Z Axis Treaded Rod
Posted by madmob
|
Re: Z Axis Treaded Rod February 12, 2014 04:58PM |
Hi Madmob,
The nut and stud provided are both zinc-plated mild steel - it's not the best combination, but mine ran for several weeks without issue before I replaced the nut with a longer one made of brass (I tapped mine myself and used johneato's replacement nut trap to house it (edit1 - couldn't find this link until I'd posted
[forums.reprap.org] - treth [forums.reprap.org] has recently sourced a brass nut from ebay in the UK).
It's not generally a good idea to have two wear components of the same material (especially soft steel), brass on steel works well, since the worn brass islubricious [edit3] slippery [/edit3 :p], hardened steel on hardened steel also works well (like the bearings on smooth rod). It's also potentially a problem to use tapped rod inside a tapped nut depending on the quality of the finish of the tapping (since burrs wear and form a grinding paste hard enough to further grind each surface, or to stick to both and cause further wear), rolled stainless stud has a work-hardened surface, with less burrs and a more regular pitch than is generally found on mild steel stud (rolled or not), so would be preferable. The zinc coating on mild steel stud is designed to prevent corrosion in static applications (it forms a sacrificial anode), not to provide a smooth bearing-surface for repeated fastening/unfastening - the zinc being quite soft may lubricate for a while but can also cause local binding, enhancing wear (edit2 - this local binding and subsequent separation is known as galling, and happens particularly if both surface are similar - stainless steel on stainless steel is particularly bad as it happens, and you'd probably get a similar result if both the nut and stud were stainless steel).
If you're using bed compensation, then the stud and nut are continuously in motion over a small fraction of a revolution (how small depending on how level your bed is) for each x or y excursion - this adds up to a lot of repeated motion over a fairly small bearing surface on the stud for each layer printed (I haven't used compensation much until some recent large prints, and I was struck by just how much work is done by Z when compensating) - if your bed is out by 0.2 mm corner to corner, then 1/4 of a turn on the thread is being repeatedly worn as the head moves over that range. The print I did today has 126 thousand moves, I guess 50% of them wound the rod one way, the other 50% wound it the other,
there were 148 layers of 0.2mm height, so each 0.2mm (1/4 of a pitch on the m5) ran backwards and forwards 850 times (of course there are maybe five or six pitches in the standard nut, so each section will get 4 or 5 repetitions of the 850 winds) - quite a lot of wear and tear for something that's made to tighten once then forget. Worse than what happens to the stud is what happens to the nut - over a print the wear zone moves a few centimetres over the stud, the 5mm or so of the nuts thread is continuously worn (the two other mentions I've seen of failure in the Z axis mechanism have been about the nut being totally stripped).
This is all ball-park arithmetic, and might be out by a factor of four or more, but nonetheless there's a hell of a lot of moving back and forth in the design - rub any two bits of metal together a few thousand times and see the difference in their appearance
I'd recommend two things to avoid it getting worse or reappearing - firstly, get the bed as level as possible to minimise the need for compensation, secondly buy or make a brass or delrin/acetal nut which will sacrificially wear/re-form self-limit the damage (you can either buy or tap your own brass or delrin, you can also thermo-form your own delrin "eva-nut" like this [bbs.homeshopmachinist.net])
Edited 3 time(s). Last edit at 02/12/2014 07:54PM by rayhicks.
The nut and stud provided are both zinc-plated mild steel - it's not the best combination, but mine ran for several weeks without issue before I replaced the nut with a longer one made of brass (I tapped mine myself and used johneato's replacement nut trap to house it (edit1 - couldn't find this link until I'd posted
[forums.reprap.org] - treth [forums.reprap.org] has recently sourced a brass nut from ebay in the UK).It's not generally a good idea to have two wear components of the same material (especially soft steel), brass on steel works well, since the worn brass is
If you're using bed compensation, then the stud and nut are continuously in motion over a small fraction of a revolution (how small depending on how level your bed is) for each x or y excursion - this adds up to a lot of repeated motion over a fairly small bearing surface on the stud for each layer printed (I haven't used compensation much until some recent large prints, and I was struck by just how much work is done by Z when compensating) - if your bed is out by 0.2 mm corner to corner, then 1/4 of a turn on the thread is being repeatedly worn as the head moves over that range. The print I did today has 126 thousand moves, I guess 50% of them wound the rod one way, the other 50% wound it the other,
there were 148 layers of 0.2mm height, so each 0.2mm (1/4 of a pitch on the m5) ran backwards and forwards 850 times (of course there are maybe five or six pitches in the standard nut, so each section will get 4 or 5 repetitions of the 850 winds) - quite a lot of wear and tear for something that's made to tighten once then forget. Worse than what happens to the stud is what happens to the nut - over a print the wear zone moves a few centimetres over the stud, the 5mm or so of the nuts thread is continuously worn (the two other mentions I've seen of failure in the Z axis mechanism have been about the nut being totally stripped).
This is all ball-park arithmetic, and might be out by a factor of four or more, but nonetheless there's a hell of a lot of moving back and forth in the design - rub any two bits of metal together a few thousand times and see the difference in their appearance
I'd recommend two things to avoid it getting worse or reappearing - firstly, get the bed as level as possible to minimise the need for compensation, secondly buy or make a brass or delrin/acetal nut which will sacrificially wear/re-form self-limit the damage (you can either buy or tap your own brass or delrin, you can also thermo-form your own delrin "eva-nut" like this [bbs.homeshopmachinist.net])
Edited 3 time(s). Last edit at 02/12/2014 07:54PM by rayhicks.
Sorry, only registered users may post in this forum.