SX2P Gas Strut Modification Issue?

Is it not as simple as needing a new rack and pinion, if you have shimmed the rack and it still had backlash surely it must point to pinion being a bad fit to the rack. Is the pinion secure to its shaft, I have never had mine apart so don't know how it's engineered.

I went with 120N on my strut only because at the time of fitting I read somewhere that 150N is a bit too strong (don't know what that was based on) so with my 120N my head is heavy re balance.

Never had the sticking issue fortunately it's as smooth as you want it though a bit heavier upwards but not arm aching so. I agree with Jason to my simple brain I would have thought the head needs to sit heavy not equally balanced especially with the amount of backlash you have in the rack.

Just a thought straight off the top of my head but if the pinion is a bad fit in the rack and you shimmed the rack until it felt rough but still had backlash then I would have thought the tips of the pinion teeth are bottoming out but leaving gaps between the sides/mating faces of the teeth. I can't see where else the backlash could come from. If that is the case I would be tempted to reduce the o/d of the tips so the pinion sits deeper and mates with the rack The rack would need shimming again of course to make up the difference.

I could of course be wrong but with your explanation of the shimming and still having backlash that's what I imagined must be happening.

rod Interesting! Just checked my junk box there’s a lot of interfaces in the system

1 the rack and pinion are well clear in their respective spaces, but the bottom of the pinion grooves seem to be rounded and could be causing interference.

2 the universal joint has about 0.04mm slack

3 the worm shaft can move in and out due to clearance prob 0.04mm

4 fit between worm and wheel ? Can’t determine

5 the dog clutch between wheel and pinion shaft 0.25mm?

6 clearance between rack and pinion unknown but it must be there.

conclusion A lot of interfaces all of which need clearance to work or clever expensive engineering to eliminate all contributing to the backlash . This makes me even more certain a z feed screw of some type is the way forward and ditch the gas strut and spring system as I did  I never lock the z , not a good idea anyway with cnc.  I have to use light cuts , nothing like Jason in his blogs, but the results are excellent and well worth the effort of converting the machine 

interesting. The rack is 25 wide. Why did they go to the expense of making the pinion gear 75 long? A bit of a waste the end could be reduced to below root dia and a bush inserted to reduce the hole diameter, or use a reduced hole..which to me seems cheaper.  I wonder???

edit.  
re sticking n my mods. I scraped and filed the casting flat, then scraped the dovetails to fit the column. This left the head mount planes out of parallel , so the top was milled true on the Bridgeport   The column was straight within 2 thou. Quite good!  I also had problems with the column  seat on the base casting. It had been attacked with a power sander and would not allow tramming  ( one corner was 30 odd thou below the others) also re machined on the big mill   See my write up of the full cnc conversion  in ME  about 3 years ago.  A lot of parts!

Edited By Zan on 05/03/2021 11:29:37

edit. Autotype errors

Edited By Zan on 05/03/2021 11:37:54

Glad you posted that Zan, because I'd totally forgotten the backlash in the fine feed drive.

So there's no point doing the split rack, - saved me some hassle, but still no closer to a solution, unless I go for a ballscrew - or what about a leadscrew like on the x-y table? Could that be fitted with less drastic modification than turning the motor?

What's the best method for getting the handwheel horizontal?

What was the concensus about a ballscrew being back driven - I thought they could be?

Thanks.

.

I don’t think we have consensus ... but I did post a couple of relevant links, yesterday

MichaelG.

Ball screws can be backdriven but that is easy to avoid. Pinch the head dovetails slightly is one option, another is to put a fixed split plastic collar on the ball screw somewhere where it can be adjusted as needed to give enough braking to hold it against unwanted rotation. An energised stepper motor also has enough torque to hold them against being back driven but you may not want to go down that sort of route at the moment.

Martin C

This has jogged my memory about ballscrews. Yes, they can easily be back-driven.

About 10 years ago we made a small platform about 600mm square, that could be raised parallel to its base. We used four ball screws, one in each corner. In order to prevent the platform lowering under its own weight, we incorporated a worm/pinion drive on one of the screws. This drove the other three via a timing belt drive. The worm was driven by a hand wheel.

We were amazed to find that despite the worm drive being specified well within its non-back-drive ratio, the platform lowered very gently and smoothly under its own weight. We ended up having a locking screw on the drive shaft.

Is it easier and cheaper to incorporate a conventional leadscrew and nut, similar to those on the x/y table. They can't be back-driven?

You are trying to avoid backlash so that the head is driven up or down predictably. Ball nuts and screws are better for this but standard leadscrews are harder but not impossible to back drive. So if you go the ball screw route you need to prevent back driving. If you go the leadscrew route you may need to be able to adjust out backlash.

Martin C

And unless I want to feed the head using a wheel stuck horizontally on the top of the machine, I'd need another gear system to re-orient the drive.

Chances are I won't bother with any of it to be honest, way too much hassle to start major modifications to the head. I just wanted a fairly quick fix that would reduce the issue if not eliminate it, but I guess if there were any out there, someone would have already done it.

Anyway, thanks all; it's been emotional. Hopefully I can now start some threads that don't have the word "issue" in the title...