Comments sought on new build: A CNC Router for RC Gliders and Planes

[kingcreaky]

These lot, make you think too much.... think less........

Just do the best you can do with the tools & experience you have. I spent ages in my build worry about this very point, in the end I just got one rail spirit level flat, and set the other one up off that. BOSH

[njhussey]

Cheers Jazz ;)
.
.
My cunning plan is to get some of my 10mm 6082 plate I have and make a plate to go across the X axis and bolt it to the carriages like so....then put some 100x50 section on it to stiffen it....
.

.
Then I can drill one hole one end of a rail and then using digital callipers (accurate enough???!!!) get the hole at the other end the same distance from the rail side, move the carriage to one end and drill a hole on the opposite rail at that end, slide the carriage up to the other end and mark the hole and drill that. Then by sliding the carriage up and down and nipping the bolts up I should be able to get them parallel to each other. I can then move the carriage near the middle and mark, drill and tap those holes. Once I've a sufficient number of holes drilled and tapped (every other hole?) I should be able to see if it slides freely or if it binds? If it slides freely......happy days, if not then epoxy it is! Sound like a (convoluted) plan?
P.S. I know the bolts should be allen head socket bolts and counter sunk but I couldn't be bothered to draw any.....sorry!

[george uk]

although the amount of deflection mentioned by jonathan may seem small, it would be noticeable over a large machine base, or if your surfacing or engraving anything bigger than 500mm.

In his build, he shows you how to use epoxy to get it level, as close as you would need it.

I also fly RC planes. If you considering cutting wing profiles, you will notice even a small deflection. All depends on what your doing.

IMHO, for the sake of £ 45 for the expox, and a few hours ( no expencive tools needed ), do the resin thing, get it as accurate as possible.

Imagine doing a glider wing profile from foam, 1000mm long each, a small amount deflection on your rails would ruin the wing.

[njhussey]

Understood George, however I think we're all knit picking here! I cheekily pointed out Jonathan's build as he has done what I was (maybe still am??!) thinking of doing and putting the rails directly on the steel. The sort of deflections I'm possibly looking at are in the magnitude if 0.5mm which is one stroke of sandpaper on a wing rib to put it in context. I've built some large and small planes by cutting out round paper templates so there would already be larger errors than what I think these rails will give.

I do think doing the epoxy levelling is worth it, after all why put in the hours on the design, the hours fabricating the frame only to not let it reach its full potential? However Jonathan's "Sufficiently strong" machine (a superb machine) was built with parts made (nested out of one large sheet of Ali) on a machine using supported round rail (I'm sure you've read on different build threads Jazz's comments on what he thinks the linearity of those rails is like!) on non levelled beams ;). Is that a chicken and egg situation??

Anyway, enough fun. I'll have a look at the rails tomorrow and poke about with some feeler gauges to explore...

[Jonathan]

The critical point to bear in mind my old machine uses round rails on the X-axis, which tolerate a lot more misalignment than profile rails. The required tolerances go up a lot when you start using profile rails.


If you can get the rails to run smoothly without the epoxy, then that's probably fine for what you want to cut. The issue is it's hard to tell when a rail is misaligned by only a small amount, as this just applies a large quite large force to the bearings, but they can tolerate that temporarily. The problem is if you don't spot the slight change in smoothness caused by this, then your bearings could wear prematurely due to the increased load. Compare that to epoxy, where unless something goes seriously wrong it guarantees better accuracy and sufficient flatness. The £45 for epoxy is not much in the whole scheme of things when building a CNC router, so I'd strongly recommend it, although it's not mandatory if you can get the rails to run smoothly without.

Excellent point and your 100% spot on.!!. . . . He's talking Bollocks.!! . . . . Measuring to those tolerences needs expensive highly calibrated equipment which I doubt he has either.

I'm not sure if I should dignify that with a response, but I will for now. Are you referring to the angular resolution I quoted? If so then it really should be obvious to you how I measured that from the photos.

However Jonathan's "Sufficiently strong" machine (a superb machine) was built with parts made (nested out of one large sheet of Ali) on a machine using supported round rail [...] Is that a chicken and egg situation??

The mounting surfaces for the Y rails were milled using the bridgeport mill at school, since it wasn't reasonable to expect the required accuracy from my machine.

[njhussey]

The critical point to bear in mind my old machine uses round rails on the X-axis, which tolerate a lot more misalignment than profile rails. The required tolerances go up a lot when you start using profile rails.

The point I was cheekily making was that we get hung up on fine tolerances when your machine is perfectly good for cutting aluminium!

If you can get the rails to run smoothly without the epoxy, then that's probably fine for what you want to cut. The issue is it's hard to tell when a rail is misaligned by only a small amount, as this just applies a large quite large force to the bearings, but they can tolerate that temporarily. The problem is if you don't spot the slight change in smoothness caused by this, then your bearings could wear prematurely due to the increased load. Compare that to epoxy, where unless something goes seriously wrong it guarantees better accuracy and sufficient flatness. The £45 for epoxy is not much in the whole scheme of things when building a CNC router, so I'd strongly recommend it, although it's not mandatory if you can get the rails to run smoothly without.

99.9% sure ill epoxy it but will prob drill and tap the holes first. As you say the £45 or so for the epoxy is insignificant compared to the total price of the machine, its a no brainier. I'll still bolt the rails straight down though...just because!! ;)

The mounting surfaces for the Y rails were milled using the bridgeport mill at school, since it wasn't reasonable to expect the required accuracy from my machine.

Again the rest of it was machined from nested parts on one sheet showing that with round rails on steel it's perfectly fine for cutting Ali!

[Jonathan]

Again the rest of it was machined from nested parts on one sheet showing that with round rails on steel it's perfectly fine for cutting Ali!

Within reason the accuracy of your rail mounting wont affect what the machine can cut - it affects how accurately the material is cut and how long the machine lasts.

[JAZZCNC]

I'm not sure if I should dignify that with a response, but I will for now. Are you referring to the angular resolution I quoted? If so then it really should be obvious to you how I measured that from the photos.

Wasn't referring to any specific area I was saying the whole measuring and graphs etc to the level your quoteing was bollocks because unless you measured and worked from a calibrated reference point with calibrated and accurate measuring devices you couldn't possibly know for sure.!

End of the day the quality of the pie is in the eating so bolt it together and see how it tastes.!! . . . . . Looking at the care taken so far I'd put a small wager on that if just left untouched no epoxy etc it will quite happily cut wood to a very high standard and do it all day long without any grumbles from those bearings.!!

[Jonathan]

Wasn't referring to any specific area I was saying the whole measuring and graphs etc to the level your quoteing was bollocks because unless you measured and worked from a calibrated reference point with calibrated and accurate measuring devices you couldn't possibly know for sure.!

The measurements I was taking were relative, not absolute. That is sufficient for the purpose, if you accept that the flatness of my surface plate as an adequate reference point.

[JAZZCNC]

The measurements I was taking were relative, not absolute. That is sufficient for the purpose, if you accept that the flatness of my surface plate as an adequate reference point.

Unless you could fit the whole machine on the surface plate so have base reference then no I don't.! . . . . . . But to be honest I don't give F@~£ all I know is that for cutting wood then with enough care and attention which has obviously been taken here it will be fine even without epoxy.

He's quite correct that If your machine and Mine for that matter can cut aluminium his will easily cut wood.!!. . . End of discussion.!!