Grinding V-profile on rail edges
 

The MechMate drawings will call for a machine shop with a milling machine to form the inverted V on the top edge of lengths of standard angle iron for the x and y rails. Pretty easy and economical - if you have access to a milling machine shop. But you may want to consider a variation on the following......


Then a quick test in a table saw.....
http://www.mechmate.com/Forum/messages/11/822.jpg http://www.mechmate.com/Forum/messages/11/823.jpghttp://www.mechmate.com/Forum/messages/11/824.jpg

Don't try this if there is sawdust in your table-saw! http://www.mechmate.com/Forum/clipart/happy.gif

This is not the way that I had my rails made (they were milled), but I am just throwing this in here as food for thought.........

As you?re pushing the plate through and the grinding wheel is wearing down; how do you keep the edges parallel and square? Do you have to adjust the fence? Or is the wear on the wheel so minimal as to not matter?

It was a thin "cut-off" wheel only 3mm thick. Not designed for feeding material from the side. And the grit is quite coarse and fairly loosely bound in the wheel. Unfortunately our table saw construction is such that a thicker blade/wheel doesn't go past the nose of the (left-hand) threaded shaft.

With the first couple of cuts, the wheel seemed to wear a lot because the cut was quickly getting lighter. Realise that all four edges are done with one setting (the plate is flipped over and end-for-end). Therefore 2 meters of edge per setting.

For the first couple of cuts I only lifted the wheel and didn't move the fence. When it got close to final shape I re-measured the wheel diam to do a final calibration. To my surprise, there was only 3mm lost on the diameter.

The last cuts were very light and all 4 flipped passes seemed consistent. The original edge was down to 1mm wide and looked absolutely consistent side-to-side and end-to-end. The fence was only moved twice, the wheel was raised less than ten times, therefore less than 10 passes per edge. (Had to do it during lunch-break when they allow me in the workshop)

Realise that I am not trying to get a 100% perfect slide by NASA standards. It must be an economic, wear-resistant, stiff slide suitable for a short-stroke z-axis on a router. Because this slide is harder and thicker than our SB's one, and because the wheel diameters are bigger, I will be able to pre-load the eccentrics quite a lot without causing excessive wear.

As I've thought about it and reading your process I see now how it could be quite accurate (more precise than SB's) If you scribe four layout lines one on each side along each edge then grind and file to the line, it might help in ensuring accuracy. The pictures you've posted look great. It appears you left a small (about 1.5mm) flat face along each edge there by ensuring the edges remain at least as parallel as the flat stock was when purchased. I'm going to have to rebuild one of my Z axis in the near future, now I have to decide between using a linear bearing slide as on the Original Mechmate and this V-Bearing slide. I really appreciate all you've shared with us.

Evan, try it on some scrap flat bar stock first. I did the draw filing only to remove the right-angle marks left by the wheel. The amount removed by the file is very tiny and I wouldn't try to file to a scribed line. I don't even think the draw filing is necessary at all if the stone is finer. My draw filing started on the SB years back, and I regard it as my personal input - much like scraping bearings and beds (don't expect everyone to know what I'm talking about here http://www.mechmate.com/Forum/clipart/happy.gif hint )

Hi all,

I dont own a SB but I'm assuming this project
is a completely seperate machine all it's own?

I've built 5 small fixed bridge desktop sized
machines, But I'd like somthing a bit larger.

I'm interested in how you're working on the flat
bar. I'd be concerned that the fine grit will
do some damage to a table saw.

What about a piece of angle iron bolted upside
down alongside a cheap grinder. That provides
a 45 angle, then an adjustable block to provide
enough tension to slide the flat bar through
in several passes.

Here's a crude sketch...Hope it make sense.

John

http://www.mechmate.com/Forum/messages/11/1893.jpg

I cut aluminum on the tablesaw without worry about damage all the time.

I always clean out ALL of the wood dust before starting. I also turn off my main dust collector and switch to just a plain shop-vac.

I also cut aluminum on mine. Mine is a 150.00
table saw. I imagine some of us use 5-6-7 hundred
dollar saws.

Relative to an aluminum shaving, the fine dust generated from an abrasive cut-off wheel (aluminum
or silicon oxide) WILL damage your motor bearings.
I know this from experience.

What I'm interested in is a way to create the necessary angles for bearings to ride on in an
on-demand, as needed fashion.

I believe a setup like I described may work very
well, accounting for a trial & error test or two.

I have seen some surface hardening compounds in a
can, but haven't tried any.

Thanks for the tip before I runined my $$$ saw!

The grinding in the table saw was surprisingly easy (20 minutes) A 175mm wheel protruding 26mm above the table surface gave the 2x45 degrees, albeit very slightly concave. Then some draw filing with a fine file and a bit of emery stone (20 minutes). There is scope for refining this process, but already the results are an A-class "pass".

If I had to do a lot of these, I would use a thicker(wider face) stone of a finer grit, but the 3mm thin cutting disc (coarse) was not too bad. The biggest job was to make a centralising bush for the disc in the table-saw.

Can I suggest you guys try it with any old piece of flat bar before you believe me? http://www.mechmate.com/Forum/clipart/happy.gif

You wouldn't likely ruin it just doing it a few
times, but it's definitly Not Good for the motor
in the saw. This aluminum (or silicon) oxide
dust is harder than the bearings. and so small
that easily gets into places one can't see.

I'd do it on my cheap saw...it's on the way out
anyhow. But I'd hesitate on a nice expensive
shop saw.

Anyway, It'd be nice to be able to make up a section of rail, maybe several feet long, as it
was needed.

John

The bearings on my grinders and table saw are identical in terms of seals - I can't see why grind dust will be a bigger issue for the table saw bearing instead of the grinder bearing. We are not talking of using the table-saw as a production 40 hour/week grinder - this is a quick light job. No worse than using a belt-sander near your table saw.

The biggest damage to the saw is caused by the burr that the grinding makes on the bottom edge of the rail. This burr scratches across the table. A bit of thin plywood under the rail will stop that.

I Absolutely Agree.

I would like to set up a small scale "production"
type of arrangement, just to see and test out how
well it'd work.

Here's what comes to mind. Imagine if this "V" track
were about as cheap as drill rod material, with
about the same tolerances. And the bearings were
well..not as expensive as they are now.

That combination would VERY useful to a lot of people. And there's really no good reason such
a combination doesn't exist. It's a big world full
of people trying to figure out what to make so they can survive, but strangely enough, nobody
makes what there's a demand for?

It can't be that hard to make a relativly straight
bar that is angled on the edges.

I'll throw away a $35.00 grinder if it fails.

What do you think of the grinder idea anyway?
I see you're the only one who's actually tried to do this.

John

I should have mentioned from the start that
I'm VERY impressed by the section you made
already. I haven't seen anyone else try this.

John

John, in principle, your scheme is solid. Adjustment for progressively deeper cuts would be tricky.

Also, I would add a support edge for the back of the rail (bottom left corner in your diagram). Imagine what happens down there if you are putting 4 bevels on a flat bar.........

Little $35 grinders are often very under-powered - that isn't a problem if you have time. Taking lighter cuts will force you into a good finish. Also, the larger the stone the better - I don't know if $35 grinders can safely take bigger stones.

The experimenters must watch the legal speeds on stones, wear eye protection, maybe even gloves for that sharp burr...

For long lengths, one is going to need a long table to carry the weight - you don't want the rail to droop over a table, lifting the center section away from the wheel.

Well then a $100.00 grinder. I became Very
interested when I saw you'd actually done it.

If attention and care were put into a tool/jig intended to do only beveling edges, it may be worth the trouble. It could even be stepper driven very easily.

To try and do a "quick & dirty" on such an important part of the machine is no good.

I saw your part you'd made and that's what got
me going. I'm working on a Z-axis for my machine
as we speak and I wished I could use the "V"
bearings. PITA to get them though!

I gather you're the guy who's designing this
machine? E X C E L L E N T !

John

What's stopping you from making your own V-rollers? http://www.mechmate.com/Forum/clipart/happy.gif

I tried to figure out how to make the rails
first, and whether it was worth the trouble.

But to do it Precisely was going to require
a jig and I just sorta put it off.

When I saw your section the light came back on.

I'm sorry I haven't done my homework correctly here but if I understand it your project is a
stand-alone machine all it's own?

You're offering the plans alone and the end user
does the best he can.

You don't offer any parts.

I'm pretty sure I'm right with that but I will
ask directly.

I'd like to be a useful contribitor, that's why I ask. I think this may be one of the very best idea's I've seen.

I don't know about SB, but I've been rig-matizing
machines for many years and may be useful in some way.


BTW those are impressive too!

John

"If attention and care were put into a tool/jig intended to do only beveling edges, it may be worth the trouble. It could even be stepper driven very easily." Apparently, that is how ShopBot make their rails. I've heard they use belt sanders instead of stone wheels, and they drive them down the rail with steppers. I guess that they control pressure and time in contact with the abrasive rather than the exact dimensions of the finished rail. Nothing really wrong with that, the exact dimensions of the rail are not critical - they just have to be consistent (parallel).

Yup, I am not supplying parts. I would really like to see guys like you standing up and offering parts to others in your area while I take a back seat. It would be nice to get a self-supporting community going.

My machine operator has around two days of free time starting from tomorrow. I had a chat with him yesterday and he confidently said that he may be able to do the rails with the Shaper. But it is going to be a very tedious and time consuming job.I will keep you guys posted of the developments. In the mean time your suggestions on this issue are welcome before we start on the job (to make my life less miserable).

Relax fabrica, go and watch some cricket! http://www.mechmate.com/Forum/clipart/happy.gif

I believe that your "shaper" and my "milling machine" are the same type of tool (different terminology in your country?). Just let your machinist do one rail first. Can you take photos for us?

Did you see how this thread developed 2 days ago?

In my country when our team is involved in a cricket match most of guys do not turn up for work (including me).

Definitly I will keep you guys posted with the photos on a regular basis.

This is the sample V rail that we did with the existing shaper which ahas a stroke of 600 mm. We did this just to get some confidence. I will post a better picture tomorrow since I am still to learn the finer point of using the micro mode of my dig camera.

http://www.mechmate.com/Forum/messages/11/1972.jpg

To do this rail we used a shaper which has a
600 mm stroke.

http://www.mechmate.com/Forum/messages/11/1976.jpg

The rail came out fairly accurate and the dimensional accuracy was Ok. The hitch is the fact that we can only do 24" inches at a time. We cannot align the 24" pieces accuratly to get the final lenth. Tomorrow we will be trying out another method which will give us a continous run (around 11 feet)

We plant to fix a 1000 watt hand drill to the moving head of the shaper and mount the L iron on to the wise and move the bed horizontally to do the profile cut. The bed could move horizontally around 36" (motor Driven). Once bed comes to the end we would unlock the wise and push the L iron foward to do the next cut.

We are planning to use a end mill cutter mounted on the drill to get the 1" height cut then use a bevel cuuter to do the sides.

Your comments are most welcome.

That is indeed a shaper and not a milling machine. It appears to be well-kept and your machinist must be proud of it. But, I don't believe you can make a rail longer than the 600mm stroke with this style of machine.

Oops, I posted too quickly - while you were typing.....

WE are trying to get the bed to do the feeding and the drill mounted on the head to do the cutting by using end mill and bewell cutter. Looking foward to your comments.

I am impressed with your enthusiasm and ingenuity for this project. For cutting the bevels, a 1000 Watt hand-drill should have enough power, but I am worried that you have to remove a lot of material to get to the 1" and that the hand-drill won't cope with that. Can you get very thin grind discs to cut off most of the edge of that rail?

Also, the vice will only carry a short length of rail inside its jaws, while you will be cutting a long path of 36". Most of your cut will be outside the vice and the rail will bend. Unless you put a strong/long support under the rail. I have the unfortunate feeling that you may be cutting a series of 36" long curves.

No Gerald, I think you have got me wrong. We will be moving the bed not the Wise. The bed is holding the wise from underneath. The bed movement would be perpendicular to normal cutting head movenemt of the shaper. The wise would stick to the same position in the bed during the cutting process.

To cut the angle iron from 2.5" inches down to size (a little above 1") we would be using modified chop saw which will enable us to feed the angle iron parellel to the cutting disk of the chop saw.

Sorry one more thing to mention. By using the chop saw we will cut the angle iron down to a size which is little above 1". The final cut would be done on the shaper with a end mill cutter.

The bed will be moving. The vice is screwed to the bed (as per photo, but will be rotated 90 degrees). So the rail, bed, and vice move as a single unit....over a distance of 36". However, the vice is only about 12"(?) wide, and it will move completely past the cutter position. This is the part that worries me. If the vice was 36" wide then everything will be okay.. In your shaper picture there is a long plate lying in the bottom of the vice. The summary of my comment is that a similar long support will be good under the rail

The chopsaw is good. Don't be concerned about the rail curving after it is cut like this - it is flexible enough to bend back again when the screws hold it permanently in place on the MechMate.

I got your point, our assumptions are wrong. The other option is to use the shaper in it's normal cutting direction. If this is so, we could do 24" inch pieces at a time and but join them together. When the rollers start to run on the rail, at the welded joints will it give a jerk to the gantry and spoil the show. If so, how are we to overcome this.

What you might do is put a some channel iron under the part you are cutting. Channel down.

Greg thanks for your input. Can you please elaborate on what you were trying to suggest to me.

Well, where you can get the 36" cut and you are worried about curving, take a length of channel iron or aluminum and place it channel down for your angle iron to rest on. You can make it 4 feet long and that will give you support for 6 inches on either side of your 36" cut.

Fabrica, I suggest that you test your original plan tomorrow, and see if the drill part works. If that is okay, you can concentrate to solve the vice problem....

Maybe take the vice right off, and put a long solid piece of material in its place. Using your end-mill, cut slightly into this long piece of material to get a nice reference bed. Then clamp/screw the rail to this bed....?

Similar to what Greg said.

It is possible to accurately mill or grind very long pieces with even a small mill/shaper. You just have to set up an indexing reference:

http://www.mechmate.com/Forum/messages/11/1990.jpg

You first need to bolt, or otherwise clamp, a 1/2" thick steel straightedge along the back of the machine's table and exactly parallel to the longintudinal travel of that table - this is relatively easy to do by using a dial gauge chucked into the spindle.

Then it is a simple matter of setting the rail on the table, first clamping it to the straightedge and then clamping it to the table using step blocks and clamps. Then remove the clamps holding it to the straightedge and mill away with a 90 degree V cutter. Mill or grind 24" or whatever the travel of the table, unclamp, move 23", reclamp, mill/grind and repeat until the entire length is completed.

Depending on the mill/shaper you have, you could mill one entire side of the profile in one shot, or sneak up on the final dimension by gradually lowering the quill. If possible, I'd recommend the former unless you have an accurate Z-height readout or stop.

As accurate as a 12' bed grinder? No, but you should be able to come to within several thou, assuming that your mill is capable of that accuracy.

Brian

I will keep your suggestions in mind and have a nother discussion with my machinist and decide on what best to do. In the mean time if anymore ideas come to your mind please let me know.

Right now I have a 4 axis controller I need to finish up and test. Then assemble that into my foam cutting machine. The old controller is unipolar and this one is bipolar. The controller I'll use for the new machine will be servo driven and I think I'll just buy the Gecko G320K drivers.

Fabrica, during supper now I realised that your shaper probably does not have a continuous motor driven sideways movement of that table. I remember that shapers traditionally have a "clicker" (ratchet) drive from the movement of the main cutter slide. When the cutter moves forward, the table stands still, when the cutter slides back a linkage moves the table a little bit.

Your idea of joined rails is not that bad. Realise that you can place the joints so that only one wheel at a time crosses a joint. Only one wheel out of four.

Do you have a friend with a milling machine who owes you a favour? http://www.mechmate.com/Forum/clipart/wink.gif