Page 1 of 8 12345 ... LastLast
Results 1 to 15 of 112

Thread: The "Ultimate Tabletop Machine" for Woodworkers

  1. #1
    Join Date
    Oct 2009
    Location
    Marquette, MI USA
    Posts
    519

    The "Ultimate Tabletop Machine" for Woodworkers

    There are a lot of guys here that use CNC machines for woodworking or signmaking and I would like to know what some of your "wish list" features would be. Most currently manufactured table top (~24 x 36) machines use NEMA23 motors and smaller components in able to fit a price point for the size class. This forces successful hobbyists and commercial users to buy larger machines than their material dictates to obtain higher end features than tabletops offer. This uses up precious real estate in the shop, no matter if it is commercial or non-commercial.

    For example, this machine: https://youtu.be/qtzkAExTA00 was purchased by a signmaker that sells product online from his home business. After 3-4 months of use (and his Christmas rush) he sent an unsolicited letter full of praise for the machine and enumerated his production increases. I had offered up (pre sale) that even tho he currently had a larger machine, that these features would allow him to double or triple his output. He has happily informed me that those results were underestimated. His machine, loaded with function and feature, costing 3-4 times what a typical machine that size does, will pay for itself in 6 months.

    So, for those that are either small commercial or discriminating hobbyists and wish to have a small footprint machine, with "big" features that would allow longer duty cycles, faster cutting............ What would those features be?
    Gary Campbell
    CNC Replacement & Upgrade Controllers
    Custom 9012 Centroid ATC

  2. #2
    Join Date
    Jul 2014
    Location
    Norfolk, UK
    Posts
    94
    For me, the biggest upgrade I could do to my machine other than make it bigger at this point would be add an ATC - however, looking at ATC spindles from China that's likely to be a US$10k investment.

    One of the best upgrades I have done to my router was go to leadshine servos for X/Y and a leadshine closed loop stepper for Z. This has massively increased my resolution and accuracy, as well as given me about 5x the rapid speed without having smoking hot motors. For one of my drilling toolpaths, running the same feedrates, this took me from about 6 mins down to 3.5 mins - just time saving travelling between holes. That toolpath is about 300 drills, 10mm deep (no peck drilling), two offset rows of drills per piece, 5mm apart. The acceleration (850W for about 0.1 seconds) and max travel speed (1000ipm) saves so very much time. I struggle to swap pieces of wood out of the jig fast enough to keep continuous production going.

    The closed loop stepper on Z was a huge upgrade for reliability - when cutting aluminium or brass I used to lose Z steps as the cutting force of the endmill overcame the holding torque (230oz in) of the stepper and pulled the cutter deeper. So eventually I'd break every endmill haha. I cut very aggressively with my router - push it to it's absolute limits and sometimes beyond.

    That being said, what I looked for in my router when buying it was rigidity and precision - all ballscrew (nothing else is acceptable on a small router), all high quality proper linear rail (unsupported rail has no place in any machine, and chinese supported round rails are nowhere near linear enough), thick and solid frame (because I'm aggressive when cutting, anything less and i'll see the effects of the frame deflection). I'd prefer a steel frame, for thermal expansion reasons, but the Aluminium is OK. I originally bought it for doing really detailed moulds for composites and casting. Now I'm using it for a lot of wood carving (3-5miles of travel per axis a day when carving), a lot of drilling (18,000+ holes per day when drilling), doing a lot of pockets/mortises, and tooling wax positives for casting silicone negatives from. I make a pretty limited range of products...

    I absolutely refuse to use any router built entirely from extrusion of any kind - I've done quite a bit of work designing extrusions for electronics enclosures in the past, and even the best manufacturer's tolerances are nowhere near what a CNC router requires.

    For now, I think my next upgrade for this router is going to be extending the bed/ballscrew/linear rail for that - it's a lot cheaper than making the gantry wider (a lot of 2 sided milling for that). My plan is to buy a 5x10' ATC this year, that will give me infinitely more productivity as I can leave it running for 3-4hrs without needing to change tools or swap out pieces. Currently, when doing drilling/mortising operating the router is a hectic full time job. When carving I get about 25-45mins between tool changes depending on the product. When making tooling wax moulds I get about 3-10 mins between tool changes... so can't really get started on anything else. Having the router be able to go through tool changes (6-8 for some jobs) and work on a lot more pieces at once while I do other stuff is like having an extra 3-4 days in the week!

