Getting the right step size for 0.55NA

[Koorosh]

Hi everyone,
I have trawled through a good 20 or 30 posts trying to find information on how to get tiny step sizes, but they have either neen attached to microscope focus blocks or using Zerene to fool the stacker. 1 step on the stackshot unit is too small and 2 steps is a little too large.
Is there a way to change the size of steps without using Zerene?
I tried 1um and I think it was just a bit too large for some details using a 50x 0.55 objective.

[ChrisR]

You just go into Custom Config and tell the Stackshot that each rotation of the motor is some large movement of the rail. Then it rotates the motor shaft correspondingly smaller angles.

Eg HERE

[Koorosh]

Ah that was the post I'd seen before! I didn't read beyond the first couple of replies thinking it was the wrong one. Thank you for the link.
So has the accuracy improved since the introduction of high precision mode? And would I change it to a number where I was using the recommended 5 step increments to get the desired measure? Do I have to use mm instead of steps to get the desired effect? By that I mean the units in autostep and not changing the revolution distance.

[rjlittlefield]

I have trawled through a good 20 or 30 posts trying to find information on how to get tiny step sizes, but they have either neen attached to microscope focus blocks or using Zerene to fool the stacker. 1 step on the stackshot unit is too small and 2 steps is a little too large.

You just go into Custom Config and tell the Stackshot that each rotation of the motor is some large movement of the rail. Then it rotates the motor shaft correspondingly smaller angles.

Eg HERE

Ah that was the post I'd seen before! I didn't read beyond the first couple of replies thinking it was the wrong one.

I think it is the wrong one. That post is in the middle of a thread that talks about using StackShot with a microscope focus block. The trickery was to inform the StackShot controller about the different in mechanics: 200 microns per turn with the focus block versus 1587.5 microns per turn with the StackShot rail.

Koorosh, if you're using the StackShot rail with its standard motor and no other modifications, then the minimum step size is nominal 0.496 microns and there's nothing you can do about that. (The number is computed as 16 turns/inch (screw pitch), divided into 25.4 mm/inch to get 1.5875 mm/turn, divided by 200 full steps per turn, divided by 16 microsteps per full step.)

Worse, due to friction and nonlinearity in the drive electronics, the actual step sizes may be significantly larger, as documented at What does StackShot "High Precision" mean? "In this run, the largest single microstep is 1.38 microns versus 0.496 microns average, and the largest double microstep is 2.06 microns versus 0.992 microns average."

I tried 1um and I think it was just a bit too large for some details using a 50x 0.55 objective.

In that case, the next thing to do is to go into Configure mode on your StackShot controller, switch the mode to "Steps" instead of "Metric", and set the focus step distance to 1 step (actually 1 microstep) of the motor.

That will get you the smallest steps that the rail is able to produce -- again, nominally 0.496 microns but in practice some smaller, some larger.

If that turns out to be OK, then you might try setting the focus step distance to be 2 steps and see if that still works OK. "2 steps" is very close to "1 micron", but due to rounding a setting of 1 micron might occasionally bump the motor by 3 microsteps instead of 2.

--Rik

[Koorosh]

Thank you as always Rik

And so with modifcations, can that include adding a new stepper motor, or will that produce the same limits but with more accuracy due to stiction no longer being an issue? I read further down about removing the stock motor and using a belt driven one. Is that a straightforward approach?

Apologies for the late response. Last year of uni currently with impending deadlines (doom).

[rjlittlefield]

And so with modifcations, can that include adding a new stepper motor, or will that produce the same limits but with more accuracy due to stiction no longer being an issue? I read further down about removing the stock motor and using a belt driven one. Is that a straightforward approach?

Belt drive would not be the right approach for modifying a StackShot rail. That's because what's critical is motion control of the screw. Remember that for a 0.5 micron step, the screw turns by only 1/3200 of a rotation. Belt drive is completely appropriate with a microscope focus block, which was the original context for the thread that I think you're referencing.

It appears straightforward to replace the standard motor with a geared stepper, but we don't know for sure what effect that will have. It seems unlikely to hurt, but no guarantees how much it will help. Based on the fact that the physical rail movements seem to track what we see in the drive currents, I am hopeful.

As it happens, I've recently been talking with Cognisys about this, and I have some parts on order to run the experiment. Some of them are coming by "slow boat", and between that and other commitments, it's unclear when I will actually get around to producing anything useful.

If you get to it before I do, let us know what happens, please.

--Rik

[Koorosh]

Hi Rik,
any progress with your initial work with Cognysis on the use of different stepper motors? I am finally, for the first time (I finished uni!), getting the chance to try out my purchase of a 50x from Lothar I put it on 1 step, so I'll see how it goes. 967 images.... Not entirely certain the flash is going to last long enough battery-wise. I think I'm using an 8th of power with Eneloops (the model down from the most powerful ones).

[rjlittlefield]

Sorry, no progress on that front. Too many other balls in the air (and a big pile waiting their chance to fly).

I've used the standard rail at 50X with good results. Just be sure to put it in HIGH PRECISION mode, or you'll get focus banding for sure.

--Rik

[Koorosh]

Yes I haven't had it on anything other than high precision so far. I mostly use a 10x so I'm used to that humming now. I'm in the process of getting the stack done now. I under-exposed I think but it didn't clip so I might be OK with some levels adjustment.
I'll post it up if it's good enough. It's stopped down to 25x to look a little sharper. Truly shocked at how small an object must be! Had to find a wasp with a small enough face!