Metrology Meeting in Stuttgart/Germany

[hwj-d]

Hello,

what do you expect?
-lots of people staring at 8.5 digit displays looking wether the last digit will flip (or not)?
...

Andreas

Not me 

[hwj-d]

Yust arrived at home (Herne), sunday 07:30 local time 
(On weekends, Deutsche Bahn is even worse than building a german airport.)

Thanks a lot. Got to know a lot of really competent and nice people. The meeting was a great success in many all the ways.

 

[dietert1]

I am very satisfied, too. Thank you André and to all others involved.
I could learn from the talks and participated in the measurement session. Time passed rapidly.
Did some check measurements when returning to office. Appears like i brought back a +/- 2 ppm (plus uncertainty of Fluke reference) calibration for voltage standards and lab DMMs.
We also measured some resistor boxes and wrote long lists of readings. In a preliminary evaluation resistance uncertainty at 1K and 10K appears to be +/- 4 ppm.

Arjens parts helped increasing redundance for bringing the standards back to our office. They will ship tomorrow, together with my evaluation.

Regards, Dieter

[branadic]

Thanks to everyone for coming, bringing all the equipment and making it such a successful event. It was you giving life to the event.
Special thanks to Andreas and Frank for their talk and thanks to Michel for having a gold standard in the lab.
We had a wonderful atmosphere, nothing got broken, everybody helped each other and there was no time for being bored.
Thanks also to Mr. Hadlok from Keysight Böblingen for being with us, showing some equipment and giving some details on the new 3458A in his presentation as well as bringing the fresh calibrated 3458A as a special service.
And last but not least, thanks for the feedback.

It was great to see the faces behind the names and a pleasure to meet you all.

-branadic-

[Echo88]

"New 3458A" as in successor of the 3458A we already know? We need juicy details. 

[hwj-d]

However, I would have made a small suggestion for improvement next time, but I would have said that things couldn't have gone better on Friday and Saturday.

Next time, as far as possible, we should sit down at the end of a short and constructive concluding round, where we have the opportunity to present our subjective impressions and concrete insights within the group and to reflect briefly.

[Logical]

Arrived today back home and from my side it was a big succes and a big thanks goes to Branadic for organazing this "Volt Nuts" Metrology meeting in Stuttgart.   

For the people that are interesting in the accuracy of the 3458a (from branadic's company) that was freshley calibrated by Keysight.
Before i left to go to my hotel i did a queck measurement with the fresh calibrated 732a (10.000010) and measured on branadics 3458a 10.000017x and that would be 0.7 ppm difference.
So it would be interested if branadic could tell us what the call lab at keysight measured during the 10V calibration last week.

Best regards,
Logical

[try]

Hello folks!

Everybody who has given me gear for calibration should have received an email in his mailbox with all my values recorded in an attached excel file.

I am happy having met all you guys from Mid-Europe. There is nothing like direct face-to-face communication.
It was really worth arriving the day before and having dinner with some of you.

My special thanks go to branadic for setting everything up, to Hahn-Schickard for providing us with space and food, for ab-precision to bring in freshly calibrated gear, to Keysight for sending in a representative and products.

Furthermore I'd like to underline that without the support of the PTB gang (Henrik, Harald, Dennis and team) who sacrifiy valuable week-end summertime for promoting the PTB and distributing electrical units at the Maker Fair I would not have gotten to the point to more or less accurately measure the references of my fellow voltnuts.

Next Maker Fair is on the 17./18.8.2019 in Hannover, Germany - voltnuts should not miss showing up at the PTB booth!

Regards
try

[branadic]

So it would be interested if branadic could tell us what the call lab at keysight measured during the 10V calibration last week.

