Need help--how to remove Nikon focus block fine focus knob?

Have questions about the equipment used for macro- or micro- photography? Post those questions in this forum.

Moderators: Pau, rjlittlefield, ChrisR, Chris S.

Chris S.
Site Admin
Posts: 3549
Joined: Sun Apr 05, 2009 9:55 pm
Location: Ohio, USA

Post by Chris S. »

Barry and DaveW, thanks! Solid, straightforward advice.

Now I have to do some thinking. . .

--Chris

elf
Posts: 1398
Joined: Sun Nov 18, 2007 12:10 pm

Post by elf »

The belt approach is actually simpler to build because the tolerances for misallignment are much higher. The stepper motor will most likely have some vibration in it while just holding the position. Try to get a stepper controller that allows you to kill the power to the stepper when it's not moving the stage. If you use microstepping and are killing the power, you may want to stop the motor on either half or full steps, otherwise the motor may jump to the next full step. This is entirely dependent on the stepper driver, so it may be unnecessary.

Are you going to be using limit switches to make sure you don't crash at either end of the stage travel?

Chris S.
Site Admin
Posts: 3549
Joined: Sun Apr 05, 2009 9:55 pm
Location: Ohio, USA

Post by Chris S. »

Elf, the belt approach sounds like the way to go. Questions, if I may:

1. Suppose I simply put a gear on my fine focus drive shaft, and turn this with a belt. My guess is that the belt would be placing a sideways force on the drive shaft that it was not engineered for, and would probably not stand up to over time. Does that make sense? So I envision needing to mount, coaxially with the ff drive shaft, a stronger shaft bedded in bearings. Then place the gear on that bedded shaft, and use a flexible coupling to connect the two shafts. Maybe I am overcomplicating this--is there a better way?

2. If I mount the step motor in a manner mechanically isolated from the main rig, will significant vibration be transferred to the rig through the drive belt? If not, such an isolated mounting might let me avoid having to switch off the motor?

3. I understand the need for limit switches, and intend to include something along that line, but know nothing about them. Had thought I'd cross that bridge when I came to it--perhaps at first with something pretty crude. Am very much open to instruction on this.

By the way, what I am currently working with is the "Stepper Pack" with "Stepper Pack option 2" from this site: http://www.pc-control.co.uk/osc/index.php?cPath=22 I realize that this is probably not the ideal stepper motor or controller, but I like the fact that they package all the components needed for PC USB-interface stepping control as a matched set (step motor, controller board, power supply, cords, software). I figure I'll learn from this set, and use that knowledge to upgrade the components as I gain experience.

Advice very welcome.

Best,

--Chris

elf
Posts: 1398
Joined: Sun Nov 18, 2007 12:10 pm

Post by elf »

Chris S. wrote:Elf, the belt approach sounds like the way to go. Questions, if I may:

1. Suppose I simply put a gear on my fine focus drive shaft, and turn this with a belt. My guess is that the belt would be placing a sideways force on the drive shaft that it was not engineered for, and would probably not stand up to over time. Does that make sense? So I envision needing to mount, coaxially with the ff drive shaft, a stronger shaft bedded in bearings. Then place the gear on that bedded shaft, and use a flexible coupling to connect the two shafts. Maybe I am overcomplicating this--is there a better way?
The only force should be torque, not sideways. I think you're overcomplicating it. Even if the bearings wear over time, they're replaceable.
Chris S. wrote: 2. If I mount the step motor in a manner mechanically isolated from the main rig, will significant vibration be transferred to the rig through the drive belt? If not, such an isolated mounting might let me avoid having to switch off the motor?
You may be able to get by with vibration mounts on the motor. The belt shouldn't transmit vibrations. The directdrive would be more problematic for this.
Chris S. wrote: 3. I understand the need for limit switches, and intend to include something along that line, but know nothing about them. Had thought I'd cross that bridge when I came to it--perhaps at first with something pretty crude. Am very much open to instruction on this.
Limit switches themselfs are pretty simple, if you're lucky the stepper driver/controller may include support for them.

rjlittlefield
Site Admin
Posts: 21120
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

There will be some sideways force due to tension on the belt. However, I would expect the finest of these belts to be pretty flexible so that the required tension will be small, something like the weight of a human hand which the focus block is engineered to handle in normal use.

