#1000: My (hi)story of the Weston cell, of the Volt, and of being a volt-nuts

[HighVoltage]

That is impressive with a cell from '43
Some of these cells are just amazing, how long they stay stable.

What is the input impedance of the advantest DMM in the mV range?
Also interesting that the advantest shows above 1000 mV in the mV range.

[beanflying]

It is a 2 million count meter and 10^¹⁰+ Ohms on the lower ranges as standard. The cells were removed from a local Uni Physics Lab (RMIT Melbourne) in 1975 and haven't been used since.

Just a cool thing to add to the collection and I am rolling an 3DP/Laser Enclosure to have a clear Perspex side for the un cased Tinsley one for display.

[tggzzz]

That is impressive with a cell from '43
Some of these cells are just amazing, how long they stay stable.

I have 1949 cell that is similarly in spec.

Saturated cells last but have a 40uV/C tempco around 20C. Unsaturated cells have a lower tempco, but don't last anywhere as well.

Here's some information from the NBS ("The construction and characteristics of standard cells" http://digicoll.manoa.hawaii.edu/techreports/PDF/NBS84.pdf ):

[beanflying]

Just because it is a little different I cracked the case on the Cambridge Cell to see what made it not tick.

While it is a single tube outer it appears to have a second inner glass with the base of it containing liquid Mercury and then into what looks like some sort of flock barrier and back outside the inner tube to the Amalgam.

$0.02 Teardown and also so it doesn't get lost in TEA

[ArthurDent]

Some may find this interesting.

The story of standard cells has some footnotes. Eppley tried what they thought was a innovation in cell enclosures and temperature control with their model 119 three cell enclosure. Instead of being heated, it used peltier effect coolers to maintain the temperature just a few degrees above zero degrees C. The theory was that, if you look at a graph of the voltage vs temperature of a cell, rather than heat the cells up, the enclosure cools the cells to the turning point in the temperature curve where the change is basically a flat line, or zero. What they apparently found, from what I heard, was that there were condensation problems and the idea was scrapped. I don’t know how many of these units were made but there is at least one which I own.

Attached is a temperature curve showing the zero point and a photo of the outside front of the 119 enclosure that has two heatsink enclosures with the power supply and peltier effect controllers on each side plus a large heatsink on the back of the center unit. The inside has a beer can size cylinder holding the 3 cells surrounded by rigid foam. There is a thermometer well hole in the middle and a matching hole in the cover plate. 

[ArthurDent]

Here are some photos of a more conventional standard cell enclosure/oven, my Guildline 9152-R4. You can see that the leads from the cells (inside the clear tubing) goes around each of the three concentric housings one complete turn to prevent heat creep on the leads. The leads are small and you’d think they would want to keep them as short as possible to minimize drops, but when measured using a bucking voltage to balance the cell voltage, you are drawing no current so there is no voltage drop on the leads. 

[ArthurDent]

Here is a photo of a dead Muirhead saturated cell. although it doesn't say, I'm pretty sure it is a saturated cell. 

[beanflying]

Interesting info on the cold option.

There is a couple of Heated Cabinets on evilbay for more than I would consider so I will make do. The curio factor and stability over decades is still fascinating.

I still have no real plans for my lot and if more come up for not much I may add to the fleet just because they are a cool toy 

[ArthurDent]

The problem with buying a standard cell enclosure now is you never know the condition of the cells inside. When I bought my Guildline 9152-R4 a few decades ago, the seller said one of the cells was iffy and the voltage on that cell is a little low. I just remeasured the internal resistance of all four cells and 1, 2, and 3 are between 900 and 1000 ohms, which is near the high end of acceptable, but the 4th one reads almost 4000 ohms.

The way the internal resistance is measured is indirectly by checking voltage drop with a known load resistor. Although some references say use a 1 megohm resistor, a 10 megohm resistor is better. I use my HP735A as a bucking reference for the cell under test and an HP419A as the difference indicator. I also monitor the HP735A voltage with an HP3457A for a double check. You can use this formula to calculate cell internal resistance. 

[schmitt trigger]

Subscribing to a fascinating thread.

I have visited Newport, RI previously. Surprising that a quaint resort town better known for its ultra-wealthy mansions and Jazz festivals, is also home to Eppley Labs.

[oz2cpu]

lets pump this thread back to life :-)
I just been at one of the best vintage parts collectors in my contry, truely a walk thru a time-machine,
and see what i was lucky to bring back home..
I did bring my keysight and test them all :-) ha ha.. all deliver the written voltage 1.01859

my two muirhead standard cell d-550 got a date handwritten on the back : 21-4-70
so they are 52 Years old !

