4-wire short - 2020

[felixd]

Howdy   

4-Wire Short replaces:
* Keithley General Purpose 4-Wire DMM Shorting Plug Model 8620 and 8610
* Fluke 884X Short 4-Wire Short
* HP, Keysight 34172B Calibration Short for Digital Multimeters

Features:
* Gold plated PCB - ENIG (RoHS)
* 18 mm long, 4 mm gold plated brass banana plugs with twisted springs from Stabuli (MultiConnect) with 0,2 mOhm contact resistance(springs are made from rigid hard-drawn copper alloy sheet)
* M4 screws / holes in PCB have 4.3 mm diameter
* materials are selected to have minimal thermal EMF voltage
* holes separation: 19,05 mm x 19,05 to 34,50 mm (should fit almost any bench DMM)

It's just great, universal 4-wire DMM input short. I am looking for people who might be interested in buying one of those. Main advantage of this project are better bana plugs that are going to be used. They are longer (18 mm) and have slightly better contact forces/lower resistance (twisted spring)






Thank to Jay_Diddy_B we were able to test 3D printed prototype on many different DMMs
More pictures are in his response to this thread: https://www.eevblog.com/forum/projects/4-wire-short-2020/msg3259760/#msg3259760

Example from HP 34401A:








Edit: We have v1 release
This project is "under design" since I am waiting for Your requests for PCB variants. Should we make one universal PCB or ones with exact plug separaion? What is your opinion?

[ArthurDent]

One modification I've made on banana pins to be sure there is good contact is to slide the springy sleeve thingy to the end where the wire connects and solder the sleeve just at that one end. It is important to slide the sleeve to one end before soldering or the sleeve may not compress properly. I found with some pins that may be of lesser quality or contaminated that there wasn't the best contact between the sleeve and the inner pin. I first noticed this on some leads that were rated 20A (and had 254 strand wire) That I was testing at rated load that had slightly higher drop than I expected. If I were using these leads with a power supply I'd use remote sensing to compensate for any drops but I wanted to see how the leads reacted to carrying 20A.

In this application and with good quality gold plated pins like those shown, and basically no current flow, this probably isn't a real problem.

[eblc1388]

I first noticed this on some leads that were rated 20A (and had 254 strand wire) That I was testing at rated load that had slightly higher drop than I expected.

You would think they would use brass for the inner pin?

Check if the inner pin is not made of iron, as it will cause excessive voltage drop.  I just binned some of mine which stuck fast to a magnet.

[ArthurDent]

...You would think they would use brass for the inner pin?...

When I say the leads had “slightly higher” drops I was talking of the scale that I’m measuring on my LTZ1000 10.00000VDC voltage reference where I had to be concerned about which end of a 3” long wire I sensed the voltage from. For average use I’d never notice the difference but I was curious to know what the lead resistance and voltage drop of these 20A leads were.

The high current leads I bought have banana pins that are not steel and use 254 strand silicone insulated wire that is very flexible. Most of the lower power leads I have are Pamona which are high quality and do not have steel in the banana plugs.

Long ago I had some ultra cheap alligator jumpers that had 7 strand untinned copper wire that wasn’t even soldered to the clips. The insulation on the wire was just stripped and the bare strands bent back over the insulation and then lightly crimped into the clip. The boot over the clip was stiff plastic and you needed pliers to pinch hard enough on the clip to open them. If I pulled on the wire while holding the clip the wire would easily pull out of the clip. As you might imagine the resistance of these short leads was quite high. I got rid of those leads the same day I picked them up at a flea market.

Back on topic, for the board used for the shorting plug, I like the one with the slotted holes that would be universal. I would like to see the 'HI' and 'LO' screened on that board along with the broad lines between holes to visually indicate that all pins are  connected together.

[bdunham7]

I'd like a version that is not actually shorted so that other things can be made, or it can be shorted if you wish.

I'd also second what Arthur Dent said about the rotating bits, I'd rather have a good-quality fixed banana plug like the old Pomonas.  I always get a little bit of resistance and noise from the spinners after a while, even good ones.  I solder them as well, but I always feel I might be giving up a bit in thermal quality when I do that, especially with gold plated.

[felixd]

Should do the job. Contacting manufacturers for gold plated version.

[felixd]

Updates: PCB holes separation have been corrected a bit. They should now easily handle:

X = Vertical
Y = Horizontalt
* X = 19.05, Y = 19.05 mm (0.75'' x 0.75')' - HP 34401A, 34410A, 34460A, Keithley 2000/2001/2002/2010/2015, Advantest R6581T, Rigol DM3068, Siglent SDM3065X
* X = 19.05, Y = 22.225 mm (0.75"x 0.875") - HP 3478A
* X = 19.05, Y = 25 mm (Datron 4708)

EDIT: * X = 19.05, Y=34.5mm -> Datron 1072 (thanks "doktor pyta" ) ... now it's time to think if we can fit those all sizes on one PCB

I have also contacted Stabuli, they give 1 year warranty for their products. Catalog states also that they should survive 1 million mating cycles Well...for sure I should receive some free samples for all those ads I am doing for them  

[Jay_Diddy_B]

Hi,

This is a picture of an official Keithley 4W-short:



Notice how the tracks are placed on the board. This is arranged to measure the resistance of a wide track that is very short.

This is from this message:

https://www.eevblog.com/forum/testgear/restoration-glory-of-keithley-2001-dmm/msg999334/#msg999334

In this message I built my own version.

Jay_Diddy_B

[Jay_Diddy_B]

Hi,

When you build the short like Keithley did there is a right way and wrong way to use it.

Wrong way




Right Way




Detail

[ Specified attachment is not available ]


Regards,
Jay_Diddy_B

[Cerebus]

It's a 4-way short for establishing a zero input for offset calibration, not a 3000ppm/C resistance standard.

Assuming the worst case 23.5 resistance, and a (worst case for a decent meter) 1 nA leakage current heading in exactly the wrong direction, that will result in a 23.5 nV offset error when using it as a short. That'll be less than 1 LSB on anything less exotic than an HP3458A. I personally wouldn't copy the Keithley style layout, I can't see any good reason for it^*, and would personally opt for a solid pour (faster thermal equilibrium would be my theory for this), but I also can't see the Keithley design realistically causing any practical problems as a 4-way short.

^*If someone does know of a good reason why Keithley did it that way, please speak up. It looks like a deliberate choice, and I've seen a lot of very fussy, careful PCB layout in Keithley kit, so it'd be unlike them to just chuck any old layout at the board. Or, it's so non-critical in design that they just let the layout junior get on with the job with a spec of "just short all four pins together".

[thermistor-guy]

...
^*If someone does know of a good reason why Keithley did it that way, please speak up. It looks like a deliberate choice, and I've seen a lot of very fussy, careful PCB layout in Keithley kit, so it'd be unlike them to just chuck any old layout at the board. Or, it's so non-critical in design that they just let the layout junior get on with the job with a spec of "just short all four pins together".

Fluke has a good discussion of shorting devices for high-resolution DVMs here:

https://au.flukecal.com/literature/electrical-calibration/watch-out-those-thermoelectric-voltages-cal-lab-journal-reprint

two design goals for the device are low thermal EMFs, and low thermal mass. The latter means the temperature difference between the shorting device and the DVM quickly settles, but not to 0. From the article, discussing a setup with an 8.5 digit DVM:

It should be noted that the terminals of the DMM are at a higher temperature than room temperature, so there will always be a thermal condition to be aware of. In this case, the input thermals measured about 29 ˚C. Due to the relatively low thermal mass of the shorting PCB the short temperature comes to temperature equilibrium within about a minute. After thermal stabilization a zero is performed on the DMM. Excluding measurement environment noise, over a twenty minute period, the stability of the zero should be six least significant digits in this particular setup.

Fluke recommends gold flash on PCB traces, to prevent oxidation (hence lower thermal EMFs). The Keithley and Fluke approaches differ greatly.

[felixd]

Since I haven't provided public link yet to the project sources and notes I will do it now:

This is my dirty development and notes branch! WIP - Work in Progress one but hold some more information and good links.

https://gitlab.com/metrology101/hardware/4-wire-short/-/tree/felixd

Once project is finished (first release, PCB manufactured) it will be merged into master branch.
Right now I am more focused on physical board shape. We will focus on traces once, first step is done I have some nice idea btw for this project. Stay tuned

BTW, https://www.keysight.com/en/pd-1108290-pn-34172B/calibration-short-for-the-34401a-34410a-34411a-l4411a-digital-multimeter?cc=US&lc=eng

Does it also shorts Current input with LO? What's the good (any?) reason for doing that?

[Cerebus]

...
^*If someone does know of a good reason why Keithley did it that way, please speak up. It looks like a deliberate choice, and I've seen a lot of very fussy, careful PCB layout in Keithley kit, so it'd be unlike them to just chuck any old layout at the board. Or, it's so non-critical in design that they just let the layout junior get on with the job with a spec of "just short all four pins together".

Fluke has a good discussion of shorting devices for high-resolution DVMs here:

https://au.flukecal.com/literature/electrical-calibration/watch-out-those-thermoelectric-voltages-cal-lab-journal-reprint

two design goals for the device are low thermal EMFs, and low thermal mass. The latter means the temperature difference between the shorting device and the DVM quickly settles, but not to 0. From the article, discussing a setup with an 8.5 digit DVM:

It should be noted that the terminals of the DMM are at a higher temperature than room temperature, so there will always be a thermal condition to be aware of. In this case, the input thermals measured about 29 ˚C. Due to the relatively low thermal mass of the shorting PCB the short temperature comes to temperature equilibrium within about a minute. After thermal stabilization a zero is performed on the DMM. Excluding measurement environment noise, over a twenty minute period, the stability of the zero should be six least significant digits in this particular setup.

Fluke recommends gold flash on PCB traces, to prevent oxidation (hence lower thermal EMFs). The Keithley and Fluke approaches differ greatly.

Yeah, I perhaps should have been clearer. I was thinking in terms of reasons for the specific trace layout that Keithley choose, with that anaemic trace connecting two much heavier traces. I can't see any logic behind that particular choice. In terms of thermal mass, the pins are considerably more significant than the traces so simply reducing the mass of copper on the board can't be the driving force in the decision, and they'd have made all the traces very thin if it was.

[felixd]

Updates, updates, updates.
There was a request in the beginning of this journey to fit Datron 1072 inputs with this project. I believe it has been done!

By a mistake I took 25 mm as vertical separation (it's from Datron 4708). Real distance is 34,5 mm. This is being reflected in this update. Final mechanical layout I truly believe 

Now it's to do quickly redraw it in KiCAD. I will post an update soon

Board will fit HP 34401A. If you have any objections, You think that this PCB won't fit your DMM please let me know while there is still chance to make changes.

Stay tuned.

[Jay_Diddy_B]

Hi,

Here are some files for those playing along at home.




I drew 1/4 and then mirrored it twice to get the full design.




I have attached the Fusion 360 file and the STL file.

I have 34401A, 3457A, K2001 and Datron 1281. I will print and try.

Regards,

Jay_Diddy_B

 eevblog_felix_4W_short.zip (68.79 kB - downloaded 64 times.)

[Jay_Diddy_B]

Hi,

Here are some photographs from a non-functional prototype.

Assembly








Datron 1281 26mm




HP Agilent 33401A 19.05 mm




Keithley 2001 19.05mm




HP 3457A 22mm




It looks okay to me, at least on the front terminals.

Regards,
Jay_Diddy_B

[felixd]

Jay_Diddy_B Thats great    Thanks a lot. Can you also check horizontal installation in 34401A (front and rear)? Rear is specially need as we are "tight" there

My F360 files:
https://a360.co/3jpQ9Rr - board
https://a360.co/3nfaBH2 - aseembled
https://a360.co/3in6mpd - assembled - sandwich

[Jay_Diddy_B]

felixd and the group,

I just did aa test on the rear terminals of a HP 34401A:




It looks okay.

I also joined the two Fusion 360 models together, to check that they are the same:




They are the same.

Regards,
Jay_Diddy_B

[felixd]

Thanks a mill! So my caliper and sheet of paper measurements were correct 

Let's call this mechanical drawing as done. I will post more updates in the morning (it's 1 AM in Poland now).

Cheers and thanks again for help.

[felixd]

FCC, CE, RoHS marks placed. Now we can go for mass production.

Anybody in need of such a solution we are fighting here for last few days ?

[Jay_Diddy_B]

Hi felixd and the group,

I took a little time today make a couple of prototypes using my LPKF Protomat c60.

3D Model from KICAD




Boards

I made two boards. One has the obvious X pattern. The other is in the Keithley style H pattern.




Which do you think will be better?

Regards,
Jay_Diddy_B

[coppercone2]

Why not just get a thick copper block, then drill and tap 4 holes in it to make the short, without the PCB?

If you use thermal shielding compound (its like wet tissues), you can braze the banana jacks.

I think it would be better because then you don't need to buy a PCB to make it. It should work as one piece and spread heat better also, from touching it, then you can add bakelite handles to make 'handling' insulators. I think clamping on epoxy is always going to be less stable then solid metal, I am just not sure about how thermal EMF looks like, but if its built heavy, it should equalize really well?

[Vgkid]

@Coppercone2
That would work as a low resistance, but due to its massit would never reach thermal equilibrium, and would induce a large thermal emf.

[mawyatt]

Early this year we developed a few of the direct plug-in type PCBs for use with our HP34401A, Agilent 34401A and KS 34465A. These are 42.3mm square, have 19.05mm plug spacing in both directions, and fit the front and rear terminals of the mentioned DMMs.

Best,
 

[Jay_Diddy_B]

Hi,

I am going to share some measurements that I made using an IET LOM-510A Micro-Ohmmeter.

Link: https://www.ietlabs.com/lom-510.html

This may be one of the best instruments available for measuring very small resistances.

The 4W-shorts are directional. The best results are obtained if you connected

Source + to Source - with a thick trace

Sense+ to Sense- with a thick trace

a short, wide trace connecting the two together.

I am going to call this the preferred orientation. The other orientation I will call non-preferred.

Keithley Design

Preferred Orientation




This reads 0.000 m

This make sense, you are measuring a very short, very wide track.

Non-Preferred Orientation




This measures 20.23 m

You are measure a long, thin track.

Universal Design (Keithley Style)

Preferred orientation




This is a very wide, very short track.

Non-Preferred Orientation




28.17 m\$\Omega\$. This makes sense, the track is a little longer than the design above.

Felixd Design

Preferred orientation




4 u (4 Micro Ohms)

Non-Preferred Orientation




1.389 m

The 'X' pattern is not as sensitive to the orientation as the Keithley 'H' style design. But, it is not as good as the Keithley 'H' in the preferred orientation.

Note:

The Datron 1281 requires a different 'H' design than the HP 3457A. The 'H' has to be rotated 90^o.

Regards,
Jay_Diddy_B