[yezard]

Hi guys, I havent done alot with the machine since I have had my year 13 jan exams but have managed to fix the X and Y axis to the main frame. Will upload some pics in the next couple of days.
Im planning on using a 3mm mild steel plate with numerous holes drilled and tapped with 5mm threads for my milling bed since this seems like the cheapest way to go. however I was wondering whether there is another way to mount my work pieces to the machine. I need the bed to allow the coolant to drain into the moving X axis. I know the traditional way to go would be to use a T table design but its seems so expensive.
Anyone got any other ideas or any places where i can find cheap beds?
Thanks

[handlewanker]

Just my 2 cents worth....the 5mm holes will be a devil to keep unblocked from swarf, even if'n you plug all the unused ones, and tapping them will give you quite a bit of experience in either an aching wrist joint or just plain drilling and tapping.

5mm tapped holes in 3mm thick plate is not very practical for holding as you will have to drill for 75% thread engagement and this in 3mm plate will not give you many threads to resist stripping.

Just my opinion, but I'd go for the built up T slots as in a piece of thick steel plate and 2 lots of strips screwed or riveted on top of each other to form T slots.......you only need to face the top when the plates are all assembled and riveted.

The other way would be to get or cast a piece of alluminium thick enough to resist bending and flexing to form the X axis bed and either cast in or mill slots in the surface....then just bolt strips of steel to the top of the plate to form the top face and Tee slots and have them face milled if'n you can get it milled, or ground if'n you can get a surface grinding job done on the cheap.

Anything is possible if'n you've got access to equipment to do it.
Ian.

[handlewanker]

On second thoughts, having read back over the posts to get an idea of the scale, to cast the table in one piece of ally would probably be something like having a go at building the Titanic.

You'll probably get more joy by completely fabricating the table with a composite frame structure of square steel tube sections, diagonally cross braced, and faced with 3mm steel plate......off cuts welded together would make a table top and the welds just ground flat.

The top section can be rectangular alluminium strips faced with flat steel strips to form Tee slots, and then surface finished with a straight edge, level, angle grinder and belt sander.

I think you mentioned a surface area for the table of 1 metre by 500mm....that probably would need a 50 x 50mm-3mm wall steel tube frame and 3mm steel plate facing, top and bottom.....it's going to need to be a bit heavy to resist twisting, and it's too late when the build needs a bit more density later on.

The top and bottom 3mm plate would need drilling through with 8mm diam holes to allow through welding to attach the plate to the frame across it's entire face, and the welds ground flush aftrewards.....this would ensure a rigid top and bottom to the table without any deflection on the faces, which is vital for mounting the carriages onto.
Ian.

[yezard]

Thanks Ian for the fast reply, didn’t think about the swarf blocking the threaded holes. Ahh i was toying with the idea of building a T slot table but never thought of using multiple box sections with steel plate welded to them, seems like a good idea. Thanks . Atm I’m trying to keep money spent to a minimum so I will probably build the machine up with just a standard 3mm steel plate for the bed with 6mm or 8mm taped holes and then use any money remaining to go about building the T slot like you have said. Will get the machine to drill its own holes into the bed be a lot easier and probably a lot more precise.

As for the current state of the machine, i have bought the various linear rails, ballscrews etc and have grinded, sanded and painted the frame. The X axis seems to roll into the open bearings fairly well however still need to adjust the height and position of each of the open bearings so they slide in perfectly. The y axis seems to roll along just fine, had slight issues with the locations of the holes on the y carriage since they were slightly out and had to grind a bit of aluminium off within the y carriage to allow the bearings to slide freely but other than that it’s all good. The machine itself seems extremely sturdy there is no play in any of the axis even when there are no screws to secure some of the parts in place. The aluminium box section that comprises the gantry bridge and the supported linear rails seems to slot into the gantry wall very securely and the parts are within 1mm tolerance to the drawing files. (Which should improve when i remill some of the parts)
Atm I am milling out the ballnut holders in HPDE and will then remill them in aluminium when i get the machines up and running.

I haven’t started working on the z axis yet, the idea was to cast a thick piece of aluminium to make the z plate however I have found a 6mm aluminium plate lying around that seems to be the right size and am planning on backing this up with a 6mm steel plate. Do you think this will provide enough stiffness?
I also need to buy a small pre machined ballscrew (about 25cm) for my z axis, zapp automation has none in stock that meet my requirements and haven’t come across any anywhere else. Does anyone know of any places where i can purchase small premachined ballscrews?

