Nugz published a new build:

UPDATE (12-1-14): I was going to build this machine geared toward milling aluminum plate. After much thought and discussion with some knowledgeable people here I have decided that it would be more trouble than it's worth to get this design to machine aluminum accurately enough for my needs. Therefore I have decided to scale back this design a bit and use it for cutting plastic and wood. It will still be "slightly beefier" than the original OX design but a lot less "beefier" than I was initially planning.

In the mean time I will be designing and building another machine for my aluminum milling adventures. This machine will use Linear rails, Ballscrews, and beefier extrusions.

Read more about this build...

 

Travis,

Wow! Thanks for the extensive reply! Your post confirmed many of my suspicions about my set-up. The more I have played with my model in Sketchup the more I keep thinking "still not strong enough to mill aluminum". At this point I'm invested only in the OX plates and some small odds-n-ends. There's two things I know a conventional milling machine can never have enough of, rigidity and density. The end goal being to reduce tool chatter as much as possible.

I would love to see pictures of what you have done to upgrade your OX. And thank-you again for speaking up! You may have just saved me a bunch of money and time that would otherwise be wasted trying to "make" it work. The funny part is the timing. I just found some pre-built Z axis' yesterday on ebay that use linear rails and thought "now that’s what I should be using". I think my build is about to take on a whole new direction, so please do share your experience. Knowing what has and hasn’t worked for you will definitely save me time and money so I am very interested my friend.

Thanks,
-Nugz

 

Thanks Travis! I greatly appreciate it! I'm heading over to your blog to check it out.
-Nugz

 

@Travis

Hi Travis
I'm assuming you used Nema 23 Motors? It says "24" on your blog. And I don't think you meant Nema 34. Just a heads up.

I've been checking out Automation Technologies lately. I'm assuming that's Keling's new site? They have some great gear!

Your Z-axis is more like what I had in mind for my build. It's very similar to the pre-built units I just found on ebay the other day. The one thing I would do differently, and the ebay ones don't have it, is to use a ball-screw for linear motion. The Chinese ball-screws look promising for what they cost. Do you have any experience with those?

You and I must share a brain. I had the same idea about using a PC case as a dual purpose PC/Control Box. I build a lot of PC's so it naturally occurred to me to do this. Edit: Turns out this isn't such a great idea. I hadn’t given any thought to the "noise" generated between these components and the possibility of that noise causing interference. It is considered best practice to isolate the PC from the Driver components. I'm digging those Aviation Connectors. That's a great idea. Nice job on the cable sleeving. I'm pretty handy with a soldering iron so that shouldn’t be a problem. Where is the best place to source those connectors?

I did the same thing with my X-Axis (you call this axis your Y-Axis) in my model in Sketchup. I was going to "sister" (mechanically fasten) three 20x60 V-slot rails together and run a triple wheel set-up across the top and bottom using 12 of the newer Xtreme Solid V wheels from the part store. But after hearing about your experience with that set-up I am definitely rethinking my plan. That's $72.00 worth of wheels alone. Another idea was to use a Flitch Beam to stiffen up the X-axis. It has been talked about here before but for the uninitiated it simply means to sandwich two extruded aluminum rails around a steel flat bar and bolt them together. If I were to go that route I would just use steel V wheels across two steel rails, basically sandwiching two steel rails around a piece of extrusion. But then the whole point of the belt and pinion drive gets tossed out the window. Speaking of the belt and pinion set-up. Can you provide some details about the steel pulleys and belting you used to modify your machine? I know you said you got them at Mcmaster-Carr but I'm not sure what you used that fits inside the V-slot rails. Do you have the part #'s for those by chance?

I am definitely using a 220v VFD and 2.2 KW water cooled spindle for my machine. I also want to implement a flood coolant system or mister for cutting aluminum so I will be using an aluminum hold down table. This will add significant rigidity to my Y-axis (this must be what you call your X-Axis) as it will be fastened directly to the Y-axis frame. I know this still doesn’t compete with all of the cast iron in an actual milling machine but I figure it can't hurt either.

Just a quick note on the axis' nomenclature that I use. I was taught that "Y" is toward/away, "X" is "across" (side to side), and "Z" is up/down when viewed from the front of the machine. This works best in my brain so I stick with it. I've noticed that some people here switch the Y and X for some reason and I find it confusing.

We both had the same idea for strengthening the X-plate connection to the rear plate by using the existing extra holes to connect the two plates with long bolts and spacers. I even had Chris Laidlaw machine me a custom taller X-plate, based on Marks (kram242) design, so that I could add more bolts/spacers if needed.

So, after reading your blog, I am left with the conclusion that I need to go with a redesigned Z axis using 2 supported linear rails and a Chinese ball-screw. I also need to re-design my X axis to either a Flitch Beam type with steel wheels riding on steel rails and a Chinese ball-screw for motion, or supported linear rails with a ball-screw. Either way the X-axis needs to be torsionally stiffer than what 3-20x60 rails alone can offer. That leaves the Y axis (long axis) with it's one true weakness being the belt drive. My thought there was to try a dual belt system with the bottom belt acting as rack (as in rack and pinion) as has been talked about here. With the belts facing each other so that their teeth engage each other across the length of of the axis and only separate at the pulley.

Hmmm....lots to think about!

Thank you so much Travis. I really appreciate the time and effort you have put into helping me out here. It's great to hear from someone who has already "been down that road." Please feel free to share any updates you have in the future. I would love to hear more about what you come up with. I went back and highlighted my questions for you if you get a chance to answer those I would appreciate it.

Anyone else with suggestions please feel welcome to chime in! I'm always open to other thoughts and ideas.

-Nugz

 

@Travis

Wow, again, thanks for all the info! I didn't know that about the C7 ball-screws vs. Precision lead-screws with anti-backlash nuts being considered roughly equal as a linear motion solution. That's good to know.

However, something you said in your last post caught my attention: "As for the x axis (shorter axis) the beam is super solid with the three 20x60 rails fastened together, the issue of the flex comes from the delrin solid v wheels. The beam itself is super solid when fastened together, and your idea with the steel v rail and the steel v wheels, is the right one, that is a much more rigid set-up."

That was my biggest concern. I must have misunderstood your original post. I thought you meant that the 60x60 (3-20x60 rails fastened together) X-axis still flexed under load. I didn’t realize that you meant that the flex was coming from the Delrin wheels.

I know you built your OX as an early adopter but many of the issues you have mentioned have been worked out in the forum here in one way or another since then. There are many things you may have missed. So, before I totally give up on my current plan I have a few more questions for you:

I intended to use all Polycarbonate Xtreme V wheels through-out my build as I was aware of the deflection with the original Delrin wheels. Have you tried the Polycarbonate Xtreme V wheels to see if they would eliminate the remaining flex in your Z-axis after your linear rail rebuild?

Did you use the 5mm wide "GT3" belt (GT2 with a 3mm pitch) and the matching "GT3" (GT2 with a 3mm pitch) pulleys on your build originally, or was it the original 6mm wide GT2 2mm pitch parts? The "GT3" 3mm pitch stuff is supposed to be much better at preventing belt slip and at 5mm fits inside the V-slot track better.

It may also be worth mentioning that I have a source for a TR8x8 Precision Lead-screw with matching anti-backlash Brass Drive nuts that I intended to use for my Z-axis. That should eliminate your concern about backlash with the Delrin nut from the part store...that's never in stock when I look anyway. I also have a modified taller X-plate with custom made 4-wheel aluminum spacer blocks.

What I have been doing for several months now, is picking apart the OX design, looking for weaknesses, and thinking up ways to fix them without spending too much money or totally building from scratch. My end goal was to have the ability to do some light duty but accurate milling of aluminum plate for my own projects. Think of it as more of a prototyping scenario than a mass production one.

So here is my current thinking:

• I am confident in my Y-axis frame design, especially once the solid aluminum plate is bolted down to the extrusions to create the "table". This will tie all of the Y-axis rails together. There will be absolutely no chance of "racking". That's why I left out the cross-braces. I think this is the weakest link with your OX after the Delrin V wheels. My design is MUCH more rigid, you just cant see it in that crappy screen grab teaser pic I posted.
  
• I am now confident in my X-axis solution of bolting three 20x60 V-slot's together as you have told me that yours is very rigid and is not the source of deflection at the tool tip.
• I am fairly confident in the ability of the Polycarbonate Xtreme Solid V-Wheels to not deflect under load. They are much harder than the Delrin wheels I see in your pics. This is why they were added to the part store.
  
• I am fairly confident in my stepper choice at 280oz/in. with a relatively flat torque curve.
  
• I am very confident in my choice of spindle. (2.2kw/water cooled/ER20 collets with VFD)
• I am fairly confident in the anti-backlash lead screw I have in mind.
• I would be much more confident if I had a Z-axis like yours with the linear rail. I think this would greatly reduce the chance of deflection at the tool tip. I can source one if need be.
  
• I'm stuck in the unfortunate situation of not having machine tools to build my machine with. If I had access to a large milling machine and a lathe I wouldn’t even bother with this approach.

Thanks,
-Nugz

 

Speaking of that...what is your Z travel now with the Linear Rail Setup?
-Nugz

 

It can't hurt right? Let me know how it works out.

As far as the GT3 belt and pulley go, I wonder if your steppers are just too strong for the belt and pinion set-up? At 425oz/in. you are 250oz/in. over the OpenBuilds Nema 23's rated for 175oz/in.. I'd love to see what the belt-on-belt solution would do with your steppers. Another idea that came up here in the forums was a rack and pinion set-up, with the rack mounted in place of the belt in the V-slot channel (gear side facing up) with the stepper gears meshed directly to the rack.

 

I'll see if I can dig up the thread where it was being discussed. It may be buried in the OX build thread somewhere, I can't remember where I read about it, but it was here somewhere. I do remember @The Dude was a big part of the discussion. I don't know if it ever went anywhere, as I've been a little sidetracked lately.
-Nugz

 

@Travis
Found it: http://www.openbuilds.com/threads/belt-rack-and-pinion.275/

And more here: http://www.openbuilds.com/threads/attaching-rack-to-v-slot.172/#post-704

You might want to talk to @Prauk about the dual belt configuration. I believe he came up with a "servo belt" type configuration on "The Buffalo". http://www.openbuilds.com/builds/the-buffalo.1157/

 

Hi Gray,
What are you on about mate? You answered your own question. I guess that's what a "Pendant" does though. I had to look up "Pendant" to understand what you were talking about.

I agree. Going heavier to mill aluminum, or aluminium as you guys call it , is probably the best solution for me. I'm going to downsize my "Angus" build to a 750x1000mm machine and use it for acrylic, wood, and maybe some Garolite here and there. Its still going to be "slightly beefier" than the original Ox design, but I'm not going to go crazy with it.

I'm going to build a leadscrew or ballscrew machine with linear rails for machining aluminum. I'm in the design process now.
-Nugz

 

@Travis
Thanks for the update Travis. Sorry it didn’t help much, I was hoping it would. It sounds like your X-axis deflection may be coming from elsewhere. Any thoughts? You have a much more substantial Z on your machine so we know that wouldn’t be the issue. If you apply force, by hand to the tool tip, is the deflection in the Y direction (front to back) or does it deflect in the X direction (side to side)?
-Nugz

 

@Travis
I see. Inlay work huh? Cool. Whatcha' makin'? Got a website?
-Nugz

 

They are very distracted by shiny things aren’t they?

@Travis
Good luck with the engagement my friend. A guy who's willing to put that kind of work into asking for her hand should have no problem getting a "Yes"
-Nugz