Official Luthiers Forum!

I'm looking to replace my 9 year old Techno servo gantry machine because of excessive sag in the rails along the long axis. The basic table design is lousy and there is no quick fix to bracing it without nearly starting over. Techno has not been helpful. Their successor machine looks too similar for me and they are treating this like it was radioactive. Otherwise the Techno has been a good machine. Their PC based controller has never misbehaved and they have a Z-axis brake other companies don't seem to do.

The machine is in my basement shop with my other power tools so it has to go down a flight of stairs and a 32" door. This limits my choices, but otherwise I would need to buildout space in the garage. I don't relish all the trips back and forth to the basement or the dust near the paint booth out there. Due to expansive soils out here, the basement is a wood floor over piers, so I need to watch my weight and distribution a bit.

I've been looking at the Camaster machines. The Stinger I looks a little on the lite side, the Stinger II looks better. The small servo Cobra looks good except the weight which is probably a show stopper. Wincnc PC controller. I see that Bobc just bought one of these.

The small Shopsabre 3636 was my second choice when I bought the Techno. They were just getting started back then and now have a lot more machines in use. Also Wincnc PC controller.

Freedom Machine Tools is a relatively local company and their machines look very stout. I'm not thrilled about the Fagor controller or it's need for 3 phase power which would need a converter. It is pricey but they can do a tool changer which would be really nice to have.

K2 is way down the list although they seem to have improved their machines and have their rails on the sides now. I think it looks much lighter duty than the others.

Some of you guys have the XZero's and seem to do wonderous things with them. Their design looks OK and the price is nice. I wouldn't like delivery or support shenanigans. I might be able to reuse the servos and controller from the Techno.

Any suggestions or thoughts ? Thanks

Given all you've said, I'd be really tempted to fix whatever geometry problem you're having. Even if it's a long fix it's still cheaper than $20K for a new machine.

What size is your existing machine? That'd give me an idea of what would need to be done to have a full replacement frame fabbed. If you find a good welding shop and a good machine shop with a big mill it might be worth looking into having a new frame made. My buddy here has a mill with a 60"x30" table...if I were making a small machine, I'd just TIG up the frame and have him face mill the whole thing and drill all the holes for my rails right on it.

_________________
Bob Garrish
Former Canonized Purveyor of Fine CNC Luthier Services

I was thinking the exact same thing.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Fixing/Improving/strengthening it was my original plan and I thought surely that Techno had developed kits or could retrofit it if I wanted to send it to them. Supposedly there were thousands of these sold and I don't believe my situation is unique.

The table itself is 750mm x 1350mm (29.5x53), overall size is 42x53.5" . It is a frameless design: the table (3 30mmx250mm semi-hollow extrusions) with two aluminum rails each with dual 12mm hardened steel rods bolted directly to the outmost T-slots under the table, and 5mm endplates held together with 3 extrusions. There are 5 rows of lateral supports across keeping the table extrusions together across. There is 1" of clearance from the bottom of the gantry to the table to add things to.

So I threw together a simplified model for SW simulation just to see what affect a little reinforcing might have. The model isn't accurate because SW won't mesh the little chambers of the hollow extrusion and I haven't accurately modeled the rails because I'm sure they won't mesh as they exist. The deformation plot does however match what is happening on the router, the magnitude isn't accurate because of the simplicity of the model. It's a long table without any frame to rest on. A couple of additional steel rods under there help < 15%. Making new endplates of 1" AL takes it to 45% if I can trust the model ???

As Bob suggests a frame for the table to sit on is a good idea. The 1" clearance isn't much to work with. Increasing that means lowering the gantry (clearance) and changing the geometry. These machines are a bit notorious for going through ballscrews/nuts due to leverage as it is. The ballnut and carriages would need risers. I've never welded anything. An undertable frame wouldn't have to be faced flat but would have to be decent.

The guitarguy in me wants an adjustable truss rod to push up on the center of the table. The current endplates are too flimsy and I'm not sure I can leverage it. A 3/16" rod probably isn't nearly enough. Just another crazy thought.

Thanks for the ideas guys. I appreciate it !

I need a few more pictures of this thing to visualize it. Is the gantry sitting on top of the rails or does it go under the table as well? Are the rails on the side or the top? I think I just need a few more angles to understand.

I swear there's got to be a solution involving a few pieces of tubular steel and some screws to work as 'jacks' to level out the table.

_________________
Bob Garrish
Former Canonized Purveyor of Fine CNC Luthier Services

Bob likely it has a center ball screw in the middle of the table with the gantry connected from either side with a plate, angle or something attached to the screw. It will be hard to put jacks under the table if it is indeed configured that way. Another option might be to go to a dual screw system mounted outside the gantry driven with a center mounted servo, then nothing would be going under the table.

Mike

Here's some more pictures. The gantry is suspended on the rails that bolt to the underside of the table. Mike has it right the ballscrew is down the center between the endplates and the ballnut is on the lower cross support of the gantry. There is a little over 1/2" clearance from the ballscrew to the table bottom.

The tension bar, jack screw, table center pusher-upper is where I went first, like the jack-screw on a pool table. There isn't much structure or space between the table and gantry to implement that.

Rand how much clearance is there between the uprights and the edge of the table? At least 3/8"??? Could you mount a 3" to 4" wide piece of aluminum plate .375" thick, on the edge of the table and that keep everything where it should be? Seems like a much less expensive fix if it will. You could model it, then use SW to tell you if it will work couldn't you?

Mike

I'd cut the plate on the underside of the gantry to remove the bits between the ballnut and the bearings, then put a 'W' shaped piece underneath it to reconnect them all. That'd leave you room to run two pieces of 3" steel square tube down the length of the machine.

Alternately, you can cut yourself clearance to get it in and put a piece of U channel open-side-down over the ballscrew so the nut can pass through. If you don't have enough space to get something good in there, then cut slots into the extrusion on the bottom and reinforce it from below (again).

Really, the only problem is that extrusion on the bottom of the gantry is in the way of everything. The best solution is to replace it with a 'W' shaped piece that'll give you clearance to put in reinforcement bars. That could be as simple as cutting out the extrusion between ballnut and linear bearing on both sides and bolting an aluminum plate to what's left to give you an extra 1-1.5" (however thick the extrusion is) of clearance for reinforcement.

I remember when I was first shopping for a CNC (way back when the only people on luthier forums who had them were me and John W!) and I was talking to one of the industrial router companies about my options. One of the other manufacturers actually brought up the sag issue in the Techno machines; he said you could measure the deflection with a tape if you leaned on the center of the table.

_________________
Bob Garrish
Former Canonized Purveyor of Fine CNC Luthier Services

Yea, do something like this:



But you could make the W shaped thing full width.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Thanks for the good ideas guys. I need to marinate on it and try to draw stuff up.

That gantry extrusion isn't the hollowcore type like the table, it is solid and heavy-duty, and very much in the way of progress. The X-axis extrusion/assembly is pretty lightweight stuff. I need to make sure that I don't weaken that with a change.

The problem is that all the weight focuses on the 5mm endplates which deform inward making the already sagging table and rails try to fold over like a PB&J sandwich. It needs enough beams/subframe to support the bottom of the table so that all the weight sits on that and the strong steel base table, and the endplates only support the ballscrew.

@mike - there's only 1/4" between the table and gantry uprights. The side of the table has a t-slot. I can try that in SW FEA with Fe and AL. Even with counterboring there isn't much room for the bar and bolts.

@Andy - The image isn't showing on your post. I would appreciate seeing it.

The image is from this post on CNCzone: http://www.cnczone.com/forums/cnc_wood_ ... post891444 Look at the one that shows the underside of the machine being built

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Yeah, that's what I meant. The W shaped thing (yeah, that's its name now!) will be the new structure under the table. It's a little fab work, but in the end it's only four pieces of laser or waterjet cut steel welded or bolted together. The W piece could be plate, but you'd need three perpendicular flanges to screw on your bearings and ballnut.

_________________
Bob Garrish
Former Canonized Purveyor of Fine CNC Luthier Services

Still a heckuva lot cheaper than a new machine!

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Looking at the structure. I would build a larger frame from the 80/20 aluminum. add a new bottom plate to the gantry and extend it out to new linear bearings. By using the original end plates you will not change the bed to gantry height. These would be also be fastened to the new and improved frame which would add rigidity to those as well. The bottom of the gantry cross piece could be cut and bolted to the new lowered cross unit. this would allow you add some full length frame struts to strengthen the bed as well..

Something like this. It's a crude picture.


Mk

_________________
I'm outside looking in, just farther from the window than most.

I thought that was what spoil boards are for...just flatten it out.

Sorry to hear about that problem. And to think I almost bought one of those.

_________________
Ken

Something like this to replace the solid extrusion. I should be able to cut that on the machine from 1" 6061 or perhaps even garolite will be enough. Need to find a way to maintain the lateral braces. The 3 table extrusions are not interlocking so those keep everything together and the rails from spreading under load. I was hoping not to have to disassemble the thing.

@mikeK I think I follow you. I'm not sure that even the big 8020 will be strong enough to not droop over 53" with a few hundred pounds on it. I'll look at that. If I'm going to move the rails I want them on end.

@Ken Cardinal rule is rails are supposed to stay straight (flat). Good machines have them mounted on end, on a strong frame. If the rails were straight, facing the spoilboard would true up the machining surface to the rails (reference) even if the table underneath wasn't very flat. With the rails like a roller-coaster, a facing operation makes the spoilboard get cut with the same dip, actually worse because of the diameter of the bit carving out more material. If the rails were straight I'd be happy with this machine. Techno used stuff from Isel Automation on this machine, some of their other machines have steel frames. Now they are selling custom-made very expensive Chinese machines. Fortunately there are plenty of CNC companies these days. Did you buy a CNC ?

Thanks

Yes, I am building this one. It fits with my philosophy and the size will allow me to route a double bass plate.

http://www.grunblau.com/PlatformCNC.htm

_________________
Ken

Rand, I have had a similar issue with my Techno and was able to straighten it up using a pair of 1/2"X1 1/2" CRS bars bolted up under the table using the T-slots on the underside. The bars run parallel to the X-axis. I do notice one major difference, however, between your table and mine which is the T-slot extrusions run parallel to the X axis on yours but perpendicular on mine.
Just a quick off the wall thought that popped in just now: Any way of filling the T-slots on the underside of your table with stiffening members which would have some depth to them? Carbon fibre flats epoxied in?

Hi Nelson - thanks for the info. So your table t-slots are perpendicular to the long axis (I call my long axis Y, short X) ? This probably means you have a different spacing for the rails etc since you wouldn't have to line up exactly to the 50mm t-slot spacing. Do you have the little U-channel stiffeners under there ? I would love to see a picture of what you came up with.

On my machine the rails are in the 1st available t-slots 50mm from the edges and with the carriage width, I have the room for 1" W x .75" H rod in the next available t-slots. If I counterbored the bolt holes I could go as much as 1x.875H. If I offset the holes - off center to the bar I can go wider. I had tried FEA runs with that configuration and it had not made a large improvement but my model was only intended to give a sense of direction. Replacing the thin endplates made a big difference. Since it is outside of the travel area I thought about also bolting steel bar on the top outside t-slots, I would have to grind off the sharps

The extrusion has little pockets that I considered trying to fill with Fe as a tension bar or CF but I would not expect the dimensions to be very accurate so that is back burner.

My gut tells me I can improve it without major surgery. Other bracing approaches cause collisions that cause reduced travels or other major changes. It's a light duty machine. I appreciate all the ideas everybody has contributed !

The table on this little Techno is approx 30"x30". The 001.jpg shows the 1/2" 1 1/2" bar on edge bolted with counterbored SHCS's. What's that mess under the table?
I think we just need to swap machines.

Edit: Rand, another big difference on our Techno's is that the t-slot extrusions on mine are only about a 1/2" or so thickness.

Thanks Nelson. You've got a very different machine. The 1.5" on end CRS would help a lot. It looks like you also have fairly heavy angle iron on the inside of the rails. The split bottom on the gantry connected with plate to make room for the ball screw. You don't have the endplates and connecting extrusions. Being shorter it wouldn't want to sag as much. Do you recall what kind of sag you started with and ended up with ?

I'm working on several different design scenarios and may try simple before launching to radical.

Rand, I checked the size of the table extrusions and they are 15MM thickness x 250mm width times three. I don't remember the exact numbers but it was somewhere around .015" runout when tramming perpendicular to the extrusions. Seems like the stiffeners straightened that up to about .005" and a little of that was "ripple" in the extrusions. Tramming with a precision straightedge across the extrusions improved that down to a couple thou. The extrusions are by no means perfectly flat but not bad considering they are not machined. The initial runout prior to bracing just about drove me nuts as I've worked with machine tools most of my life where at least .001"/12" was the norm. My pet peeve is cnc router manufacturers bragging about positioning accuracy and that's fine but let's talk about geometric alignment of X,Y & Z also.

I've heard that aluminum stressed and deflected over a period of years will eventually take a set. That's something to remember when parking the gantry on these machines. Seems silly, I know. I don't recall if you mentioned how much sag you're getting but it really surprises me that there's much at all given the thicknesss of your table extrusions and the fact that they run parallel with the ways.

All things considered, it appears to me that the only thing these extrusion machines have going for them is the light weight. Getting them into tight spaces such as a basement is possible where a steel frame machine would be a little more difficult. I suppose that they are much less expensive to manufacture due to the bolt-together feature and the fact that table top and bottom t-slots are part of the extrusions.

You mentioned the angles under the frame and they are indeed angle shape with the addition of a thickened area at the outer end of one leg. This lets a 6mm screw pass thru the endplate and threads into the end of the angle. The 6mm fasteners are, I believe, the largest fasteners used on the entire machine.

I'm confident that you will be able to improve your machine greatly given what I'm hearing.

The positional accuracy one is still going strong, and being 'confirmed' by customers that don't understand enough about measurements parts to realize otherwise. It's relatively harmless for the most part except when someone a little green buys one of those machines and actually needs precision parts for their business. That's the tragic moment when one realizes why a mill 'with the same specs' is so heavy and expensive

I'll believe it on the aluminum. That's the reason aluminum components in planes have a service life of ~20K hours and then must be replaced. Aluminum has a crystalline structure that breaks down a little every time it's deflected, unlike steel and titanium that can be deflected (to a point) and bounce back forever without wear. Most parts never get enough use to hit that wall, but nobody wants to be flying in one when it does!

Re: extrusion...they'd make them out of concrete blocks if it was cheaper and people would still buy them! The guys selling extrusion and extrusion-based machines have pulled a DeBeers level marketing campaign of convincing people that the material is high-grade structural engineering magic-sauce. It's stiffer than steel, lighter than titanium, as stable as granite, and smells of fresh baked bread. It's a lot like the positional accuracy thing, though: nobody singing its praises has actually put a dial indicator under a piece and leaned on it.

_________________
Bob Garrish
Former Canonized Purveyor of Fine CNC Luthier Services

See, there's precedent for my suggestion

I think that's still the simplest solution, if it'll gain you enough room to get some square channel in there.

_________________
Bob Garrish
Former Canonized Purveyor of Fine CNC Luthier Services