PDA

View Full Version : cnc for marble



yasser
23-12-2015, 12:38 AM
Hii All i am trying to build a big cnc router to use for stone engraving ,i build my first one one month ago but was full of mistakes ,i need advice on solving the problems i am facing right now like loosing steps ,achieving higher speed and higher precision building a rigid machine i will be working on a 3m *1.5 m machine

Chaz
23-12-2015, 02:36 PM
Hii All i am trying to build a big cnc router to use for stone engraving ,i build my first one one month ago but was full of mistakes ,i need advice on solving the problems i am facing right now like loosing steps ,achieving higher speed and higher precision building a rigid machine i will be working on a 3m *1.5 m machine

I suspect before anyone can help you, you need to provide more details. Your question is a bit like 'how do I solve world peace'?

Pics, specs? What speeds can you do, what are the issues etc ...

yasser
12-01-2016, 12:23 AM
Dear chaz thank u for your reply and of course u r right here are some photos for my first machine and the major problems are
1- the max speed i can use is 400 mm/sec after that i loose steps dramatically and also this consumes a lot of time as the max step down is from 2-4 mm, i use the steppers directly i think i should be using speed reduction method ,i use nema 34 steppers 11n.m
2- the aluminium sections are very weak for this application i think i will be using steel as the new machine will be 3m in th x axis and 2 m in the y axis so i am searching for a stiff beam design for the length 3m
17204172051720917203172061720717208

Rufe0
12-01-2016, 05:55 AM
Is this machine to be used only for marble? I would have thought the only way to machine marble would be very slowly. In which case the maximum speed of the machine would be of little importance. So yes it maybe a good idea to use a gear reduction however all gear systems have backlash so you will need to account for that.

Clive S
12-01-2016, 09:11 AM
1- the max speed i can use is 400 mm/sec after that i loose steps dramatically and also this consumes a lot of time as the max step down is from 2-4 mm, i use the steppers directly i think i should be using speed reduction method ,i use nema 34 steppers 11n.m
Yasser first welcome to the forum. Are you sure about the 400mm/sec as that is 24mtr/min and is very fast. you need to give the specs of the controller and drives also the voltage of the output of the power supply.

yasser
12-01-2016, 03:57 PM
would using a belt system be a good idea

Rufe0
12-01-2016, 06:42 PM
The problem is at 3m long your going to need quite large screws so that they aren't flopping & wipping around. Large screws generally have a large thread pitch, plus you'd want to use a large thread pitch anyway for rigidity. With a thread pitch of say 20mm your only going to get steps accuracy/resolution of 0.1mm with a stepper motor(with no gear reduction). Which may or may not be OK in terms of accuracy but it means the slow speed movement of the machine would be horrible. So I would suggest you find a way to gear it down or use whatever system to achieve a high resolution. 0.001mm would be good I guess. That would make slow feeds very smooth.

In terms of the machine design I'm sure we can help you design a machine that will be alot more rigid. Do a drawing and we can comment on it. Firstly you need atleast 2 rails per axis, 2 carriages/bearing blocks per rail and 2 screws per axis. I recommend you use thick wall steel box section tube for the construction, however some parts may need to be even heavier duty and so may need to be custom machined for you. I would recommend a dual gantry design. Large seperation between carriages/bearing blocks probably 400 or 500mm. That means you machine will have to be 2000x3500 to get 1500x3000 travel.

I assume that the marble will almost always be a similar thickness and that it won't be very thick. That is a good thing because you won't be able to have much Z travel, in fact I recommend you make it as little as possible.

17215172141721317216

phill05
12-01-2016, 08:03 PM
Hii All i am trying to build a big cnc router to use for stone engraving ,achieving higher speed and higher precision building a rigid machine i will be working on a 3m *1.5 m machine

You will not do much good with the type of bolt together frame that you are showing.
You do NOT need higher speeds, but you do need a RIGID machine and plenty of water, rough out leaving 4 to 6 mm for finish cut, using Diamond and carbide cutters depending on the type of stone your cutting, marble / limestone is soft and will cut well with carbide, for granite use PCD (poly crystal diamond ) cutters, sandstone use brazed diamond tooling or pcd.


