Author Topic: 4 wire resistance measurement  (Read 4604 times)

0 Members and 1 Guest are viewing this topic.

Offline cowasakiTopic starter

  • Frequent Contributor
  • **
  • Posts: 605
  • Country: gb
4 wire resistance measurement
« on: July 23, 2018, 09:10:56 pm »
I have a Keithley 2100 which I have been using happily with my decent normal standard probes. 

I recently picked up a 3/4 wire PT100 and that is now set up and working and a few days ago picked up some 4 wire resistance probes.  The 4 wire resistance probes came with no instructions and have no labels on any of the shielded banana plugs.

There are two crocodile clips with a red and black lead running to each.  Now I thought that two wires of the same colour should go to each clip which become input and sense or am I wrong?

So based on the colours if I plug in

clip1 red into red input
clip1 black into black input
clip2 red into red sense
clip2 black into black sense

The CONT doesn't work and 2w & 4w show OVLD

If I select 4 wire resistance I get "OVLD"

----------------

If I totally ignore the socket colours and plug them in as follows:

clip1 red into red input
clip2 red into black input

Then I select 2 wire resistance I get 000.096 ohm
Then I select 4 wire resistance I get OVLD
Then I select CONT I get 0000.11 and a beep

If I now plug in

clip1 black into red sense
clip2 red into black sense

I get the same readings

----------------

If I take clip 1 and stick the red into red input and the black into black input there is no continuity.

 

Offline glarsson

  • Frequent Contributor
  • **
  • Posts: 814
  • Country: se
Re: 4 wire resistance measurement
« Reply #1 on: July 23, 2018, 09:21:48 pm »
There are two crocodile clips with a red and black lead running to each.  Now I thought that two wires of the same colour should go to each clip which become input and sense or am I wrong?.
That eBay special is not usable for four wire measurements. I don't know what  purpose (if any) they are designed for.
 

Offline cowasakiTopic starter

  • Frequent Contributor
  • **
  • Posts: 605
  • Country: gb
Re: 4 wire resistance measurement
« Reply #2 on: July 23, 2018, 09:49:04 pm »
Thanks. Might have to re-wire them
 

Offline ArthurDent

  • Super Contributor
  • ***
  • Posts: 1193
  • Country: us
Re: 4 wire resistance measurement
« Reply #3 on: July 24, 2018, 12:06:27 am »
Here is a picture from a Keithley manual that should help.
 

Offline cowasakiTopic starter

  • Frequent Contributor
  • **
  • Posts: 605
  • Country: gb
Re: 4 wire resistance measurement
« Reply #4 on: July 24, 2018, 09:10:52 am »
Here is a picture from a Keithley manual that should help.

Thanks, thats the way I thought I had wired it.  Looks like the actual wiring of the clips isn't as expected.  I checked continuity between the wires going to each clip and there was none.  I will see what the supplier says and possibly just re-wire them.
 

Online tggzzz

  • Super Contributor
  • ***
  • Posts: 21225
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: 4 wire resistance measurement
« Reply #5 on: July 24, 2018, 10:18:32 am »
I had a problem that sounds similar.

First test: try swapping each black lead and see if the display changes. Ditto red leads.

Second test: use a multimeter to measure continuity from each connector to each side of the croc clip. Choose several points on each side of each croc clip.

Then see whether the results are "random" - but could be explained by a insulating layer on one side of the croc clip's jaws. If so, demand a refund. Optionally file the croc clip teeth to remove any insulating layer - but that will have obvious disadvantages.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline cowasakiTopic starter

  • Frequent Contributor
  • **
  • Posts: 605
  • Country: gb
Re: 4 wire resistance measurement
« Reply #6 on: July 24, 2018, 10:30:25 am »
I had a problem that sounds similar.

First test: try swapping each black lead and see if the display changes. Ditto red leads.

Second test: use a multimeter to measure continuity from each connector to each side of the croc clip. Choose several points on each side of each croc clip.

Then see whether the results are "random" - but could be explained by a insulating layer on one side of the croc clip's jaws. If so, demand a refund. Optionally file the croc clip teeth to remove any insulating layer - but that will have obvious disadvantages.


Thanks for your reply.  There is no continuity between the red and black lead going to each clip.  It appears that only the two red leads go to the clips and the black leads are not connected at all!   I am going to check this further where I get a moment though.  I'm happy with 2wire measurements for virtually everything but I thought it would be worth doing seeing as I have such a nice meter!
 

