Cannot make one wheel touch C-profile

Is this normal:

when I had mounted all 8 wheels between the 2 gantry plates, and when I had introduced the C-beam Alu profile, I only managed to get the soft contact to 7 wheels, whereas one wheel - without excenter - was still free running. And the span of the excenter was on wheel at maximum, another was at minimum position.

I have the feeling that there should be only 2 fixed wheels an two axis and 6 adjustable wheels, otherwise the system is "overdetermined" if the gantry plate are not of absolutely identical dimensions. The tolerances of the hole distances are in the 0,2 mm range, I measured.

Ciao Schwipp

 

Hello helping people,
many thanks for instant reply!
I use drill presses, always.
I dismounted all gantry plates and already paired the plates with nearly similar distances of the holes, measured with a vernier of 0,05 mm resolution.
I also measured the diameters of the wheels, all are of the same diameter.
I have the feeling: when I activate one excentric nut it deflcets he whole ensemble of the 2 gantry plates - because the wheels in it are more or less elastic.
On the z-motion there are only 4 wheels with single side fixed screws and the wheels are all in contact all the time, however the stiffness/elasticity will be less because of plastic.

Metallic V wheels are more rigid. Are their contact surfaces polished ore just machined with a lathe, on some fotos I see the grinding lines ?
I think of just replacing 8 delrin wheels by 4 metallic.

thanks for your time when replying
ciao Schwipp

 

In a list of a portuguese OB reseller I read that Tronixmaker in Begium, where I bought, is an official Openbuilds reseller.
a) Is this correct? I am wondering because Tronixmaker advertizes this, but you cannot find am impressum.

Indeed I got fake delrin nut block (each 3,5€) from him which I returned because unsuable.
He also offered me a fake expensive spindle holder which I returned.

I digitally measured the width of the C-beam profile. I is 79.8 mm with an reading unaccuracy of about -+ 0,03 mm.
b) Shouldn'd it be 80 mm? This would explain a little bit.
c) The polycarbonat wheels are about 23.9 mm diameter.

Thanks for reply!
ciao Schwipp

Later:

I think besides the tolerances manufactured too large the problem is:
when I rotate an excenter, it also influences the position of the two gantry plates in respect to to each other, because the screw is deflected. Thus all wheels are altered more or less in position.


As I already wrote, the Z-motion with only 4 wheels, each on a screw which is attached single sided on the one gantry plate, works without problems, all wheel are in contact to the C-beam profile.


I see only one possibility to get rid of the coupled functionality of the excenters and the screws leading through them:

connect the two gantry plates with four large extra spacers, so that the gantry plate are connected very stiff to each other;

and then attach each wheel which a with a seperate screw to one gantry. When I use regular nuts (without blocking O-ring) there is just a 1 mm gap in between two screws, what a good OB-design °°.
I illustrate this in fotos.

 

1) What is the exact distance between the gantry plates;

or more precise: the distance from the inside of the left gantry plate
to the inside of the right gantry plate when used with 8 wheels and the 80x40 mm C-beam profile?

It would be nice to get this value!!

2) Also there should be a recommended torque when the screws connecting the gantry plate (with wheels in beween) are tightened because the polycarbonat

wheel material should not be compressed to much, otherwise it could deform nonrevsal.

 

Hello Rick 2.0,
Thanks a very very lot!
So the wheels contain a metallic spacer in their center along the axis, that is fine.

I think I will fix the gantry plates together without wheels in between and fix the wheels single sided, because I have good experience with such single sided mounting, e.g. I build a wall fixed saw + a very accurate press drill with this, I showed already in my other thread here concernig the nut-blocks.. Here I show in more details how I fixed the single wheels, strong but elastical (including this elastic washers), excenter "spacers"are in each wheel
except one or two.
(Wheel bore is for 12mm wheel axis, screw through this bore is M8, two 1 mm thick plastic stripes press the screw and a piece of aluminium tube, 8mm inner dieameter, 10 mm outer diameter, excentric in the wheel bore against the bore rim, see fotograph, screw has large head diameter to produce large forces when it is inclined by the wheel, there is no thread in the iron which holds the screw).
Works perfect, constant pressure on the wheels all over the total range of travel (1,7 m).