It was calibrated the day before the event


DC Volt Offset
Range   Input   Minimum      Measured      Maximum      Uncert.
10V      0V      -0.0023mV   0.0003mV      0.0023mV      0.00032mV

DC Volt Gain
Range   Input   Minimum         Measured      Maximum         Uncert.
10V      10V      9.9999340V      9.9999987V   10.0000566V      0.000025V
10V      -10V      -10.0000618V      -10.0000046V   -9.9999392V      0.000025V

-branadic-

[hwj-d]

Someone asked me about Jason's (cellularmitosis) ltz1000 pcb's, of which I still have a few left to give away for nothing (oshpark ones). Please report the one who asked me. But in total there are only 2-3 pcb's.

[Andreas]

Hello,

To calculate the actual deviation you have to calculate the nominal value of the calibrator as:
nominal = (minimum + maximum) / 2

see above example for the Wavetek 7000 from me.

with best regards

Andreas

[Sprock]

Hello,
that was me. Please PM. Just to keep here the noise low.
regards Sprock

[hwj-d]

Hello,
that was me. Please PM. Just to keep here the noise low.
regards Sprock

So, do it.

[HighVoltage]

Thank you branadic for making this happen so successfully.
It was really great to meet you all and finally putting some faces to the names.
Here are some gear pictures.

[Zucca]

Glad to see it happened.

Sorry guys I was a lazy ass and I decided to go mountain biking in Italy instead.

[BU508A]

Glad to see it happened.

Sorry guys I was a lazy ass and I decided to go mountain biking in Italy instead.

Which is, imho, a very good idea, too. 

Thanks to branadic  for all the organisation and your company for giving us all the room and lab space.
And also a big "Thank You!" to all other participants. I learned a lot and it was really fun to see all
these children engineers playing around with all this nice test gear. 

I'm really hoping, we can do this again.

Thanks,

Mounty

[Henrik_V]

"New 3458A" as in successor of the 3458A we already know? We need juicy details. 

Well, I have to byte my tongue for a while now   And it should be new thread and there are others more qualified to fill that thread.
There will be a new rev. of the 3458A. (I think KS will even keep the same order number (the 3458 is qualified in many applications!)
They rebuild the 3458A, board by board with new (without unoptanium)  parts. You can swap any old to new board, do a calibration and run it.
A former PTB fellow has verified that even inofficial specs are met (or better ).  (better internal EMC, SMD parts, about 5K lower internal temperature , higher input impedance (more space on the board) ). Linearity (don't nail me on that one, maybe selected boards)  down to 0.01 ppm?   


   

[Echo88]

Sounds nice. Will be interesting to see if/how they solved the U180-unobtanium-problem. Would be nice if the mentioned involved persons could share their wisdom about the new 3458A in a new thread.  

[hwj-d]

...
And also a big "Thank You!" to all other participants. I learned a lot and it was really fun to see all
these children engineers playing around with all this nice test gear. 

I'm really hoping, we can do this again.

Thanks,

Mounty

I've heard someon loudly speakin' out in the corridor in front of the door to the lab (to himself):

"Endlich unter 'normalen' Menschen ..."
that means "finally among 'normal' people ..."

Do you know who that was?     

H.-W.

[Echo88]

Did you invite Tom Gerhardt? 

[hwj-d]

Didn't even know who held the copyright.
But that came from the deepest soul. 

[Dr. Frank]

From calibration protocol (of yesterday):

for +10V the 3458A reading was 3.4 uV too high against calibrator actual value.
for -10V the 3458A reading was 4.1 uV (amount too high)
Calibration uncertainity is 2.5 ppm (+/-25uV) with 24 hrs we get 2.6 ppm (+/-26uV)

so for 10V + 56.8uV we have to correct to 10V + 53.4 uV
for -10V -58.5uV we have to correct to -10V - 54.4uV

the difference to 10V + 41uV from Fluke 7000 is 12.4 or 13.4 uV which is within the +/- 26 uV +/- 2.5 uV uncertainity.
But which device is showing the truth? -> we need more references/DMMs   

Edit: what where the results between the Fluke 7000 and the other Flukes (732)?

