  • Posts: 7547
  • Country: 00
  • +++ ATH1
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #39 on: March 11, 2013, 08:45:35 pm »
I'm not Rufus  ;), but you got it right. That is essentially the concept used in the design.
The signal comming out of Pin13/X is buffered and the signal on the other side of the range resistors (going to the inverting input of the opamp) is buffered as well.
These buffered signals are than fed into an instrumentation amplifier.

Martin

Hey Martin, thanks for the confirmation ! As an EE wannabe, I learned a lot and got quite few tricks from his sleeve posts, and one of my fav from many members around here as well. ;)

And thanks again for the additional info on handling the output, really, I haven't give a thought on that part yet.  :-+
« Last Edit: March 11, 2013, 08:49:07 pm by BravoV »
 

Offline Rufus

  • Super Contributor
  • ***
  • Posts: 2095
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #40 on: March 11, 2013, 09:37:53 pm »
Scratched my head quite hard thinking what you said here, made some rough sketch, something looks or similar like this Rufus ?

I don't think you quite have it. They call it a U/I converter but it is easier to think of it as a simple op-amp buffer as below.

The op-amp forces point X to the same potential as the +ve input, the +ve input is earthed so point X is forced to be earth also. Point X is called a virtual earth. If you push current in or out of point X the op-amp will compensate and keep it at earth.

If you add resistance in the op-amp feedback path it works the same but the op-amp output voltage has to change (a lot) more. The current in the feedback path has to match the current pulled or pushed from the virtual earth. The voltage across a resistor in the feedback path will be proportional to that current.

As long as you measure the voltage across a stable part of the resistance (Rref) it doesn't matter if there is additional unstable resistance (Rswitch) in the path.


Oh, and if the meter has high input impedance it doesn't matter if there is a small unstable resistance in series with it. One half of the 4052 feeds the op-amp output to the required resistor, the other half feeds the voltage from the required resistor to the meter.
« Last Edit: March 11, 2013, 09:47:37 pm by Rufus »
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 37661
  • Country: au
    • EEVblog
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #41 on: March 11, 2013, 10:03:54 pm »
When You have silicon chip in your own country why do you need elektor ? .Other magazines beg borrow and steal articles from SC .A Person having a goldmine never understands the importance then a person who does not .I fancy that semtester they had few months back and now epe mag uk has it hee hee  .That was one awesome project .If you come to india stay with me  ;D

I offered my uWacth project to Silicon Chip first as a world exclusive, they turned it down too.

Dave.
 

Offline BravoVTopic starter

  • Super Contributor
  • ***
  • Posts: 7547
  • Country: 00
  • +++ ATH1
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #42 on: March 12, 2013, 06:26:51 pm »
Rufus, thanks a lot !  Another new entry again from you written in my notebook.  :-+

Btw, mind point out what is the flaw of my implementation please ?

Offline Rufus

  • Super Contributor
  • ***
  • Posts: 2095
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #43 on: March 12, 2013, 06:46:47 pm »
Btw, mind point out what is the flaw of my implementation please ?

Your circuit looks more like a switchable gain inverting amplifier. The leftmost resistor isn't there, and the op-amp +ve input needs to be ground.

I don't know if you have seen the elektor circuit, if not the RLC-2 one here http://kripton2035.free.fr/Resources/RLC-2_rev.1.2.pdf is very similar.

 

Offline BravoVTopic starter

  • Super Contributor
  • ***
  • Posts: 7547
  • Country: 00
  • +++ ATH1
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #44 on: March 13, 2013, 09:58:47 am »
No, I haven't seen the circuit and not interested anymore.  :P

Thanks, and any comment on that RLC-2  circuit ? I mean the design wise for such LCR measurement ?

Offline kripton2035

  • Super Contributor
  • ***
  • Posts: 2572
  • Country: fr
    • kripton2035 schematics repository
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #45 on: March 13, 2013, 11:57:26 am »
well my first comments are : it's a gigantic puzzle ...
there is no kit available, only a main board pcb, everyone makes his own pcb for the supply and the display.
this board is smd quite only so it's very small. sure it's not a project for beginner.
you must find the good components to fit on the board and there are many versions of each one.
after I make mine I will make a review of this adventure !
also people that made it are mainly russians so the comments about it are difficult to translate.
the proradio web site has more than 450 pages of forum about it... I didnt read all ;)
 

Offline BravoVTopic starter

  • Super Contributor
  • ***
  • Posts: 7547
  • Country: 00
  • +++ ATH1
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #46 on: March 13, 2013, 03:20:58 pm »
Yeah, honestly I'm a bit overwhelmed just by looking at the circuit, its has quite high component counts.  :-\

Since you're making your own pcb, are you using the existing one or you're redoing from the scratch ?

Offline kripton2035

  • Super Contributor
  • ***
  • Posts: 2572
  • Country: fr
    • kripton2035 schematics repository
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #47 on: March 13, 2013, 04:28:32 pm »
I've been making the main rlc smd board that is on my repository
then I plan to install it in a hameg 8000 empty module so there is already a power supply at the back
only have to make some capacitor-inductors filters.
I will only make a pcb for the lcd and buttons front plate. that one will be made at home.
the main rlc board fits fine in an hameg box.
 

Offline BravoVTopic starter

  • Super Contributor
  • ***
  • Posts: 7547
  • Country: 00
  • +++ ATH1
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #48 on: March 14, 2013, 01:07:14 am »
Which program are you using to redraw the circuit for the pcb ?

Offline EEVblog

  • Administrator
  • *****
  • Posts: 37661
  • Country: au
    • EEVblog
Re: 500 ppm LCR Meter at Elektor March 2013
« Reply #49 on: March 14, 2013, 01:50:55 am »
Yeah, honestly I'm a bit overwhelmed just by looking at the circuit, its has quite high component counts.  :-\

My "simple" uSupply USB has over 50 resistors on it!
Some designs just take what they take.

Dave.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf