L&N 4393 Universal Ratio Set

[Vgkid]

Leeds and Northrup 4393 Universal Ratio Set
Background, since it is in the mail:
What is a Universal Ratio Set?
A Universal Ratio Set or URS(abbreviated) is best thought of a a slide wire. A 3 terminal restive voltage divider. One that has a constant input resistance, and a varying output resistance(simplified).
A more descriptive version is to imagine 2 resistor decade boxes connected together, except the switches are wired in a way so that as one turns one way, the other turns the opposite way. As the resistance is increased in the first decade, the total resistance decreases in the second. A 20 step 100 Ohm decade connects the two decades.

Now there some upgrades since this basic version was created, as shown below.

Here are the specs for the newest variant, note the number of extra binding posts, and additional switches. I will go into detail later. For this version, that is almost $19.1K today. And you thought that that the 3458A was expensive, when It comes to L&N factory prices they can only go up...
Expect a massive info dump, until I get my unit.

[Vgkid]

For those wanting to read ahead look into these:
A Versatile Ratio Instrument for the High Ratio Comparison of Voltage or Resistance
nvlpubs.nist.gov/nistpubs/jres/70C/jresv70Cn3p169_A1b.pdf
Practical Methods For Calibration of Potentiometers
https://archive.org/details/practicalmethods172rama
Precision Measurement and Calibration Selected NBS Papers on Electricity — Low Frequency
nvlpubs.nist.gov/nistpubs/Legacy/SP/nbsspecialpublication300v3.pdf
^^*A must read*^^

[Vgkid]

Well it came, the info dump will be explained as I go.

With this, we can see that the number of binding posts doesn't match up with the more modern unit that is in the advertisement. One thing that is similar, is the weight. This thing has to easily be the densest piece of equipment that I own. The shipping weight to me was 43 pounds, with about 3 pounds of it being cardboard, It was packaged very well.

Popping off the cover we are greeted with a insulated box, and another black Bakelite box holding all of the goodies. The insulation used is a thick layer of cork, I was fearing a fibrous type of insulation. The Bakelite box is lined with copper sheeting.
Spoiler ye olde voltnut voodoo

 I haven't had a chance to tear it down properly yet, so I will add some better pics tomorrow..

[lowimpedance]

I do like looking at the way construction was done in the days past by different companies, have not seen cork used before in other gear of this vintage.
Rather interesting plastic resistor carrier around the switch.
Thanks for the pics.

[Vgkid]

It is interestinv how different companies assembly there gear. L&N seems to like these specialty fasteners which do not require a tapped hole to fasten into, just a lip on 2 sides. They are a pain to get lined up to remove. The cork as an insulator is neat.
Just wait until I start posting the teardown pics.    

[Assafl]

What are the black hook shaped wires pointing upwards? They seem to all over the place.

[Vgkid]

Those black wires are interesting. In pictures of a modified unit( will post later, pretty neat) the wires appear in those pictures as well. I thought they were fine trim wires, but they are not connected to anything. I do have a hunch what they are for.

[TheSteve]

They are pretty interesting. The only thing that comes to my mind so far would be for some sort of thermal balancing.

[Vgkid]

That is one of the ideas what they are for. The other is part of a 4 wire test during assembly. As these resistors are "thermally" matched.

[Vgkid]

After removing the external case, and its cork lined thermally lagged enclolsure. We are greeted with the large black Bakelite structure that forms a shielded box around the double decade resistors. I haven't seen a ton of Bakelite in conditions as good as this. It almost looks like a black Plexiglas.

When I unscrewed the inner case from the chassis, I noticed, that it had an impressive weight to it. When I flipped it over, I was greeted with the red gleam of copper. With some oxidation in parts. I was thinking that it is from the reaction with mercury, as I never found and mercury in the unit. The thermometer was broken inside the unit.

The first decade we are greeted with is the decade containing 20, 1milliohm resistors. Remember each switch actuates 2 sets of movable rotary contacts, that select resistors in increment/decrement pairs. Since these resistors are silver colored I will assume they consist of the alloy Constantan(Correct me if I'm wrong please.) When It comes to how the switches are wired up, that will come in a much much later post. That might be part of another thread .

This shows the 10milliohm decade. These resistors are designed the same way the smaller range is, except that these are covered in varnish. I'm not sure why these are, and the others are not. I will need to look and see if the smaller decade has any parts filed off.

When we arrive at the 100milliohm decade we are greeted with a common sight. The same resistor technique that is used to create the resistors used in not only my 4232 Wheatstone bridge, but those same resistors(different value) as those in the 4385 Shunt box.

With the 1.0 Ohm range , these resistors are wound a bit differently than those in the 4232. These are wound on a segmented ceramic bobbin, as opposed to being wound on a ceramic bobbin.This is of the design that I have seen, as opposed to being straight wound.

With the 10 Ohms range, the resistors have the same appearance as those in the 4232, so no use in explaining these.

With the 100 Ohm range there aren't any differences with those in the 4232. The interesting thing with these is that they are mounted so close to each other that rather than having wires connecting each resistor to the switch, than having the trim on another section of the switch for that decade. The trim resistor is mounted on the base resistor, and a wire connects that resistor to the switch.

Since L&N doesn't make a 20 position rotary switch, they employ another manufacturers switch. This picture also illustrates the explanation in the previous paragraph.
More to come...

[Assafl]

My bet on these black pieces of wire is handles. They seem to be in places they don't want to touch and/or clean. So each resistor has one - except ones which are daisy chained. Or it may be for calibration - but not all resistors have them. So perhaps it works in addition to connections to the front panel at specific switch locations?

That wafer switch with those "Carefree" lengths of white/gray wire (as opposed to the Teflon/Copper ones mounted symmetrically with "Exceptional Care" elsewhere)... As if "carelessness" (with purpose!) was an attribute they were looking for with the Most Significant Decade. Looks like their emphasis for accuracy and precision is clearly the low-ohms end rather than the high ohm end. In that respect it is an inverted version of a KVD in terms of accuracy. (please note I am not saying LN was careless - just that this is not a potentiometer - albeit being wired as one - Whereas for a potentiometer one would care for precision and linearity, here it is precision and resolution - with reduced care for large scale linearity).

Fits in with the idea that one would use this to calibrate the high sensitivity settings of a complex potentiometer rather than to calibrate a standard low-resolution potentiometer. You would standardize the potentiometer around any multiple of (more or less) 100 Ohms - and then start checking the high resolution controls against the high sensitivity controls of the potentiometer. 

Very curious device. I wonder to what purpose they still sell these. Are there still potentiometers around? Maybe for calibrating inertial platforms?

[Vgkid]

Until I can find a decent pdf maker , that doesn't increase the file size 10x.  Here is a compressed phone scanned version of the manual.
When I get a proper book scanner , I will make a better version.