But this special type of handmade excenter is not needed with OB designs of course, don't understand me wrong.
best wishes Schwipp

 

I will try:
6mm spacer + 1mm precision shim + 11mm wheel + 4mm M5-nut = 22 mm for a single sided wheel mounting + the same arrangement on the opposite gantry plate, thus there is a 0,7 mm gap between both single sided mounted wheels.

If I do not use OB precision shims but ordinary shims of the same DIN number,
their thickness (from my supplier) is 1,1 mm, so I could compensate a little bit some deviations.

Rick 2.0 wrote:

"1. Depends on the configuration but the typical configuration of 6mm spacer + 11mm wheel + 9mm spacer + 11mm wheel + 6mm spacer = 43mm. And under normal circumstances you should get there with a tolerance of -0.1/+0.4. (Note those are not official tolerances just what I have typically found to occur.)"
​

In my configuration it counts like this:
6mm spacer + 1mm precision shim + 11mm wheel + 9mm spacer + 11mm wheel + 1mm precision shim + 6mm spacer = 43mm + 1,5 ... 1,8 mm, average about 44.7 mm,
I think the wheels are a little bit less wide than 11mm, can this be?

By the way : from wheel center to wheel it must be 20 mm, because of the aluminium C-profil geometry
==> 0.5*11mm wheel + 9mm spacer + 0.5*11mm wheel = 20

best wishes Schwipp

 

I found the reason for the too small distances between the gantry plates, should have put grease below the excenters and tighten the screws less tight:

I will try a single ended mounting of the wheels with a disk spring 1,7 thick when not pressed flat, minimum flat thickness 0,9 mm at very high pressure, several hundred N. The disk spring replaces the 1 mm thick precision spring, though it is now placed below the excenter and in contact to the gantry plate.

 

Oops, lots of time and pleasure to improve the adjustment of the wheels. It works, all wheels get contact to the c-beam profiles, that was the aim.

Each wheel has its own screw and can be adjusted transversial to the wheels axis without influencíng the 7 other wheels position. The wheel distances in z direction are adjusted with a tools caliper rule (? google translation). The gantry are very stiff and parallel connected with distance tubes.

Now the excenter still moves the screw as before, I will change this, and reverse the excenter so that it only acts on the wheel, the position of the screw will be stable than, which is better.

However I have to use 688 RS bearing (with 8 mm inner diameter) in the wheels than.

I first adjusted the four wheels in direction to the motor (side1).
Afterwards I started adjusting the axial distance of the other four wheels (side 2). Doing so this wheels became loose (because their distance was incorrect some tenth of mm) - but the wheels at side 1 did not alter their contact to the C-beam profile - as it should be but wasn’t the other day with the original recommended OB mounting. I think the main advantage now is that the 2 gantry plates are very stiff connected with the 4 extra spacers, (b.t.w.: of 46 mm mm length), what a big difference!

 

Hello Jeffmorris,
1 mm shim + 6,2 mm gantry plate + 46 mm to the other gantry plate (including an extra 1 mm shim) + 6,2 mm gantry Plate + 1mm shim is about 60 mm,
60 mm m5 screw + M5 threaded bush from Rampa
The final "parallelism" of the gantry plates edges depends very critical on the performance of the spacer tubes. If their ends are not absolutely parallel AND/OR not of equal length you will notice this when you put the mounted gantry plate ensemble upright (both plates touching) on a polished plane, e.g. smooth glass plate during thightening the screws the last turn, the ensemble becomes oblique a little bit; I will try to make the spacers length equal with precision shims of 0,1 and 0,2 mm tickness which I ordered. But the ensemble is VERY stiff.

 

Now I found a mounting with which I get a fairly well balanced force on each wheel.

First I clamped the 2 gantry plates together including 4 spacers, each about 45 to 45.1 mm long; I trimmed the overall length of all spacer to about 0.1 mm, manually with sandpaper of very fine grain. During fixing the screws I pressed the two gantry plate to a plane machined surface to avoid any inclination between them when fixed, this works and is a real improvement.