Offline cowasakiTopic starter

  • Frequent Contributor
  • **
  • Posts: 605
  • Country: gb
Re: 4 wire resistance measurement
« Reply #7 on: July 29, 2018, 07:30:01 pm »
Right after a quick 2 second tear down here are a few photos.

The items itself is "UNI-T Kelvin Multimeter Leads Probes Clip Meter Test Line Cable for UT612/UT611" from ebay at about £9

So after a quick removal of heat sink and a closer look here goes....

  • Each clip is physically attached to two banana plugs
  • Each banana plug has a wire leading from it to a joint where a single cable comes out and that cable is an shielded single core.
  • The single core connects the RED banana to one of the sides of the clip
  • Each side of the clip is insulated from the other.
  • The BLACK bananas only actually go through to the shield and have NO electrical connection to the clip at all !!


Now I've spoken to the seller who is sending me a second set on the basis that these were somehow a one-off.  It would seem rather strange to do this if he thought that I would find the same again.
« Last Edit: July 29, 2018, 07:34:43 pm by cowasaki »
 

Offline rherber1

  • Newbie
  • Posts: 9
  • Country: au
Re: 4 wire resistance measurement
« Reply #8 on: August 08, 2018, 09:49:24 am »

The items itself is "UNI-T Kelvin Multimeter Leads Probes Clip Meter Test Line Cable for UT612/UT611" from ebay at about £9


The leads in the photos are definitely not "Kelvin" leads. Also, the UNI-T UT612 does NOT have Kelvin inputs - as I discovered only after receiving my meter today.

The leads in the photos are simply "screened" leads (looks like RG58U) with the outer screen cut short of the component test clip and ending in an open circuit. The banana plug connected to the screen is plugged into either one of the guard sockets on the UT612  - both guard sockets are connected to the common or ground plane inside the meter.

Kelvin leads have both conductors on each of the +VE and -VE test leads ending in a short circuit at the component test clip. One wire in each of the leads is used to drive the test signal into the component under test and is usually of heavier gauge conductor than the sense wire. The sense wire carries virtually zero current (and can be much lighter gauge) and merely "senses" the voltage across the component (for processing in the meter) and therefore imposes no additional loading across the component.

Some Kelvin leads advertised on Ebay have a fairly substantial alligator clip on the component end with the drive wire connected to one jaw and the sense wire connected to the other jaw. When the jaw is opened there is no electrical connection between the two wires but when the jaw is closed or clamped onto a component being tested both drive and sense wires are electrically connected together.

The Ebay advertisements falsely claim that they are suitable for use with the UT612/UT611 meters. Well, they can be used, as long as you don't plug any leads into the guard sockets - which makes them worse than single wire test leads due to the capacitance of the unused lead wires dangling in the air. So if you feel it is necessary to go 4 wire with the UT612, the best alternative is to use screened leads as per the photos.
 

Offline larsdenmark

  • Regular Contributor
  • *
  • Posts: 139
  • Country: dk
Re: 4 wire resistance measurement
« Reply #9 on: August 08, 2018, 11:05:30 am »
I bought these leads earlier this year.

Upon learning that they did not work at all I asked for a refund that I received without any complaints in less than 12 hours.
 

Offline Specmaster

  • Super Contributor
  • ***
  • Posts: 14483
  • Country: gb
Re: 4 wire resistance measurement
« Reply #10 on: August 08, 2018, 12:17:16 pm »
This is how your Kelvin leads should look like, a separate core from each banana plug runs through to the clip, 1 core to each half of the clip. The red clip has 2 red plugs and so on.
 
Who let Murphy in?

Brymen-Fluke-HP-Thurlby-Thander-Tek-Extech-Black Star-GW-Avo-Kyoritsu-Amprobe-ITT-Robin-TTi
 

Offline ArthurDent

  • Super Contributor
  • ***
  • Posts: 1193
  • Country: us
Re: 4 wire resistance measurement
« Reply #11 on: August 08, 2018, 01:14:06 pm »
I bought a similar pair of kelvin clips to those shown above but the ones I found on Ebay that were closest to what I wanted had the wrong connectors at the DMM end. The ones I found did have 2 separate wires going to each clip which was correct but the HP DMM required banana plugs.  I wanted it to be simple to just plug the leads in and have the wires in the correct position.

