Maker Pro
Maker Pro

what am I

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,505
Joined
Jan 21, 2010
Messages
25,505
It's pretty cool actually.

WP_20160118_12_38_24_Pro.jpg
 

hevans1944

Hop - AC8NS
Jun 21, 2012
4,739
Joined
Jun 21, 2012
Messages
4,739
OMG! You have a dual Dekapot! I am so envious! Be careful not to put too much current through it. We used these to set precision test-set parameters back in the day.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,505
Joined
Jan 21, 2010
Messages
25,505
That was just too quick!

The wirewound pot for the inner right control is not in great shape. However it's also the easiest part to replace.

I googled it to figure how it worked, and when the penny dropped it was like "that is just so elegant".

This says it has a maximum input voltage of 250 (Volts, presumably), but I've only tested it with the 3ish volts from my multimeter on the diode check range.
 

Osmium

Jan 28, 2013
67
Joined
Jan 28, 2013
Messages
67
OK. Dekapot now added to list of cool devices I "need" to own...

Thanks for the info. And here's a cool vid with a different version:
 

hevans1944

Hop - AC8NS
Jun 21, 2012
4,739
Joined
Jun 21, 2012
Messages
4,739
The Kelvin-Varley's main claim to fame as a precision voltage divider is unequaled., I think its best feature is constant input resistance, because each decade is switched in such that it always presents the same impedance to the previous decade. Very handy when used with precision voltage sources. The output resistance varies with the divider setting of course, which usually isn't a problem for most metrology setups.

These things have been built with up to seven decades of precision (see image below) which rivals modern 24-bit delta-sigma analog-to-digital converters. The ones I used in the Air Force were only four or five decades (I forget which) plus a variable resistance (potentiometer) with graduated marks mainly useful for nulling. Of course A/D converters have largely replaced K-V dividers, but they were once used with a null-meter and a precision voltage source as manually adjusted digital voltmeters. IIRC, there was even a digital voltmeter that used stepper switches to obtain a null against an unknown input voltage, although one version required manually adjusting each decade for a null. Interesting bit of history Dekapots. Easily abused however, and old ones that have been on the shelf for many years may not be accurate because of long-term drift in the wire-wound resistors. Good find, Steve.

Kelvin_Varley_Divider.PNG
 

davenn

Moderator
Sep 5, 2009
14,040
Joined
Sep 5, 2009
Messages
14,040
The Kelvin-Varley's main claim to fame as a precision voltage divider is unequaled., I think its best feature is constant input resistance, because each decade is switched in such that it always presents the same impedance to the previous decade. Very handy when used with precision voltage sources. The output resistance varies with the divider setting of course, which usually isn't a problem for most metrology setups.

These things have been built with up to seven decades of precision (see image below) which rivals modern 24-bit delta-sigma analog-to-digital converters. The ones I used in the Air Force were only four or five decades (I forget which) plus a variable resistance (potentiometer) with graduated marks mainly useful for nulling. Of course A/D converters have largely replaced K-V dividers, but they were once used with a null-meter and a precision voltage source as manually adjusted digital voltmeters. IIRC, there was even a digital voltmeter that used stepper switches to obtain a null against an unknown input voltage, although one version required manually adjusting each decade for a null. Interesting bit of history Dekapots. Easily abused however, and old ones that have been on the shelf for many years may not be accurate because of long-term drift in the wire-wound resistors. Good find, Steve.

Kelvin_Varley_Divider.PNG


thanks Hop

enlightening :)
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,505
Joined
Jan 21, 2010
Messages
25,505
I think it was $40 well spent, if only for what I've learned about it so far.

And yeah, one of the things I've been toying with for a while is building an ovenised voltage reference. I've managed to get some reasonably high spec zeners with integral heaters similar to those used in HP's older multimeters (including the 3458).

The amusing thing is that once I've built something like this I wont have any way of really testing it. :D
 

shrtrnd

Jan 15, 2010
3,826
Joined
Jan 15, 2010
Messages
3,826
davenn, if you've still got your own teeth, you're younger than me and hevans1944, and you HAVE seen a dino before. Just take a look at hevans1944's avatar. I don't use an avatar because it'd be even more scary.
I am presently working in a metrology lab that services (among other things) new and old air force gear that hevans1944 used to work with.
You might be surprised to know that, as hevans1944 claimed, a lot of the old dinosaur test and measurement instruments have not been improved upon.
People probably won't believe that. Every generation thinks they're the cat's meow, the newest, best and brightest.
We talked about this before. The oldest instrument I have here was made in 1957. Nothing we've bought since then is more accurate.
We're talking cold-war, pull out all the stops minds with the money to develop the best they could, and build it to last forever.
Sure, as Hop also said, components age and drift, but the remarkable thing about the old stuff was, that it was specifically designed to be repaired.
The engineers that designed it just didn't realize that one day the replacement components would disappear.
I used to go to garage sales in a city that was chuck-full of retired WWII machinists. The tool steel I bought from them is in the most beautiful, hardest precision tools I've ever found.
Think about it. We know about dinosaurs, because we're still unearthing their remnants. New engineers are smart, but the old ones weren't stupid either. And there was a time when money was no object.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,505
Joined
Jan 21, 2010
Messages
25,505
After a little mechanical "adjustment", the pot is now in working condition, and to the limits of the resolution of my 4 1/2 digit multimeter, it's spot on in accuracy.

It really is an impressive piece of equipment.
 

hevans1944

Hop - AC8NS
Jun 21, 2012
4,739
Joined
Jun 21, 2012
Messages
4,739
Your Dekapot will last forever. Just make sure you don't exceed its input voltage limitation and (for best accuracy) use a very high impedance load on the output. Your 4-1/2 digit multimeter probably loads it enough to cause a one or two digit "error" in the divider ratio, which is probably insignificant for hobby applications. The most accurate way to use the Dekapot is to feed the output to a null-meter and provide an adjustable voltage source to the other side of the null meter. When the meter is nulled, the adjustable voltage source will be at the same voltage as the Dekapot output, but with the advantage of very low source impedance. Add a precision meter display to the adjustable voltage source and... voila! You now have a calibrated variable voltage power supply! Enjoy.

Umm. You could include the Dekapot as the reference element in a variable output power supply with just a little more engineering effort.

Hop
 
Top