[tggzzz]

It looks like I've stumbled on a working unsaturated Weston standard cell.

I picked up a couple of Solartron DVMs with the giveaway "CAL 1.019" and "CAL 1.0185" buttons and trimpots. After being on for a while, and twiddling the trimpot so the display matched that of my Agilent 34410A, one read 0.04 (i.e. clearly dead), and the other 1.019. What? It isn't dead?!

I'll ignore the dead one, and concentrate on the apparently good one.

It is inside a Solartron 1420.2BA. A datasheet is stamped "received 2 Feb 1966". The service manual states "internal Weston standard cell (unsaturated)" and "Weston Cell Temperature Coefficient +-0.0005% per degree C", but does not state the specific type. There are no ICs nor valves to hint at manufacturing date; the datasheet even lists the number and type of diodes and transistors! However, the nixie displays are marked 69xx, so it looks like the meter is ~1970 vintage, i.e. over 50 years old.

When spelunking, I used my Agilent 34410A on Hi-Z setting to measure the standard cell voltage, and it was 1.018725V when the unconditioned room temperature was 25C. According to Conrad Hoffmann, the voltage should be 1.0190V @ 20C, decreasing by 5µV/C as the temperature rises, so if "new" my cell should be 1.018975V. Cells tend to decrease voltage by 25µV/year (with quite a wide range), so my 250µV fall equates to 10years. Not bad for something with a 7-18 year lifespan.

The cell is inside the black plastic case in the centre of this picture; I can't see any identifying information and it looks like it is simply a box to protect the glass phial. I'm not going to attempt to desolder it to read the other side.

[mikeselectricstuff]

A few in my collection :

[dseyst]

I measured my Eppley model 100 standard cells which are naturally all unsaturated. The highest six voltages I recorded were:

1.018979
1.018996
1.018996
1.019030
1.019061
1.019328

Strangely enough, the highest voltage came from the oldest cell - serial #748240. My cheap thermometer was reading 21.8 degrees C. The voltages were all measured with an Agilent 34401a set to 10 gig input inpedance. I have checked all of them with a Fluke 8505a and a Keithley 196 also, and all meters read within 10 microvolts of each other. The other four cells I have all read around 1.018300 or so.

Dave

[Conrad Hoffman]

I had a collection of Eppley model 100s and what you need to do is cross compare them on a regular basis for a year or three. Even though mine appeared to be in reasonably good health, their long term stability was poor. Saturated cells can go nearly forever if properly cared for, but trusting unsaturated cells is risky. IMO, one of the nicest looking cells is the Weston model 4.

[dseyst]

I do plan to check these Eppley cells on a somewhat regular basis but I don't have a real use for them other as a kind of curiosity. I was born and raised in Newport, R.I. and at one time lived about a two minute walk away from Eppley Labs. I wasn't much into electronics years ago or I think I would have bought some of their saturated standard cells when they were still being manufactured.

Dave

[artag]

I've been after a Weston cell for a long time, and when I saw one offered for peanuts (thanks to mikeselectricstuff!) I snapped it up. It's the Muirhead D-845-F as commonly found above and has a 1971 cal sticker from Ferranti's Wythenshawe cal labs. The voltage then was 1.01924. Mike mentioned that he thought it was still in spec. Serial is 461647 so that was probably close to the date of manufacture, judged from the dates of other people's cells.

I dont (yet!) have a 3458A and the DVMs I do have (HP34401A and K2015) are out of cal, but they do agree to a few uV that it's 1.018636V, albeit at 26C. So if correct, that would be only 11uV per year fall. Too good to be true so I guess it's likely that my meters are actually reading a bit high (they were set to be equal about 5 years ago).

[dseyst]

Well, I was interested to find out what my Eppley standard cells (all unsaturated cells) would measure at 20 degrees C.  I measured them with a Agilent 34401a and the thermometer was just a cheap digital from Amazon.The first eight cells are model 100 cells and the last two are miniature cells.

1.019216
1.019029
1.019062
1.019000
1.019029
1.019325
1.018854
1.018943
1.018341
1.018193

The third cell and the two miniature cells seem to vary the least with temperature. Amazing if you ask me.

Dave

[HighVoltage]

You have 10 pieces Eppley standard cells?
And they are all above 1.018V and stable.
That is impressive after so many years.

One of my standard cells has taken a sudden dive and went from 1.0181V to 1.0164V within 6 month.
All the others are also very stable.

