[joeavaerage]

Hi peteeng,

Epoch is for plastic, foam and timber so the pushback is much less then metal.

Don't be duped....you need as much resisting torque as you can get. I got lucky and had 'enough', but only just 'enough'.....more would be better, and you'll need more than I as the
radius at which you will be working is much greater than mine.

marketing txt it says no backlash but in the tech specs some have 5arcmin and some have 10arcmin..

I think that is wrong, or a misprint. If they are genuine HD (the original Japanese patent holders) then that should be 5 to 10 arc seconds, and they do not describe it as 'backlash' but 'lost motion' which
reflects the flexure of the wave cup at rated torque. In any mans language very good indeed.

I did squish the bearing this morning and it got better the harder it was squished. I used an M8 bolt and some large washers but I'm bending the washers so need to make a better rig... But its looking good so far...Peter

I think you've got the best of it. When a stub axle gets put through such bearings followed by a splined flange that mounts the wheel studs, and gets torqued 'until your eyes bulge',
that is s technical term for doing it up as tight as you can, the two inner races are hard up against each other and the bearing assumes its designed pre-load/clearance as the case may be.
If you have clamped them up until the washers start to deform, you've already probably got the inner races hard up against each other and the bearing will assume its designed properties
or as close to that makes no difference.
Craig

[peteeng]

Hi Craig - I'll reread the manual may have been arcsecs that would be more like it. The planetary is definitely arcmins. I rebuilt the squish rig with M12 bolts and used long spanners. I think the crush is about 0.1mm. Spins nicely. I've designed the rough axle for the trunnions and will use a 1.0mm pitch locknut. I'll laser cut a custom spanner and torque it "hard". Peter

[joeavaerage]

Hi peteeng,
yes good planetaries are arc minutes, my mini-mill had < 2 arc min lash planetaries. My current trunnion gear reducer is <2 arc min...which surprises me because you'd swear that would show up as lash in the trunnion,
but I can 'zero it' to less than 5um per 75mm or 13 arc seconds. My current C axis gear reducer is <1 arc min and I cant measure any lash in that either. Not complaining mind!

Would not get to carried away with torquing the wheel-bearings, the duty that you are calling on them for is negligible.

You did comment in an earlier post that you'd quite like to use the same bearings as the C axis/platter bearings. I'm less sure that you would be happy with them in that role.
I suspect even when done up tight there will be a little clearance, being typical of wheel bearings, not much mind, but some. I suspect if you had a platter in just one
of the wheel-bearings you might expect the platter to rock a bit, I'd guess an arc min or more which would correspond to a motion of 0.1mm or more at the periphery of the
platter, which you may find objectionable.

You might get around that by carefully removing one of the inner races and grinding a few um off the mating surface and re-install it. What might have been a little clearance would now vanish
and result in genuine preload. Might take a little experimentation to perfect, but if you did it would mean a very useful and cheap source of platter bearings.
It was not an uncommon occurrence when fitting these things to accidentally knock one of the inner races out of a unitized bearing. When installing them in a car often requires tens of kN's and a sh******t ton
of swearing. We had pullers that were based on a 30mm diameter ACME thread, and fairly fine pitch at that and hardened and blued to all get out and could still strip and bend them. They are not easy to fit.

Craig

[peteeng]

Hi Craig - I'll only know the true clearance when I build an axle and mount it up and be able to magnify the wobble. So onward. I like the solution, we shall see how good it is in practice... Peter now I have to get 10mm fine metric bolts, should be easy. But I may re-tap them coarse so bolts are easy to get... Oh by the way when done up real tight the bearing still swings under the self weight of a spanner so I think that's great...

[joeavaerage]

Hi peteeng,
I think those are JIS threads, 10 x 1.25, whereas ISO fine is 10x 1. You'll have to rat around in the local garage to find some bolts.

Craig

[joeavaerage]

Hi peteeng,
I think if you look closely the threads have been formed and then induction hardened. The discoloration around the bolt holes suggests that is the case. If that is correct
there will be no re-tapping them to accommodate ISO course.

Any local garage that is accustomed to servicing Japanese cars will no doubt have boxes of surplus bolts. The Japanese tend to use good quality bolts for this sort of thing, maybe not 12.9, but probably
10.9 and would expect them to exceed 8.8.

