Measuring voltage Conrad Hoffmann style - with a Weston cell

[try]

Hi all,

there is an interesting story about the restauration of a Weston cell based multimeter.

The following link is about the restauration (German language),

https://www.mikrocontroller.net/topic/450690?goto=5467046#5467046

but the linked thread contains a blog link in English language

https://entertaininghacks.wordpress.com/2016/12/28/a-weston-standard-cell-an-introduction-to-voltnuttery/

which describes how to measure voltage using a weston standard cell with simple equipment:
A cheap 2% precision multimeter and 1m of resistance wire.

When reading the story it instantaneously remembered me Conrad Hoffmann's pages.
It's a good example how to measure stuff with an equipment next to nothing.
The only restriction today is that Weston cell are not widely available anymore.

Furthermore the short article outlines the symptoms of an infection. 

Regards
try

[tggzzz]

Furthermore the short article outlines the symptoms of an infection. 

A barely contained[1] infection

Here's using that Cambridge Instruments Saturated Cell as a thermometer. It shows an uncalibrated Solartron 7081 measuring the voltage over a couple of days.

Measurements are in a spare bedroom, taken every 50s, so 1500 readings are 24 hours. The two peaks are approx 24 hours apart, as are the two troughs. The first trough to second peak is 85uV, and the temperature range was very roughly 18.2C to 20.1C, i.e. 40uV/C.

Overall the standard cells are nowhere near as stable as my old £14 Standard Reference Labs Trancell 2 transportable zener diode, but the standard cells are useful for evaluating the noise in the Solartron: ~0.3uVrms on the 1V range iff my calculations are correct. The Trancell 2 is very good w.r.t. noise: zero popcorn noise and ~1.2uVrms on the 10V range iff my calculations are correct.

[1] I can claim that due to the lack of calibration and gross approximations

[try]

Hi tggzzz,

Furthermore the short article outlines the symptoms of an infection. 

A barely contained[1] infection

Here's using that Cambridge Instruments Saturated Cell as a thermometer. It shows an uncalibrated Solartron 7081 measuring the voltage over a couple of days.

Measurements are in a spare bedroom, taken every 50s, so 1500 readings are 24 hours. The two peaks are approx 24 hours apart, as are the two troughs. The first trough to second peak is 85uV, and the temperature range was very roughly 18.2C to 20.1C, i.e. 40uV/C.

[1] I can claim that due to the lack of calibration and gross approximations

The 40µV/K is the figure I find in literature, too. The sinusoidal curve remembers me my own measurements when the sun is shining.
In the morning the sunbeams hit my measurement equipment.

I don't believe you when you say that your Solartron 7081 is uncalibrated. 
At least one range is!
Am I wrong in assuming that you connected your calibrated Trancell to your Solartron after calibrating the Trancell here (picture attached)? 

By the way, the next maker fair in Hannover is on the 15 and 16th of september 2018.
I hope you'll be there again!

Regards
try

[AG7CK]

In the interest of history, general education and metrological methodology, I allow myself to suggest that the OP and readers of this thread possibly could find it beneficial to read: https://en.wikipedia.org/wiki/Potentiometer_(measuring_instrument)

[Edit: Shortened.]

[try]

Hello AG7CK,

it's good to have somebody like you here who preselects content and decides what's worth reading and what's not.
Maybe Dave can provide you with a new forum function so that you can automatically filter unworthy content that others should never see?

When did you start representing "history"?
Is it possible that you missed the point of the linked article with regards to metrology?

Regards
try

[tggzzz]

Hi tggzzz,

Furthermore the short article outlines the symptoms of an infection. 

A barely contained[1] infection

Here's using that Cambridge Instruments Saturated Cell as a thermometer. It shows an uncalibrated Solartron 7081 measuring the voltage over a couple of days.

Measurements are in a spare bedroom, taken every 50s, so 1500 readings are 24 hours. The two peaks are approx 24 hours apart, as are the two troughs. The first trough to second peak is 85uV, and the temperature range was very roughly 18.2C to 20.1C, i.e. 40uV/C.

[1] I can claim that due to the lack of calibration and gross approximations

The 40µV/K is the figure I find in literature, too. The sinusoidal curve remembers me my own measurements when the sun is shining.
In the morning the sunbeams hit my measurement equipment.

I don't believe you when you say that your Solartron 7081 is uncalibrated. 
At least one range is!

Well, it is a bit better than completely uncalibrated, with a tracable calibration - but not officially!

I reckon it is about 20ppm out at 10V.

The combination of a 7-decade Julie Research KVD and the 7081 is linear to typically 1.5ppm, 2.5ppm worst case.

Am I wrong in assuming that you connected your calibrated Trancell to your Solartron after calibrating the Trancell here (picture attached)? 

Aargh! My cover is blown

By the way, the next maker fair in Hannover is on the 15 and 16th of september 2018.
I hope you'll be there again!

I won' t unfortunately, but I do highly recommend it to everybody! They are a great bunch, and the gear there will have any TEA addicts drooling.

I've advised them to frisk people before they leave the stand

[PA4TIM]

Overall the standard cells are nowhere near as stable as my old £14 Standard Reference Labs Trancell 2 transportable zener diode,

If so, I want one, my best  standardcell  decreased 600 nV in 14 years, so better as that for 14 pound would be great.
http://www.pa4tim.nl/?p=2141 pictures of my Guildline 4 Weston Cells cabinet

[pelule]

If so, I want one, my best  standardcell  decreased 600 nV in 14 years, so better as that for 14 pound would be great.

If I take a cell voltage of 1,0181V, drifted by 600nV in 14 years == ~0,6ppm in 14 years == 0,042 ppm/year.
1: it proofs - you handle your cells with the right care
2: in my understanding thats an excellent drift value
3: I daubt a zener diode based reference will give similar performance
I own a Guildline 9152T/4 since 2 years - I measure the cell voltages each quarter with my 3458A in High Impedance Mode.
So far the drift is less than 0,1 ppm/year (taking the uncernity into account).
/PeLuLe