How Do I Fix Table Droop?

[Gs.spencer]

After a week or so with my new Shark Pro Plus HD, I've messed up a few jobs because the blue table top wasn't totally flat. My Shark system out of the box was easy to set up, but the 12 attach bolts (6 each side) for the gantry took nearly an act of God to insert because the tolerance of the pre-drilled holes was something a little shy of perfect. The bottom line however (that took me over a week to figure out), was that the Z axis elevation between the extreme left side of the table versus the far right side of the table was off about 0.050". This I finally figured out is what was causing my engravings to be deeper on the left side of my projects as compared to the right.

After reaming out the 12 gantry attach point holes to make them larger and induce some slop, I was able to raise the right side of the gantry to even up the Z axis so everything matched elevation wise on the blue tabletop. Next to my amazement, I now discovered I have a .015 gap in the center of my table even though the left and right sides are now aligned at the exact same Z elevation. After aligning a straigt edge across the X axis I can clearly see the tabletop has a trough in the middle. Even though the naked eye looks as though everything is flat and even, it really isn't.

My problem initially began with a slant error of 0.050" and further compounded with an additional droop of 0.015" if I aligned a 12" wide board from the far left side of my table. Total induced slant error (across 24") now of nearly 0.080" (that's like 1/12th of an inch).

This has taught me one valueable lesson: Check and compare your Z axis elevation between the left and right side of the table.

Does anybody have any recomendations on how to raise the center of my table 0.015" in the center?

Or, is this just something I'm going to have to pay attention to when clamping my boards where I must always check left and right Z elevations and shim my projects as required before running my G code?

At least my table is even (except for the middle) and if I always perfectly center my projects across the X axis I should be fine.

Greg
www.wood-gsco.com

[drueth]

Greg

I found my top also had a wave in it. What I did was remove the top and found two plastic rails support the top, with only two bolts holding them to the frame. Why this is not something a little stronger is a guestion to me.
What I did was verify my Shark was level on the table it was on. Then I leveled the two rails that hold the top. Drilled two holes about 2.5" from the center of the rails to add more support to the rails. The holes that support the rails did have some play in them so. Put the top back. Now I put a pice of .25 dowel in the router and rounded off the end. With the dowel in the router I move the router from side to side and end to end and verified the table was good to the router. When I had all the measurements from the tip of the dowel to the top. I raised or lowered the top to get it all even. This did take a little bit of playing to get it right but it will fix the problem of the cut being deeper on one side.

Good luck.

Dave Rueth
Shark Pro Plus HD

[rungemach]

Hello Greg

There are some very useful tips in a number of threads on this forum regarding the table flatness issues.
You are certainly not alone in having a table that is not flat or square to the machine.

It seems that the plastic bars used to hold the table and the number of holes and fasteners used leave something to be desired.

In my case I used 1.5 x 1.5 x 1/8th inch thick aluminum angle to hold the extrusions to the machine and 1/2 inch rectangular aluminum bar to make the center bars that hold the sections together at two places down the table. if you have strong flat bars holding the table sections at the front and back, you can have oversize holes that will hold the angle to the machine, and then have adjustability. With the 4 corners adjusted for even z height you can shim up the center extrusions if you like, or mount a spoil board and plane it flat with the router. It is good to have the table adjustable , in case you hit a clamp or an end stop hard and shift things out of square.

Just search for table flatness in the search box, there are several threads like the one below that may be helpful.

http://www.cncsharktalk.com/viewtopic.p ... ness#p4364

Hope this helps

Bob

[Loudspeakerboy]

I replaced my top with 80/20 1030. It's not cheap, but it solved the problem. You can buy the stock on ebay..

Best, M

[Gs.spencer]

Does anybody know of a way to contact the Shark manufacturer and get them to give us all 2 replacement support beams made from sturdy metal stock? I think for spending $4K for a machine that is delivered with a flawed plastic support system is unacceptable. It sure wouldn't take or cost much to get a pre-drilled piece of stock to fix this problem.

I read the other threads about table flatness that was also quite interesting. I noticed how one guy complained of sagging as the router passed thru the center of the X axis giving the false effect of a high crown in the middle of the table (I have just the opposite with a low trough in the middle). Now I wonder if my gantry is sagging in the middle due to the weight of the router? However, that's why I bought the Pro HD because I read that model had less of a sagging problem.

Either way, I haven't noticed gantry sag but will at least check that to see if it's causing any other problems.

Thanks for the replies, at least I know I'm not alone...

Greg

[jeb2cav]

Sure - you can call the NWA support phone number listed on their web site and in the manual.

Tim Owens has also posted his email, and as a member of the forum, you could also PM him.

[Gs.spencer]

Now that I've applied some forensic analysis to this problem I've figured out why this is happening.

The main problem is from the use of the base material being constructed from acrylic (plastic, plexiglass, or whatever you want to call it). The plastic is not rigid and will bend if struck by a soft mallet. It will also bend (or collapse) when too much pressure is applied when tightening the under table attach bolts. During assembly, any torque beyond what is required to prevent table wobble will actually start collapsing the main support vertical plexiglass sides. As in my case, the 4 inner bolts were tightened too much causing a low spot in the center of the table. Another CNC user reported his table was actually high in the middle, this wasn't caused by gantry droop but rather from the outboard table attach bolts being over torqued.

For a simple solution I ran my V groove bit down (the Z axis) and located the highest point across the plain (I inserted a small block of 3/4" oak between the bit and table) so as not to run by bit into the table. I made sure the bit was lowered until it began to lightly scratch the oak board. I then moved my router across the X axis and stopped at each of the other lower 9 table end attach points. If I discovered a table section with a gap, I loosened the t-bolt until the table was loose enough to fully insert a metal shim until the table rose enough to lightly touch the router bit, then I tightened down the t-bolt and pressed on to the next table section.

I didn't have much for shims, so I ended up using either a straight edge razor blade or a spacer shim from my stackable dado tablesaw blades. The shims I used ranged in thickness between .005" and .015" (the razor blade was exactly .010" thick).

I was now able to shore up my table top where my worse gap now in only one point is .008" off where all other 9 attach points are all withing .003" of each other. This is much better than my original problem where the difference between the highest and lowest points of my table were of by .065".

So now some of those finer carvings I was doing at 0.1" depth look totally unform across my entire project.

Another tip before carving: since my table tolerence is now within .010", I check the X axis of my projects before carving and if need be, I insert a razor blade under one end to shore it up and match the Z elevation of the other side.

Greg

[Gs.spencer]

My final solution was to replace the 2 half-inch plastic table support pieces with 3/4-inch square aluminum stock 1/16-inch thick. My entire table is now virtually flat across the entire surface. You have to pay attention before drilling the 5/16" holes into the aluminum stock. These were the steps I followed to replace the 2 table supports:

Note: Before performing step 9, I always use a piece of scrap oak to check my Z axis and step it down 0.001" at a time until my V bit lightly scratches the surface. This way I don't have to worry about driving my bit into the table or an actual project.

1. Remove the 4 bolts that secure the table to the base and remove the entire table as one assembly and set upside down to reveal the attach bolts.
2. Cut the aluminum stock to match the width of the table and set along side the existing plastic support.
3. Mark the new drill holes using the T-track groves as a reference and not the holes already cut in the plastic bar.
4. Disassemble the 10 T-track pieces and place only 2 tracks across the middle portion of the CNC base.
5. Loosely bolt each aluminum bar to the bottom of the tracks.
6. Attach only the remaining 4 of 8 tracks to the bar so 2 missing tracks are left on both the left and right side.
7. True the table and tighten the bolts on the 6 tracks.
8. Using 2 clamps, raise the aluminum bar (only on one end) approx 1/16" above the plexiglass base and tighten.
9. Run the router down the Z axis until it touches the far left surface and zero the control panel Z axis.
10. Run the router down the X axis to the other side and check the difference in Z elevation.
11. Loosen the clamp and adjust table height until the V bit lightly scratches the surface.
12. Repeat steps 9 and 10 until the Z axis is identical and then tightly clamp the aluminum bar to the plastic CNC base.
13. Using the predrilled holes in the CNC base, drill horizontilly thru the aluminum bar and attach the 2 bolts.
14. Recheck Z elevation across the entire X axis and adjust table end height until the Z axis is the same.
15. Repeat steps 8 thru 14 for the other aluminum bar ensuring the Z axis from the 1st bar is the same as the 2nd.
16. Install remaining T-tracks.

These steps made my table perfectly flat verified by a straight edge and my Z axis almost perfect across the entire surface. There is however slight droop of the router/gantry when it's in the center of the X axis. My gantry sags about 0.005" which is a lot better than the original 0.065" I was getting before with a crooked table.

Greg
www.wood-gsco.com

[baby15]

put a piece of bowling alley hard maple on the top. drilled holes an used 1/1/2 t bolts to lock it to the bed. now have no holes in bed. still got to surface plane it an put the t bolt lines in every 4 inches. I used the aluminum bar to pre drill all the holes but might not use them all. will put the t slots between the holes

t bolted to bed

[Sallye]

Has anyone attempted to create a software "shim" in aspire to correct for the uneveness in the table? I like the aluminum angle fix, and I'll do that for sure, but as a quick fix, I am going to use a 3d "topo" model to mirror the uneveness of my tapletop in Aspire, then just "project toolpath onto 3D model". It's a down & dirty fix, forcing the software to automatically compensate.