Craig

[peteeng]

Hi all and sundry - Went around to the local fastener shop and they have a good range of M10x1.25 bolts which fit the bearing. So no need to over think it. Perfect. Speaking of units, I'm in a deep conversation with Autodesk about material properties in Fusion. If you pick mm as a unit in preferences the density is spec'ed at some weird N.s^2/mm^4... Then if you update the units they don't apply until the next model!! So I have put forward a list of suggestions. 18/20 of them have been put in the development list. The shear modulus issue and the density issue seems not to be translating well. I wait for the next round of conversations with Pradip... Peter

[peteeng]

Hi All - I have had several discussions with Fusion support regarding density and masses displayed. The guy even made a nice little video for me to run me by the entries. But the short story is; 1) there are default values to enter for each environment 2) there are global preferences to enter 3) there are document units to enter (design simulation and drawing settings) then there are menu settings like when you enter txt mass. So the long story is that it is overly, way overly complicated to set up the densities and masses easily in Fusion. By the way its now called Fusion they flicked the 360 some time ago apparently... So I have said other CAD systems have a one entry table for density units and they work easily and well and that Fusion is way too complex in this area so fix it (please)... Peter

[peteeng]

Hi All - I have been purging the workshop of old stuff and old jobs. Decided to turn my trial surface plates into stepping stones... I can recover them if needed into the future I suppose, weathering them will be good for them. Peter

[peteeng]

Evening All - Question for all you 5 axis gurus. I'm figuring out the relationship between the rotary table and the Y axis. Should the XY axis be able to reach ABC&D or only required to reach half way across (plus a bit) as the platter turns? Then if you turn the trunnion up 90degs the XYZ should be able to reach all projected surfaces on that axis? ie 300x500mm in this case. I plan on a Dia500 and 300mm high envelope... Any thoughts? Peter

[joeavaerage]

Hi peteeng,
my mill has axis travels of 350mm x 350mm x 350mm. These travels were pretty much determined by the C5 ballscrews that I could get, and the capacity to cast and heat treat the axis beds.

The center of the machine, ie where the X axis has 350/2=175mm travel both to the right and to the left, and where the Y axis has 175mm of travel forwards and backwards id the geometric
center of the table.

When I designed my trunnion and platter I sought to place the rotational center (of the fourth and fifth axes) at the center of the table....or at least as near as I can.
Thus, with the trunnion table at zero degrees, ie dead flat the center of the platter is at the center of the table and thus I retain the same travels, namely +-175mm in both X and Y directions.
The center of rotation is 20mm (approx., 22.521mm... near as I can measure) above the center of the platter. With the trunnion at 90 degrees I have still +-175mm of travel in X but +155mm and -195mm
in Y.

I definitely wanted the center of rotation above the platter that I might devise tooling that allows me to touch off to the rotational center, as it seemed to me the logical way to setup your machine, with the least fuss and bother,
and greatest repeatable accuracy. However, neither did I want it too high above the platter as that increased the moment of cutting forces but also skewed the available Y axis travels, ie I wanted to retain a largely
symmetrical machining envelope about the rotational center irrespective of trunnion angle.

The Z axis bed of my machine is bolted to the upright of a column frame. I can unbolt and shift the Z axis bed at will. I have not done so yet and consequently I've lost significant Z axis travel by virtue of the Z height of
the trunnion/platter combination. The travel above the platter and vice is only about 80mm at the moment, which is enough for most jobs but over the next month I will raise it. I have yet to decide how much by.
I'm thinking that with the trunnion at 90 degrees I will make the rotational center +175mm and -175mm in Z travel. While this seems logical with the trunnion at zero degrees I could lower the Z axis to be well below the platter,
ie I'm wasting Z axis travel....so my +175mm-175mm is being rethought.

Craig

[peteeng]

Hi All - I want to build an enclosure for E5 to control dust. I have tried to integrate a enclosure into the last machine but motors and things got in the way. So a "outside enclosure" maybe the go. Looks like a fridge! But is doable... Plus I want to integrate the spindle controls and machine controller into the facia... a few things to think about. I think I may make the Dia500mm platter a little smaller Peter

[joeavaerage]

Hi peteeng,

I think I may make the Dia500mm platter a little smaller Peter

Yes, I certainly think making the diameter smaller will make for a much easier design and build task.

