[gary_808]

We use granite/eg bases on our grinding machines... from 1 tonne to 40 tonne.

We don't use any steel re enforcement, it causes issues, Same as it does with concrete.

Concrete buildings would last 10times longer with out rebar, (assuming no natural disaster etc hit them)

The rebar corrodes even when incased in another material, grows in size and causes cracks and weak spots.

[G59]

The rebar corrodes even when incased in another material, grows in size and causes cracks and weak spots.

Different story when in EG mix. First the Epoxy ensures no oxygen. It takes oxygen to OXide... The EG grows pretty much at same rate as steel. None of us will be around in 70 years and by that time all these machines will no longer be needed so I wouldn't worry too much about developing cracks..

[LeeWay]

I have seen epoxy coated rebar in various sizes too. They used it in an elevator shaft at a Hospital I helped remodel back in the 90's. They had to use special concrete too because the bottom of the shaft was below the water table. That is where special engineering comes in. I have no doubt it was very expensive.

[RotarySMP]

Thanks for explaining your insert plan. Not sure what benefit you are expecting verse the way the industry does it, but good luck with it. I like your air cylinder plan.

Mark

[RCaffin]

Balancing out the weight of the Z axis is a very good idea, but using an air cylinder seems way too complex. Just get some decent air springs as used on car bonnets etc. They do work: I have one on mine.

Cheers
Roger

[sco_27]

Balancing out the weight of the Z axis is a very good idea, but using an air cylinder seems way too complex. Just get some decent air springs as used on car bonnets etc. They do work: I have one on mine.

Cheers
Roger

Air springs tend to wear out, that is why pros (commercial machines) use air cylinders. The system of course needs to be filled with Nitrogen, because of the thermal expansion of air.

Additionally, air springs used in car bonnets have "brake" inside, which is totally unacceptable for CNC machine.

Regards
Lubos

[John1987]

Any progress?

[epineh]

Very interesting build, though I have to admit I skipped through to see the first castings, now I have to go back through and read everything to get up to speed.

Cheers.

Russell.

[jnack95]

Any updates? This thread is an inspiration. One other option for counterbalancing the z-axis that I don't think anybody has mentioned is the use of simple constant force springs (aka clock springs). I have used several of these on my large CNC router and they work great. The benefit is that there are no seals, pneumatics to contend with, low mass so they add little to nothing to the inertia the motor must overcome. The problem is the biggest one I could find is ~33lbs (Century Spring). I believe you said you needed to offset nearly 90kg? that would require quite an array of springs. Anyway, looking forward to seeing the next update.

[JimFritzMI]

Has this build gone silent. I don't see that there have been any posts at all from petruscad in over 6 months?

[shadowjack]

Is there need to go to such length ensuring squareness of axis on CNC machine? Can't it be compensated in software, leaving only task to buld bind-free and slop-free linear motion?

[Chazaxl]

Is there need to go to such length ensuring squareness of axis on CNC machine? Can't it be compensated in software, leaving only task to buld bind-free and slop-free linear motion?

No, you can make up for some elements such as play in axial (along the ballscrew) with software but if the machine is not square, you will always have issues.

[shadowjack]

... but if the machine is not square, you will always have issues.

Can you elaborate, what kind of issues?
If machine is out of square in XY plane for example, when you command movement in X direction software can add small movement in Y direction and vice versa.

[Chazaxl]

Can you elaborate, what kind of issues?
If machine is out of square in XY plane for example, when you command movement in X direction software can add small movement in Y direction and vice versa.

Suppose it could be done but I dont see the value of taking the risk.

As it is, good machines struggle to do 'perfect' circles. This will add another dimension of issues and tuning. I think its difficult enough to make a good machine work well and reliably if it is square (think of high speed machining), I cant see the value in taking something not square and then somehow mathematically trying to fix it.

Plus, not all errors are linear. If the X 'snakes' a few micron left and right, how would you compensate? You would need a massive calculation to know, your time is better spent just getting the mill square in the first place. Within reason, its not that difficult if you have the correct tools.

My own mill will probably have 0.05mm error only as I have a limitation on what tools / ability I have to do at home. But for me, that's fine.

[ebrewste]

Mostly, yes. But, most machines aren't great at moving near zero velocity, due to friction and backlash. So, lets say you are moving back and forth in the Y direction. The X will have to make very small moves with changing velocity direction, causing motion problems. These will be very small problems, but will likely show up as strange surface finish issues. If tiny surface finish issues are ok, then compensation will likely work for you.

[miqtidar]

nice

[nateman_doo]

nagging? its been a year, and I asked once. How is that nagging?

[nateman_doo]

see? and people thought I nagged :P
going on 2 years now.

[rocketflier]

If you left click on the person's name, then select 'view profile', you can see their last activity date. June of 2017 in this case.

[skrubol]

He may have gotten the wrong type of epoxy. Some epoxies have more shrinkage than others. I think I remember reading somewhere on the huge epoxy granite thread on here that VOC's are the thing to look for. Any VOC's will evaporate and cause shrinkage.