I agree with elf that this is probably not a problem.

--Rik

DaveW
Posts: 1702
Joined: Fri Aug 04, 2006 4:29 am
Location: Nottingham, UK

Post by DaveW »

If it were a cog belt it would need less tension than something like a "V" belt and the drive would be more precise like gearing. From Wikipedia:-

"Timing belts

Timing belts, (also known as Toothed, Notch, Cog, or Synchronous belts) are a positive transfer belt and can track relative movement. These belts have teeth that fit into a matching toothed pulley. When correctly tensioned, they have no slippage, run at constant speed, and are often used to transfer direct motion for indexing or timing purposes (hence their name). They are often used in lieu of chains or gears, so there is less noise and a lubrication bath is not necessary. Camshafts of automobiles, miniature timing systems, and stepper motors often utilize these belts. Timing belts need the least tension of all belts, and are among the most efficient."


http://www.directindustry.com/industria ... 69452.html

As a thought, you may be able to find a miniature cog belt and pulleys out of an old scrap computer printer which often use these to move the print head backwards and forwards.

DaveW

AndrewC
Posts: 1436
Joined: Thu Feb 14, 2008 10:05 am
Location: Belgium
Contact:

Post by AndrewC »

I have some modest experience in drive systems.

Belt drive vs direct coupling: potentially more backlash and less precise than direct drive. But, when driving in a single direction (as in a focus stack) the backlash is less important.

Precision: you can't just keep dividing down through gear trains or pulleys and think it gives you more precision. At the end of the day you have a pinion gear coupling onto a rack (in the uscope focus block case). Your effective movement isn't always how much you turn the knob ! Imagine you put a 100:1 reduction gear on your fine focus - you don't magically get a 100x improvement in resolution. The advantage of gearing is that the motor end of your drive system can move further and so you reduce errors coming from there - that is mainly why people microstep - smoother more accurate motion but not necessarily higher precision. It would be really interesting for someone with the $$'s to put a linear encoder position sensor on a microscope stage and see how accurately it does control - I might be obliged to eat humble pie !

Coupling joints / mounting: you do need axially flexible couplers, otherwise your motion becomes imprecise as you force the stepper motor out of internal alignment, and wreck it's bearings ! Don't mount the stepper on flexible mounts - that way the motor moves rather than the shaft. The only case you might consider flexible motor mounts is if you expect the motor to make several rotations for each "focus step", which you might do if you had a massive gear/pulley reduction.

Have fun :)

Chris S.
Site Admin
Posts: 3549
Joined: Sun Apr 05, 2009 9:55 pm
Location: Ohio, USA

Post by Chris S. »

Thanks, Andrew, for your voice-of-experience input--just what I'm looking for.

A couple of comments:
AndrewC wrote:(snip) potentially more backlash (snip)
This gives me a chance to bring up an observation: My two focusing blocks have far less backlash than my Newport model 462 linear stage. Backlash has been so far non-noticable with my Nikon focusing block, and very noticable with the Newport stage. There are of course much more expensive linear stages out there, and some of them are no doubt much better. Perhaps the experiences of those using linear stages and those using focusing blocks (all (of varying levels of quality) may vary enough to be worth noting.
Precision: you can't just keep dividing down through gear trains or pulleys and think it gives you more precision. (snip)
Right--I didn't mean to suggest that these mechanical devices have infinite capacities for precision if simply geared down enough. But I do have a subjective impression that the focus stage I'm using has a bit of capability remaining beyond what I can reliably dial in by hand. If my stage is marked in 2-micron divisions, I suspect I could get down to maybe .25 microns with it--or even .125. I won't know, of course, until I test. Like you, I would highly doubt there is, currently unused, anything like two orders of magnitude.

Cheers,

--Chris

Chris S.
Site Admin
Posts: 3549
Joined: Sun Apr 05, 2009 9:55 pm
Location: Ohio, USA

Post by Chris S. »

Thanks, DaveW. And the Internet does seem to be awash in a vast array of toothed belt offerings. . . .

