Author Topic: 4-wire connections to reference resistors  (Read 4875 times)

0 Members and 1 Guest are viewing this topic.

Offline leonerdTopic starter

  • Regular Contributor
  • *
  • Posts: 167
  • Country: gb
4-wire connections to reference resistors
« on: July 11, 2016, 03:46:24 pm »
I'm planning to make myself a reference resistor; basically just putting a Vishay Z201 in a box with some banana sockets. E.g. this 10k

  http://www.mouser.co.uk/Search/ProductDetail.aspx?R=Y145310K0000V9L

I'm going to make two - a 100R and a 10k (because those are the values mouser has easily - I don't see a 10R or 1k). Less so for the 10k, but the 100R at least likely benefits from having 4-wire connections on it, so do two onto each leg in the case.

I wonder - if I mount the resistor on stripboard, I can attach two sets of wires to it, one on each side. For minimising the error, should I "cross over" the force vs. sense flows? I.e. should I put the force on two diagonal corners and the sense out of the other two, or doesn't that matter?

Relatedly, if I'm using a larger scale SMT resistor on a PCB (such as a 2010wide) and adding 4-wire sense support, should I pass the two flows in parallel, or cross them over?
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4543
  • Country: ua
    • xDevs.com
Re: 4-wire connections to reference resistors
« Reply #1 on: July 11, 2016, 03:54:32 pm »
Just twist sense+/- together into one pair and current+/- into second pair, and you likely be okay to go for hobby needs. I'd recommend 4 wire connection for both values.
Make sure you use fresh cleaned copper wires and copper terminals if you after good accuracy.
YouTube | Metrology IRC Chat room | Let's share T&M documentation? Upload! No upload limits for firmwares, photos, files.
 

Offline leonerdTopic starter

  • Regular Contributor
  • *
  • Posts: 167
  • Country: gb
Re: 4-wire connections to reference resistors
« Reply #2 on: July 11, 2016, 04:12:02 pm »
Ah, just free mount it in air? That might be better actually, because looking at the data sheet, the pin spacing is 0.15" - who thinks that's a good idea? :( Won't fit in 0.1" stripboard. Well, *maybe* I could angle it diagonally; 0.15 is very nearly 0.141... but I don't like the idea of the stress.

Speaking of - what would be best for physically mounting it within the case, if it's not mounted on stripboard which is then glued to the case? I don't know what I can sensibly do to the resistor without risking ruining its high precision.
 

Offline TiN

  • Super Contributor
  • ***
  • Posts: 4543
  • Country: ua
    • xDevs.com
Re: 4-wire connections to reference resistors
« Reply #3 on: July 11, 2016, 04:18:28 pm »
Have the suitable box, which allow air-tight seal (could be metal to provide more shielding).
Mount good copper banana plugs on case securely, and then use copper wires (e.g. strands from CAT5 LAN cable work great) to do the wiring.
There should be no tension to resistor legs or body. Good idea to drop in some humidity desiccant bags to dampen humidity effects (resistors in plastic package are not hermetic and subjected to humidity/resistance changes).

Here is worklog on one of my resistance boxes I've build, but it's more specific to my needs with triaxial ports instead of banana. But idea is the same. I'd recommend also having separate box per resistor, so you can manage it easier.
YouTube | Metrology IRC Chat room | Let's share T&M documentation? Upload! No upload limits for firmwares, photos, files.
 

Offline kripton2035

  • Super Contributor
  • ***
  • Posts: 2727
  • Country: fr
    • kripton2035 schematics repository
Re: 4-wire connections to reference resistors
« Reply #4 on: July 11, 2016, 06:16:05 pm »
Have the suitable box, which allow air-tight seal (could be metal to provide more shielding).
Mount good copper banana plugs on case securely, and then use copper wires (e.g. strands from CAT5 LAN cable work great) to do the wiring.
There should be no tension to resistor legs or body. Good idea to drop in some humidity desiccant bags to dampen humidity effects (resistors in plastic package are not hermetic and subjected to humidity/resistance changes).