  3. #3
    Join Date
    Mar 2003
    Location
    SE PA - Central Bucks County
    Posts
    65,685
    Gary, the impression I've gotten throughout my own research is that throughput of a particular machine is very material to doing "real work" in reasonable time frames. Quality obviously counts, but if two machines of identical size can produce a product of equal quality, the one that can do it more efficiently is going to have a much better ROI, even with a higher cost of entry. (That's no different from any other business, including telecom that I recently retired from) So in my mind, that's what would truly differentiate a table-top solution for folks who want or need to do production for money off of a machine. Better motors (stepper or servo), better spindles and the robustness to sustain the higher "speeds and feeds" will pay off for that kind of buyer, even when it's a physically small machine.

    Another thing that's clear to me is that a pro shop with a big machine could also benefit from having a smaller machine to run those jobs that are physically smaller so they don't take up time on the bigger machine. Those shops will likely prefer and benefit from the things you are asking about.
    --

    The most expensive tool is the one you buy "cheaply" and often...

  4. #4
    Join Date
    Oct 2007
    Location
    Dawson Creek, BC
    Posts
    1,033
    I think the benefit of your products is what I assume is how close you pay attention to details. Why do I think that? I have read a ton of your posts. I know none of what I am about to add is news to you Gary, and I only add it to offer some info for others. The thing most of us do not understand when we first buy these is how important the quality of the build is. I sure did not, but now I do after chasing some build problems and spending a ton on measuring tools used by machinists. The table of the big machines are milled in some fashion to give a person an accurate reference point, most small machines are not. Why is that a problem? Steel and aluminum parts are not perfect. I knew that from my profession, but for some stupid reason I did not put together how important the accumulative effect of that might be. Each piece of steel, aluminum, or extrusion has a tolerance spec (defined by different governing bodies), and the specs vary for different sizes and shapes. Tolerances on steel or aluminum are based on the cross section size and the area of the piece. All these little errors add up. Without milling the surfaces, how do you measure the location of your machine parts to say .005 - 0.010? Renshaw makes some very accurate measuring tools, but it looks to me one could easily get a $100k into measuring tools if you go down that road. For most it comes down to balancing errors with off the shelf steel or aluminum parts. That can be fine if you are using machinst grade equipment, but the time to do that properly adds up quickly and you need a very patient person. Build time impacts production so it is easy to cheat there. Some may think none of that is a big deal and that is true if you are fine with an accuracy of 0.020"+ tolerance. That is greater than 1/64", and I can see that on a tape measure. Here is an interesting article on machine accuracy. http://s3.cnccookbook.com/MTLatheCNCAccuracy.htm The author of that blog is a steel guy, but the principles are the same. Some suppliers get around the topic of accuracy by confusing consumers by talking about repeatability. Why is that? Because that largely boils down the errors to the electronics. The quality control of companies like Mitsubishi, Sanyo and many of the other electronics suppliers are very strict. They build things for machines working in the aeronautical or other industries where they are building parts requiring very strict tolerances. They know what they can produce very very well.

    After the build quality, table size, spindle v. router, ATC the next step for me would be to improve the stiffness of the gantry and frame. To reduce chatter without slowing things down ridiculously the machine stiffness is so important. The problem is that is not a very salable talking point. This is the next thing I might change on my machine. Problem is I will quickly end up having too much into this machine to ever re-sell it to buy the one I would like now.
    Last edited by Brad Shipton; 01-21-2018 at 3:33 PM.

  5. #5
    Join Date
    Jul 2014
    Location
    Norfolk, UK
    Posts
    94
    I'm not sure on the build quality side of things for small business/serious hobbyists - there are a lot of chinese laser cutters out there, and the build quality on those is atrocious - hobbyists buy them over a trotec/epilog because they can either deal with the issues caused or improve it themselves (or don't know better). Once I tell businesses how much effort I put into my chinese lasers to get them to the spec they were at when I sold them, they generally opt for paying 6x as much for a trotec, which as we all know, are absolutely sensational. To them, the time spent in labour isn't worth it or they don't have the inhouse skills.

    For me, this is why I'd be buying a chinese 5x10 ATC for US$25k, rather than say a cheap US/EU model for $75k+. I have the time and skills to tear it all down and rework all the missing/missized bolts, poor grounding, bad electrical connections, get it all levelled and squared up, but I don't have the extra US$50k.

    All that being said, I think it also depends on what the customer wants it for. A friend bought a 4x4' router for machining foam to make cores for composite layup -it was $2k, ran on belts/v bearings and isn't much more rigid than a wet noodle in my opinion. But for what he is doing, it's more than sufficient as he has virtually no cutting force. Sure, there's visible ringing from the belts as he changes direction, but he just set the acceleration lower to mask it. You'd never be able to make anything precise in a reasonable time and have reasonable tool life if you were cutting walnut or maple all day with that thing - the belt stretch would be awful.

    Likewise, I'm guessing a sign maker (at least those just cutting out something visual only) can deal with +/- 0.1" of tolerance, whereas someone using it to make parts that have to fit together precisely, or who is making multi part moulds probably won't get away with it? I know some guys who make very tiny models, and machine gears from plastic with a 1/64" endmill - you can't do that with a typical palm router's runout...

    Personally, I think rigidity is a saleable point, and it's not something enough companies advertising to hobbyists/small business talk about enough. I've seen way too many machines even in woodworking shops with just a bit of laser cut steel sheet metal for the gantry sides, and then some vacuuformed plastic over it making the side look quite bulky, but the reality is you can deflect the gantry 0.1" or more just by leaning on it.

  6. #6
    Join Date
    Oct 2007
    Location
    Dawson Creek, BC
    Posts
    1,033
    Very valid points Mark and I agree that many will not find the accuracy as important. Myself, I want to be able to give a customer a part that is easily measured with a tape and they should not be able to detect an error using that tape they have in their pocket. That means less than 1/32" for a lot of discernible customers. I also want to be able to build an elevation of doors and have that 3/32" design gap I set out for. We are all wired differently and have different needs. My father was a fussy fabricator, and I clearly picked up his traits. The point of my discussion was to point out aspects that might be important to some. I got caught up in the typical research before I bought mine and in the process made some silly assumptions regarding accuracy. I am not faulting the supplier. I should have bought a much more expensive machine or went to see them building machines. That said, the knowledge I have gained is very valuable to me.

    A sign guy may not care, but I know of one guy that is having problems creating 3D signs he wants and I suspect part of his problem is the tightness and overall setup of the machine.

    How salable the stiffness may be seems to be something many find out after the fact. I have read a lot of "help buying CNC" type threads, and machine stiffness or build quality rarely are mentioned. The common focus is speed, features and cost. Cost is the typical leading factor, and I get that. That was a big part of what drove my decision. Those priorities are fine if you are buying a 6000lb monster where they have a group of in-house engineers doing the design, but it gets tricky once you get into the sub 2000lb market.

    I will leave the China discussion to another thread. This thread is Gary doing research to find out what people may or may not find important.
    Last edited by Brad Shipton; 01-21-2018 at 3:26 PM.

  7. #7
    I'm approaching all this from a hobbyist perspective, so mostly waiting for stuff to trickle down from the professional machines the majority of folks here seem to use.

    There is an opensource effort to develop an Automatic Tool Changer: https://github.com/xpix/XATC

    and there's a reasonably low-cost commercial option for Kress spindles: http://www.cnccat.com/index.php?id=2...catid=&lang=en

    The closed loop endorsement is especially interesting --- there was one opensource CNC which did used positional encoders, the Lobo CNC --- apparently it used surplus parts from a discontinued video game controller and when those ceased to be inexpensively available became untenable pricing-wise.

  8. #8
    Join Date
    Dec 2007
    Location
    Rochester NY
    Posts
    147
    This is a very interesting thread to me, I'm hoping to step up to a 2' x 3' machine within the next year, and spending a lot of time trying to understand and evaluate the options.

    I think there is a huge middle ground between high productivity commercial shops and hobbyists. There are all sorts of small professional shops concentrating on quality rather than quantity- custom/art furniture makers, artists and sculptors, musical instrument makers, prosthetic device makers, etc, etc. These would be businesses that would want a user friendly machine that is accurate and reliable, but not necessarily blindingly fast, especially if you are making one of a kind or small edition stuff. Saving 10 minutes in part time is pretty pointless, it's getting the part made perfectly that really counts.

    I am a semi-retired studio furniture maker, now making one of a kind banjos. I down-sized my shop from 6k feet 6 years ago, still have most of my big industrial wood and metal machinery. I took the plunge into cnc a year ago, with a little Techno Davinci machine, only 12" x 10" workspace, but built insanely well and heavy, with ball screws, proprietary double rail linear ways, casti alum gantry sides. I picked it up a auction of a big millwork outfit, they used it to make templates for a pattern grinder to make moulder knives. I wanted to try pearl inlay and engraving, and cut fret slots in fingerboards for starters. Had to replace the drivers and controller with Gecko 540 and Mach 3. Took way longer than I could have dreamed to claw my way up the learning curve (I'm 70 yrs old), but the results were great, I'm hooked on cnc and have all sorts of ideas for cool stuff to do with it.

    So, here is what is important to me in a small machine (thinking 3' x 2')

    1. Accuracy - I like to work to thousandths, even if it isn't always necessary. I made very technically sophisticated furniture, and worked to metal working tolerances on the shaper, lathe and Bridgeport mill. I assume that means ballscrews, altho the Camaster Stinger with rack and pinion claims .001 repeatability, yes?

    2. Reliability - I am often machining parts or assemblies that have a great deal of work in them already, a screw-up is very costly. I assume this is a function of the controller, plus correctly sizing components to avoid overloading any part of the system. Again, I'm not in a big hurry,could afford light slow passes if necessary.

    3. Quality of cut - I work a lot with highly figured hardwoods, especially maple, and need to avoid tearout. I expect to have to sand every surface, but would prefer not to have to start with really coarse grits or a file.

    4. Minimal spindle run-out - I use mills down to .015" for cutting pearl inlays and pockets in ebony, need runout of no more than a few tenths. The Kress router on the Davinci was a few thou out, was able to cobble together a setup and grind it in with a little die grinder on my lathe. Wasn't expecting much, but it worked out great, just 1 tenth out!

    5. Indexability - The bed should also work as an indexing plate for location of fixures and work. On the Davinci I made a 1" grid of reamed 1/4" holes, with 1/4-20 threads underneath on the t-slot plate. Did this on a friends Hass - nice machine! Maybe theres an easier way to approach rhis on cnc machines, but i know so little so this is what I did.

    6. Affordability - I know this is at odds with all the above, but I'm hoping theres a way to keep such a machine within reach. The compromise I'm willing to make is speed, don't need to crank out product.

    I think there are probably a lot of shops that could have the same requirements I do. These are the guys who might be looking at the Axiom Pro machines, the Stinger 1, or Shopsabre 23, or kits and wondering if they will get what they need from them.

  9. #9
    Join Date
    Nov 2013
    Location
    Leland, NC
    Posts
    476
    Gary, interesting discussion. Historically, people getting into CNC tend to focus on the machine. Yes, a good machine is very important, I would never argue it wasn't. However, where the ball gets dropped way too frequently is on the software side of things. I have yet to find the "perfect" solution. It seems each package has pluses and minuses. As you know, I run Aspire and it is really, REALLY good on the artistic stuff. It is also great for a hobbyist who wants to do a lot of different types of work, artistic and mechanical. But it does have a few drawbacks that go with it. BTW, there is no way I am bad mouthing Aspire, I love it, I use it.

    Recently I took a dive into Fusion 360. It is a true 3D parametric solid modeler. Whooppee fizz. But I got my eyes opened on the CAM side of F360. Adaptive cutting. No more cutting at one half the diameter of the tool slowly stepping down into deeper pockets. Full flute length cuts with no bit breakage. Bits last much, much longer. But, alas, F360 is not perfect either. It is not real swift with artistic endeavors. Nor does it handle an STL file without a lot of fiddling around, and then the results are less than spectacular. I have been doing mostly mechanical stuff lately, so that parametric modeling is a dream. The cutting strategies for mechanical work are incredible. Wrapping 4th axis artistic work? Nah, not good at all.

    My point is that one can buy an expensive high quality machine and then hobble it with poorly suited software. The machine guys who usually represent the larger investment seem to brush aside the software as inconsequential and the buyers also think the place where they spend the most bucks deserves the most interest. Here is some food for our collective thinking: A lower cost machine can easily outperform an expensive machine if the software is not matched to the users needs.

    Here is an interesting factoid about adaptive machining. Since the machine is engaged to full capacity, the feed rate is a lot slower, but the cycle time is shorter. So much for the "need for speed" treasured by so many. As you know, being a machine builder kind of guy, that slower feedrate solves a lot of nasty problems.

    Just thought I would throw this into the mix. Ideally, in the future, the gaps in capability I mentioned above will become much smaller. I do not think it is going to be very long before we see software that addresses both mechanical and artistic equally well.

  10. #10
    Join Date
    Sep 2006
    Location
    Deep South
    Posts
    3,970
    The two things I wish I had bought with my Stinger 1 is a real spindle and a larger work surface area. With the kind of work I do, a tool changer would not be worth it but I can see how that feature would be invaluable in other circumstances. I can't say I would rather have NEMA 34 electronics because that is not what is limiting my throughput at the moment. If I ever upgrade to the (alleged) 3.5 hp Milwaukee router that you advised me on, I might see an improvement. Another thing I just thought of is a sophisticated work holding system for small machines. That is a real challenge for small items. I am saying all this as a wood and PVC cutter/carver. If I were working with aluminum, it would be a different story.

  11. #11
    Join Date
    May 2014
    Location
    Santa Fe, NM
    Posts
    259
    Lots of very good comments so far. It's interesting to me that nobody has mentioned a vacuum table. It's my top project for our router, after a proper dust shroud.

  12. #12
    Join Date
    Jul 2014
    Location
    Norfolk, UK
    Posts
    94
    Interesting comments from Richard, I feel the same way about a lot of that except speed - often people run their routers WAY too slow, and cut quality can also suffer from that as the bit rubs and gets hot - tool life definitely suffers from the machine running slowly. Likewise from Ted - with HSM I actually run my tools far faster and harder with fusion's clearance toolpaths than I ever could with Aspire, as I know my tool engagement will almost always be exactly what I say it should be (except for a few odd bugs in fusion). With aspire I can only use the bottom portion of the flute, and the engagement varies so much I have to run the tool as if it's getting full engagement. With fusion I'm running a 1/4" endmill 1/2" deep at 6-8000mm/min (236-315ipm) in MDF when making jigs, something I couldn't dream of in Aspire. Tool life as a result has gone up considerably.

    Speed is often times not much more expensive to add, if the frame is rigid and solid, using bigger motors to support the high acceleration/deceleration shouldn't cause any issues - something I think Gary was hinting at in his initial post with the nema23 comment?

    I'd really love to see a hobbyist priced machine (about US$3500-5000 - this is what i've seen price wise in woodworking stores) that:
    * Is about 2x3' working area
    * At least 4" of gantry clearance (Once the spoil board is on!!), so you can run a 3" chuck 4th axis
    * Has closed loop control (lots of vendors here, i like leadshine personally but there are plenty of others). At least 300W on X/Y.
    * Solid, machined metal frame at least 3/4" thick
    * Ballscrews on all axis - C7 standard, C3 or better as an optional upgrade when people want more accuracy
    * Hiwin/TBI/other name brand linear rails on all axis
    * Solid bed - not extrusion. Should see under 0.001" deflection if you lean your whole weight in the center.
    * 2.2kW watercooled 4 bearing chinese spindle/vfd (HY ones that can be RS485 controlled are awesome)
    * Optional ATC, or ATC future upgrade - probably not something you'll sell much of immediately, but after a year or two I bet you'll get people come back for it if you email them a reminder!
    * Running Mach3/4 with Ethernet SmoothStepper, OR running LinuxCNC with Ethernet MESA board.
    * Optional industrial CNC controller (Syntec is fine)
    * Under gantry LED lights (I have about 120W of LED strip under my gantry)
    * 4" Dust collection shoe with LED lights built in (its awesome to see the cut in progress perfectly lit through the shoe, and it makes tool changes and setup easier with lights on the head)
    * Optional air assist - this makes a MASSIVE difference to cut quality and tool life. (see my video: https://youtu.be/nGIBv1w-Lbs)

    If I was able to find something like this on the market, I would have recommended it to at least 10 people who were trying to find something that meets those requirements in the past year. I think this would also be a big hit in the education market - if there was an option for an enclosure. A lot of high schools all around the world are getting laser engravers and cnc routers and 3d printers these days - its a great way to teach kids CAD and automation and such. The kids who are studying today are likely to be doing more programming of machinery to do what is currently manual work than the manual work in their lifetimes. This is something Canada's government is really pushing, taking people from skilled trades that are being more automated and teaching them to program the automation using all their extremely valuable experience of how the processes work.

    The hobbyist stuff seems to swing from either very cheap and poor mechanics (shapeoko) to extremely expensive industrial machines... thats not a lot in the middle. Those i've seen priced in the middle are still really lacking in either the mechanics (leadscrews, or belts, or chain - v bearings, or rail), or lack rigidity.. they haven't been designed by experienced CNC people, usually by people who have some experience with a shapeoko/similar and wanted something bigger and better then decided to sell what they came up with.
    Last edited by Mark Canada; 01-22-2018 at 7:05 AM. Reason: wrong video link

  13. #13
    Join Date
    Jul 2014
    Location
    Norfolk, UK
    Posts
    94
    Quote Originally Posted by Mick Simon View Post
    Lots of very good comments so far. It's interesting to me that nobody has mentioned a vacuum table. It's my top project for our router, after a proper dust shroud.
    The problem with vacuum tables, at least like larger cnc routers use, is the 5-10hp vacuum pump. Its not something you can do with a shopvac or venturi as air leakage when cutting is just too high. Its really handy when cutting sheet goods, but I wouldn't use one in my day to day operation as all my stuff is clamped up in specialised jigs...

  14. #14
    Join Date
    Dec 2007
    Location
    Rochester NY
    Posts
    147
    The machine Mark describes sounds ideal to me, I'd love one. But I can't imagine it selling anywhere near his desired price range, certainly not if made in the US or Canada.

    Perhaps from China, but isn't that what the Axiom people are doing? But their machine lacks some of his features, and costs considerably more.

  15. #15
    Join Date
    Oct 2009
    Location
    Marquette, MI USA
    Posts
    519
    WOW... I can't say I was blown away with the comments, as most were expected, but more in the contradictions in some of the individual lists. Items that if you select one, the other cant be achieved. I actually was surprised about the rigidity comments. Very few pre-purchase users pay any attention to this, probably most important of all, specification. Most important, but the hardest to define in layman's terms, which would be necessary.

    Rigidity, or the ability to hold position during cutting. Deflection of the frame components. Applying cutting loads that force weak steppers into a stronger torque detent (or microstep level) Rigidity, or the ability to hold a given tolerance under cutting load is what determines the price class of all machines. All OEM's know and speak internally of it, it is seldom mentioned publicly. What is mentioned is step resolution and repeatability.

    Step resolution is simply, based on the settings we use, the smallest amount a single electrical pulse could move the machine. FYI, very few machines under $100K will actually be able to provide an actual movement that small. And just so you know, with most stepper drives, step resolution can be made 2, 4 or 8 times as high with a flip of a DIP switch with more likely a decrease in accuracy at a given load. Repeatability is a simple, usually rounded or guessed tolerance number that the machine can reach under no load. In other words if you had a needle point bit, made a small puncture in the table, drove the machine around and then back to those coordinates, the needle point would be plus or minus .002 (common number) from that point. Repeatability is the combined total of the drive lash, play in bearings, etc, etc. Under no load! The combination of these means something. For example, .0005 step resolution and .002 repeatability means that the machine could command 4 steps before any motion is detected with a sensitive measuring device. This positional accuracy has little or no effect on those not cutting for NASA.

    What, in the end, turns out to be important to most users is "deflection under load". This is for the most part controlled by horsepower and chipload. Chipload is 100% in the hands of the operator and reducing chipload reduces "accumulated deflection" and increases accuracy, or the relationship between the actual size of a cut part to its design size. I don't use a shop full of $10K measuring devices, but I do have a simple test. Using 250ipm, 12.5k rpm with a 2 flute 1/4" bit (.010" chipload), if I cut a <6" square both on the inside and outside and compare, what is the measured difference? The deflection is half the difference. Half in on the outside and half out on the inside. These are also cut climb and conventional. Most small machines will result in 4 different sizes. To gain accuracy on small machines chipload must be reduced from the bit mfgr's recommendation to match the inherent deflection in the machine. And since chipload can be reduced by the operator to achieve normal accuracy (.005') for wood products, they give themselves a "pass" on not publishing those numbers.

    That deflection can only be reduced by "cubic dollars". Heavier, more expensive components, milled steel/iron frames, bigger motors, drives and power supplies, better bearings and much greater time proofing the assembly process, verifying accuracy along the way. The reality is that most non commercial users would rather reduce chipload to the machines capabilities than pay the extra dollars to upgrade to a class that has less "deflection under load". There is nothing wrong with this, nor is it wrong to wish for both. I assume there will be great improvement in the future as technology increases. But for now, it ain't gonna happen!

    I call my 250/250/125 (.250 bit, 250ipm, 12500rpm) test a "unit of push" its right around 50 pounds of force with a 2hp spindle. Most desktops or router fitted machines cant do it. Most spindle fitted tabletops will. I like to use the ratio between the chipload and deflection as a factor to determine how a given machine deals with deflection. If the deflection is less than the chipload, the machine is pretty stout. Those that run at half that feedrate and have deflection that equals the chipload will handle a "half unit". Same goes for quarter on small machines and up to double for those more capable. Larger, more powerful machines can cut 2 or 3 times that fast (250) and have deflection only a percentage of the chipload. 3 times the speed, 20% deflection is 15 units of push. So there is the formula, a unit of push times $15k is pretty close to the price class of the machine. It is what it is, whether you do the math forward or backwards. The deflection determines the price, or the price determines the deflection. The class price of a machine determines how expensive the components will be.

    The OEM's don't have to defend this, as most non commercial users buy by price. Most commercial buyers buy by spec, or tolerance at a given feedrate. This is normal. Most users need to have a budget, but be aware, not all machines in a given price class put the same dollars into quality components as they do into options that are made to aid sales. It's like fishing lures, same lure 20 colors. Made to catch fisherman much more than fish. What's not normal is a prospective buyer thinking that a $2500 machine will cut like a $5K, $10K or $15K machine, only slower. In part true, but do the want to save half the dollars for 1/8 the chipload?

    Thanks to all for your comments, I was being a bit vague in my initial post trying to eke out some new option ideas for potential future custom machines. My list of "standard custom" or base features far exceeds your combined lists. Unfortunately for most, so do the prices. I am exploring the option of using Asian imported frames, qualifying them, "blueprinting" of sorts, replacing the control electronics and motors and adding some of these nice to have options. The time to qualify or machine these cast and billet frame components, add HIWIN bearings, closed loop steppers, install new control, add a PC and completely test gets a ~2 by 3 table top in the $65-7500 range. And it would need a stout table as it would be over 400#. The free standing 100% milled iron version is over 600# and over $10k, $7.5k more with an ATC.

    Keep posting the high end wish list items! Thanks again, GC
    Last edited by Gary Campbell; 01-22-2018 at 1:04 PM.
    Gary Campbell
    CNC Replacement & Upgrade Controllers
    Custom 9012 Centroid ATC

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •