Rack Confusion

[Bugsy36]

This is my first post, so please bear with me. I have been playing with a CNC router for a little while now. No matter how much I learn it always seems like I know nothing and that there is that much more to learn. This time I cannot seem to find my answer by reading - so I am turning to those that know (CNCzone).

Currently, I have a home-built CNC router (pipe-rail). It is a good machine and it did a fantastic job of breaking me in. As great as it is as a starter machine, I have outgrown the table size (25x17) and the speed (12 ipm). The time has come that I: a) Build a much more suitable machine or 2) Buy a commercial machine like the Shopbot. I keep thinking that building is the better way to go.

I have completely figured out how to build my new 5x5 router. Construction is not a problem. I have a welder and all the hand tools to handle the majority of what I need. Any precision and repeated drilling I can do on my small machine for now. My problem comes from my desire to have a 5x9 machine. I figure if I build a 5x9 I will be done with it. The primary purpose will be for routing wood, foam, and light gauge aluminum.

I understand linear slides and lead screws. For my 5x5 I can do just fine with .750 leads. The 5x9 is giving me trouble. I know that on a 5x9 I would need to use leads that are at least one inch. Going to a 4 or 5 start one inch lead starts to get a bit expensive for my budget. That pushes me towards the rack and pinion, and that is where I get confused.

I have read Boston Gear's "Gear Theory" and it just does not sink in. I DO understand that the gears have to be of the same degree, but it is the sizing that is getting me. I see that many here are using some sort of belt assembly for gear reduction. If at all possible, I would like to avoid that, for the sake of simplicity.

The motors that I intend to use are NEMA 34 - 900 oz (Dual drive X, single Y). Can someone suggest a gear combination that will give me ample rapids (300 ipm ??), while still allowing me to use a direct drive, and keeping decent accuracy? Looking at the Shopbot, it appears that if they have any type of gear reduction it must be internal of the motor because the show no visible belts. I am afraid that if I go with larger motors that I will not be able to properly tune them (as Art suggests on the Mach forum)

Chances are that I should probably just stick with the 5x5 configuration but if I can get the rack figured out then I may be able to just go to the 5x9 for a very nominal cost.

You guys are great! I have learned much from you guys, and I am looking forward to learning more. Thanks in advance for all your help with this.

[Mike Nash]

You need to check out the MechMate at http://www.mechmate.com/ and also be sure to scan the forums. They've already got a motor/rack/pinion combo worked out. Plenty of info and it's free.

[ger21]

I'll second the mechmate choice. Much, much cheaper than a shopbot. And everything is all worked out for you.

But if you do want to go with your own design, then you'll want to use a gear reduction. Belts are most common, because they are probably easiest to implement. The shop does does use geared motors as you suspect. Without gearing, your resolution will suffer.

[Bugsy36]

Thank you to both of you. The MechMate forum does have a wealth of information. Unfortunately, I seem to find it harder to navigate than the zone. It also seems that 99% of the configurations are using some sort of belt driven gear reduction and I have yet to come across any without that.

I guess I am at a loss as to why it would not be possible or advisable to just use a different gear on the motor for reduction?

By the time I factor in any more than just a different gear, the build time and costs increase to where it may be just as easy to use a large diameter lead (easy being the keyword here). I also have the additional concern that any lead larger .750 may be too much without going to servos because of the additional weight.

I know to the experienced and knowledgeable that this may sound stupid or just downright silly. (I sincerely apologize for that).

It seems to me (and that is not not saying much ) that I had a misconception that adding an additional 4 feet to the X-axis would be nominal (unless of course I was already using a rack with gear reduction).

Just for the sake of clarification, the machine that I am wanting to build will be primarily manufactured from aluminum (too keep the gantry weight to a minimum)

[Mike Nash]

I haven't reread the entire thread, but the original MechMate used non-geared direct drive steppers on that massive steel gantry.

http://www.mechmate.com/forums/showt...eferrerid=1630

There's a picture here along with some rack choice info:

http://www.mechmate.com/forums/showt...3&postcount=11

I bought rack for a (future) homebrew here: http://www.stdsteel.com/gr_stock.htm

They've got the best prices I've seen, but I have no idea how much they may have gone up with the current steel pricing situation. It cost me $148 shipped back a year ago on the 17th for 5ea P/N 200011 73" 20PA 20 Pitch Gear Racks.