The four screws clamping the wheels are fixed at only 2 positions (instead of four originally), the 6 other wheels are adjustable with the excenters. These screw are clamped softer because instead of the 2 mm thick washer below the nuts I use 4 packages of 4 disc springs in parallel, 5.2 x 15 x 0,7 mm, DIN 2093, the total spring force clamping the screws is about 1000 N.

Of course I had to widen some holes of the gantry plates, e.g. to use 6 instead of 4 excenters before, all I need was a drill press.

I found that some excenters are not as excentric as they should be and sorted them out.

With this arrangement I can manage each wheel to be in much better, nearly equal strong contact to the C-beam profile.
ciao Schwipp

 

Hello, new and more flexible approach:

Now I managed to get all wheels in contact with the c-bréam profile with approximately the same force on each wheel, the procedure is nothing for beginners.

I excluded some eccentric spacers because they had an adjustment range too small.

I made 6 wheels to be adjustable with 6 eccentric spacers, only 2 wheels are fixed.

First I mounted and adjusted the wheels of each gantry plate separately, being on the c-beam profile,
then I moved them “in position” to each other, measured the distance of each pair of wheels.
I fixed both gantry plate together with long M5 screws, hereby I took care that the distances of the wheels of each pair did not alter ( < 0,1 mm range).

At last I adjusted the nut blocks with the lead screw transverse to its axis directions:
I moved the gantry plates - together with the nut block and the lead screw through it - as close as possible to the c-beam end mount.

I loosened the lead screw at both c-beam end mounts
and tried to shift the lead screw into the ball bearing of the c-beam end mount without friction exactly through its center. I managed to do the necessary adjustment of the 2 nut blocks with my fingers tips after removing the c-beam end mount preliminary. I repeated this procedure twice at both ends - left end, right end, left end, right end - of the c-beam profile. By this I found the final and best adjustment of the 2 nut blocks and the lead screw.

Of course I had to add some extra holes, threads etc. at the gantry plates and the c-beam end mount, this was no problem.

 

To end this story:

This last mounting I described works very well,
because the gantry plates are fixed to each other with 4 screws which are separated from the 8 screws used to adjust the 8 wheels.
So there are more degrees of freedom.

I also found that the width of the C-beam profile can vary from specimen to specimen, even on a single profile the width at the open side of the profile can be different from the width at the closed side, to about 0.2 mm. I noticed this when I had adjusted the 4 wheels of a single gantry plate to be in good contact with the profile at the open side and then removed it from the rails and inserted this gantry plate to the rails of the closed side of the profile, the wheels were completely loose on that side.

However, I have to state that NONE of the items I got from the official openbuilds seller "Tronixmaker" in Belgium showed the OpenBuilds logo on it - so have to assume all items were delivered to the shop by alibaba-dealers or other not known source.

 

Long time has passed.. ..
Meantime i looked up the general tolerances of extruded aluminium profile from manucfactures like ITEM, BOSCH and so on, in Germany.
They all claim 0,2 mm accuracy, not better.
I also know the best tolerance claimed is 0,1mm, used for motorized height adjustable tables with aluminum legs, but in Germany the is only a single producer who claims this for simple rectangular tubes of aluminium.
I found out:
If I adjust all wheels to have reasonale contact - not to soft and not to rigid asrecommended in the OB video - at one position on the C beam profile,
this contact can be quite different when I move the gantry plate unit at another position of the C beam, e.g 30 cm away.
I think:
This is because 0,2 mm - for instance in width accuracy - is a lot
and explains my observations.
So to my opinion it is a hopeless effort to try to make all wheels to be in uniform contact all over the full length of extruded beam profiles.

I tried a mounting of the wheels equivalent to the my post 10 above,
first image,
and it seems it work also for this small OB wheels,
if i replace the inner bearing with a bearing of the dimensions 16 x 8 x 5 mm ( the same whis is also used for the lead screw).
The wheel then is in strong but elastic contact all over the C profile, 0,2 mm change in width is no problem.

Any comment?

best wishes Schwipp

by the way: the official OB dealer tronixmaker is closed and now tells on his closed homepage that he bought his stuff from ritrag in portugal.....But advertized OB earlier.