BuildingAfloat

[BikerAfloat]

Hello All, here is my first post in what I hope will become a full build log. As mentioned in my Introduction (http://www.mycncuk.com/forums/new-me...say-hello.html) I have one quite tough requirement a bit of steel machining and am building it with a view to making the plans available to others who are in the same boat as me, well not my Narrowboat, but you get what I mean :-)

Posted here are the drawings for the machine I have in mind for my use X=600mm (Apprx 400mm useable) Y=500mm (Apprx 300mm useable) and Z=350mm (Apprx 150mm useable). I hope this design is scaleable by adding longer rails and extra rail supports, that was certainly my intention. I have also allowed space where the drive nuts are to go upto 16mm leadscrews for the bigger machines, though I have seen 8'x4' machines with 12mm, to me it just doesn't look man enough.



Build requirements are to use as many standard components as possible and any unique parts must be able to be made on a Pillar drill. I tried to design it so there had to be no sections that had to be cut to length, but I'm afraid I was only happy with a design that involved the 4 long X Axis beams to be of exact length. All other joints are on the main faces of the extrusions.

To repeat my self slightly My choice of components are governed by 3 factors 1) Cost, 2) Sizing, where possible they are Multiples of the 20mm of the sections I have used & 3) The need for nothing more complicated than a pillar drill.

The machine has come together quite chunky, but that to me is OK as I want to machine metal quite a bit of the time, say 10-20% and I have allowed the cutting area to extend over the rails at the front to give a solid base as I can for doing this.

I will put a Bill of Materials (BOM) on here soon, it's just a bit messy, as it was being constructed along side the drawing. I posted some of the major specs in my introduction, but a BOM will give a more complete story.

I may do a slight redraw to include the DumpsterCNC anti-backlash nuts as for the price I'm not sure my idea, to be posted soon can compete! I have tried to justify Ballscrews, but I wince at the prices and speed and accuracy are not my ultimate goals. I may change my mind of course once I start making stuff, but I am fairly confident I can retro fit them at a later date, without too much trouble. Mostly I want to get a machine going so I can start learning!!!!

I'm happy to provide more detailed views if you need them and really want to know if this design is viable.

Cheers,
Geoff.

[JAZZCNC]

Hi Geoff,

I see some issues that will definitely affect it's ability to cut steel and will also affect performance in Aluminium.?

The tall gantry won't be any where near strong enough.!! While it's sat directly on the bearings the what looks like 45x45 uprights just won't handle the forces steel and aluminium will put on it.

The un-supported rails on the Z axis and to be honest the whole z axis design just won't be upto the job of steel or Ali.!

Don't know if your planning on putting any braces across the frame to support the bed.? But again to cut any hard material then extra support will be required.

Please don't take my comments has pulling the machine down just comments based on quite extensive work building router based machines that can cut hard materials and still give an acceptable quality.! . . . . The thing you and others should be aware of is that router based machines can never cut steel or aluminium like dedicated milling machines can.!!

Wish you luck and keep the designs coming.

[BikerAfloat]

Hi Jazz,

I see some issues that will definitely affect it's ability to cut steel and will also affect performance in Aluminium.?

I am really not surprised by this :-) , it is why I have sought help on the design! I am starting to think that some of the videos of what to me at least look like flimsier machines cutting Ally accurately, might be using a bit of artistic licence!

The tall gantry won't be any where near strong enough.!! While it's sat directly on the bearings the what looks like 45x45 uprights just won't handle the forces steel and aluminium will put on it.

Although doubled up, the uprights are in fact just 40x20! so are as weak as soft smelly stuff in that case :-) I was thinking of joining the two Y Axis cross beams with a sheet of Ally to stiffen the entire structure, just haven't got round to adding it to the drawing.

The un-supported rails on the Z axis and to be honest the whole z axis design just won't be upto the job of steel or Ali.!

Don't know if your planning on putting any braces across the frame to support the bed.? But again to cut any hard material then extra support will be required.

To cover both of the above points, the Y Axis will travel over the cross members at the front of the X Axis and I was planning to work the material with the Z Axis as close to the top of its travel as possible. That means all loads will travel the miniumum distance to the strongest parts of the machine. Not sure if that will help enough? I am hoping to keep the centre as free from bracing as possible as I want to be able to engrave on Motorcycle engine cases, but I guess it is only a small amount more work to remove bracing to get big stuff up from underneath, when compared to removal of the Sacrificial bed and its support.

Please don't take my comments has pulling the machine down just comments based on quite extensive work building router based machines that can cut hard materials and still give an acceptable quality.! . . .

I didn't Jazz, it was what the drawings are posted for!

. The thing you and others should be aware of is that router based machines can never cut steel or aluminium like dedicated milling machines can.!!

I do appreciate that dedicated machines are the best way to go, they always are :-) and I would love to have space for more machines, but I will be working in a 6ft cubed area on my boat and I really only have space for this 1 machine outside the Pillar drill! Super accurate finishes are not going to be critical on the parts I make, they will be more art than engineering, so cutting speeds, hand de-burring and polishing, while a nuiscance aren't performance critical.

Wish you luck and keep the designs coming.

Thank you Jazz, I guess I'm always going to be in conflict with my ambitions for the way the machine HAS to be built and the way in which it will operate. As it stands the bearings take up a big chunk of the travel and I must admit I have run out of ideas how to reduce the gantry height or improve the Z Axis support without loosing even more travel or introducing the need for outside machine work! Any advice on how to achieve this and still maintain my production intent would be very gratefully received!

The only 3 rules here are:

1) I must be able to build it with just basic tools and a pillar drill.
2) There must be no physical outside help with the mechanical build - Virtual help with the design is OK!
3) As close to a 400x300 minimum working area as possible, within the above 2 constraints as possible.
3 and a bit) Sneaking in up to 100mm on X&Y is just about acceptable space wise, but I 'd rather avoid it!

That is the challenge I have set myself. Compromised performance will just have to be lived with, it is the nature of the beasty. Having said that the design can be changed to reduce the compromise!

Cheers,
Geoff.

[John S]

Geoff,
Just a heads up but any extrusion ordered from KJN at Hinkley comes cut to length very accurately and requires no more work.

KJN Aluminium Profile

Decent prices as well.

[Robin Hewitt]

Jazz is the man, you won't go far wrong if you have faith and simply do everything he says.

Incidentally, putting 2 steppers on one axis usually ends in all sorts of problems.

[BikerAfloat]

Jazz is the man, you won't go far wrong if you have faith and simply do everything he says.

I have followed Jazz's inputs on varied posts and have to say from what I have read I do have faith in his advice, however I was brought up to question everything where you have doubts, so I wont follow anyone blindly :-)

Incidentally, putting 2 steppers on one axis usually ends in all sorts of problems.

I have read that this idea can cause problems, yet one of my favourite machines out there is the Heiz from CNC-step, This was the inspiration for this feature being included in my build. They can't be that bad (well at least I hope they aren't) as their sales are in the 1000's now!

Advantages I see are:

1) Open bed design, this is important for me as I will want to get awkward shaped engine components in the machine for engraving.
2) More grunt for moving the heavy gantry.
3) Even loading of gantry to prevent, or at minimum reduce, racking.
4) Reduced component count
and 5) Possibly better accuracy when compared to a single central motor & belt system?

Disadvantages I see are:

1) Increased cost, but this should be a one-off.
2) Synchronisation/Alignment of the screws at the assembly stage, again this should be a one-off.
3) Machine ties itself in knots if one motor looses steps
or 3.5/4) Worst case scenario machine goes completely to peices should one motor stall or stop all together.

Prepared to accept any more Pro's and Con's before I freeze the design.

1 & 2 aren't my biggest worries, as a) this is an expensive tool anyway and I think (hope?) the advantages are worth it & b) this b@rst@rd is going to have so many adjustments required to get it straight and true, one more won't really matter :-)

3, 3.5 & 4) Scare the living daylights out of me!

Any advice to how best to go about reducing the risks and or providing the single motor & belt drive within my construction requirements would be gratefully received.

Cheers,

Geoff.

[Jonathan]

however I was brought up to question everything where you have doubts, so I wont follow anyone blindly :-)

Excellent

I have read that this idea can cause problems, yet one of my favourite machines out there is the Heiz

Your options are limited to having two leadscrews on the X-axis - there is no way you'll get away without two when cutting metals unless the gantry is exceptionally strong which is almost never the case.

I use two motors on my X-axis and have not had any problems due to using two motors. When testing I have had one motor stall, but the gantry is sufficiently strong for this not to cause noticeable damage, especially since once one motor has stalled the other will invariably stall a fraction of a second later.
You can save a little money by only having one motor and using pulleys, but since you want the motor and driver later for a 4th axis you could always start with two motors and if you're not happy with it link the screws with a timing belt and use the now spare motor and driver on the 4th axis.

I urge you to reconsider not using ballscrews. I would add up the cost of the 4 leadscrews you are intending to get along with all the parts they require, then send Chai on eBay a request for the price to get the 4 ballscrews (and rails) you would need with the standard end machining and bearing blocks. I'm sure you will be pleasantly surprised.

You mentioned using the 3Nm motors from CNC4You. These are a good choice since they will be more than adequate for this machine and offer the best price per Nm in Nema23/24 frame size. Currently you've got all the leadscrews direct drive which gains simplicity, however using timing belts gains a lot more than that. You can change the drive ratio (important to get a good feedrate if the leadscrew is quite a low pitch), motor alignment is no longer so critical, plenty of people seem to have problems with couplings breaking and belt drive provides damping which helps reduce resonance issues to name but a few...This is a good place to get the timing belts and pulleys.

Putting a sheet of aluminium on the back of the gantry will greatly increase it's stiffness since it will essentially eliminate bending of the gantry sides parallel to the Y-axis, thus reducing the tool deflection in the same axis.

[JAZZCNC]

I have followed Jazz's inputs on varied posts and have to say from what I have read I do have faith in his advice, however I was brought up to question everything where you have doubts, so I wont follow anyone blindly :-)

Good upbringing because I wouldn't want anyone blindly following my advice and I only offer it to help the person come to the best design for there needs and hopefully steer them clear of any issues I know thru experience will come from what I see or comment on.!

I have read that this idea can cause problems, yet one of my favourite machines out there is the Heiz from CNC-step, This was the inspiration for this feature being included in my build. They can't be that bad (well at least I hope they aren't) as their sales are in the 1000's now!

Ermm I wouldn't be so sure and they are certainly massively over priced IMO.!

Advantages I see are:

1) Open bed design, this is important for me as I will want to get awkward shaped engine components in the machine for engraving.
2) More grunt for moving the heavy gantry.
3) Even loading of gantry to prevent, or at minimum reduce, racking.
4) Reduced component count
and 5) Possibly better accuracy when compared to a single central motor & belt system?

1) preference really but open bed is generally less accurate and more time consuming to set-up accurately.
2) Yes
3) yes
4) Erm not really when you take account of electrical side, wiring etc.
5) Absolutely NOT and the reality in use is that they can possibly be less accurate due to any missed steps in one or both motors slipping by un-noticed until the error shows it's self in the work, which is far too late for me.!!

Disadvantages I see are:

1) Increased cost, but this should be a one-off.
2) Synchronisation/Alignment of the screws at the assembly stage, again this should be a one-off.
3) Machine ties itself in knots if one motor looses steps
or 3.5/4) Worst case scenario machine goes completely to peices should one motor stall or stop all together.

1) The cost difference isn't so much after buying belts and pulleys.
2) Big disadvantage and you'll be doing it all the time (every time you use the machine if you want to be sure of accuracy) either by butting up to adjusted hard stops or using home switch's to sync.
3,3.5/4) Massive potential for serious damage when it happens and unless the slaved motors are run well within there limits then it will happen for sure and it will scare the living day lights out of you at best.!! . . . Best case it will tie it''s self in knots wasting a few hours setting and adjusting backup.!!. . . . Worst case with high probability if it happens at high feed rates damage will occur leaving you crying and shacking in the corner. .

Slaved motors work fine if run within very safe tolerances regards tuning. They are very very intolerant of poor design causing any binding or miss alignment and will cause problems with stalling one of the motors while other continues on it's merry way causing chaos.!!

Belts completely remove this problem with very very good accuracy and are basicly fit and forget.!
In over 3yrs hard use my machine hasn't lost position regards screw sync more than few 10th mill and that was just in first few weeks due to pulley slippage and settling down.
In normal cutting use it has never missed a single step or lost positional accuracy. Again unless I cock up with programming and embed the cutter in something hard or try to run the gantry of the end it never stalls motor.!!. . . and if It does then no big deal just re-home the machine and I'm back on track with no damage to machine or hours straightening the machine out.!

The only down side to belts IMO is the look and the extra work designing into machine.? both more than worth the sacrifice or effort IME.

[BikerAfloat]

Thank you both Jonathan & Jazz,

Subject: Twin Motors vs Single Motor & belts

Thank you both for helping me clarify this issue. In the light of your comments and a PM from AudioAndy who also recommended twin screws and belts. Since twin motors seem to be a riskier strategy than a single motor and belts. I'll play it safe, especially in the light of the following advantages of using the belts.

1) Flexibility of gearing,
2) Damping of drive
3) Removal of the weak spot of the couplings.
4) Fit and forget!
5) Simpler to sync leadscrews,
5) No risk of motors trying to tear machine apart - a big worry of mine!

One motor would allow me to mount the belt system at the back of the machine in what is effectively a "dead" space, I could then use the reduction of sticky out bits to gain a 100mm on X Axis travel - Win! Redesigning was inevitable so no worries there, measure twice cut once, as many a wise craftsman will advise!

One less motor & driver for the entire machine = effectively a cheaper 4th Axis when I get that far! Mmmm, fuzzy logic, but you catch my drift!

Q: This leads me to another question of course, is the 1x3.1Nm motor enough, or is it worth going to something more grunty?

Subject: Trapezoidal vs Ball screw.

While my Engineering head says Ballscrews have to be the best, my goal for this design was that involved a build where as much as possible the components would be compatable with the 20mm spacing of the slots on the extrusion, the aim being to design a machine to be built with a pillar drill and odd spaced holes always seem to fall on the edges of the sections and would be difficult to drill with said drill. I am fighting with what is best for me accuracy wise and what was intended as a design for other people with equally limited workshops... Okay writing this has allowed the grey cells to work in the background.

Since the support bearings for the ballscrews are more pricey than the solution I have drawn and at odds with the 20mm spacing intention, I'm thinking of a compromise. I am using 12mm ID Axial load bearings and a simple clamp system to fix them to the uprights.

Q: If I go with 12mm Ballscrews is there anything stopping me using this method of attachment? I gain significantly in accuracy and reduction in friction, while keeping cost as close to the Trapezoidal idea as I can AND still keep my 20mm spacing.
I will of course need to drop the ID at the motor end to allow the threading of the clamping nut, I need to do some more research now for the machining of the ends, I'm guessing it will be M10 and then hoping 10mm ID bearings are available!

Subject: Plating Gantry - No brainer

Onto Jazz's points...

Subject: Advice

My strategy for advice is now more concrete!

When it comes to advice. Evaluate it. Valididate it in the light of your own understanding. If the results prove to be incorrect, the responsibility is your own and not that of the advisor!

Subject: Heiz Machines

I have been re-evaluating these in the light of my readings here and yes for what you get they are expensive! Guess I was blinded a bit (lot?) as I thought I was getting one for 400 notes!

Subject: Pros & Cons

Advantages

1) Agreed, much more of a preference than an advantage, but as I really want to be able to engrave motorcycle casings it is going to have to be something I live with! I can minimise the compromise by measuring up the biggest clutch cover I can find and fixing permant braces in the bed upto that point, I can the add removable bracing to allow the access.
2) See question about motor above.
3) Twin leadscrews and belt drive it is!
4) Agreed, one less driver and wiring is a plus!
5) Ageed, two motors is too risky for me in the light of the above and errors in the work, screwing up a casing is a real worry and reduced risk is always worth it.

Disadvantages

1) I'm committed to belts and pulleys, the cost is now the cost
2) I can see that to be true, problem removed!
3-5) Problem and fear removed by belts!

As I mention above a redesign was inevitable and I can now see so many pluses for belt drives on all Axies (What is the plural of Axis? It's bugging the hell out of me!)

Yes belts aren't very pretty, but then it is a just a machine after all. It is more important the work comes out aesthetically pleasing and not the tool that made it!

Off to play about with design in the light of these changes, might even work a way of reducing the Gantry height an beef up the Z Axis, but since I was running low on ideas for this I will happyly take advice on these areas to!

Cheers both,

Geoff.

[Robin Hewitt]

If you use 2 pulleys you need to check available belt lengths. If you use 3 pulleys to get belt adjustment, one of them will only use 1/4 of it's teeth, or less, so you will need bigger. HPC gears do a good selection of belts at reasonable prices (unlike their pulleys).

Make sure you get aluminium pulleys, the iron ones are cheaper but the weight will slow you down.