Shielded cable for DIY Kelvin clips for DE-5000 LCR meter

[omgfire]

Greetings.

I have ordered Der Ee DE-5000 LCR Meter and some Kelvin clips to mod TL-21: replace alligator with Kelvin clip (for 4 wire measurement) and replace short cables with longer ones. I planned to use speaker wire. But according to Dave's video, TL-21 uses shielded cables. I looked at datasheet for Cyrustek ES51919/ES51920 and it declares necessity of shielding:

The test leads for DUT should be as short as possible. If long extended cable is used, the guard shielding is necessary.

There is report at UT612 thread that probes without shielding did not pass calibration. Is parasitic capacity of unshielded cables big enough to fail calibration?

Question 1: How guard (5-th wire) works on Cyrustek based LCRs? Is shielding for isolating "plus" from "minus"? Is shielding for isolation force from sense? Does force and sense have common shielding or separate?

I haven't found variety of shielded cables in my cable drawer.  Only shielded stranded twisted pair with thin cores.

Question 2: Can sense and force wires be twisted or untwisted or indifferent?

4 Wire sensing ignore voltage drop at force wires, so it is possible to use thinner wires and not sacrifice measurement accuracy.

Question 3: What is appropriate resistance (diameter) for cables for Cyrustek based LCRs?

DE-5000 measures up to 100KHz. So wavelength is about 3km. Since cables will be within 0.3m-1m, I would assume that this LCR meter do not need proper coax cable with constant conductor spacing and specific characteristic impedance. Any shielded cable will work. Right?

I have looked at ebay for shielded stranded cables and I found these:

ID 1.3mm, OD 2.9mm. Area 1.32mm^2. Is PTFE rigid enough to make this cable poor choice for Kelvin probes?

H155 Core 19 x 0.28mm. OD 5.4mm. Area 1.17mm^2. Is it too thick for 5 wire measurement?

RG174 Core 7x0.16 mm. OD 2.55mm. Area 0.14mm^2.

RG178 Core 7x0.1 mm. Will 0.05mm^2 be enough?

Silicone AWG 24. It should be flexible, but overseas shipping cost cancels this option.

So, the last question 4: which cable should I use? 

PS: sorry for long post 

[PedroDaGr8]

The tweezers have a pretty large diameter wire on them. Do you want me to take measurements of wire diameters for you?

[omgfire]

The tweezers have a pretty large diameter wire on them. Do you want me to take measurements of wire diameters for you?

Thanks, but I more interested about how cable constructed than outside diameter. Perhaps, I should wait for delivery, and then I'll see is the cable twisted or not and how shielding is done (separate or common).

[robrenz]

Single shield over both wires.  Twisted is optional. Very small gage probably around 26  28 Gage wire is used.

Guard or shield  From the Agilent impedance handbook:
"The guard terminal is the circuit common of the auto-balancing bridge and is connected to the
shields of the four-terminal pair connectors. The guard terminal is electrically different from the
ground terminal, which is connected directly to the chassis (Figure 2-12 (b).) When the guard is
properly connected, as shown in Figure 2-12 (c), it reduces the test signal's current but does not
affect the measurement of the DUT’s impedance (Zx) because Zx is calculated using DUT current (Ix.)"

[omgfire]

Single shield over both wires.

Excuse, non English speaker. What does "single" word mean in your sentence? Like single woman or like single bedroom.
(i) Separate shields, like 4 coax for force plus, force minus, sense plus, sense minus.
(ii) One common shield, like 2 shielded twisted pairs: force plus and sense plus inside one shield; force minus and sense minus inside another shield.
In second case force influences sense.

Twisted is optional. Very small 28 Gage wire is used.

Thank you.

[robrenz]

The DE-5000 attachments are (ii)

[omgfire]

robrenz, thank you. 

[Electro Fan]

Surfacing a very old thread.

What is the impact of the shield (vs. no shield) in Kelvin test leads?  With the shield maybe the same leads and clips that could otherwise measure reliably to milliohms without the shield might measure more reliably to microohms with the shield, or is there some other purpose?  Thx