Thanks everyone for all the help advice on this forum has been great!!

[yezard]

Haven’t posted anything on here for a while so if there are any followers I do apologise. The machine is essentially finished, there are a few bits here and there I need to finish up but I can now use the machine to cut and machine out parts.

I have opted to coat the underside of the milling table with zinc galvanising spray to inhibit rusting, the milling table holds about 8 litres of coolant which eventually will be connected to a pump to circulate the coolant. Have been experimenting using some left over refrigerator pumps to do the job but they seem more suited to compressing gases than moving fluids.
The first thing I tried to machine was a router holder out of a home cast aluminium block to replace the rubbish temporary wooden router holder I had in place. This didn’t go too well, I knew home cast aluminium was really gummy and horrible to machine and the fact I cast the block the day before didn’t quite help either (I believe cast aluminium is better to machine after it has been sitting around for a while or even better when age hardened) the passes were shallow (about 3mm) and the feed rate was set extremely low (can’t remember the exact figure) coolant had to constantly be pumped onto the part and yet the surface finish wasn’t to brilliant and it was SO LOUD!! So i eventually gave up and thought I would machine a foam part out instead and lost foam it. Took about 5min to machine the foam part out and a further 30 to cast it opposed to the 2 hours of failed machining!!

I still want to machine out the gantry sides from either aluminium and steel to replace the cast sides I have, I also want to replace the cast y axis so will hope to machine this out as well. The machine itself seems to machine steel BUT SLOWLY! (Not too sure on the grade of steel i was machining) My router has only 700w of power and when run through my dimmer switch to slow down the router it’s more like 300! So when I try to speed things up the router seems to seize! I can however get through about 1.5mm at about 5-10 cm per min i have yet to measure the absolute accuracy of the machine since I am still missing some bearing holders (which i will machine out) and various spacers.

Another thing I would like to possibly change is the unsupported linear bearings on the z axis because quite frankly put THEY ARE ****!! I can move the z axis about 1mm to the left and right with considerable force. Would using THK rails or HIWIN rails solve this? I haven’t come across a lot of data sheets for the THK or HIWIN so do not really know what forces they can handle. If hiwins or thks rails work is there anywhere i could purchase them cheaply?

As for replacing the gantry sides could anyone point me in the right direction to the thickness of the material I would need? I’m thinking of using a 25mm aluminium (5000 or 6000 series) or 20mm steel (was looking at 1214 dues to its ease of machining) but I hear mild steel is not too bad to machine.

[handlewanker]

Hi Yez, that's forward thinking, making a pattern in easily machined material and doing the heavy hitting by casting.....this is sililar to lost wax casting by making an easily molded intricate wax pattern and casting the final piece.

The HiWin slides will cost you a bit but I think that type having full base contact will prevent any deflection under load.

BTW, 1214 is mild steel the cold rolled type......and this will give you headaches with distortion from internal stresses rolled in, but S12L14 is leaded mild steel and comes off in small chips.

I think it was mentioned previously that the slides you presently use would deflect under load.....oh well, more work.

The refrigerator pumps are piston type, (in the ones I had) and a simple pond pump would work much better being a centrifugal type, it just depends on the flow rate, and a washing machine pump can also be used.

A washing machine pump with integral motor (240 volts A/C) needs to be down at the fluid level, but not in it, whereas the pond pump can be totally immersed, and also won't cost an arm and a leg.

You'll need a strainer to prevent chips getting stuck in the impeller housing, and a design that allows the daily strainer removal for cleaning.....no plenty cleanee, no plenty flowee...LOL.

A tank for the pump should also have a baffle to divide the incoming coolant with waterborne chips and sludge so that the sludge sinks to the bottom and the clean coolant flows over the baffle to the pump side.

If your router motor draws 700 Watts and the dimmer is rated at 300 Watts, you are going to burn it out.....the motor is acting like a dead short to the dimmer.

There is a chopper circuit diagram for an add on control used on DIY Tig welders to control the primary and this gives you at least a couple of amps at 240 volts, depending on the triac you use.