Edit: And even more evil: I did not find in the calibration protocol if the calibrator has been already set for the new volt 

with best regards

Andreas

Andreas,
I have compared the 732A from Adrian, 732A from Wulf (?), 7000 from Michel, and my LTZ #2

Procedure was similar to yours:
using a freshly adjusted 3458A (to Adrians 732A), I measured each polarity @ DUT, 16 x NPLC100, with statistics.
Average of these two absolute value gives "true reading", to cancel e.m.f.
Adrians 732A (10.000 010 V_nom.) was uncertain to 0.5ppm, Michels 7000 (10.000 041 V_nom.) was uncertain to 0.25ppm, so it got double weight for the difference calculus.

I calculated corrected values for these references:


732A (Adrian) = 10.000 006 5 V (StD ~ 200nV)
7000 (Michel) = 10.000 042 7 V (StD ~ 130nV)
732A (Wulf?)  = 10.000 008 5 V (StD ~ 132nV)
LTZ#2(Frank) =  7.118 228 61 V (StD ~ 104nV)

The agreement between these 732A and the 7000 'felt' really impressive.

I had to correct my baseline (drift of my 3458A) by about +1.8 ppm. I.e. my 3458A was reading too low.. maybe that's the -2ppm Philipp was looking for.

But this is also quite impressive, as the last direct transfer was done 7 years ago, and that demonstrates that an ensemble of 4 references (LTZ and SZA) determines drifts in the sub-ppm range, and also is capable of < 1ppm/yr. drift performance (ovens at 50..55°C).
Not bad for home-grown references.


A big Thank You to André (branadic), who made that event happen.. it was perfectly organized.
It also was a great experience and a great pleasure with all participants.. I really felt deeply inside, that I'm not alone being 'nuts'.

Frank

[try]

Hi Frank,

remember you did not want to believe me when I said that I see your LTZ-2 2 ppm higher? 

Phil and me ended up searching them in his appartment a whole afternoon long till midnight one week after the Maker Fair 2018.

Your 10k reference by the way agreed perfectly to the ESI SR1010 from D. and indicated that my 34401A understated the measurements by -7,6 ppm despite being adjusted in September 2018.
That's why I told the fellow voltnuts to add +8 ppm to the 10k figures of their references.
Don't forget to bring that reference in to the next meeting. Ohms seem to be more challenging for my opion.
This time measurements fit pretty well.

The picture attached is taken from my calibration log file that I sent to you as well, names are anonymized except for the publicly known ones.

My conclusion:
Even without a Keysight 3458A (5 devices present) and Fluke 732A (3 devices present) you can squeeze plausible numbers out of low price gear.

Regards
try

[dietert1]

I remember another resistance check with Martins 100R resistor box, where we had readings of 34.. 37 ppm with an expected list value of 43 ppm. Another indication of the -8 ppm resistance calibration of that 34401A.

Our three old Lab DVMs 1x Fluke 8502A and 2x HP 3456A, that came to us about 19 years ago and that were never adjusted in all those years, exhibit offsets of -13.6 ppm, 12.7 ppm and 3.1 ppm according to the measurements in Stuttgart. (10 V DC measurements) Those DMMs did receive some maintainance like recapping power supplies and the like. The difference between the two HP 3456A was measured as 6.5 ppm in 2008 and increased to 9.6 ppm since then.
Good news that confirms Franks statement of "1 ppm/year" being within reach technically. Of course, if someone needs certificates for legal reasons, that's another story.

Regards, Dieter

[BU508A]

...
And also a big "Thank You!" to all other participants. I learned a lot and it was really fun to see all
these children engineers playing around with all this nice test gear. 

I'm really hoping, we can do this again.

Thanks,

Mounty

I've heard someon loudly speakin' out in the corridor in front of the door to the lab (to himself):

"Endlich unter 'normalen' Menschen ..."
that means "finally among 'normal' people ..."

Do you know who that was?     

H.-W.

I have NO idea.