[MOD] Rod replacement for ultra-precise 50 micron (0.050mm) Z layers! (Page 2 of 2)

Not intending to hijack this thread, but don't EEPROM settings override Configuration.h settings?

Back on topic - this is really an extremely useful thread and I really appreciate the contributions to it.

Regards,
Matt

Not sure what you mean. Original steps was 2268, new rod 4501.

The exact amount will change depending the pitch of the thread.

I can only assume my pitch is 0.7 in that case. Dial gauge confirms it is spot on.

Hm, I've just tripple checked mine with a dial gauge, looks like lawsy somehow got a different pitch (or my mind is playing tricks on me.)
A quick test reprint of a 18mm high with 4500 steps_per_mm setting gave me 19.9mm actual height. 4000 steps_per_mm gave me a 17.75 resulting height. 
May be it's not metric? Like 5/16 is close to M8 but the thread pitch is not compatible with it. (Not a big expert on anything non-metric - just a wild guess)

I will double check the sizes of everything when I get home and post if it is different.

On another note my ballscrew has arrived from China and I'm machining the end to suit the coupling this afternoon.

Take the rod_mount.stl from the very first post. Print twice and you have the coupler.
Then, see my linked post for my custom bed nut holder. It is designed for 12x12x12mm custom machined nut, but there is the openscad design at my github, so you can adjust to your liking. The linked post also has instructions for custom nut machining.

One of these as a replacement for the threaded rod. I'm very happy with the M5 replacement however, so fitting this will be a low priority.

i think this spyder type coupler is better because of the "no Z lag" of the mount

just to find one in 5mm-5mm.. cost more

This math threw me off I had to pull out excel, lol

1: the steps per rev is simple
200*16 = 3200 steps/rev  (not in my case I have 1/32 steppers so I get double

Then this is fun math I made the 4501 mistake and got the same results right and wrong way
Lets get to the default 2268 steps per mm in marlin.

Your above math is confusing because you say times .8mm/rev. Because 3200 * .8 = 2560 steps ok. I was like wait that is only marginally better then the 5/16-18 so I backed into that

First _ did 18threadspi / 25.4 convert to mm = .708

ok so the stock sd is .708 so what is 3200 * .708 = 2267.7!!!! wow ok
So same math applied to .8 step = 2560 is this right? No right but what gives?

So what if I divided the 3200 by the pitch (thinking i had inverted wrong somewhere)
3200/.707 = 4515!!! ok so maybe that was mistake above. Still wrong right

SO I converted the 18threads per inch wrong! OK so what, here is the twist.
1/18 then = 0.055 inches per thread then convert to MM

0.055 x 25.4 = 1.411 pitch for the stock rod

so lets go back to stock conversion:

3200 / 1.411 = 2267.7

3200 * .708  = 2267.7!!!

both give exact same numbers even though .708 is wrong. Since if you applied this to another rod (steps x pitch you would be off)

So stock is:
3200 / 1.411 = 2267.7

And 5M-.8

3200 / .8 = 4000

But on my sure stepper I have

6400 / 1.411 = 4535

for upgrade
6400 / .8 = 8000!!!!!

I am a dork sorry for the rant

Yeah, I did the math right but typed it wrong.

[200 steps/rev]*[16]/[.8mm/rev]=4000 steps/mm.

Dimensional analysis.

I know this is an old post but any updates on it?
I also think about using ballscrew for x and y and would love some feedback if you did it.
thanks