I replaced the 4 individual  connectors the clip set came with  and used two double banana plugs with one of the wires from the red clip going to the red double banana plug and the second wire from the red clip going to the black double banana plug.  The black kelvin clip has one of its wires going to the red double banana plug and the second wire from the black clip going to the black double banana plug. This gives me the corrections I had shown in post #3. On my HP 3457A the 5-way posts for input have different spacing between horizontal posts than it does for the vertically aligned posts so the double banana plugs can only plug in the vertical post which have the correct 0.75" spacing.

The possible problem I see with the previous photo of clips using 4 individual banana plugs is that both plugs from the red clip are red and you actually want one lead to go to red on the meter and the second red plug to go to black on the meter to have the corrections correct. If I had that clip set I would switch 2 of the plugs so the colors made sense to me and minimized the chance of connecting the plugs incorrectly.
 

Offline rherber1

  • Newbie
  • Posts: 9
  • Country: au
Re: 4 wire resistance measurement
« Reply #12 on: August 08, 2018, 01:33:13 pm »
Replies #10 and #11 are both correct examples of Kelvin test leads.

The Kelvin leads I bought are https://www.ebay.com.au/itm/LCR-Meter-Test-Leads-Lead-Terminal-Kelvin-Clip-Wires-for-UT612-UT611/332008183641?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2060353.m2749.l2649

As we have discovered from the UT612 teardown thread the internal photos of the UT612 clearly show that the inner guard sockets are connected directly to the ground plane and as such this meter is not capable of using Kelvin leads.
https://www.eevblog.com/forum/testgear/review-and-tear-down-of-uni-t-ut612-lcr-meter/msg199378/#msg199378
 

Offline cowasakiTopic starter

  • Frequent Contributor
  • **
  • Posts: 605
  • Country: gb
Re: 4 wire resistance measurement
« Reply #13 on: August 08, 2018, 02:28:31 pm »
This is how your Kelvin leads should look like, a separate core from each banana plug runs through to the clip, 1 core to each half of the clip. The red clip has 2 red plugs and so on.
 

Yes, that’s what I expected to see. They’ve sent me a second pair which are the same as the first so I’m going to rewire one pair to four banana plugs and one pair to the Philips din plug standard.
 

Offline cowasakiTopic starter

  • Frequent Contributor
  • **
  • Posts: 605
  • Country: gb
Re: 4 wire resistance measurement
« Reply #14 on: August 08, 2018, 02:31:06 pm »
Now to find some good quality banana plugs! I need quite a few as I need four for my PT100 probe as well.
 

Offline Specmaster

  • Super Contributor
  • ***
  • Posts: 14483
  • Country: gb
Re: 4 wire resistance measurement
« Reply #15 on: August 08, 2018, 02:35:02 pm »
The leads I show my post #10 are simple to use and depending what meter you are using them on it will be labelled in one of 2 ways or maybe both ways, LO and HI or maybe Red and Black, but either way the black leads go into the black or LO sockets and the red plugs into the HI or red sockets its as simple as that. There is polarity difference between either of the reds or blacks, it just simple connects one of the grips on the clip to input and the other to the sense connection. When these clips are connected to a resistor, they both connect at the same location on the resistor
Who let Murphy in?

Brymen-Fluke-HP-Thurlby-Thander-Tek-Extech-Black Star-GW-Avo-Kyoritsu-Amprobe-ITT-Robin-TTi
 

Offline rherber1

  • Newbie
  • Posts: 9
  • Country: au
Re: 4 wire resistance measurement
« Reply #16 on: August 09, 2018, 11:16:31 am »
Now to find some good quality banana plugs! I need quite a few as I need four for my PT100 probe as well.

If you don't want esoteric gold plated plugs I can recommend the Radiall 4mm insulated banana plugs shown on page 20-15 of this document.
https://www.radiall.com/media/document_library/COAX_Section20_2017_Banana.pdf

The arched spring maintains good tension over many years. I still have a swag of red and black from a batch I purchased around 1989.
 
The following users thanked this post: cowasaki

Offline GreyWoolfe

  • Supporter
  • ****
  • Posts: 3652
  • Country: us
  • NW0LF
Re: 4 wire resistance measurement
« Reply #17 on: August 09, 2018, 01:15:34 pm »
Thanks for this thread!  I made a set of Kelvin clips for my GW Instek 851A.  I decided to actually check them.  Found out by the pictures here that I had assembled them wrong and they didn't work.  Now they work as they should.
"Heaven has been described as the place that once you get there all the dogs you ever loved run up to greet you."
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf