re: Fineline Shutdown or Perpetrating a Fraud!!
Quote:
Originally Posted by
islaww
I have mounted linear rails to numerous surfaces. I have seen few cases where linear rails mounted to aluminum extrusions without machining them first works well. Same goes for most any "mill product". Aluminum or steel, they need to be machined. And it will say so in the linear rail mounting guide. If you read the specs for mounting linear rails and then look carefully at the extrusion mfgr's production specs, you can see that aluminum extrusions are not made to accept linear rails.
It's easy to tell.... Loosen the bearing block bolts, if it rolls easier, then you need to machine the surface, or shim under the rail.
I guess I got lucky, although I took an extraordinary amount of care in prep and assembly. I know it's highly unlikely that my Hiwin assemblies meet the published specs, but I can say that everything moves the same, whether individually, loose or tight. I know, because I checked it as part of my assembly setup checklist. Movement was the same in all circumstances. Stiff, but consistently smooth. Perfect? Probably not. Close enough that I cannot detect any difference? Yes. I attribute some of my success to my design and the care I took milling the extrusions for square ends and for length. I would have milled rail mounting surfaces, but my mill isn't even close to large enough for that. Also, there is no practical way to take a bolt together machine to an industrial mill. Too many opportunities going and coming to knock the frame out of alignment. Too much to go wrong. Welded steel is a completely different matter.
Morale of the story: If you can afford it and are willing to part with the money, buy a welded, stress relieved and milled CNC. I can afford it, but am not willing to part with the money at this point in my retirement. So, I either accept the compromises inherent in my decisions, or I forego a CNC altogether. I chose the former, which is apparently a choice you would never make. Good for you for being able to stick to your principles. I wish we were all so lucky.
Gary
re: Fineline Shutdown or Perpetrating a Fraud!!
Quote:
Originally Posted by
wmgeorge
Gary I think 80/20 offers a machining option would that be an good option for a DIY builder?
Bill, I do not believe 80/20 offers surface milling (like for linear rails) as an option. Misumi does, but not 80/20.
Quote:
Originally Posted by
islaww
Bill...
Yes it would, but few if any would pay for it. Besides, you still end up with an extrusion. For the most part they are used because they are 1) Low cost, 2) lightweight and 3) easy to bolt to. Extrusion machines are no where near as rigid as most branded machines, even the Chinese.
I agree with Gary. Even if 80/20 offered surface milling, probably no one would pay for it. Remember 80/20's cut tolerances, .002"/inch for square and .015" for length. If that's the closest they can get for cutting, you have to wonder what the tolerance would be for surface milling. As much to the point, the cost would probably drive off most folks. If money was no object, you'd buy a GOOD branded machine every time.
At the end of the day, CNCRP/Avid may provide a good case study for extrusion-based CNCs. 80/20's cut tolerances are pretty loose. Yet, the owners get them put together. The other tolerances are way below what they are supposed to be for linear rails, yet most have little problem getting them to run smoothly. Even more remarkable, given Avid's instructions for rail mounting - placing a jig at each end of the rail and simply tightening down the rail bolts from there. If the rails aren't perfectly straight, and there's no guarantee they will been, there are many opportunities for things to get off a little. Heck, just the act of tightening down a mounting bolt can cause more than the acceptable amount of deflection. Still, they run and folks are happy with them. Owners sing their praises. Are they as stiff as welded machines? No way. As heavy? Nope. As accurate? Maybe; maybe not. Do they do what owners bought them to do? You bet. Do they cost as much as a branded welded machine? Not in this lifetime.
My machine, except for the gantry, uses profiles twice the size of the kits. I also used the heavy stuff, versus the lite variety the kits use. Safe to say, given my design, mine is about twice the weight of the kits. Not the weight of a welded steel machine, but pretty heavy. All-in-all, way more stout than the usual kit or DIY.
Gary
re: Fineline Shutdown or Perpetrating a Fraud!!
Quote:
Originally Posted by
Erik F
Sorry for the short reply...but I'm exhausted. Im saying even if the rails are parallel to each other...if you look down the table...like you might look down a 2x4 timber to see if it's straight...the rail could possibly be toed in or out vertically like if you were to draw a line from floor to ceiling through each rail those would not be parallel even though the rails lengthwise were.
Ok I know what you mean now, it's because the steel tube dips in the middle. If you took a 123 block and put it o the tube across the width you will see what I'm talking about you will see daylight and can fit a feeler guage in the gap to see how much. What happened is the rails aren't wide enough to to bridge the gap so they wound up cockeyed. The gear rack has the same problem but worse since it's mounted closer to the edge. I already shimed my gear rack so that it's square with the pinion gear, luckily I caught it early since it was already starting to wear the aluminum pinion gear. Basically you are going to either need to buy some shiming material spend some time squaring and leveling the linear rails or mount a piece of material to bridge the gap and then either use set screws to level and square the flatbar or break the machine down and take the frame to a machine shop and have all the mounting surfaces milled. I haven't decided yet which direction I'm going to take.
BTW by loosening the bolts I told you about it will compensate for the cockeyed rails as well, my machine has the same problem and one side of the bearing blocks have a bigger gap than the other. On my machine the liniar rails are tilted inward slightly, in other words the side of the rails closest to the machine bed is lower than the side facing the rack and pinion.
I'm sure you can tell I've spent way too much time finding all the build flaws on these machines as well as how to fix them. If you have any other questions or need me to send you pictures or a video just let me know. I feel bad for anyone who is trying to get one of these routers dialed in, and I just hope that once my machine is dialed that there isn't any component failure due to the **** build quality. It does seem to have decent parts installed on the frame so time will tell.
It kind of reminds me of buying a old car that has a bunch of nice parts but you still have to redo almost everything because the guy who built it didn't know what he was doing. Yes this has happened to me more than I would like to admit and by the time I got things right and done cussing at whomever the hack was that did the original work I was like why the **** didn't I just build one from scratch.
Unfortunately sometimes trying to save some dough costs you in the end, actually most of the time it goes that way but some of us are forever searching for that damn unicorn.
re: Fineline Shutdown or Perpetrating a Fraud!!
Quote:
Originally Posted by
Ntl
I would like to know the same as far as what type of damage will be caused by the mounting surfaces not being flat or parallel.
Dan
Dan,
I've seen the often, but never an answer. To you have a lawyer like answer, it depends. For instance, out-of-parallel by .002" is going to yield a very different life span than being out by .040". It's going to be pretty machine specific. Honestly, I doubt that anyone has tried to quantify it and reported out on it. We can reasonably conclude that with such tight tolerances, any deviation is bound to have a negative effect. Otherwise, why publish the tolerances? They matter. If things are out so much that there's an obvious bind, life may be pretty short. If all seems smooth, the rails/blocks may outlast you (or maybe not). I'd think with the problem Erik describe, life would be on the shorter side. I doubt you are going to get a clearer answer.
Gary
re: Fineline Shutdown or Perpetrating a Fraud!!
Quote:
Originally Posted by
GME
Dan,
I've seen the often, but never an answer. To you have a lawyer like answer, it depends. For instance, out-of-parallel by .002" is going to yield a very different life span than being out by .040". It's going to be pretty machine specific. Honestly, I doubt that anyone has tried to quantify it and reported out on it. We can reasonably conclude that with such tight tolerances, any deviation is bound to have a negative effect. Otherwise, why publish the tolerances? They matter. If things are out so much that there's an obvious bind, life may be pretty short. If all seems smooth, the rails/blocks may outlast you (or maybe not). I'd think with the problem Erik describe, life would be on the shorter side. I doubt you are going to get a clearer answer.
Gary
That's what I'm hoping for, if I can get it close and don't feel any noticeable drag or binding that things will last until I decide to get a new machine. It would just suck do to the labor if the bearings wear out quickly. I seriously doubt the liniar rails can go bad and obviously none of machined parts aside from the pinion gears worry me. I suppose if the bearings went it would give me a excuse to have the steel frame machined since it wouldn't be much more labor once the gantry was removed. If the original steel tube gantry would have had the holes driled tapped properly I would have already had it machined flat in all the crucial areas. Since it wasn't it would be less work to just make a new one from scratch and have everything machined, that still may be a option I'm just not sure I want to keep polishing a turd if you know what I mean.
re: Fineline Shutdown or Perpetrating a Fraud!!
Quote:
Originally Posted by
wmgeorge
In your haste to "be right" you must of missed the " ground flat" in my reply.
Please show us the pictures of the machines your making, selling and exporting so we know how to do it right.
Oh and BTW my machine was not heated and stress relieved and it works fine.
ADDED: So Gary GME is building a new machine from 80/20 aluminum extrusions, as have been thousands of others done. I am sure they do not meet the .0005 in all planes criteria above. So are all these machines therefore junk? I am looking forward to see the machines your making, please post pictures. Thank you.
You have to have an understanding of what happens if you Grind CR most times the flatness is worse after it is Ground, it would have to be welded in place, stress relieved and then Ground to be of any use what's so ever
Yes the extrusions don't get anywhere near where it has to be for linear rail mounting but one company does machine the extrusion for mounting Linear Rails
Your machine may be working well but every time you use it the Linear Bearing will be wearing more than normal, who knows how long it will last nobody can answer that question
re: Fineline Shutdown or Perpetrating a Fraud!!
Quote:
Originally Posted by
GME
I guess I got lucky, although I took an extraordinary amount of care in prep and assembly. I know it's highly unlikely that my Hiwin assemblies meet the published specs, but I can say that everything moves the same, whether individually, loose or tight. I know, because I checked it as part of my assembly setup checklist. Movement was the same in all circumstances. Stiff, but consistently smooth. Perfect? Probably not. Close enough that I cannot detect any difference? Yes. I attribute some of my success to my design and the care I took milling the extrusions for square ends and for length. I would have milled rail mounting surfaces, but my mill isn't even close to large enough for that. Also, there is no practical way to take a bolt together machine to an industrial mill. Too many opportunities going and coming to knock the frame out of alignment. Too much to go wrong. Welded steel is a completely different matter.
Morale of the story: If you can afford it and are willing to part with the money, buy a welded, stress relieved and milled CNC. I can afford it, but am not willing to part with the money at this point in my retirement. So, I either accept the compromises inherent in my decisions, or I forego a CNC altogether. I chose the former, which is apparently a choice you would never make. Good for you for being able to stick to your principles. I wish we were all so lucky.
Gary
If the Bearings have enough clearance in them then you would not notice any difference when mounted on extrusion, Standard Linear Bearing can have up to .004" or .1mm clearance
re: Fineline Shutdown or Perpetrating a Fraud!!
That was why i spent extra money for good quality German made extrusion. Also the company supplying it in Australia cut it very precise...
re: Fineline Shutdown or Perpetrating a Fraud!!
Yes, it depends on the preload of the bearings. The higher the preload, the more precise the mounting surfaces need to be. The specs I mentioned above were for light preload. As the preload goes up, the tolerances get much tighter.
All the Hiwin catalog says it that the bearings will wear faster. How much would depend on how much deviation you have.
re: Fineline Shutdown or Perpetrating a Fraud!!
But if its not binding and cutting square is it still wearing? Everything wears, its part of the cost of using a machine. I doubt a hobby user would wear one out if its not binding, and if its does.... buy a new set of bearings after 20 years or so. Its amazing we have not heard these complaints over on the CNCRP area.
Gary Campbell does a fine job and I have seen pictures of his work posted, IF I had a commercial shop and needed a machine I would have one of his for sure.
Re: Fineline Shutdown or Perpetrating a Fraud!!
Quote:
Originally Posted by
mactec54
If the Bearings have enough clearance in them then you would not notice any difference when mounted on extrusion, Standard Linear Bearing can have up to .004" or .1mm clearance
FWIW, mine have a medium preload, so the tolerances are tighter than Standard (light preload). I'm not quarreling with anything you've said about linear rail wear when installed on imperfect surface. You've put out good information for any prospective buyer or builder.
I expect that mine will wear more than they than they would on properly prepared surfaces. How much? I guess time will tell.
If I had a commercial shop, I would buy a commercial quality machine. I hope mine outlast me, but since I'm just a hobbyist, I can justify buying a new set of rails and blocks, if the need arises. Still way cheaper than buying or building a commercial quality machine.
Also FWIW, it's iffy whether one would ever get extrusion-based routers to linear rail spec. They are bolted together, after all. Whether milling the extrusions themselves, or adding plates for milling, the completed frame would have to be taken to the machine shop for milling or grinding. The chances of introducing some misalignment during transport - whether coming or going - are a probably high. Then, when the machine gets back to the shop, the surfaces will be only as good one's ability to reestablish perfect alignment. I believe it could be done, but would be a tough job. IMO, the only way to ensure a proper finished product would be using welded steel, stress relief, and then milling/grinding on a suitable large format machine. But then, you are back to a commercial/industrial grade machine. I guess that's why so many hobbyists opt to compromise quality for price. IMO, what important is to recognize and understand the limitations and consequences of going with extrusions.
Having said all this, I do quarrel with the notion that using linear rails on extrusion-based machines is a total waste, which seems to be your position. Clearly less than optimal, but I'm doubt anyone really knows what to expect over time. We could get a nasty surprise, or a pleasant one. Only time will tell.
Gary