Craig

[suspension]

Hi Pippin - Just back I posted this. Its shrinkage of HE80 over 56 days. So at 28 days its 553ustrain and at 56 days its 560ustrain. So stable in reality? I plotted the shrink data and it looks flat at 60 days. The chemist said the 1800 day (4.9 years) data is same but not published yet... Got to jump somewhere... Plus if you add aggregate say 50% by V then you halve shrinkage.

Peter?

Hi Pete,

how did you test for the shrinkage? Did you use a length comparator?
My tests with the grout I selected gives me high shrinkage (like 10um per day for a sample that is 220mm long) after 28 days. I am trying to figure out if this is an issues with my test setup.

Thanks
Sus

[peteeng]

Hi Sus - I have relied on the suppliers data for shrinkage. 0.010/220= 0.004% thats very small (you say per day? so what's it at a month?)... The grout I have here is 0.025% at 28 days - Sika3350 and HE80 is 0.046% at 28days and if you add aggregate it drops to 0.017% HE80AG

500ustrain is 500/1,000,000*100= 0.050% Peter

[peteeng]

Hi All - I have been reading up on 5 axis machining geometry and figuring out Epochs range of motion to optimise it. I came across a thought about 4 axis tilting head machines and have decided that a tilting head with a rotary table has several advantages. Here's a video of one in operation. Using a trunnion means that as the trunnion tilts it moves away or towards the tool. This means long tools have to be used. Plus there are two rotary axes operating simultaneously in series. This could mean poor accuracy. We have discussed the accuracy of the gearboxes before. With the tilt on the z axis the machine can be more compact, use short tools and would seem to be stiffer to me. Anyone had experience or can comment on this change in direction? I'm morphing V17 into this config now... Peter

https://www.alibaba.com/product-deta...7e9c4fd8LIGDbx

[joeavaerage]

Hi peteeng,

Plus there are two rotary axes operating simultaneously in series.

So you have a rotary axis (platter) at the end of the X & Y axes, and another rotary axis on the end of the Z axis. Around the loop there are three orthogonal axes and two
rotary axes. No matter where you break the loop there are always the equivalent of five axes in series. Whether they be two rotary axes in series or interspersed with one or more
orthogonal axes matters not. The error in ANY one axis, be it rotary or linear propagates throughout all axes.

Using a trunnion means that as the trunnion tilts it moves away or towards the tool.

Incorrect. With the tool at the intersection of the A axis (trunnion) and the C axis (platter) you can rotate either or both axes and the tool tip remains in the same location ie the machine center.
That is after all why its called the machine center.

With a rotating head you require TCP, whereas with a trunnion you do not.

Craig

[joeavaerage]

Hi peteeng,
attached some pics, being worth a thousand words.

Set the DROs to zero with the tool tip at or near the machine centre. Then rotated the trunnion by 60 degrees, and note the tool tip remains at or near the machine centre.
Rotating the platter will have no effect either. This is the advantage of having the machine centre exposed, you can get to it to touch off, or otherwise have jigs
that can allow you to place stock in relation to the machine center...then run the toolpath....without the need for TCP.

Craig

[peteeng]

Hi Craig - I have no problem centering the trunnion rotary to zero. Machines come with fixtures to do this and I was going to make a ball fixture for this purpose suitable for Epoch. In another thread a guy was trying to make a lifting gantry to get 500mm Z and it was getting silly. This took me down the road of the tilting head.

But I have in the past looked at 4 axis machines where the head tilts to make aerofoils and hydrofoils so I can use a flat end tool for the machining. They are very common for making mass produced reliefs and panels in timber.

It occurred to me that adding a platter solves several issues. I then found a couple of machines that work this way. So the space used is much less, in fact it could be reduced to a more mill like structure vs the high wall structure, see video. I can't see any fatal errors in the config vs a trunnion. There maybe angles it can't get to.

I think the trunnion roots is its an add on to a 3 axis machine vs starting from scratch to get to 5 axes. Then the config has grown up from there.

https://www.alibaba.com/product-deta...7e9c4fd8LIGDbx

Anyways - I've just finished a mash up visual of the machine, don't know what to call it config wise, its a 5 axis machine.... My other thought is that this config can be supplied as 3 axis then upgraded to 4 or 5 axis... a very good marketing feature. Peter

[peteeng]

Hi All- Here's a Mill version of the configuration. Peter

https://www.youtube.com/watch?v=_fQOYuluW5s