[mhasting2004]

Hi Sonic.

I have not noticed any problems to date and I don't expext to as the point loading is probably quite low. If you look at the pics you see that the epoxy is getting squeezed between a fairly stiff 8020 style extusion and a very stiff I beam. These both act as pritty big washers to spread the load. If I was very concerned I had planned on removing an area of the straight epoxy just around the holes and filling this with some liquid steel or similar to make a higher compressive strength disk which the bolt would go through. So far though I havent felt the need.

Cheers

Mark

[twistedfuse]

Would adding aluminum powder to the epoxy mix do much? It quite cheap and readily available from good arts suppliers? Might effect the viscosity but letting it free flow with the help of a vibratory motor of some type might help it.

Daniel
P.S Keep up the good work.

[mhasting2004]

Hi Daniel

I've read somewhere that they use Al powder in come epoxy concrete / grout mixes as it oxidises and grows inn volume to compensate for shrink back as the epoxy sets. No personaly enperience with it though.

Still plodding along.

[mhasting2004]

Bit of an update.

Since I lucked out and now have two 2.5 m long ball screws I've decided to drop the belt drive for now and do a driven nut design instead.

Here she is:

[eloid]

Bit of an update.

Since I lucked out and now have two 2.5 m long ball screws I've decided to drop the belt drive for now and do a driven nut design instead.

Here she is:

quick question what cad package are you using inventor soild edge etc

[eloid]

Mark, I was hoping to see how the belt drives worked for you, as I'm currently designing a belt drive similar to yours. Smaller machine, but I'm hoping for 800 ipm.

what cad package are you used for your design? autodesk inventor?

looks great...

[mhasting2004]

I use Solidworks.

Cheers

[ger21]

Mine was Solidworks too.

[mhasting2004]

Hi guys

Well this is way overdue! I've been ramping up my efforts this past month and the project is getting much closer to completion. As always the advice of "if your stuck... drill a hole" has paid off and of the squillion of holes since I started this project now only a few remain.... Yeah

Here are some phone pics of some of the progress.

I'm really happy with the spindle clamps I made up. They work a treat and hold the spindle very snugly using the opposing clamps on each unit. I posted a CAD file earlier I think.

The gantry is in bits again as I finalise some of the Y axis drive as well as getting it all square. The top extrusion has been setup as a radiator/resevoir for the spindle motor and is sealed and full of ping pong balls (to reduce the volume and get the coolant to have a disrupted flow) Although the barbed fittings are both at this end, the one that is to the right has an extension hose inside the extrusion that takes it to the far end so the flow of coolant is the entire length.... we will see if this works well or not.

The Z axis is now almost complete and the counter balance air cylinder can be seen on the from as well as the cable chains (spindle mounts in the space at the bottom). I've also put in an array of threaded holes on each side for future fixtures (laser zero/scanner, air drill etc)

The X axis is finished and I'm in the process of lining everything up which is a bit tricky as there is not alot of clearace between various components. I also have included a shot of the belt drive which drives the ball nut as its not clear on the assembled unit. I had to get some flanges made for the large sprocket as the belt was wandering too much (poor allignment of the motor versus the driven pulley mount) but this has given me the oportunity to incorporate some captive slugs to have a pendulum dampener built in.

Last but not least the all important ...."? look how messy my garage is" photo.


Cheers

Mark

[mhasting2004]

Well I've decided to make another slight change to the gantry (so have to dissasemble AGAIN!). In order to stabilise the pressure in the air cylinder that makes up the Z axis counter balance I realised that having a receiver tank to increase the volume of low presure (around 60PSI probably) in the system would make the changes in volume from the Z axis movement less significant. I am using a standard pressure regulator (from the junk I had) at the moment however a precision one would be better as they have less historysis and vent with a smaller differential to the set pressure. These go for around $200 so I will wait for now.

To create an approximately 8L storage tank/ receiver I will now seal the bottom extrusion of the gantry... hence the disassemble

The regulator will be moved from the Z axis to under the water pump on the gantry and will "T" into the extrusion.