Here is worklog on one of my resistance boxes I've build, but it's more specific to my needs with triaxial ports instead of banana. But idea is the same. I'd recommend also having separate box per resistor, so you can manage it easier.
are the gloves necessary to handle these components ?? =)
 

Offline Vgkid

  • Super Contributor
  • ***
  • Posts: 2732
  • Country: us
Re: 4-wire connections to reference resistors
« Reply #5 on: July 11, 2016, 06:51:01 pm »
are the gloves necessary to handle these components ?? =)
Shouldn't be, you still need to wash the flux off of the leads anyway. If you think about it, the average meter is specced to have less than 30pA of input bias current. That multiplied by 10k gives you .30uV. That is about the same voltage potential that you would have to deal with wrt thermal emf's. If we were talking about higher resistances, then leakage currents do reslly matter.
If you own any North Hills Electronics gear, message me. L&N Fan
 

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 16386
  • Country: za
Re: 4-wire connections to reference resistors
« Reply #6 on: July 11, 2016, 07:11:40 pm »
Gloves so skin debris and oil along with sweat salts do not contaminate the resistors or the connections. Causes leakage current to increase on high value resistors and also causes corrosion of the contacts so increasing residual resistance and noise on the lower values.
 

Offline Vgkid

  • Super Contributor
  • ***
  • Posts: 2732
  • Country: us
Re: 4-wire connections to reference resistors
« Reply #7 on: July 11, 2016, 08:01:42 pm »
Yep, a good cleaning with some strong ipa is all you need.
If you own any North Hills Electronics gear, message me. L&N Fan
 

Offline RobK_NL

  • Frequent Contributor
  • **
  • Posts: 253
  • Country: nl
Re: 4-wire connections to reference resistors
« Reply #8 on: July 11, 2016, 08:24:28 pm »
What is not there does not need to be cleaned away and cannot react with the cleaning agent to form some non-soluble residue.
Tell us what problem you want to solve, not what solution you're having problems with
 

Offline zlymex

  • Frequent Contributor
  • **
  • Posts: 556
  • Country: cn
Re: 4-wire connections to reference resistors
« Reply #9 on: July 12, 2016, 01:54:03 am »
.......
Speaking of - what would be best for physically mounting it within the case, if it's not mounted on stripboard which is then glued to the case? I don't know what I can sensibly do to the resistor without risking ruining its high precision.
Speaking of precision, it can be mainly decompose to
1. Long term stability. Hermetic sealed(instead of plastic sealed) is much better, with added benefit of not affected by humidity
2. TCR. Z-foil or VHP101 is better than regular foils. It can be compensated to very small value if required(10 times to 100 times better, window TCR)
3. Deviation to nominal value, not very big deal if you know the value.
 

Offline TimFox

  • Super Contributor
  • ***
  • Posts: 9088
  • Country: us
  • Retired, now restoring antique test equipment
Re: 4-wire connections to reference resistors
« Reply #10 on: July 13, 2016, 09:12:12 pm »
I have had no problems with the 0.15" lead spacing into two diagonally-spaced holes (0.141" nominal) in 0.1" PTH breadboard PCBs.
 

Offline leonerdTopic starter

  • Regular Contributor
  • *
  • Posts: 167
  • Country: gb
Re: 4-wire connections to reference resistors
« Reply #11 on: July 13, 2016, 10:16:23 pm »
Speaking of precision, it can be mainly decompose to
1. Long term stability. Hermetic sealed(instead of plastic sealed) is much better, with added benefit of not affected by humidity

People have suggested using a hermetic sealed case to put it inside. I'm curious though - what do you do about the holes for the terminal sockets? Are there hermetically-sealed sockets available? Not much point getting a sealed case if I'm just going to drill 4 big holes in it anyway.

2. TCR. Z-foil or VHP101 is better than regular foils. It can be compensated to very small value if required(10 times to 100 times better, window TCR)

Yeah; these are indeed Z-foils; the Z201 range.

3. Deviation to nominal value, not very big deal if you know the value.

Mmm that's true; some sort of calibration and recording the value might be in order.
 

Offline zlymex

  • Frequent Contributor
  • **
  • Posts: 556
  • Country: cn
Re: 4-wire connections to reference resistors
« Reply #12 on: July 14, 2016, 02:50:59 am »
Speaking of precision, it can be mainly decompose to
1. Long term stability. Hermetic sealed(instead of plastic sealed) is much better, with added benefit of not affected by humidity

People have suggested using a hermetic sealed case to put it inside. I'm curious though - what do you do about the holes for the terminal sockets? Are there hermetically-sealed sockets available? Not much point getting a sealed case if I'm just going to drill 4 big holes in it anyway.


What I implied was to buy a ready made hermetic resistors such as VHP101, VHP202Z, VHP102Z or VHP203.
Most of(if not all) the commercial standard resistors use hermetic resistors inside.
If precision is not critical, then the 4-wire connection is not necessary for 10k after all in my opinion.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf