[c131fr]

First, I am a complete noob. I have no formal CNC training- everything I know I've learned here or elsewhere on the internet. Three cheers for the internet!

10 years ago I bought a CNC retro kit for my Enco mill/drill. Unfortunately, it never got installed, and now the Enco is long gone. I was left with four servo motors (425 oz in), four Gekko drives, a torroidal transformer, and all the bits to make it all work. I also have 20mm and 25mm linear rails I got from a surplus store.

My other hobby is radio controlled aircraft, so I decided to make a gantry type router CNC so I can build molds and parts for models. It is all made of hardwood and mdf board. My biggest obstacle has been getting the holes for fasteners dead nuts accurate. I have a farily large Y axis- 30 inches- so the machine has to be quite square or it won't be terriby useful for parts that make things fly...

Anyway, if you see an obvious error in my photos and log, please speak up. I basically designed this in Solidworks, but real life is unveiling issues that the computer didn't see.

[c131fr]

The base was harder than I expected. Getting the rails perfectly parallel was a trick. I did it by using one rail to offset the other and then drilling holes. It took some tweaking, but they are very close to perfect. It literally took me 6 hours to get them right.
Attachment 186500

[c131fr]

The Z axis was actually the first thing I built. Then I rebuilt it. Then I rebuilt it again. Getting this right was extremely difficult. Now, I'm rebuilding it one last time because the slide mounting for the router isn't quite long enough...

[c131fr]

I'm using 25mm rods for the X axis. Probably a bit overkill, but what the heck. I want it to be rock solid once everything is built.

[c131fr]

My biggest concern so far can be seen in the attached. The height of the gantry creates potential issues. First, I'll have to make it incredibly rigid because the mechanical advantage of the motors pushing agains the frame will be significant. I plan to double up the sides once everything is in correctly, and probably glue everything together instead of simply relying on screws.

[c131fr]

I have a very limited budget for this project (like ten bucks or so, if the wife gets her way), so I ordered some acme screws instead of ball screws. They came with derlin anti-backlash nuts, which seem to be pretty good. I just hope I don't ruin my servo motors because they are much tighter than ball nuts. I made the base slider out of scrap wood, plus there is a 1/8" aluminum strip screwed and glued to the bottom of the slider just in case. I can't afford any flexture at all, and the weight of the gantry is a lot more than I expected.

[c131fr]

So that's where I am so far. Now it's out to the garage to continue my build. If anyone sees anything alarming, please speak up. Better to fix it now than when it all together (and glued...)

[jckstrthmghty]

If you want rock solid you will have to build with something other than wood. The build may appear to be solid until you throw a cutter on and give it a push. However there are some things you could do to increase rigidity.

I can see the single bearing blocks on the top and bottom y axis having some play. Typically people use 2 block per rail. You may want to run some stringers inside the box between the x rails. Even consider adding skins to the top and bottom to make a torson box. Yes that gantry looks tall. I can ssure you that will deflect significantly. On this type of build I've seen people bolt on aluminum or iron angle to stiffen things up. THe way you mounted the leadnut to the gantry could use some beefing up as well.

Sorry for being critical. Been down that road with a similar type of build and want to save you some fustration before you bolt it all together.

[c131fr]

I want criticism. That's why I posted this. I planned to install a top and back to the gantry for rigidity. I agree- wood isn't ideal, but with a $10 budget, what can you do? lol If the planets align right, I'll replace it with 1" aluminum.

I have a second set of blocks for that axis, I just didn't use them because they would be right next to each other. But if you think it will be a problem, maybe I'll go ahead and install the second set. Thanks. Oh, and I definately am running stringers where you suggest. Thanks for confirming that it was a good idea to do so.

[Dr.Fiero]

I want criticism.

Fine!
You smell. Yo momma dresses you funny. And your car needs an oil change.

But seriously folks...
Other than a few bracing for rigidity things, this really reminds me of my first router I built (bunch of vids on YouTube. Look for user "drfiero").
What are you doing for motor couplers?
Are those dumpsterCNC nuts? I've had good luck with those so far on mine.

[jckstrthmghty]

You can do amazing things with wood framed cnc routers. Jason Marsha in the JGRO forum has done fantastic stuff with the base JGRO build. Your build is along the same line but your rails are a huge upgrade. I don't think 25mm rails are overkill as they are unsupported. Defintely throw on the other bearing blocks for the y. Acme screw is good stuff if you don't require super fast rapids. Dumpster, you just can't go wrong there.

[c131fr]

Thanks guys. I do smell...

I'll check out the videos!

[c131fr]

Back to the garage. I removed the plate that will hold the router and made a new one. The original one was too short. When I made the first one, I used a precision square and a pencil to mark the holes. I have four bearing blocks, so that means I had 16 holes. It literally took me days to get the holes right- and even then they weren't really right. What I did was drill the holes at the correct diameter for the bolts I was using. When I went to bolt it onto the blocks, only 3 or four bolts would start in their holes on the bearings because they simply weren't lined up perfectly. I then tightened the few bolts that I could, and aftewards used the next size up drill and enlarged the remaining holes. This left lots of slop in the router mouting plate. My thought was to after the machine was done, loosten all the bolts and tweak them around untill the router was perpindicular to the CNC base.

But then I got a crazy idea. I built my design in Solidworks before I began, so I decided to print the sketch I used when I first figured out the holes. I then taped the pater over the new mouting plate and hand started each hole. I used my trusty cordless drill and placed the bit over the paper as accurately as I could, and then hit the trigger for a half second or so. After all the holes were started, I removed the paper and put the bit in my drill press. The starter hole moved the mouting plate into position as the drill press came down, and I soon had 16 holes in my plate.

To my surprise (and shock), every single hole save two were abosolutely perfect! The plate bolted on in no time (I had to fight the old one to get on for a half hour), and it looks fantastic. It's perfectly square to the rails. I'm so thrilled with it, I'm seriously considering redrilling all the holes on the machine. Probably not, but if there are issues with the machine being squre, at least now I know a more accurate way of making the holes that using a simple precision square.

Anyway, a couple more photos:
Attachment 186834

[c131fr]

Getting ready to fabricate a bracket for the drive nut. I will probably add the other two bearing blocks, but it will probably be the last thing I do because it requires a complete dissasembly of the Z axis unit to do so.

[c131fr]

Okay, new day, and I'm starting to see some problems now.

I added the second side panels for added strength, which went well. I then added the mount for the servo motor. The motor is round, so instead of making a plate to mount it on with screws, I simply am using friction to hold it in place. It is incredibly strong. I tested it with an oil filter wrench and when tight, it breaks the wood before it turns the motor. These photos show where I was making sure everything lined up.

[c131fr]

Here is where the issue of the morning crept up. I'm buiding this by myself, so installing the back plate was quite difficult. I screwed one corer to one top side, and then started a screw on the opposite corner. I then held up the board with one hand and tried lining up the sides so they were perpindicular to the base. I got it pretty good (I used a bubble level), but when I tried to turn the acme screw by hand, it wouldn't turn. I then loosened one of the corner screws and moved the gantry back and forth until I found the spot where the screw moved most freely. I then screwed the back on there. The issue is- the gantry is crooked! It's tilted to the left (you can sort of see it in the photo.

I was about to take everything back apart when I decided to check to see how far the Z axis was out. Much to my shock and surprise, the Z axis is perfectly parallel with the base. Or, I should say, the linear rods are perfectly perpindicular to the base.

So now I'm at a impass. Do I take everything off to make the sides straight? IF I do that, it might take me hours to get the Z axis right, so my thought is to just leave it alone. What do you guys think?

It's pretty solid right now. Adding that 10" back plate really stiffened things up. Once I get the Z axis servo installed, I'll run a brace across the top as well.

[wizard]

If you want rock solid you will have to build with something other than wood. The build may appear to be solid until you throw a cutter on and give it a push. However there are some things you could do to increase rigidity.

True that rigidity may be a problem for some uses of a router machine but lets not forget that Airplane parts are often made of balsa. This design isn't perfect but I believe it is very salvageable for at least some of his uses.

I can see the single bearing blocks on the top and bottom y axis having some play. Typically people use 2 block per rail.

This is a very good catch.

You may want to run some stringers inside the box between the x rails. Even consider adding skins to the top and bottom to make a torson box. Yes that gantry looks tall. I can ssure you that will deflect significantly. On this type of build I've seen people bolt on aluminum or iron angle to stiffen things up. THe way you mounted the leadnut to the gantry could use some beefing up as well.

Sorry for being critical. Been down that road with a similar type of build and want to save you some fustration before you bolt it all together.

Actually throwing a lot more wood at the machine will lead to stiffness assuming of course it is installed in the right way in the right place. He probably won't go far with the machine in aluminum but that doesn't appear to be a goal.

[wizard]

I want criticism. That's why I posted this. I planned to install a top and back to the gantry for rigidity. I agree- wood isn't ideal, but with a $10 budget, what can you do? lol If the planets align right, I'll replace it with 1" aluminum.

Many a decent router has been built out of wood, so don't knock yourself because of wood. is it ideal, no few materials are. The ultimate metal for machine building is still cast iron. Few people go the route of casting and machining cast iron so they compromise. The trick is finding good solid structural designs for the material you have on hand.

I have a second set of blocks for that axis, I just didn't use them because they would be right next to each other.

If you have them I'd most certainly put them to use. Even relatively close together they will help counter act the torque on the axis.

But if you think it will be a problem, maybe I'll go ahead and install the second set. Thanks. Oh, and I definately am running stringers where you suggest. Thanks for confirming that it was a good idea to do so.

[c131fr]

More progress. Why does the finishing tweeks take three times longer than the building? Do the building gods realize we have wives. My wife keeps asking "why did you quit working on your machine?"

I used a dial indicator with a large square that was clamped to the frame to measure flexure. Amazingly, I'm not seeing any. Like none. There's a ton of flex in the base, but at the moment I've only got 6 screws holding it together, and it's about 1/4 inch out of alignment on one end. So hopefully when I get the base glued and screwed to the frame, and the top glued and screwed to the frame, it will be pretty darned ridged. I wanted plenty of Z-travel, and I'm pretty happy with the Z-axis mounting plate height. The plan is to add a 3/4" top, and then screw another 3/4" plywood top on it for my working surface. That will bring the base of the Z-axis mounting plate to 2". The router's collet will be about 1" above the base when it's installed then. I hope that's about right. I should have about 7 1/2 inches to work with in the Z-axis, which is what I hoped for. I can make good molds with 7 inches or so of depth available. If it doesn't work out, I'll just chop a couple inches off the gantry sides and try again. lol

[c131fr]

Finished squaring up the base and placing the top. I had an issue with one of the servo motors- I dropped it and the brush keeper broke. I managed to fix it by hotgluing the old keeper in place. It seems like it worked.

It's pretty darned ridged. I added stringers in across the base under the top which did wonders. I think screwing on the top and bottom will make it like a rock.

The next step is to adjust everything so it's perfect. I'm sure this is going to take me a while. It's a pretty big machine- 44 X 28, and just a hair off on one end will be miles off on the other. Let the tedium begin!