The other aspect that will help with maintaining a constant counter balance pressure is that the air cylinder is acting thru a pulley system that will mean it only moves half the distance of the actual Z axis slide.

All theory at this point but am pretty confident it should do the job.

Cheers

Mark

[mhasting2004]

A question to anyone out there following my build....

Whats the average steps per mm (or inch if you prefer) that most builds aim for?

I have been forced into my setup due to the parts on hand (if you just joined this thread I basically salvaged 90% of my machine) which has me running my X axis at 222 counts per mm (10uSteps per step) or 9mm per motor rev.

This all means that the maximum jog speed right now is around 5000mm per minute (~200"/min). I can live with this for now but may upgrade to servos later if the beast makes any money.

So what is the average everyone else has?

Cheers

Mark

[Matty Zee]

Hey Mark,

Good to see you're making some progress! It all looks great too. Well Done.

I'm not sure there is an ideal or average steps per mm. It will vary for each person depending on their motors and the performance they want from their machine (both resolution and speed).
My machine currently is set at about 61steps/mm but this is soon to change to ~100steps/mm. Currently i can rapid way faster than i'm confortable going at >40000mm/min(~1575IPM). With the new ratio belt drive i reckon that will drop to ~25000mm/min (~984IPM) which is where i have it set at the moment. My first machine was around 393 but was painfully slow to rapid. Max out at around 4000mm/min.

Your setting of 222 will give you great resolution but you may get frustrated over time with the slow rapid movements. Depends really on how big the stuff you are cutting and how many rapid movements you have. But seeing as you salvaged much of your machine i think you've done really well. Upgrading to servos is a good upgrade path.

[mhasting2004]

Thanks Matty

Yeah I reckon painfully slow will be the order of the day to start off with. 25000mm/min sounds like a good target top speed and fits with some of the cutters I've seen on the Onsrud chip loads site.

Its just the X axis thats an issue as the others are easily geared up (presently they are geared down to match the X). The largest single stage belt gearing I could fit was a 72:40 HTD5 which brings the 5mm pitch of the ball screw up to 9mm. Not crazy about doing a multi stage belt drive so servos may be the only answer.

Once I get it up and running I may re-visit the omega belt drive idea as it still intrigues me.

Now the fun bit... getting everything aligned and running smoothly.

Cheers

PS least I know the spindle control works well

[twistedfuse]

Mark,

Looking fantastic as always, and always light years ahead of others with ideas and changes.

I think your target speed of 25,000mm/min would be great to achieve for such a sized machine. Although I personally think the rapids shouldn't be too much of an issue, a decent rate of around 12,000-15,000mm/min would suffice but your major one on the machine will probably be acceleration. When machining a 3d file, if your acceleration takes longer to ramp up and down to the speed, your machining times would be a lot longer.

Cant wait to see your machine running.

Daniel

[mhasting2004]

Thanks Daniel

I have to credit alot of of people both here and all over the net for most of the ideas. No such thing as original thoughts ... just alot of applied plagerism.

Given my lead screw pitch of 5mm I need to aim at between 3000 and 5000rpm of the nut... hmmm better recheck the specs on my angular contact bearings and get the ball nut spec some where too. So a motor speed from 1667rpm to 2778rpm using the current maximum geared up belt ratio.

Seeing as one of the jobs this machine has to do once we start making chips is a kit for some cats (the ones that like water) it had better no run like a dog! I'll keep plodding along with the gear i have for now and not get too hung up about speed for now and concentrate on accuracy and alignment.

Pretty anxious to see it moving and assembled too as its where the rubber meets the road and all the theories either work of fall in a heap!

Be a bloody big paperweight if that happens!

Tested out my air tank/gantry extrusion yesterday. I cut some 19mm MDF plugs and hammered them in each end and further in by 5mm or so. I then put copious amounts of Sika flex polyurethane sealant in this gap and covered each end with gasket paper. Reassembled to the gantry side plates and presurized to 40PSI to force the sealant into any nooks and crannies and left to set. Can find no leaks so am happy that this will work.

I was going to put sand in all the hollow bits of the frame when I first started but didn't get around to it. This may be a good thing as now I have two cavities filled with water and air ... if the frame had "earth" that would have left fire as the last element to go in the remaining void.... the electrical cabinet :O

More updates should follow next week.

Cheers

Mark

[mhasting2004]

I've now put the gantry back together and leveled out the top rail using an engineers level. The lower one will be adjusted when the z axis is mounted to ensure parallel and distance.

The rear shot shows the spindle coolant pump, air regulator and VFD all mounted on the control side of the gantry. Water and glycol in the top beam and compressed air in the bottom. I mis-calculated the volume of the coolant resevoir however (must have been 8l without the pingpong balls .. opps. as it filled to the brim with only 2.5l in the system. I'm still ok with this as the radiator effect of the beam will more than compensate (I hope) given the limited testing and low increase in temperatures measured (it will get hotter under load I know). I can always remove them latter but want to avoid dis- assembly again if I can avoid it.

The rear checker plate panel has doublesided foam tape between it and the extrusions to avoid any possible "rattles" and it also has a strip of 3m dampening tape on the inside (shiny bit between the extrusions in 2nd photo).

Cables are routed already in the nylatube chain and ready for the z-axis mounting.

Been looking at these 250w servos and drivers.. anyone have experience with them?

http://www.dmm-tech.com/Pricing.html

I need to do the calcs on what inertia my gantry has and what I need power wise but think these are in the ball park

Cheers

Mark

[mhasting2004]

I think I spoke too soon on those servos. They actually look kind of small. The steppers I have now are NEMA 34, 550 oz-in old stle Superior electric motors (round body not square). Remember these were not selected but just what I happen to have.

For the sake of ease I would prefer the same Nema 34 chassis in a servo ... as long as all the other specs fit and a max rpm of 3000.

At the moment the max feedrate I can get comfortably without losing to much torque is 3000mm/min. There is heaps of torque given the slow pitch of the ball screw (5mm) and I am driving the gantry from just one side at the moment ... look out when I get both motors running!

So at the moment I have built a Landrover Defender... heaps of grunt but a bit slow... and what i hope to get to is more like my other car.... Mazda SP23.. ie Zoom Zoom


Anyway time for bed. I'll do all that maths this weekend to figure out what spec motor I need.

Cheers

Mark

[epineh]

Hi Mark, nice work so far!

While I haven't had first hand experience with the servo's you mention, here is my 2c...and please this is only my opinion I don't claim to be any kind of servo "guru"

I have a mate that does large style retrofits and new builds for a living and he reckons that a general rule of thumb is that if you go by oz-in ratings alone (don't use the wattage too much for ratings, the torque is what you need to know) then to get identical performance a servo can be a tenth the torque rating of a stepper. Now this is only a rule of thumb and with so many variables it won't work for everything...

Having said that the servo's you are looking at have a NM value of 0.8 so this is about 113 oz-in. Using the good old one tenth guide, this would equate to a 1130 oz-in stepper, roughly twice the size of your current steppers. So you should be able to double your speeds, all things being equal

The servo's on my machine are 150W, I am pretty sure I am driving them at half their rated voltage (they were second hand, datasheets are not available, long story) and the force they produce is brutal, a hand in the wrong spot on my machine would come out second best, mind you I am using threaded rod as a leadscrew so this helps a lot for torque.

I crashed my machine a few months back and the result was the Z axis assembly buckled and misaligned and the brand new carbitool 60deg cutter snapped like a toothpick, the remains left a really nasty gouge in the table before anything tripped, not one of my brighter moments...(chair)

I would hate for you to buy those servo's and not get the speed you want but in my opinion they should be up to the job, and they are certainly cheap enough, esp being AC servo's. I will keep that page bookmarked for future reference.

Cheers.

Russell.

[mhasting2004]

Is it me, or is this boy seriously lost


Following too many threads is my guess.

[mhasting2004]

That last post has been deleted... better point this out as I'm not taking a dig at Russell. Thanks for the info BTW

I'm looking at these blokes too (after all they are more local)

http://zealcnc.com/

Bit pricier though.