Crown in Table?

[wardsa]

I posted about this awhile ago but never got it resolved. I can surface my spoil board, check it with a straight edge or string and it's flat. When I put a sheet on it to cut and it cuts .050 deeper in the center of the table, as if the table is crowned or the gantry is sagging??? Even when I put a sheet on it and turn on my vac system and check it, it shows that it should be flat. It's a perfect or gradual deeper transition to the center of the table. I've had this problem forever and can't figure it out. Please , HELP!! Thanks,Terry

[Brady Watson]

it cuts .050 deeper in the center of the table, as if the table is crowned or the gantry is sagging???

Bingo.

I'm going to take a wild guess & say that you have a PRT with aluminum extrusion & steel gantry (silver & blue).

The easiest way to remedy this would be to get some 1/4" B7 specification threaded steel rod and run it thru the 2 Y extrusions. Then nut off both sides and 'true' them like a wheel spoke to get the sag out. This will function as a torsion member and fix your issue. Use a dial indicator at the center of the table as a reference. Move to Y1 and back to center to help gauge how taut the rods need to be.

McMaster.com part number 98957A733 or 98957A029 should do the trick. Remove one of the bolts on the gantry to verify size before ordering. You only need one rod per side of the gantry, 6 or 7 feet long each depending if it is a 4' or 5' tool...probably 5'

-B

[wardsa]

I understand what you are saying, but I'm not sure it is sagging. After surfacing my table, I left my surface bit in, lowered it to -.005, turned my router on manually and with the key pad ran all over the table. Their was little difference in marks. If their was a sag, wouldn't be there all the time? With a straight edge I checked the gantry at 0,0 ,then I checked it at X48,Y24 (48X96 machine) still nothing noticeable. I think the rods are a great idea to stiffen up the gantry, but I have no idea what to "0" it to? Right now everything "0's" fine but I know that as soon as I put a sheet on there it will cut deeper in the center. This doesn't bother me for sheet goods but I'm doing more signs and depending on where I "0" it will V-carve deeper towards the center.
Terry

[gerryv]

A couple of guesses: Maybe a combination of clamping along the edges only and the type of bit you're using pulling up on the sheet? Maybe your gantry only dips when under load?

[Brady Watson]

Have you closely observed the 8 v-roller bearings on the Y car to see if they are riding correctly? They all have the capability to raise the tool if they are not riding exactly center on both top Y rails and side Y rails.

I'm assuming that the spoilboard was adequately slathered with wood glue, and properly clamped from the inside out to the edges. I'm also assuming that your Z-axis pinion lash, as well as t-rail, are free of slop - For the heck of it, grab the bottom of the router, with the control box on, but router OFF(!), and move it all around, up, down, side to side, every possible direction, listening & feeling for ticking, slop & backlash. Move the Z up and down with keypad to areas that appear worn on the t-rail - wiggle it all again.

-B

[harold_weber]

Thought of this while reading these posts: Are you using a spiral bit? A spiral bit will result in some force in the Z direction when cutting. The Z force from a spiral bit will result in more Z deflection as you move towards the center of the center of the table in the Y direction.

If you are using a spiral bit, try using a straight bit and see if you get the same result. I'd like to hear what your results are.

[ken_rychlik]

Here is my guess, but is based on Brady's thoughts.

With a big bit you are pushing up on the gantry while surfacing.

With a smaller bit when you are cutting your sheets, you are not pushing up so much.

It is flexing the gantry up when surfacing to some extent and possible sagging a little while cutting.

It can't be slop in the z, because it would be the same all over the table.

Assuming the spoil board is glued down and not moving, it has to be the gantry flex.

Kenneth

[wardsa]

Ok, it seems as though everyone is leaning towards the gantry. Brady, so do I totally remove the 5/16 bolts holding the gantry together and then just replace with thur-rod? 4 pc's total, 2 each side? I'm still not convinced but I'll try anything at this point. You guys are the experts! Thanks, I'll post how it works out.
Terry

[harold_weber]

What Kenneth said above makes sense --- some surfacing bits (well, mine anyway) have a bit of negative rake angle, and this would put an upward force on the gantry when surfacing. Then if you use an upshear spiral bit for cutting parts, you would have a downward force while cutting.

[ed_lang]

Here is a quick test to show if the gantry is in fact flexing.

Place a heavy chunk of metal in the center of the table. I used channel iron. Mount a magnetic base dial indicator on the metal and zero it to the bottom of the gantry, in the center. Do this with the Y car to one side. Now move the Y car to the other side and watch the dial indicator. Is the gantry flexing as the weight of the Y car and router/spindle is moved from one end of travel to the other? You can also push down on the gantry a little and watch the dial indicator.

Another good test to run would be to set up as explained above and turn on your vacuum pump. Watch the Trupan spoilboard change thickness over the next half hour to hour. I expect you will get a lot of movement in the first 1/4 hour and the movement will slow after that..


The tests above test for two different things. Either one or both together will cause your cut depth to change.

If you find the gantry is flexing, fix that first.

You should verify that the machine is adjusted and in proper working order before running these tests.

The PRT is a good machine and will provide you with years of great service. I think ShopBot makes the best kit router package out there.

[Brady Watson]

so do I totally remove the 5/16 bolts holding the gantry together and then just replace with thur-rod? 4 pc's total, 2 each side?
Terry

Yes. Remove the 5/16 bolts and run the 1/4" thru it.

The only way you are going to chase down exactly where the deviation in Z is occurring is to get a bit more scientific than a woodworker's straightedge or a piece of string. A dial indicator with holder and mag base, which by the way is dirt cheap, is the simplest and most reliable way to see what is going on.

-B

[curtiss]

Could you make a jig out of a small camera tripod and some threaded rod that would just clear the bottom of the y car in a static mode

and then run the file (router on) and see if the bottom of the y car shifts (down) and bumps the top of the rod ???

That would tell if you have some movement or flex there

[Gary Campbell]

Terry...
Hopefully you are taking Brady's advice on the stiffening. He is beyond a doubt the goto guy for the PRT. Notice how he guessed what machine you had by reading only the problem?

Should you need convincing that the gantry flexes down under load, cut 2 identical shapes near the Y center of the table using somewhat aggressive speeds. Do the first with a down spiral bit and the second with an upspiral using a .010 cut thru. When done mic the depth of the slots in the spoilboard.

[wardsa]

OK I was wrong! I does look like it was a sag in the gantry. I tightened it up with the threaded rod like Brady suggested and he was right on. Thank you so much for sending me in the right direction Brady!!! I'm still getting alittle difflection but it's 99% better. I would suggest anyone with a PRT with the aluminum rails to do this if you haven't. The machine just seems so much more solid. Thank again to everyone.

[Brady Watson]

No problem. Glad you got it sorted out.

-B

[curtiss]

Which of the bolts did you replace with the rod on the gantry ?
The top or bottom or does it matter ??


Is it better to move the y car / router to the side overnight rather than park it in the middle ?

[Brady Watson]

I usually do the tops - one side at a time.

No, you don't have to alter where you park the Z. It's deflection we are dealing with, not permanent sag.

-B