Looking for 4 terminal 100 milli-Ohm precision current shunt

[perdrix]

The exact value as usual isn't hyper-critical, but as it will be used to calibrate a calibrator its value must be known to a very high level of accuracy.  Its thermal behaviour also.

I know that I can spend a huge bag of money and get one, the question is: Are there any reliable sources whose prices aren't quite so painful?

Thanks
Dave

[quarks]

depending on your budget a Burster 1240 could be a good choice

https://www.burster.de/en/products/p/detail/1240/

[HighVoltage]

Yes, I totally agree with quarks, the Burster 1240 is a really good choice.

There is one on ebay Germany right now for 95 Euro
http://www.ebay.de/itm/222599516413
That one of course will probably not have a calibration certificate.
But if you order a new one from Burster, they can supply a full calibration certificate with it.


Here are a few pictures of my 1240 100mOhm

[alm]

Define very high level of accuracy? Tempco?

[Kjelt]

I don't know the inside of the Burster but two of these precision measurement resistor boxes I have contain inside only an Isabellenhutte precision resistor.
That might be a "cheaper" alternative, buy the resistor with the specs you need and do the mounting of the connectors and boxing yourself.

http://www.isabellenhuette.de/praezisions-und-leistungswiderstaende/produkte/uebersicht/

[alm]

But then you would also have to do the calibration yourself, and it sounds like that is what Dave is after. Sending it to a cal lab may not be cheaper than buying a calibrated one.

[Kjelt]

My resistors had a calibration value on them, but you're right adding the wires and the connectors will influence the total resistance with these small values.

[alm]

And even just the soldering if you are looking at ppm-level changes (calibrating a calibrator of unspecified accuracy).

[2N3055]

Dave says:

...
The exact value as usual isn't hyper-critical, but as it will be used to calibrate a calibrator its value must be known to a very high level of accuracy.  Its thermal behaviour also.
....

That means characterised and calibrated by people with necessary equipment and knowledge.

Burster 1240 is very inexpensive (compared to other) way to that type of equipment. I've seen old resistors in dubious state go for more on E-Bay that what Burster will charge you for a new one.. If it's accuracy grade is sufficient, it's an easy choice..

Regards,

Sinisa

[ap]

Burster is 10ppm/K, Isabellenhütte is 1ppm/K. Not sure how one could use a 10ppm/K resistor in a real precision application.

[HighVoltage]

Isabellenhütte is 1ppm/K.

Thanks for this info.
Do you know, if Isabellenhütte makes lab ready 4 terminal resistor with binding posts?
Or may be they deliver only the resistors?

 

[quarks]

afaik Isabellenhuette offer IKL and RUG-Z as their calibration series.

So far from my own experience I do like Burster resistors very much.
Price, accuracy and long time stability is most likely very hard to beat.

[lukier]

Burster 1240 is very inexpensive (compared to other) way to that type of equipment.

So far from my own experience I do like Burster resistors very much.
Price, accuracy and long time stability is most likely very hard to beat.

OK, I get that it is very inexpensive, that the price is very hard to beat, but how much exactly Burster (not eBay) asks for one? No prices on their website which is always worrying.

[2N3055]

afaik Isabellenhuette offer IKL and RUG-Z as their calibration series.

So far from my own experience I do like Burster resistors very much.
Price, accuracy and long time stability is most likely very hard to beat.

Looking at datasheets for both RUG-Z and Burster, RUG-Z can handle more power and be had with lower tempco, but Burster is specified for better long term stability 0,01 % vs 0.05% for RUG-Z...
IKL-F, that is comparable one from Isabellenhuette is not better than Burster 1240 on any parameter, and worse long term stability..
Of course, those are datasheets...
In reality, i think both are good choice if they are good enough for the purpose, it only depends for what money they can be had.

[quarks]

@ lukier,
you can find prices i.e. on
https://www.calplus.de/burster-1240.html
a 1240-0,1 is 240€ plus 19% tax + shipping cost

[lukier]

@ luckier,
you can find prices i.e. on
https://www.calplus.de/burster-1240.html
a 1240-0,1 is 240€ plus 19% tax + shipping cost

Thanks! That is a good price. I might consider getting one at some point.
For up to 0.01 Ohm / 15A I have the L&N 4385 shunt box, but I've ordered a second hand 20V/50A PSU and usually something like Guildline 9230/100 is recommended for calibration. With 50A the and the Burster 1240 0.0001 Ohm resistance the dissipated power would be only 0.25W, that means 2.75 deg C increase in temperature (11 K/W). A bit pushing it, but might be doable.

[tszaboo]

Isabellenhütte is 1ppm/K.

Thanks for this info.
Do you know, if Isabellenhütte makes lab ready 4 terminal resistor with binding posts?
Or may be they deliver only the resistors?

No they dont. Also, I worked a lot with their resistors, they do require calibration often. That 0.2% drift on their RUGZ resistor is not a joke. We had a bunch of these, mounted on heatsinks, measuring up to 50A, but they were very drifty. Calibrating these are also not cheap. If they are not constantly loaded, it is even more difficult to predict, when to send them to calibrate.
That being said, the new 34461A and others have a Isabellenhütte SMV shunt, 0.1 Ohm. 0.5% after 2000h is the spec by them, and the meter said 0.23% for 1 year stability. Agilent uses it at <10% of its load capacity.

[alm]

Keysight could have aged or even selected those resistors. Especially if they specify their own part number on the BOM (which is probably not public).

[HighVoltage]

Isabellenhütte is 1ppm/K.

Thanks for this info.
Do you know, if Isabellenhütte makes lab ready 4 terminal resistor with binding posts?
Or may be they deliver only the resistors?

No they dont. Also, I worked a lot with their resistors, they do require calibration often. That 0.2% drift on their RUGZ resistor is not a joke. We had a bunch of these, mounted on heatsinks, measuring up to 50A, but they were very drifty. Calibrating these are also not cheap. If they are not constantly loaded, it is even more difficult to predict, when to send them to calibrate.

Thank you for clearing this up.
I will stay with the Burster 1240 series resistors for now for low resistance values.

That being said, the new 34461A and others have a Isabellenhütte SMV shunt, 0.1 Ohm. 0.5% after 2000h is the spec by them, and the meter said 0.23% for 1 year stability. Agilent uses it at <10% of its load capacity.

Good to know.
That is then how they increased the current rating from 3A in the 34401A to 10A in the 34461A
Do you know the Isabellenhütte part number for this one or link to a datasheet?

[tszaboo]

Keysight could have aged or even selected those resistors. Especially if they specify their own part number on the BOM (which is probably not public).

I've tested ten of these as samples. The Tempco was much much better than the datasheet said.
On the aging part, I'm not sure. The shunts I've worked with were already few years old, and they still drifted a lot. Some condition seems to reset the aging process. I never had the equipment to investigate this futher, since these were "the" reference.

Good to know.
That is then how they increased the current rating from 3A in the 34401A to 10A in the 34461A
Do you know the Isabellenhütte part number for this one or link to a datasheet?

These are the shunts:
http://www.isabellenhuette.de/fileadmin/content/praezisions-leistungswiderstaende/SMV.PDF


You are right, the 34465A measures 10A. But that uses a different shunt, R231, between relays.
https://www.flickr.com/photos/eevblog/16791788975/

The 34461A only goes for 3A and uses the Isa. SMV
https://www.flickr.com/photos/eevblog/9090323576/in/album-72157634225126018/https://www.flickr.com/photos/eevblog/9090323576/in/album-72157634225126018/

[HighVoltage]

Thanks for the links to the parts.

The older 34401A, 34410A and 34411A all had only the 3A input.

All the new models, 34461A, 34465A and 34470A have two separate current inputs.
- One for 3 A (Probably to keep the compatibility to the 34401A)and
- One for 10 A

So, may be they use two different Isabellenhütte resistors for the two different current paths?

[tszaboo]

You are right, only the 34460A has 3A range, the 34461A and bigger models have 10A range. Doh!

The other shunt they are using, I could not definitely ID them. The SMV is easy, because the shape, and I've used them, and the marking said so  . The others, I think they are CSM2512S or similar CSM from vishay foil.
http://www.vishaypg.com/foil-resistors/power-current-sensors/
But the 10A range tempco is 50ppm/K, which is not that impressive. The device I've worked on (designed) had much better spec.