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

Quote:

---

Originally Posted by EddyCurrent

Hooray ! it's done, the build will speed up now that hurdle is out of the way, looks good and a plus for Reactive Resins

---

Cheers, yes the build will speed up now....looking forward to cracking on with this in the new year!!

Wow, that's one heck of a nice job you've done there. I will most certainly be asking for guidance when it's my turn to run the epoxy gauntlet.

Cheers Graham, as with everything the key is in the preparation.

I was looking at the leaks before I left work this afternoon and all the leaks are actually where the mdf was joined. Next time I'll have more time and use wood glue and then seal the joints with sealant. Will also use more silicone sealant under the moat as it's easy to get off the metal.

Sat down after Christmas dinner [emoji490] [emoji486] [emoji485] and I really want to pop into work to see how the epoxy is doing.....there's no hope for me I know!

I have decided to use the 20mm thick Ali plate (don't know the grade) as a bed for it. I'll buy the bits for the Y axis gantry and the Z axis in eco cast, especially as work are paying for it [emoji6] ill get it on order when I get back to work so I can crack on ASAP.

http://tapatalk.imageshack.com/v2/14...6c73d05bd8.jpg

Just popped in to work as I was round the corner recycling some of the Xmas paper, boxes etc. It's set nice and hard, shiny and smooth. It's going to be a pain getting the moat off and the silicone sealant but overall I'm happy with it.

http://tapatalk.imageshack.com/v2/14...102f418c23.jpg

http://tapatalk.imageshack.com/v2/14...0f8d44a831.jpg

http://tapatalk.imageshack.com/v2/14...5c09c9bcde.jpg

http://tapatalk.imageshack.com/v2/14...97e9509360.jpg

Great job, looks like that epoxy came out perfectly! I've had some good luck removing sealer with a hard wire brush.

Ok so the MDF moat is off, wasn't too bad to get off in the end.....but in hindsight I'd make sure the MDF was not oily (been hanging around in the machine shop for a couple of months) and put brown parcel tape on it as it doesn't stick to it and comes off cleanly. I'd also put parcel tape on the steel where the moat was going so that the silicone came off easier too. It's nothing that a bit of elbow grease won't sort. I've found the best tool to scrape off the MDF is an indexable lathe tool...not scraped it fully in the below picture

http://tapatalk.imageshack.com/v2/15...06a3876bfc.jpg

Here you can see the meniscus, the epoxy is about 7mm thick.

http://tapatalk.imageshack.com/v2/15...4909bf3f4b.jpg

Here you can see the difference between the wrong ratio mix of epoxy and the correct...the incorrect is on the right and although you probably can't see it the epoxy is a bit grainy looking and opaque whilst the correctly mixed epoxy is far more translucent and shiny...

http://tapatalk.imageshack.com/v2/15...16d19e150d.jpg

One rail is done and the other is half done. More MDF stuck to the epoxy on this side so it's taking longer to get off...

http://tapatalk.imageshack.com/v2/15...1ab47fddbd.jpg

http://tapatalk.imageshack.com/v2/15...a90577f6fd.jpg

http://tapatalk.imageshack.com/v2/15...cfdbeeaf7a.jpg

http://tapatalk.imageshack.com/v2/15...2c464c55ef.jpg

Quick video of me getting a section of the MDF and silicone sealant off today.

Also filing down the meniscus on both sides.

http://youtu.be/pnQsH_rOY1E

If its not to late in the clean up game, get yourself down to the nearest UPVC setup, ask them for a bottle of PVC cleaner, you want the clear stuff not the "jiff like" stuff, it's great I use it for everything it's designed to desolve silicon sealant and it dose so like magic, it doesn't turn it into a pasty mess like you would think it just somehow frees the silly up enough that you can just peel it off.

Seriously, get some anyway it removes all sorts really well. I use ISOPROPYL for some stuff but the PVC cleaner works great for allot of things, £2.50 here for a one 1ltr bottle.

.Me

Cheers Lee, I'll have a look at that!

Finished off the last of the cleaning up of the epoxy this afternoon.

Worked out how I'm going to get the rails square. We have a 24" Mitoyo caliper which just happens to be the correct length for the spacing of the rails. I've drilled and tapped one hole (after shortening one rail to the correct length) on one rail as the base reference and ill clamp down the other ends using the caliper as a measuring stick. I'm then going to use the milling machine to mill a piece of 10mm Ali plate down one long edge and both ends, then ill drill mounting holes for the carriages and using squares off the reference face of the carriages make sure all is square before tightening down the plate. I can then use this to slide up and down to get the rails parallel and then mark and drill the holes.

Anyone see any issues with this method?

http://tapatalk.imageshack.com/v2/15...adf1db8a75.jpg

http://tapatalk.imageshack.com/v2/15...d64764feb6.jpg

http://tapatalk.imageshack.com/v2/15...7377cf33df.jpg

Depends on whether you have a gantry to use as a gauge? I had built the structure of my gantry in advance for this purpose, so I bolted down one rail along the box section centre line, placed the second rail on its box section, and attached the carriages to use to align the rails. I think, in essence, you are doing the same thing but you will then pick off the rail spacing dimension to finalise the gantry. Hiwin recommend aligning the rail against a machined step but where this is not possible you align the second rail from the first. My only reservation is how you guarantee the straightness of the first rail but I think you can finally adjust this by test cutting. I hope so, anyway!

Looks a nice epoxy job.
To fit the first rail I clamped a good wide straight edge on top of the beam then butted the Hiwin rail up to it, like Neale said about the machined step. Then I mounted the gantry and used the bearings themselves to line the second rail, just nipping the bolts up slightly. Before fully tightening the second rail down I put the straight edge up to it to make sure there were no kinks. After that I also checked the distrance between them at points along the length just too make sure they were parallel.

Quote:

---

Originally Posted by Neale

Depends on whether you have a gantry to use as a gauge? I had built the structure of my gantry in advance for this purpose, so I bolted down one rail along the box section centre line, placed the second rail on its box section, and attached the carriages to use to align the rails. I think, in essence, you are doing the same thing but you will then pick off the rail spacing dimension to finalise the gantry. Hiwin recommend aligning the rail against a machined step but where this is not possible you align the second rail from the first. My only reservation is how you guarantee the straightness of the first rail but I think you can finally adjust this by test cutting. I hope so, anyway!

---

Quote:

---

Originally Posted by EddyCurrent

Looks a nice epoxy job.
To fit the first rail I clamped a good wide straight edge on top of the beam then butted the Hiwin rail up to it, like Neale said about the machined step. Then I mounted the gantry and used the bearings themselves to line the second rail, just nipping the bolts up slightly. Before fully tightening the second rail down I put the straight edge up to it to make sure there were no kinks. After that I also checked the distrance between them at points along the length just too make sure they were parallel.

---

Thanks guys.....

I'm going to do a similar thing to both of you. I don't have a good straight edge that I can butt the rail up to so I'm going to use the first rail as a reference and use the dummy gantry to get the second rail square to the first. Regarding striaghtness of the rails I might look at using a small laser pointer that I have to shine along the length to check for any bend in the rail, not sure if this will work but don't see why not in principle?

Got some more cutting discs (and G clamps to clamp the rails) on the way in to work this morning so I can cut the second rail to length. Looks like it's going to be lots of drilling and tapping today, hope there's not too much proper work to get in the way of doing this...!

But how do you know the first rail is striaght.???

I wouldn't waste any time setting the second rail without the gantry on the machine. Set the master rail straight and parallel to frame then run set slave rail using gantry running up n down feeling your way along as you tighten up the rail. . . Feel for binding is a very good guide in setting up.!

Quote:

---

Originally Posted by njhussey

Got some more cutting discs (and G clamps to clamp the rails) on the way in to work this morning so I can cut the second rail to length.

---

I rarely use metal cutting discs so I bought these after reading the description, I was amazed how good they are; quick, little mess, cuts rails like butter. !00 times better than the thicker type.

http://www.toolstation.com/shop/p39737

Quote:

---

Originally Posted by EddyCurrent

I rarely use metal cutting discs so I bought these after reading the description, I was amazed how good they are; quick, little mess, cuts rails like butter. !00 times better than the thicker type.

http://www.toolstation.com/shop/p39737

---

I've got some 1mm thick ones, as you say they cut through the rails like a hot knife through butter. I've got the Y & Z axis rails to seperate (curently two rails 1100mm long with 4 blocks on each soon to be split.....) but now thinking of making the Y axis rails longer to get the carriage spacing wider. Just re-drawing it to see if I can get away with slightly less Z axis travel.

Quote:

---

Originally Posted by JAZZCNC

But how do you know the first rail is striaght.???

I wouldn't waste any time setting the second rail without the gantry on the machine. Set the master rail straight and parallel to frame then run set slave rail using gantry running up n down feeling your way along as you tighten up the rail. . . Feel for binding is a very good guide in setting up.!

---

Dean, how do I make sure one rail is straight without buying a proper striaght edge? My frame is not straight enough to align the first rail, or at least I don't think so as it's just plain 100 x 50 x 4mm steel, with any sort of accuracy?

I was thinking of getting some of my 10mm thick Ali and using the milling machine machine as much as I can down one edge and then turn it round and then using my edge finder set it parallel again and machine the rest to give a straight edge. Would this do?

Quote:

---

Originally Posted by njhussey

Dean, how do I make sure one rail is straight without buying a proper striaght edge? My frame is not straight enough to align the first rail, or at least I don't think so as it's just plain 100 x 50 x 4mm steel, with any sort of accuracy?

I was thinking of getting some of my 10mm thick Ali and using the milling machine machine as much as I can down one edge and then turn it round and then using my edge finder set it parallel again and machine the rest to give a straight edge. Would this do?

---

How about using some fishing line stretched very taut along the length. ..Clive

Quote:

---

Originally Posted by Clive S

How about using some fishing line stretched very taut along the length. ..Clive

---

Wouldn't that be a little flexible? Plus I've got an end stop on one end so can't wrap it round the rail, would have to accurately drill a hole through the steel.

Just looking at a Veritas straight edge http://www.axminster.co.uk/veritas-steel-straight-edge which I could clamp to the rail to keep straight?

Quote:

---

Originally Posted by njhussey

Wouldn't that be a little flexible? Plus I've got an end stop on one end so can't wrap it round the rail, would have to accurately drill a hole through the steel.

Just looking at a Veritas straight edge http://www.axminster.co.uk/veritas-steel-straight-edge which I could clamp to the rail to keep straight?

---

You could clamp the line to the rail with a spacer say 1mm between the rail and line then do the same at the other end with spacer. With the line tight you can then measure with a feeler along the length after fixing one end first. ..Clive

As Jazz said, sliding a carriage along is a pretty sensitive test of whether or not the rail is fairly straight - you quickly feel the graunching sensation if it isn't! My design uses an aluminium block with two carriages bolted to it (gantry then bolts to these later) and having two carriages fixed in line is even more sensitive. No way am I going to buy a 1.8m straightedge for alignment purposes.

Just bolting it down by eye (I loosely bolted one end of mine, then wiggled it a bit to let it find its own line) is probably good enough for a first pass, with a sanity check of running the carriage up and down. It's not going to be many thou' out, I reckon. I then set the second rail to this, knowing that it's probably not absolutely straight but good enough to work, and will then fine-tune it by cutting test pieces once the machine is in a state to actually cut something. It's another example of the "you can't build it accurately enough but you can tweak it afterwards" home-workshop approach...

Well as I said earlier I've a low power laser (that i was going to use on the mill for alignment) that I've played with. Taped it to a DTI magnetic stand and aligned it so it shone along the rail. You can tell any high spots as the line gets brighter. The spots are dust/dirt. Will tweak this tomorrow but the rail looks straight enough. The carriages certainly slide easily with no binding, I'll try 2 bolted on a piece of Ali tomorrow after machining a reference edge to line the carriages up to.

http://tapatalk.imageshack.com/v2/15...9694a69032.jpg

http://tapatalk.imageshack.com/v2/15...521e48d924.jpg

http://tapatalk.imageshack.com/v2/15...5f9925ab75.jpg

Neil, fix a nice plate to both bearings on your choice of master rail, then pop a clock dial on via a mag base featuring a long reach arm/boom to get to the other rail with the clock, then traverse the master rail via the plate and bearings dialing in your second rail as you go.

Ideally you would want to do this with out the bearings and plate and so just moving the mag base along the master.

That's how I always do it for the machines I build from wood [emoji23] .

.Me

I've decided to cut the 20mm piece of Ali I currently have to the correct width for the bed which will leave me a 100mm (ish) wide strip 1000mm long (ish) which I can use for mounting blocks. I'll cut a piece 200mm long and machine a carriage block...

Attachment 14339

Then after attaching the carriages on the master rail I'll move it up and down and clamp the rail ready to drill.

I've made a small drill guide to go in the rail so the drill bit is kept at 90° (or as near as damn)

Attachment 14340

Attachment 14341
I've put a piece of 10mm Ali plate on the bed and moved my Pillar Drill on to it to drill the holes, a bit like below...

Attachment 14342

Should make drilling easier (but slower) and I hope not to snap any more drill bits as I snapped one in an end hole on the rail :hopelessness::hopelessness:

Neil, It's obviously best to drill the holes spot on but if they are tapped at an angle it's wasted effort, I like your drill guide so maybe a similar thing is reguired for the tap or maybe you could just open out the drill guide hole slightly.

Sliding carriages along the rail gives some indication of the straightness, but very little indication of the twist. Fortunately for us these rails have very low co-efficients of friction, but when it comes to aligning them this makes small errors in straightness harder to detect as they don't produce much force.

Using an indicator mounted on one rail carriage to measure and fix the distance between the rails is good once you've got one rail straight. If you do if before then, you'll just copy the error - so if one rail is banna shaped the other one will be the same banna shape. To be fair that's still better than having a banna and some other fruit...

From a practical point of view, if you've got the rails aligned well enough that the bearings run smoothly then the bearings should last a long time, so it's arguably OK to use but you will need to have realistic expectations of the accuracy.

Quote:

---

Originally Posted by EddyCurrent

Neil, It's obviously best to drill the holes spot on but if they are tapped at an angle it's wasted effort, I like your drill guide so maybe a similar thing is reguired for the tap or maybe you could just open out the drill guide hole slightly.

---

Funny you should say that....I've just made a tap guide as I had that very thought. Will be tapping them by hand, shant be impatient and use a drill!

Quote:

---

Originally Posted by Jonathan

Sliding carriages along the rail gives some indication of the straightness, but very little indication of the twist.

---

He's right, I forgot about that. Neil just go and buy a straight edge :concern: I remember putting one across the top of my rail bearing beams to find they were leaning out slightly. This was due to the welding process but the epoxy job brought it right again. I did however put the straight edge across the top of my rails once fitted to ensure they were still level. Obviously if they are out and you bolt down the gantry cross member there will be a lot of stress put on the Hiwin bearings.

Quote:

---

Originally Posted by Jonathan

Sliding carriages along the rail gives some indication of the straightness, but very little indication of the twist. Fortunately for us these rails have very low co-efficients of friction, but when it comes to aligning them this makes small errors in straightness harder to detect as they don't produce much force.

Using an indicator mounted on one rail carriage to measure and fix the distance between the rails is good once you've got one rail straight. If you do if before then, you'll just copy the error - so if one rail is banna shaped the other one will be the same banna shape. To be fair that's still better than having a banna and some other fruit...

From a practical point of view, if you've got the rails aligned well enough that the bearings run smoothly then the bearings should last a long time, so it's arguably OK to use but you will need to have realistic expectations of the accuracy.

---

Surely the twist will be sorted when the rail is bolted tightly down, that's assuming that the epoxy is perfectly flat of course?

I'm going to use the sliding carriage method to drill and tap the holes, there is a clearance of 0.5mm either side of the bolt in the rail hole so plenty of room for adjustment. When I get the tooling/eco plates I'll be machining a reference edge to mount the carriages to the bottom piece of Ali, then the gantry will have a reference edge to mount the carriage plates to and thus I hope the gantry will be straight/perpendicular/parallel etc. etc. so that when I bolt the carriages to it the second rail will be referenced from the first which will be as straight as I can get it without having a machined reference edge in the epoxy to go on.....if that makes sense?

I think I'll get a straight edge, that will help with twist and straightness....plus the above and it should all be as acurate as it's possible for a bodger like me to get :thumsup:

Quote:

---

Originally Posted by njhussey

Surely the twist will be sorted when the rail is bolted tightly down, that's assuming that the epoxy is perfectly flat of course?

---

Gezzzz People have too much trust and expectations of epoxy.? You can't beat checking so buy or make a straight edge to confirm if you really think you need that level of accurecy in a router.!!!

Quote:

---

Originally Posted by njhussey

When I get the tooling/eco plates I'll be machining a reference edge to mount the carriages to the bottom piece of Ali

---

Why waste money on Eco-cast if your going to mill the ground surface away.? . . . The whole point is the accurecy of the ground plate and no matter how careful you won't mill to same level as grinding gives.!

What I do with ground plate when I need a ref edge is mill pockets on the ref line and insert aluminium dowels 3mm proud of surface. Then I machine half the dowels away to the plate surface leaving me machined ref edge without disturbing the ground plate.
Like this.
Attachment 14343

Have to agree with Dean, can't see the point in going to all trouble of designing your machine, getting feedback from the experienced members and then making a horses arse of aligning the rails at the last hurdle.
No offense intended.
Regards
Mike

None taken Mike, im all ears with regards to how its done. That's why I keep saying how I think I'm going to do things then listening to the advice given by the more experienced members.

The last thing I want to do is make a horses arse of aligning the rails having got this far :)

Glad about that Neil
I assume you have tried the reference edge of your rails when placed face to face? can you get a feeler between them, not that that is a definative test as they could both be bent the same. I know the squares and straight edges are a substantial investment that doesn't seem justified on possibly building a couple of machines,are you near any engineering or tech colleges that have this sort of gear a suitable deposit might get you the loan of some suitable kit.
Good luck with the build
Regards
Mike

Mike, I'm ordering a straight edge and a couple of engineers squares tomorrow along with the tooling plate.

I'll try the feeler gauge tomorrow and see how that is but I do remember when I first got them that I had them together every which way and couldn't see any light between the rails.

That sounds promising if you cant get a feeler gauge between the reference edges you could use the rail as a straight edge. have a look @ Silyavski's post before you buy any kit he did a lot of researching looking for straight edges and squares, you want the stuff that's made to a decent spec can't remember the BS number of hand but it's there in his thread.
Regards
Mike

Quote:

---

Originally Posted by Jonathan

so it's arguably OK to use but you will need to have realistic expectations of the accuracy.

---

Bingo he's Hit the nail on the head.! . . . . . . . People are getting silly on this forum regards the level accuracy they are trying achieve or THINK they are achieving with what is essentially Cave man tools in engineering terms.

Be realistic and remember it's a router based machine that will probably spend most of it's time cutting materials that have high variations of in-stabilty that far excede the accuracy machine can cut too.!

If you take your time to align the rails with sensible means IE: Clive S suggestion of taught wire along with feeling of bearings etc you'll be very close if not spot on for the use of the machine.
Yes rail alignment error will affect performance and live span of the rails but with care and patience the level of error will be so small to make little difference at DIY levels. Even in small/medium scale production enviroment it will take good few years to have any affect on wear if care is taken on setting up and alignment.
Just throw them on and you'll pay the price but with careful setup and sensible attention to detail you won't go wrong.!! . . . . Stop over thinking it and take sensible precations with checking straightness/twist along with feeling your way and you'll be fine.!! . . CRACK ON

Neil your build is looking great. I bet you can smell the wood chips already :-)

Please take what I’m about to say with a huge pinch of salt as I just fitted my first rails ever on my first build, so no real experience here but I’m putting it out there hopefully to get corrected or affermed. Look, I don’t know, maybe my expectations of final accuracy is just low or I got very lucky with the batch of 15mm Hiwin rails I received but those reference edges are as straight as Charles Bronson. Keep in mind my build is relatively small so I guess on longer rails it might be a totally different story. Few of us has the right equipment but I believe Jazz’s way with machining the dowels is the best approach assuming the mill you cut it on is accurate over that distance. I’ve seen builds where straight edges are clamped to the rail before drilling but I don’t think a straight edge is still straight after clamping it to something that’s potentially not straight?

I did not use a clamped straight edge for my reference rail. To ensure I didn’t bend it out of shape I clamped it to the epoxy base, drilled, tapped and fastened the ends and then repeated the process by moving the clamps for the rest of the holes always making sure I only drilled next to a clamp if that makes sense. On my rails the holes are 4.5mm and the bolts are 4mm which leaves a huge amount of adjustment as far as these things go assuming one does not stuff up drilling and tapping the holes.

For drilling I used a method proposed on here, I think Silyavski : Using the rail holes as reference, I took a 4.5 mm drill and drilled though the epoxy to the point where it made a good tapper dimple on the steel below and then drilled the rest of the hole with a 3.3mm drill for the tapping. That 3.3mm drill has no choice but to run to the centre of the dimple. Touch wood, I’ve done 40 something holes so far and not one out of centre so it appears this method is working really well. The important thing is that there is no binding whatsoever when you put the bolt in otherwise it will push the rail out of straight or best case you won’t any room for adjustment later. The steel on the inside of these rails where the holes are, are softer than I imagined so I guess the outsides where the bearing run are only case hardened. Not ideal but I’m totally prepared to enlarge the hole in the rail a little to ensure the bolt is not touching sides due to bad tapping or drilling the tap hole at a slight angle. I then used my straight edge and a feeler gauge (ok, it’s not really a straight edge but it’s a good quality stainless steel ruler and I know the one edge is relatively good) to check that the rail. I believe I have enough adjustment in the rail holes to fine tune it somewhere in the future if I get a real straight edge but I’m pretty confident it’s already better than my expectations of accuracy.
I then used my gantry after aligning it perpendicular to my first rail etc to align the second rail and then repeated the clamping drilling, tapping and fastening exercise on that side. As pointed out by Jonathan, I realise that these rails are quite slippery and it would require a huge force to actually get them to bind, but I did the best I could to move it around while aligning the second rail to feel for any sort of friction and I’m very happy with how smooth it’s running.
Even with this potentially dodge method I followed it still took me more a full day to get my x rails on so no quick wins here :cheerful:

Best of luck with fitting your rails.

Well I've stopped buggering about and started drilling and tapping the first (reference rail) rail. I put the two reference sides of the rails together , as suggested, and I couldn't get a 1 thou feeler gauge to even start to go between them.

I've drilled and tapped about 7 holes so far with my set up as below and all the holes are nicely central with approx 1/2mm each side of the hole so plenty of room for any adjustment if needed.

http://tapatalk.imageshack.com/v2/15...e993f30e81.jpg

http://tapatalk.imageshack.com/v2/15...98492109ca.jpg

http://tapatalk.imageshack.com/v2/15...159e91c2c7.jpg

Tooling plate is on order but will be 5-6 days [emoji17] and now going to order a straight edge from the same place as Silyavski in Germany to make sure that there isn't any (or negligible) twist in the rails.

Thanks for all the comments and suggestions guys!