I bought a 12" gear blank from McMaster P/N 6847K22 http://www.mcmaster.com/nav/enter.as...2&pagenum=1014 for $33.62 + shipping to make a bunch of pinions from. If you have the tools (or a friend who does ), it's way cheaper than buying individual pinions.

I see you are in the Southeast, tell us where and you might find someone volunteering to help.

[dmoore]

Originally Posted by Bugsy36 Thank you to both of you. The MechMate forum does have a wealth of information. Unfortunately, I seem to find it harder to navigate than the zone. It also seems that 99% of the configurations are using some sort of belt driven gear reduction and I have yet to come across any without that.

I am only aware of a single belt-reduced MechMate:

http://www.mechmate.com/forums/showt...&postcount=354

By the way, with the right parts (SmoothStepper) and 30 tooth pinions, you should be able to hit 1,500+ IPM on a MechMate:

http://www.mechmate.com/forums/showt...58&postcount=7

The remainder are direct drive or, now it seems most are going with geared motors (out of the box). It's just "bolt on" - that's the way mine is and I have resolution down past the thousands.

Originally Posted by Bugsy36 I guess I am at a loss as to why it would not be possible or advisable to just use a different gear on the motor for reduction?

The MechMate guys then use 7.2:1 geared motors with 20, 30 or 35 toothed sprokets to get the speed/accuracy they want. I have 30 tooth pinions.

Originally Posted by Bugsy36 By the time I factor in any more than just a different gear, the build time and costs increase to where it may be just as easy to use a large diameter lead (easy being the keyword here). I also have the additional concern that any lead larger .750 may be too much without going to servos because of the additional weight.

I'm pretty sure you'll have problems when you get to 9' with screws (actually they should be closer to 10-11ft long). Nearly all commercial machines in that length are rack. The rack has zero backlash as they are pre-loaded.

Originally Posted by Bugsy36 It seems to me (and that is not not saying much ) that I had a misconception that adding an additional 4 feet to the X-axis would be nominal (unless of course I was already using a rack with gear reduction).

It is trival with a rack system. Here is a 20'x6' MechMate:

http://www.mechmate.com/forums/attac...1210820894.jpg

Adding length to x and y on a Mechmate is pretty simple and doesn't raise the cost that much.

Originally Posted by Bugsy36 Just for the sake of clarification, the machine that I am wanting to build will be primarily manufactured from aluminum (too keep the gantry weight to a minimum)

You might want want to reconsider that for the following reasons:

• 80/20 Al is hard to come by in the lengths you are talking about (12-13ft)
• Unless you are a TIG expert, you'll have to go with 80/20 which limits your choices
• Weight is a GOOD thing. Ask a football player. With a light weight unit, your system will never be as accurate as a heavy one. Think about any "good" table saw - they all have case iron tables, there is a reson for this
• Steel is cheap. My 100"x52" table for steel was only $950 and I paid "retail" prices.

[Bugsy36]

Again, that you all for your help. Your patience with my lack of detailed explanation is greatly appreciated.

I am only aware of a single belt-reduced MechMate

My mistake. I thought that is the way the MM people were going and that Shopbot was using the Vextra with the gear reduction built in. I have attached a pdf to make sure that I am understanding correctly. If I am, then gear to gear should not be that difficult or expensive. I am trying to avoid the bolt-on gear reduction motors due to cost. Using the 900 oz. complete 4-axis setup I am at approx. $1200 vs. something like $275 each + drivers + power supply. (This does not include any gear reduction)

I'm pretty sure you'll have problems when you get to 9' with screws (actually they should be closer to 10-11ft long). Nearly all commercial machines in that length are rack. The rack has zero backlash as they are pre-loaded.

Which is why I am trying to better understand the rack systems. Screws are easy for me to comprehend but at that length (and proper diameter) the cost is what is hard to swallow.

It is trivial with a rack system. Here is a 20'x6' MechMate

Unfortunately it will not fit

You might want want to reconsider that for the following reasons: * 80/20 Al is hard to come by in the lengths you are talking about (12-13ft) * Unless you are a TIG expert, you'll have to go with 80/20 which limits your choices * Weight is a GOOD thing. Ask a football player. With a light weight unit, your system will never be as accurate as a heavy one. Think about any "good" table saw - they all have case iron tables, there is a reson for this * Steel is cheap. My 100"x52" table for steel was only $950 and I paid "retail" prices.

I believe that I have a source for the 80/20, but have yet to verify. I will know more about current metal pricing on Monday or Tuesday. I actually like steel, and do not have too much issue for MIG welding and cutting (need to get MIG out of storage) but any drilling will be another issue. I do not have a drill press, nor the help to sling the large pieces of steel around. I am contemplating using steel for the X and 80/20 for the Y. I am concerned about the gantry weight and the motors being 900 oz. Not sure if they are too much or not enough. Please give me your opinions.

I see you are in the Southeast, tell us where and you might find someone volunteering to help.

I am in Orlando

You guys have been great. Thank you!!

[dmoore]

Originally Posted by Bugsy36 My mistake. I thought that is the way the MM people were going and that Shopbot was using the Vextra with the gear reduction built in. I have attached a pdf to make sure that I am understanding correctly. If I am, then gear to gear should not be that difficult or expensive. I am trying to avoid the bolt-on gear reduction motors due to cost. Using the 900 oz. complete 4-axis setup I am at approx. $1200 vs. something like $275 each + drivers + power supply. (This does not include any gear reduction)

All I can say is that "you get what you pay for". If you want to push through hardwoods at 600+ IPM, you need that much torque. No freed lunches.

Originally Posted by Bugsy36 I believe that I have a source for the 80/20, but have yet to verify. I will know more about current metal pricing on Monday or Tuesday. I actually like steel, and do not have too much issue for MIG welding and cutting (need to get MIG out of storage) but any drilling will be another issue. I do not have a drill press, nor the help to sling the large pieces of steel around. I am contemplating using steel for the X and 80/20 for the Y. I am concerned about the gantry weight and the motors being 900 oz. Not sure if they are too much or not enough. Please give me your opinions.

I think by the time you get 10+ feet of 80/20 still enough for really accuracy, you'll be past the cost of steel. You don't have to use a drill press to drill the MechMate parts - in fact, the vast majority of my drilling on my my MechMate was done with a standard corded drill. On the other hand, a drill press is only $200-$300. The steel weight isn't a major deal. I built my entire MechMate by myself - including moving around the steel. If you can handle 3/4" sheets of MDF, than the steel shouldn't be any problem.

The weight of the gantry is only somewhat important (other than to ensure better quality cuts), for example, my gantry, Y-car, Z and router clocks in at over 250 pounds. Despite that, I can easily push the gantry with a single finger. Keep this in mind - large steppers are bad and steppers generate the most torque at lower speeds. See this post:

http://www.mechmate.com/forums/showthread.php?t=255



Maybe I should ask a few other qualifying questions:

• What size materials will you be cutting?
• What are the maximum speeds you need to cut?
• What is your budget?
• Is this hobby or commercial work?
• How much space do you have for the CNC Machine?
• How much accuracy do you need?

[Bugsy36]

* What size materials will you be cutting?

3/4" Hardwood Ply, 3/4 MDF, 1" PVC, 2" HDU and .080 Aluminum (maybe .125 Aluminum on rare occasion)

* What are the maximum speeds you need to cut?

I am trying to target the machine to be have rapids of 300 to 400 ipm. The majority of my cuts are made at .030, some a little deeper. This is what started the idea of a new machine.

The current machine runs at 12 ipm, and only has 90 oz 23's. I am going to start running deeper and much slower (2 to 3 ipm), but my table is still too small (which is why I am not opting to modify the current machine).

I have seen Art's post, which is why I have considered 900 oz 34's. Larger definitely seems to be a no-no, and I have no objection to 600 oz 34's, if that would be a better choice. (David) What size are you running on your machine?


* What is your budget?

Please don't laugh. $2000 to $2500 (not including a new Porter). From the calculations I have gathered so far, it is possible.


* Is this hobby or commercial work?

Both. Primarily commercial but as you know the option of CNC opens the door to all sorts of hobbies. If I had the cash I would be going after a 5x9 Shopbot Alpha that they want 10K for it (spindle, phase converter, etc.) and just be done with it but I do like the world of debt.


* How much space do you have for the CNC Machine?

I have a space of 8 x 16. Enough for a 5x5 and really pushing it for a 5x9. Stepping up to a larger 4x4 or 5x5 is a need, while a "full sheet" machine would make some wood working hobby projects easier.


* How much accuracy do you need?

I am trying to get close to .005 but in reality, for what we do we could get away with .015 and never know the difference. What we do not need is machine metal accuracy for moving parts like you would find in motors and engines (but that would be nice).

[dmoore]

Originally Posted by Bugsy36 * What size materials will you be cutting? 3/4" Hardwood Ply, 3/4 MDF, 1" PVC, 2" HDU and .080 Aluminum (maybe .125 Aluminum on rare occasion)

Nearly all these ship in standard 4' widths (maybe excluding the Aluminum), are your finished parts greater than 5x5 or 5x4?

Originally Posted by Bugsy36 * What are the maximum speeds you need to cut? I am trying to target the machine to be have rapids of 300 to 400 ipm. The majority of my cuts are made at .030, some a little deeper. This is what started the idea of a new machine. (David) What size are you running on your machine?

Clearly not a problem for a MechMate with a basic router setup. My motors are the geared Oriental Motor PK263A1A-SG7.2 - at $257 each. I think trying to quote a single torque number is pointless based on how they are wired and the speed at which the torque is measured. They are "listed" at 43 lb/in (NOT oz/in) at just above 0 RPM. The proof is in the pudding - even jogging at 400+ IPM, I can't physically contrain the gantry with my hands. There is clearly plenty of power to push just about any bit through pretty much anything (which moves you to a spindle anyway...)

Originally Posted by Bugsy36 * What is your budget? Please don't laugh. $2000 to $2500 (not including a new Porter). From the calculations I have gathered so far, it is possible.

From what I've read in the past on Joe CNC's 4x4 forum - that is possible, though for a 4x4 machine.

Originally Posted by Bugsy36 * Is this hobby or commercial work? Both. Primarily commercial but as you know the option of CNC opens the door to all sorts of hobbies. If I had the cash I would be going after a 5x9 Shopbot Alpha that they want 10K for it (spindle, phase converter, etc.) and just be done with it but I do like the world of debt.

I guess the reference to the 10k Alpha is used? A new, 5x8 is listed at $15,795 (http://www.shopbottools.com/prsalpha.htm) and a 3hp spindle/VFD is an extra $3,395 - or a total of about $20,000 with shipping. Your CNC business must be on the smaller side? A few hundred a month? Do you have specific ROI based on future orders you can handle with the larger machine? I'm willing to bet that if commerical guys can make a profit on a 15-20k shopbot, it is surely possible to make a ROI of 5-8k. There are a number of posts on the MechMate forum about paybacks of just 6 months for a complete build - not bad...

Originally Posted by Bugsy36 * How much space do you have for the CNC Machine? I have a space of 8 x 16. Enough for a 5x5 and really pushing it for a 5x9. Stepping up to a larger 4x4 or 5x5 is a need, while a "full sheet" machine would make some wood working hobby projects easier.

The 8x16 sounds like a single bay garage - no? Of course this opens all kinds of other cans of worms - space for finishing of goods, space for dust collection, space for materials storage, space to manipulate sheet goods on to the table, etc. I think if I were in your place, that I'd go back to the maximum size of the items you will be cutting. If you arn't doing full sized outdoor signs, 8ft lengths of moulding or other things that just "have" to be cut all at once, then I would fall back to the smaller 4x4 or 4x5 machine.

Another possiblity is to build a MechMate in a 5ft Y and 4ft X. Then, if you get more business that requires a 5x9 or 5x10 or 5x20, you could simply pull the gantry off the rails, build a new base (I'm guessing maybe 300-400$ to move from 4 to 8ft) and you are there. All the "guts" of the MechMate are located on the gantry and the y-car (proxies, motors, wiring, etc).

Originally Posted by Bugsy36 * How much accuracy do you need? I am trying to get close to .005 but in reality, for what we do we could get away with .015 and never know the difference. What we do not need is machine metal accuracy for moving parts like you would find in motors and engines (but that would be nice).

That shouldn't a problem for pretty much any machine. Just for reference, the MechMate, with 7.2:1 motors and a 30 tooth pinion is a resolution of .0003"

[Bugsy36]

Thank you David. The space mentioned is the amount of space that is allotted for the machine (one garage out of three). The is other space for storage, manipulation, and finishing.

As of now, the CNC portion is small. The plans are to make it bigger than what it is. On occasion full size routed signs are needed, but not so much that it could not be handled without indexing. A rack driven 5x5 is probably the feasible way to go. I guess I could also index "home honey-do projects" also.

You are 100% correct about the ROI and the Bot being used (actually repo with 4hp spindle). In regards to the "pay as I play philosophy", I have been down Credit Rd. before. I like the idea of not having to worry about bills being paid (once owing more than 500k for equipment). On the other hand if I start to approach $5k then it would be to my advantage to just buy the Bot, new or used, although hating the thought of doing it.

The design that I am thinking about is very similar to what the MM is but also using the top and bottom of the X-rails as Joe did with his 4x4, I believe that I can put it together in such a way that I can literally build it in less than a week, and preferably two full days of work. If I could do it out of aluminum, like Joe's new 4x4, I could build and assemble it in a day (with proper planning of course).

Went to the steel supplier today. Here is what they gave me for pricing (no 80/20 through them):

• 6"x2"x11ga tubing is $320 per 24'
• 4x2x11ga tubing is $315 per 24'
• 2x2x11ga tubing is $137 per 24'

I am okay with those prices.

My biggest expense seems to be coming from the motors/electronics. compared to your machine, with the Oriental steppers, it seems that 900 oz. may be too much. Yours come in at 688. My guess is that you have about $2k in motors and drivers. Am I correct? Was I understanding the gear reduction properly? If I was, what would you recomend for gears (hopefully a source too) for the 20 rack?

Hopefully I will be able post my design idea later tonight or tomorrow.

Again, Thank You!

[dmoore]

Originally Posted by Bugsy36 The design that I am thinking about is very similar to what the MM is but also using the top and bottom of the X-rails as Joe did with his 4x4, I believe that I can put it together in such a way that I can literally build it in less than a week, and preferably two full days of work. If I could do it out of aluminum, like Joe's new 4x4, I could build and assemble it in a day (with proper planning of course).

It sounds more hobby than commerical and that build time is more critical than some of the other driving factors. The modified Joe CNC 4x4 design sounds like a good match - fast to build, off the shelf parts, no welding, etc.

Originally Posted by Bugsy36 Went to the steel supplier today. Here is what they gave me for pricing (no 80/20 through them): 6"x2"x11ga tubing is $320 per 24' 4x2x11ga tubing is $315 per 24' 2x2x11ga tubing is $137 per 24'

You didn't say but I can only assume that is for aluminum - which comes to a total of $772. The EXACT same thing in steel (which is already stronger to start with and you could go up a few steps, is $273 at .80$ per pound (about the retail going price).


If you have a 5x9 - you'll be crawing ON it - just something to keep in mind.

Originally Posted by Bugsy36 My biggest expense seems to be coming from the motors/electronics. compared to your machine, with the Oriental steppers, it seems that 900 oz. may be too much. Yours come in at 688. My guess is that you have about $2k in motors and drivers. Am I correct? Was I understanding the gear reduction properly? If I was, what would you recomend for gears (hopefully a source too) for the 20 rack?

You can see EVERY single dime I spent on mine by going here:

http://www.dmoore.com/MechMate/bom.xls

I spent $147 x 4 for drivers = $588 (actually I got them cheaper at the end-of-year "wholesale" sale.
I spent $257 x 4 for motors = $1028
Total of: $1,616 Plus, I could have always sold (ebay) or transfered them to another machine in the future, so I feel they are a good investment.

Check my BOM (above) for rack and gear suppliers.