A 300 Watt dimmer is only about 1-1/4 amps max, whereas you can go to 240 volts @ 10 amps with a DIY chopper circuit.....you'll need Triacs rated at 10 amps +, or whatever your router motor needs.

If I can locate it I'll post it.
Ian.

[HorridHenry]

The machine itself seems to machine steel BUT SLOWLY! (Not too sure on the grade of steel i was machining) My router has only 700w of power and when run through my dimmer switch to slow down the router it’s more like 300! So when I try to speed things up the router seems to seize!

Man,I didn't think it was possible to machine mild steel? with a 700w wood router? I mean I can vision/hear the bearings breaking up good style in my mind,I suppose this will happen eventually if used often.

I just got a treadmill motor that would suit that machine down to the ground,loads of torque and can be slowed down a lot for machining.

[HorridHenry]

I'd like to buy this one for the treadmill motor as its 180v DC out but Im thinking it might be just to cheap and nasty for the price?

Plastic Shell Input AC 220V Output DC 180V Motor Speed Controller | eBay

[yezard]

Thanks for the replies, been doing a few calculations using beam theory to work out the deflection my gantry sides will undergo (modelling them as cantilevered beams fixed to my frame).

Using the equation of deflection at load being:

WL2/3EI

Where E is the modulus of elasticity
I is the moment of inertia
W is the lateral load
L is the length of the beam (or gantry sides)

Assuming the cutting forces when milling steel is about 75N (I know this is dependent on feedrate, depth of pass, dullness of tool, power of spindle etc) and then taking the load to equally distributed between the two sides (the two sides of my gantry are connected using 4 threaded rods which essentially clamp the sides together with the aluminium box section between them and I am assuming these rods will deform very little and undergo no plastic deformation i.e. always be under the same tension from clamping the two sides together) so force on each gantry side being 37.5N

Taking the cross sectional area of my beam to be rectangular the moment of inertia being

I=bh3 /12

Where b is my width of gantry
.h being the thickness of the gantry sides

When the equation is applied to a gantry side thickness of 20mm and made from mild steel:

37.5 x 0.43 / 3 x 210x10+9 x (0.18x0.023 /12)

Giving a deflection of 3.1746x10-5 or 0.0317mm
I also modelled the sides on solid works and got a deflection of 0.0368721mm

A 25mm aluminium gantry side gives: 4.876x10-5 or 0.04876mm (solidworks: 0.0562488mm)

A 32mm aluminium gantry side gives: 2.3251x10-5 or 0.023251mm (solidworks: 0.0269242mm)

On the 44mm cast aluminium sides I currently have I get 0.0222782mm however my sides are home cast so the young’s modulus of the material is probably a lot lower than the value used in the calculations.

So it seems the 32mm (1.25 inch) aluminium sides have the lowest amount of deflection. My initial goal was for the machine to hold a tolerance of 0.05 to 0.1mm, of course without a FEA or measuring the actually value I will not know if the machine can hold this. But from the calculations the gantry sides do not deflect too much to make this goal unreachable.
I will have to look into the prices of the metal I require and then make my final decision on the material choice and thickness

As for the dimmer that controls my router it is rated at 1000w the reason i said my cutting power is about 300w is because i believe the dimmer works by chopping my ac current (chopped up sine wave) this effectively reduces my voltage (lowers the rms voltage) and since voltage is roughly proportional to motor speed a lower motor speed which i need when milling through metals results in a lower power output.

I’m no engineer so if I have made any errors or have misunderstood anything please let me know.

[yezard]

And a 700w wood router is pretty much next to useless when it comes to steel, but im out of money so will have to keep my current router, dont get me wrong it does the job but fairly slowly and with swallow passes.

and regarding the z axis bearings.....the unsurrported bearings were soo cheap compared to the other options it was so tempting. i guess i listened to the advice on changing all the other unsupported bearings to supported and thought i could go skinny on the z axis...i can live with the deflection though it isnt to bad but will deeply consider changing them.

[fgc]

Interesting build. I'm amazed how far you come! Hope you can give updates again as you continue to refine your machine.

Cheers
Fgc

[steve123]

I love your aluminum castings, and want to try that some day!
I was worried about your z axis. I just saw your thread today, and just finished reading through.
I think you should look at doubling up your z-axis bearings. There should be one bearing above the other on each side. They should not be touching, but with some space between. I found this site to be very helpful Building a CNC Router Step 4: The Z-axis assembly See the D2 dimension. Cheers!

[yezard]

ahh thanks guys for the replies, haven’t done all that much with the machine since my last post as I have my year 13 exams coming up. I see your concern with the z axis linear rails I managed to get myself a pair of 12mm hiwin rails however these only have one bearing block per rail will see if i can purchase additional bearings for them. I know 12mm is fairly small and will limit my cutting forces but they seem sturdy enough and lie within the moments that my z axis will experience.

I plan to mill a replacement y axis and z assembly from 1 inch or thicker aluminium which should stiffen up my machine.

I have used the machine to attempt to mill out the x axis ball screw bearing in steel however i had to do it so slowly taking about 1mm passes at a time since my router is so underpowered. Another problem I faced was some of the newer and precise mill ends seem to dull and the flutes seemed to chip fairly quickly most probably because of the small deflection in the z axis. I did try at one point to go through at about 3mm per pass although it did mill fairly quickly the bit chipped and dulled fairly fast. The tolerances achieved are adequate but it should improve as things get stiffer. I couldn’t mill all the way through the 20mm bar stock since I had only a 13mm flute length so used an angle grinder to cut off the rest

As for the gantry side i looked into prices for the different materials. The 20mm steel came in at 130 pounds and the 32mm aluminium came in at 250pounds, the obvious choice is the steel however will take forever to mill out with my current setup. The aluminium would mill fairly easily however the cost is not justified. I may inquire into the price of 25mm aluminium however the deflection would be considerable more, i may bolt square box sections to the gantry side to compensate for the thickness.
I also milled out some bearing spacers that space out the x axis linear bearing from the frame, the machine was able to mill out these fast 6 took about 30min

I actually have the updated cad files and will post them up here soon. And the other various photos off the build so far.

[yezard]

oh and looking over some of my calculations in my previous post the equations didnt come out as expected and some of the numbers should be in fact be to the power off rather than multiplied

[digits]

Looking good so far! I really love your DIY castings!

How big are the bits you're planning to do in 32mm aluminium? £250 sounds like a lot to me - what do they weigh?

Also, if you just want to stiffen the Y-sides, couldn't you just bolt some angle-iron onto them?

What sort of rpm does your router run at? I would have thought it is probably 5-10x too fast for steel, and so is just cooking and so blunting your end mills. Have you tried any carbide end mills for cutting steel? I've bought a few on Ebay-UK for £10-20ish, and they seem to be better value than 3 cheapo HSS ones IMO.

Anyway, best of luck and please keep posting pics

[yezard]

Sorry for the long delays between posts, my exams and all have kicked in and have had little time to work on the machine.

The parts I want to do are the gantry sides, they weigh about 10kg each, 250 is mad so I may try recasting them or milling them out in 25mm and doing what you said and bolt some angle irons or box sections to the sides of them. As for my router it runs at 30,000rpm (no load), when run through my dimmer I can lower it to about 15,000rpm (a guess) although at this speeds the torque produced really takes a hit and the router is next to useless. The setup is not for steel or aluminium I really need to upgrade the cheapo router to a proper spindle. I was surprised though when the setup milled though steel and did a decent job of it.

I have milled out some of the parts for the new y axis. although I have to run my router quite slow and at low passes mainly because I’m using a wood carbide bit to do my ali cuts (I have ordered a HSS end mill though so hopefully that will improve things ) The router makes a racket when milling and the bearings have really taken a beating. (Had to replace them recently) the parts however have come out alright and the Hiwin bearings fit very snugly in the z axis plate.

I have posted the updated CAD images of the machine, these are the CAD files I work with, just never had the time to upload them up here. According to the model the machine should weigh in at around 100kg although it seems more like 120kg when I lift it (although I havent added all the parts to the CAD file yet)

I may upgrade my spindle so I can use the machine to cut and mill out more beefier and complex parts. Are there any good decent priced spindles out there that will do the job? I want to routinely mill out aluminium parts and occasionally steel so I assume those Chinese spindles out there that look ever so tempting will be useless right?

[wizard]

I might suggest carbide end mills if you can't get your spindle speeds down.

By the way nice looking castings. Maybe you can inspire more people to go that route. Given a little engineering it ought to be an easy path to a ridged structure. You certainly will have fewer things to rattle loose.