[dseyst]

Yes, I do have 10 Eppley standard cells. Three of them are very stable. The other 7 are stable but are more temperature sensitive. The three most stable only vary by microvolts as the temperature rises and falls. I also have 4 Weston model 4 standard cells but they all have problems.

Dave

[iMo]

Infected by this topic I got my first black box:

Muirhead D-845-D
Reference Cell
No: 459872

I did a measurement immediately after its arrival (34401A in HighZ) and got 1.018418V. That is on the low side from what I see here. It was shaken during the transport and also it was pretty hot outside, so the voltage will change a little bit after the cell stabilize itself, imho.
Is there a chance to estimate the production date based on the serial number?

[grizewald]

This is my Muirhead D-402-A Standard Cell - A saturated Weston cell in a beautiful black Bakelite case which I picked up from eBay several years ago. It was a bit of a gamble really, probably totally illegal to ship and when I went to collect it from my local post agent, I wasn't looking forward to having to tell them that they needed to call a hazmat team if I could see that the package was wet.



Thankfully, it survived an international journey through the postal system. I can see a few drops of mercury on top of the cadmium sulphate crystals in the negative leg, but other than that, it is unscathed. These particular cells were designed to withstand being transported in any orientation, given suitable packaging, so it's not really that surprising that it was still OK.



The cell itself is just two years younger than I am and is probably in significantly better condition!



From time to time, I measure the cell voltage with my calibrated 7.5 digit Solartron meter and the reading is usually around 50µV higher than the temperature adjusted expected reading and well within the meter and cell's specifications. The thermometer that I keep in the thermometer hole in the cell's case isn't calibrated, so there is also some uncertainty in my calculation of the temperature adjusted voltage.

It has shown no signs of the voltage changing during the just over three years it has been in my possession, so even after some 55 years, it is still performing remarkably well! The latest entry in my log is:

Date         Temp    Expected   Actual     Delta
2022-05-22   25.25   1.018351   1.018400   0.000049

While it might be more useful these days as a thermometer rather than a voltage reference, it's still a great piece of history which I am very happy to own and I'm amazed that it still seems to work just as well as it did when it was new.

[tggzzz]

This is my Muirhead D-402-A Standard Cell - A saturated Weston cell in a beautiful black Bakelite case which I picked up from eBay several years ago. It was a bit of a gamble really, probably totally illegal to ship and when I went to collect it from my local post agent, I wasn't looking forward to having to tell them that they needed to call a hazmat team if I could see that the package was wet.

I used that to beat my supplier down in price to £0 (but maybe £5 for a couple of D-402-Bs, some Na, some Mg, some KMnO₄, and some NaClO₃. A local 16-18yo college was shutting and selling off all its contents. A local clearance company was dealing with it, and the young girl knew a bit about disposing of hazardous materlals.

I gave them a reasonable but low price for a couple of very useable stereo microscopes; normally I can't stand the things. I also gave them tips about flogging boat anchors that didn't interest me. (That was back in 2016, before I became a *nut, abd before the TEA thread had exposed/developed our vices)

Thankfully, it survived an international journey through the postal system. I can see a few drops of mercury on top of the cadmium sulphate crystals in the negative leg, but other than that, it is unscathed.

See pictures of mine in reply #35

These particular cells were designed to withstand being transported in any orientation, given suitable packaging, so it's not really that surprising that it was still OK.

I haven't seen any info to that effect, but that is absence of evidence, not evidence of absence. Any refs?

[grizewald]

These particular cells were designed to withstand being transported in any orientation, given suitable packaging, so it's not really that surprising that it was still OK.

I haven't seen any info to that effect, but that is absence of evidence, not evidence of absence. Any refs?

In the attached document from Muirhead, they say: "The construction of the cell elements is such that, with suitable packing, they may be transported in any position."

I remember seeing your much older examples of the same cell. Are they still as stable as mine?

[tggzzz]

These particular cells were designed to withstand being transported in any orientation, given suitable packaging, so it's not really that surprising that it was still OK.

I haven't seen any info to that effect, but that is absence of evidence, not evidence of absence. Any refs?

In the attached document from Muirhead, they say: "The construction of the cell elements is such that, with suitable packing, they may be transported in any position."

I remember seeing your much older examples of the same cell. Are they still as stable as mine?

Thanks for the PDF; saved.

I certainly haven't done any "formal" measurements of stability, but I have informally showed it is a good temperature sensor, e.g. https://www.eevblog.com/forum/metrology/weston-cells/msg4093978/#msg4093978 and https://www.eevblog.com/forum/testgear/muirhead-kvd-and-internal-corrosion-ignore-it-or-fix-it/msg2894446/#msg2894446