No question I'll go with a timing belt approach. Now, my pondering has moved on to the issue of how to mount the motor such that it travels with the focusing block (which can be adjusted about 18 inches fore and aft) but is mechanically isolated from it (for vibration reduction). I can think of a few ways to do it--a fun problem to noodle on--but need to pick one and go with it.

Best,

--Chris

AndrewC
Posts: 1436
Joined: Thu Feb 14, 2008 10:05 am
Location: Belgium
Contact:

Post by AndrewC »

Chris S. wrote:.... but is mechanically isolated from it (for vibration reduction). ...
Why ? The motor won't be moving when you take the photo and you will have to wait for any vibration from the focus block + camera moving to die down anyway. That's why you microstep - 8 little steps are a lot smoother than one big one and if you wanted to be really smooth you would ramp up and down again, or just wait a little bit :)

Belts and gears are more fun to watch but direct drive is much more accurate.

Andrew

DaveW
Posts: 1702
Joined: Fri Aug 04, 2006 4:29 am
Location: Nottingham, UK

Post by DaveW »

One way to stop backlash is to provide some friction on the focusing block pulley to damp it so it does not move back slightly when the motor stops.

Perhaps just a bit of thin spring steel bearing lightly on the belt over the pulley at that end, or on the side of the pulley. Or even a spring washer between the pulley and the block, as I think Charles mentioned you had to push the knob in against something similar when refitting earlier?

DaveW

AndrewC
Posts: 1436
Joined: Thu Feb 14, 2008 10:05 am
Location: Belgium
Contact:

Post by AndrewC »

DaveW wrote:One way to stop backlash is to provide some friction on the focusing block pulley to damp it so it does not move back slightly when the motor stops.

Perhaps just a bit of thin spring steel bearing lightly on the belt over the pulley at that end, or on the side of the pulley. Or even a spring washer between the pulley and the block, as I think Charles mentioned you had to push the knob in against something similar when refitting earlier?

DaveW
In this context "backlash" means the free travel you get when you reverse direction, some also use it to mean the initial (rotary) movement you have to make before anything actually moves (linearly). When using belt driven systems I have heard of energy getting stored in the belt and released when the drive stops but you would need a stretchy belt for that !

Andrew

g4lab
Posts: 1434
Joined: Fri May 23, 2008 11:07 am

Post by g4lab »

Backlash can also be taken up by backing up more than the backlash and approaching from the same direction each time. Like we used to do on old worn out machine tools. :lol:

You can find a lot of good info on microscope automation here a website that has several hours of good reading about things like servo versus stepper driverrs , encoder feedback , Z axis drives etc. They are probably the state of the art and are bundled by most of the major scope manufacturers.

DaveW
Posts: 1702
Joined: Fri Aug 04, 2006 4:29 am
Location: Nottingham, UK

Post by DaveW »

Even theoretically a long shaft twists in it's length when torque is applied to one end before moving the device driven at the other end, and then unwinds when the torque applied stops, but the amount is infinitesimal.

I would not think with a cog belt, which I think are usually hardly stretchable nylon or some similar synthetic that you would get the same springiness as in those containing rubber like "V" belts?

G4's remarks remind me of a tale my friend who is in engineering told me. They had a lathe at their firm that was getting a little worn and on which only it's operator could turn to the tolerances required. Evidently he used to lean on the tool holder whilst turning to remove any backlash or judder! :lol:

DaveW

g4lab
Posts: 1434
Joined: Fri May 23, 2008 11:07 am

Post by g4lab »

G4's remarks remind me of a tale my friend who is in engineering told me. They had a lathe at their firm that was getting a little worn and on which only it's operator could turn to the tolerances required. Evidently he used to lean on the tool holder whilst turning to remove any backlash or judder! Laughing
There have been a couple of occasions in my excuse for a life where I was the only one that could coax the required performance out of cantankerous old machine tools and also analytical instruments.

Ya gotta let em know you've got a screwdriver and you are NOT afraid to use it. :lol: :lol:

Post Reply Previous topicNext topic