My soviet mikrocator has two strips twisted together, 9 turns per side. I have the technical manual pdf somewheres if you want reference. Mode of failure was oxidation of wire due to moisture inside the instrument. Dessicant paper was not sufficient. 😅
3:46 Rollers. We use two rollers to flatten drawn wire into flat ribbon industrially the machines that draw the wire through vary (naturally) with some having a drive roller and free roller, two driver rollers, and others having the wire pulled through by a spool or a continuous captson acting like a spool. The last is what I made tools and dies for most recently in the heavy industry side and you can get extremely precise with them but it is a little bit of trial and error translating roller force into ribbon width, as even well worn in machines running smoothly there are too many sources of variables for setting and forgetting it
I would try not annealing after forming the wire flat. Otherwise it is very easy to stretch the ribbon when applying tension and this stretching will likely act to straighten out the twists.
I agree with others, the stiffer the material likely the better it would work (you want it to twist, not stretch). Carbon fiber is already twice the elastic modulus of copper, about on par with steel, tungsten is 4x. If you could find a flat ribbon of carbon fiber it'd probably work well, I think anything stiffer would be hard to source and worse to work with.
You could try balancing the needle by adding tiny drops of superglue. Probably easier to do without imparting excessive mechanical shock to the assembly than trying to snip the glass fiber. This seems to be how many precision d'Arsonval meter movements are balanced (I think they use shellac resin or a hard wax in that case, which actually would also be a good option). Snipping anything brittle with flush cutters creates a massive peak stress wave of very short period that travels down it and can damage anything attached. This info comes from documentation on reed switches, where cutting the leads is best not done at all, and if done should be done with a shear plate rather than flush cutters. I have to imagine that as you snip the fiber, a shock wave travels down it, and probably plastically deforms the strip right where it attaches to the superglue. Why are you twisting one side at a time? It seems like the system really wants to be twisted after having been attached to the final measurement device, by winding up the central pointer. I guess my other question is: the anvil you're using as a rolling surface is flat, but is it smooth? It doesn't look like an optical grade surface. I'm not sure this really matters, but having the ribbon smoother on one side than the other seems like a potential issue. Also I'm guessing that even with a large diameter roller such as you have, you're imparting a curvature to the ribbon by rolling it between a moving roller and a flat stationary surface? Again I'm not sure this really matters. Very cool, this is going to be great if you get it working!
more twists, more better? also a metal that has less dampening effect would be better, something like a hard wire, which might be tricky to flatten, but would have better elastic response
In the Johansson I have the ribbon runs through a loop filled with a thick oil / thin grease near where the pointer is attached. Pretty sure this is to dampen the pointer while the ribbon is optimized for elastic response.
Omgggggg I have been obsessed with solving this problem, the way the patent is drawn it looked like the strip was micro perforated through its middle which I thought might allow more twisting, but It doesn’t look like that is implemented in the actual mikrocator. I have also tried straightening cheap watch hairsprings to some success but have not tried repeated annealing, and some double wire versions a la @genabazarko . I’m so glad you are taking this on!!
So fun to watch this snowball into a bigger project ❤
My soviet mikrocator has two strips twisted together, 9 turns per side. I have the technical manual pdf somewheres if you want reference. Mode of failure was oxidation of wire due to moisture inside the instrument. Dessicant paper was not sufficient. 😅
3:46
Rollers. We use two rollers to flatten drawn wire into flat ribbon industrially the machines that draw the wire through vary (naturally) with some having a drive roller and free roller, two driver rollers, and others having the wire pulled through by a spool or a continuous captson acting like a spool.
The last is what I made tools and dies for most recently in the heavy industry side and you can get extremely precise with them but it is a little bit of trial and error translating roller force into ribbon width, as even well worn in machines running smoothly there are too many sources of variables for setting and forgetting it
As a fellow metrology nerd, I absolutely love all of these videos on insane precision. Keep it going!
I would try not annealing after forming the wire flat. Otherwise it is very easy to stretch the ribbon when applying tension and this stretching will likely act to straighten out the twists.
I agree with others, the stiffer the material likely the better it would work (you want it to twist, not stretch). Carbon fiber is already twice the elastic modulus of copper, about on par with steel, tungsten is 4x. If you could find a flat ribbon of carbon fiber it'd probably work well, I think anything stiffer would be hard to source and worse to work with.
You could try balancing the needle by adding tiny drops of superglue. Probably easier to do without imparting excessive mechanical shock to the assembly than trying to snip the glass fiber. This seems to be how many precision d'Arsonval meter movements are balanced (I think they use shellac resin or a hard wax in that case, which actually would also be a good option).
Snipping anything brittle with flush cutters creates a massive peak stress wave of very short period that travels down it and can damage anything attached. This info comes from documentation on reed switches, where cutting the leads is best not done at all, and if done should be done with a shear plate rather than flush cutters. I have to imagine that as you snip the fiber, a shock wave travels down it, and probably plastically deforms the strip right where it attaches to the superglue.
Why are you twisting one side at a time? It seems like the system really wants to be twisted after having been attached to the final measurement device, by winding up the central pointer.
I guess my other question is: the anvil you're using as a rolling surface is flat, but is it smooth? It doesn't look like an optical grade surface. I'm not sure this really matters, but having the ribbon smoother on one side than the other seems like a potential issue. Also I'm guessing that even with a large diameter roller such as you have, you're imparting a curvature to the ribbon by rolling it between a moving roller and a flat stationary surface? Again I'm not sure this really matters.
Very cool, this is going to be great if you get it working!
more twists, more better?
also a metal that has less dampening effect would be better, something like a hard wire, which might be tricky to flatten, but would have better elastic response
In the Johansson I have the ribbon runs through a loop filled with a thick oil / thin grease near where the pointer is attached. Pretty sure this is to dampen the pointer while the ribbon is optimized for elastic response.
What is the alloy used in the patents? Iirc it was something uncommon.
Some kind of bronze IIRC.
So cool!
👍
@@realcygnus no nifty 😔
@@cylosgarage gotcha ... nifty AF 🤣
This is cool ua-cam.com/video/2zEeAzJq-CQ/v-deo.html
Omgggggg I have been obsessed with solving this problem, the way the patent is drawn it looked like the strip was micro perforated through its middle which I thought might allow more twisting, but It doesn’t look like that is implemented in the actual mikrocator. I have also tried straightening cheap watch hairsprings to some success but have not tried repeated annealing, and some double wire versions a la @genabazarko . I’m so glad you are taking this on!!
Also have you seen “Kadimendium 8PCS LCD Screen Separation Wire Tungsten Alloy Steel 0.028 0.03 0.035 0.04 0.05 0.06 0.08 0.1mm Cutting Wire Cell Phone Screen Separator” it’s pretty amazingly fine and strong
I have a Johansson and the wire is perforated one hole per twist.