Tips/Help for getting Z axis 100% accurate?

[aschutsky]

Hi Guys,

After making a few dado cuts last weekend, I noticed that my Z depth cuts were off quite a bit - from .06 to as much as .125 in some places. Overall it is cutting too light/shallow. I was also not cutting all the way through my .73 thick ply with a commanded depth of cut of .76 so that raised some eyebrows right away. The only recent changes to my setup were the addition of my vac table and a new spoilboard. I'm using a freshly surfaced piece of UL MDF which is secured via foil tape to the plenum as per Brady W's suggestion. It is NOT moving around and seems to be a fairly flat (see comments below) surface overall.

Here's what I've done and observed so far:

Machine: 2004 PRT Alpha w/ Colombo spindle, all original parts and controller. Latest firmware applied and otherwise working fantastically. The unit values for X/Y/Z are all the same, standard setting of 1273.2385.

-I noticed the table was out of level a bit so I fixed that and retested with some small cuts right near the 0,0 center at the bottom left hand side of the table. It was still off a bit around .03 too shalllow IIRC. I've checked the z-zero plate and manually 0'd the Z just to make sure that is OK - and sure enough that was fine.

-I've also checked the commanded Z depth in the control software and all looks great there as well. I'm using Aspire for toolpathing, version 4.509.

-I've checked for any play in the Z car and there seems to be none.

OK - what else can I check? Are my expectations too high? Thanks in advance for all suggestions.

Andrew

[Kyle Stapleton]

Had the same problem last week
http://www.talkshopbot.com/forum/showthread.php?t=19552
This helped a lot.

[aschutsky]

Had the same problem last week
http://www.talkshopbot.com/forum/showthread.php?t=19552
This helped a lot.

Hi Kyle - thanks for sharing. What was the actual thickness of your material and where did you have it set?

[aschutsky]

Hi Kyle - thanks for sharing. What was the actual thickness of your material and where did you have it set?

Never mind - saw the earlier posts where you were setting it a bit thicker than actual material thickness. I'm sure I was doing the same thing. I had 23/32 material set to .76 thinking that would force it to cut deeper but that was not the case! Let me try that.

Any other suggestions are welcomed. :)

[Kyle Stapleton]

yup, sounds just like mine.

[gundog]

Measure your tool bit to collet face depth and make sure the bit is not moving in the collet.

You could check this also by putting a piece of blue masking tape on the surface of the material you have zeroed to and bring the bit down to .001" from hitting the surface with the router spinning. Blue masking tape is .002" to .005" thick so it should cut the tape. After you have made your cuts repeat the test. This will tell you if you are losing height then you need to figure out what is moving.

Mike

[Brady Watson]

Carefully inspect your table. It has more to do with Z inconsistencies than anything else. Always flatten it before doing any critical work. If running a vac system - comb over it well, flatten with vac on to help keep it down - although this is not 100% guaranteed to result in a completely flat bleeder board because it can bubble up in the center where you cannot see it - especially if you screw down the perimeter and have nothing in the center holding it down, besides gravity.

As long as the Z doesn't have backlash you can feel while doing your daily 'shake down' - then you're either off on your programming or the table is whacked out.

-B

[aschutsky]

Thanks for the suggestions guys. I've 'upgraded' to a Trupan bleeder that was delivered yesterday so I'll be putting that on this weekend and going over everything again.

So I'll be doing:

1) Double/triple check levelness of table and correct as needed
2) Install new Trupan bleeder board without screws using foiltape around edges to hold to plenum (this worked great before). I also have a hardwood trim around the edges of the plenum/bleeder to hold/seal them
3) Fix/correct aspire material thickness in files - measure point where material is being z-zerod with calipers

[aschutsky]

OK so I've ruled out any programming issues but still remain quite a bit short on depth. At this point I've ordered new pinions and will try that - stay tuned. After taking some more measurements I'm more accurately around .08-.11 shallow on all cuts.