17217 17218

Head being cut is Limestone and cut with carbide to a depth of 16mm, the eyelet is in sandstone and is cut with PCD.

Phill

yasser
12-01-2016, 09:17 PM
thnx rufe0 doible girder is a good idea , i thing of using rack and pinion instead of screws and belt driven system for gear reduction i will make the drawing and post it for your feed backs

yasser
12-01-2016, 09:46 PM
You will not do much good with the type of bolt together frame that you are showing.
You do NOT need higher speeds, but you do need a RIGID machine and plenty of water, rough out leaving 4 to 6 mm for finish cut, using Diamond and carbide cutters depending on the type of stone your cutting, marble / limestone is soft and will cut well with carbide, for granite use PCD (poly crystal diamond ) cutters, sandstone use brazed diamond tooling or pcd.


17217 17218

Head being cut is Limestone and cut with carbide to a depth of 16mm, the eyelet is in sandstone and is cut with PCD.

Phill excellent work just how long did it take to finish the head nad what is the shape of the tool used ...these are the results172191722017221 i reached till now
the hexagon tile 50 cm * 50 cm took around 3 hours maximum depth 5 mm step down 5 mm

Rufe0
12-01-2016, 10:25 PM
The distance traveled per revolution with rack and pinion is the pitch circle of your pinion gear diameter * pi, for example 1inch 25.4mm pitch circle diameter pinion gear, 25.4*3.145 = 79.883mm distance traveled per revolution. Divide that by your stepper motor step count which is normally 200. 79.883/200=0.399415. So the resolution is 0.399415... so to get that 0.001 resolution I was talking about you'd need a 300:1 gear ratio, which is crazy. 0.001 might be a little optimistic... but the point being that the problem of smooth slow feeds will be much worse with rack and pinion.

Don't get me wrong rack and pinon is great for soft material routers that need high speeds but I don't think it's correct for your application. Of course you could probably make it work but it wouldn't be the ideal solution.

yasser
12-01-2016, 11:39 PM
The distance traveled per revolution with rack and pinion is the pitch circle of your pinion gear diameter * pi, for example 1inch 25.4mm pitch circle diameter pinion gear, 25.4*3.145 = 79.883mm distance traveled per revolution. Divide that by your stepper motor step count which is normally 200. 79.883/200=0.399415. So the resolution is 0.399415... so to get that 0.001 resolution I was talking about you'd need a 300:1 gear ratio, which is crazy. 0.001 might be a little optimistic... but the point being that the problem of smooth slow feeds will be much worse with rack and pinion.

Don't get me wrong rack and pinon is great for soft material routers that need high speeds but I don't think it's correct for your application. Of course you could probably make it work but it wouldn't be the ideal solution.

ok then i will stick to the screw system but i am worried about the length 3 m , and i think there is a problem in my setting of the driver i am using ,this is the setting i am using on the driver
17222

Rufe0
13-01-2016, 12:49 AM
What specifically do you think is 'wrong' with it? As is do you think there is malfunction with the driver or are you wondering if another setting would be more appropriate?

Rufe0
13-01-2016, 12:53 AM
Just to reiterate on earlier 0.001 resolution was a bit optimistic. 0.01 is more realistic and would be fine.

yasser
13-01-2016, 02:03 AM
i was wondering if another setting would be more appropriate

Rufe0
13-01-2016, 02:22 AM
Maybe, try a few out. It depends on your steppers, drivers and how well they work together, there is probably a sweet spot. I would stay clear of the really high ones though, mircosteps aren't always as good a thing as you might think they would be for a number of reasons. Have a read about it http://www.micromo.com/microstepping-myths-and-realities . Obviously remember if you change your microsteps you will need to adjust the settings on your computer otherwise it will come out the wrong size.

JAZZCNC
13-01-2016, 10:13 AM
Don't get me wrong rack and pinon is great for soft material routers that need high speeds but I don't think it's correct for your application. Of course you could probably make it work but it wouldn't be the ideal solution.

Sorry but your wrong R&P is the perfect choice in this application. Just about every industrial Stone cutting machine uses Helical R&P.
Your also Wrong about the Resolution and Micro steps because no one who runs steppers would run at FULL step 200ppr. WHY.? . . . Because Resonance and the affects it has on the motors is crippling to motor performance when full stepping. Infact you'll be hard pressed to find decent quality drives that provide Full step mode. Half Step(400ppr) is usually minimum. In real world use then between 800 and 2000ppr are used to give best balance of performance/smoothness.

In this application cutting hard material and Steel framed machine the resonance will be high so using higher micro steppping would be used to help smooth motor performance. Micro stepping however shouldn't be thought of as way increasing resolution, though it does to some degree dependant on motor quality.

Resolution should be calcualted based on pitch. R&P will minimim ratio of 3:1 (typical 5:1) dependant on several factors like Pinion size, motor Rpm, desired Rapid speeds etc.

Lets look at it again with realistic figures using same pinion size. Lets start by applying a minimum Ratio 3:1 with minimal 400ms setting and seeing the speeds, resolution this would give.?
With an affective pitch of 79.833 / 3=26.611mm pitch which based on usable RPM of typical motor used being 700rpm this would give 18627mm/min or 18.6Mtr/min which is ridiculous speed for cutting stone that would never be used.
The resolution would be 26.611 / 800=0.066 which is still more than enough resolution.

Now lets go to more likely setup that would be used. 5:1 ratio still giving a speedy 15.966 Pitch and using 1600Ms which is often optimal for motor smoothing.

Rapid speed: 15.966 * 700= 11,176mm/min or 11.1Mtr/min still a nice healthy speed which you'll never use cutting stone other than for positional rapid moves.
Resolution: 15.966 / 1600= 0.009 far more than needed.
Now the Resoultion figure isn't exactly true figure but rather theoretical figure and in reality there will be some mechainical loses but the higher it is to start with the higher it will balance out to in the real world.

Also the higher ratios have increased torque which will be needed with R&P because it's quite inefficient compared to ballscrews so will need more power.

Helical R&P it's better than straight cut and provided you protect it from debris and keep pinion tensioned into rack properly then it's the perfect solution for long machine like this.

Rufe0
13-01-2016, 12:18 PM
Just about every industrial Stone cutting machine uses Helical R&P.
That doesn't necessarily mean that R&P is 'the ideal solution'. Hate to say the obvious but if everyone jumped off a cliff would you join them? R&P is great for a number of things, I think primarily where you have a very long travel. Is it ideal for a machine where you primarily need smooth slow speed movement in a travel length acheivable by a screw? I don't think so but thats just my opinion.


Your also Wrong about the Resolution and Micro steps because no one who runs steppers would run at FULL step 200ppr.
Yes sorry if it came across as me suggesting you use fullsteps. Of course you should use microstepping. What I should have mentioned is... referring to my link http://www.micromo.com/microstepping-myths-and-realities

"The real compromise is that as you increase the number of microsteps per full step the INCREMENTAL torque per microstep drops off drastically. Resolution increases but accuracy will actually suffer.

The consequence is that if the load torque plus the motor’s friction and detent torque is greater than the incremental torque of a microstep successive microsteps will have to be realized until the accumulated torque exceeds the load torque plus the motor’s friction and detent torque.

Simply stated, taking a microstep does not mean the motor will actually move!"

A microstep doesn't guarantee movement. For my point about wanting a high resolution for smooth slow speed to be valid and useful the resolution has to be real world accurate resolution.

So how many microsteps can you guarantee? Well ofcourse your motor could stall completely, but lets put that to a side for a moment. 1 microstep/full steps you can guarantee them. 2 yeah thats fine. 4 probably OK. 8 maybe I wouldn't be confident. More than that I would say you definitely couldn't call that guaranteed accurate movement.

From the point of view that over doing it is better than under doing it, using 200steps/rev for this calculation means you definitely will get that accuracy. However realistically 400/2microsteps or 800/4microsteps is probably the right thing to base the calculation on.

When the machine is built you can still run it at whatever microsteps you like, nothing stopping you doing that.

I'm suggesting something like 20mm pitch screw with 1/3 belt drive. 20mm/3 = 6.6mm / 400 = .016mm or / 800 = .0083. 0.016mm - 0.0083 real world accuracy. Or better yet a 10mm pitch screw with a belt drive.

JAZZCNC
13-01-2016, 03:29 PM
That doesn't necessarily mean that R&P is 'the ideal solution'. Hate to say the obvious but if everyone jumped off a cliff would you join them? R&P is great for a number of things, I think primarily where you have a very long travel. Is it ideal for a machine where you primarily need smooth slow speed movement in a travel length acheivable by a screw? I don't think so but thats just my opinion.

You obviously never built a large machine.? If you had you'd realise the complications involved with using long ballscrews and why R&P is more suited so therefore Ideal for this application and why it's used. Got nothing to do with following the crowd it's a practicle solution that does the job.



Yes sorry if it came across as me suggesting you use fullsteps. Of course you should use microstepping. What I should have mentioned is... referring to my link http://www.micromo.com/microstepping-myths-and-realities

"[I]The real compromise is that as you increase the number of microsteps per full step the INCREMENTAL torque per microstep drops off drastically. Resolution increases but accuracy will actually suffer.

Ok here again we have typical Theory not matching real world usage. These things your reading while true have no real world affect on how the machine performs, well not that you'll see while using.
There are so many variables that come into play that affect the performance so much more than these technical technical details. Quality and spec of drives and motors along with voltage will have much greater affect, as will Machine stiffness and Resonance.

Motor Ms theory is based on Fixed parameters for a particular motor/drive/voltage/ combination which mostly goes out the window in real world because of too many variables make the machine setup a moving target.

The simple fact is that what your quoting from the article regards Ms/torque etc doesn't have any noticble affect in real world use because of all the other variables come into play. But running the motors in MS range that lets them get affected by resonance certainly does. This is a 100% fact not theory with 95% chance of happening if you run at full step with poor spec components.!



When the machine is built you can still run it at whatever microsteps you like, nothing stopping you doing that.


Sorry but again this not true. There is a limit you'd run at. Running too high causes all sorts of undesirable results which have a negative affects on performance and system.
The simple fact is that there is NO IDEAL setup and each combination of components and machine design/Stiffnes needs to be setup to suit that machine. . . .But there is a Limit.!!

We are not going to agree on this probably but I'm sure you'll come to realise what I'm saying when you have built a machine.!

Rufe0
13-01-2016, 04:00 PM
Re:JAZZCNC

Your not saying anything that I haven't said. I said he should try out different settings to find what works best.

I'm saying on top of that he should design his new machine with high resolution in mind for smooth slow speed assuming that you won't get better than 2 or 4 microsteps accuracy. That way if you set your machine to 8 or 16 microsteps and it achieves that accuracy then great you've not lost anything. If it can only achieve half that accuracy then thats OK because you designed it with that in mind from the start.

yasser
26-01-2016, 08:06 PM
hii jazz i have to agree with you on the point that all stone machinery producers use R&P so i after reading previous threads on large machines and seeing a lot of nice designs i made a preliminary design ,
i made a adjustemnt screws for the y linear guides to adjust parralel and leveled axis,this machine will work on marble and graniteso we nned a heavy body as the wieght of the slaps can reach 5oo kg
waiting for your input
17374173751737617377173781737917380173811738217383