How to Improve Accuracy?

[johnny]

Does anyone know if there is a way to tighten up the .015 cutting accuracy that is stated in the SB specs? I ordered a PRT 120x60 w/steel table, is the table a factor or is it in the drive?

[bjwat@comcast.net]

The accuracy is limited to the drive. In order to get a higher degree of accuracy, you would need to spend thousands of dollars for precision linear slides and servos. Just as a side note, most materials fluctuate greater than .015...Which means that even if you did improve the accuracy of the drive, you would still have dimensional fluctuations due to temperature and humidity. Some cutting bit diameters can fluctuate that much too...especially as they begin to wear and some right out of the box.

If you are planning on doing fine jewelry molds or cutting aerospace spec jobs, you aren't going to do it on a PRT.

[bill.young]

You might be able to change the pinion gears to increase accuracy, but you'll sacrifice speed.

Bill

[john]

Good point about the material fluctuation. I'm just a little paranoid. But this was the information I was looking for - no simple fix. Thanks gents.

[jimmapson@netscape.net]

Not so quick. You CAN improve the accuracy of the PRT series machines significantly. Whether or not it is 'a simple fix', depends on how far you want to go in improvement. Several of the items below are rather simple to accomplish and some can be ordered from Shopbot at additional cost. If this is a hobby machine, and/or you are making signs or cabintry, be aware that +/-.015" may be much better than you can accomplish with any normal woodshop machines. The Shopbot will also be extremely repeatable, assuming proper holding of the material to the table and proper cutting speeds etc. But if you are curious, here's some things we implemented at our facility:

1) Replace the router motor with a real spindle, ie. Colombo 5hp. This virtually eliminates bit runout. Also increases maximum shank size to 5/8.

2) Use minimum 1/2 shank bits. This reduces bit flex significantly

3) Use a Phenolic 1" thick table instead of wood. This eliminates warpage/size change due to humidity variations.

4) Add 2 additional legs in the center if you have an 8 foot or longer table. This reduce xaxis sag in the side rail supports.

5) Add an additional spacer and bolt between each of the furnished positions along both xaxis guide rails. Currently they are maybe 30 inches center to center, which allows a sag as the trolley rides down the rails. Adding the additional support between the factory positions virtually eliminates this sagging.

5a) Once the Yaxis rails are aligned parallel, drill and ream a hole in all four corners where the rails attach to the xaxis bearing holders. Install slip fit shoulder bolts sized so that the shoulder maintains alignment between the sandwiched angle irons, with a locknut. This allows you to easily adjust squareness without having the entire assembly flopping around.

6) Add 1/4" aluminum triangulation plates at each leg to crossmember junction. This reduces racking in the yaxis direction of the table.

7) Order machine with two yaxis stepper motors. This does two things. First, it adds pressure to both sides of the carriage, which holds all four guide bearings down into contact with the rails at all times. Second, the carriage is now driven symmetrically about the center of cutting, eliminating racking of the yaxis movement (with twice the horsepower - like the xaxis).

8) Lastly, and more difficult, replace the entire angle iron zaxis trolley with a 1 inch plate aluminum trolley, with 1" aluminmum bearing support plates screwed and doweled to the trolley main plate. Machined by a proper machine shop, this will provide a stable platform that maintains bearing alignment so that all four bearings always track straight. It also provides a rigid platform upon which the zaxis may be mounted (via 1 inch plate aluminum pillow blocks) and squared.

Without the Aluminum trolley, you can improve the accuracy to approx +/- .0075", or half the factory specs. Once we got to this point, we built and installed the plate aluminum trolley, and were able to half again the error, to approx. +/- .0035". Drilling a 24" by 24" grid (1 inch centers) of spot drills (using a centerdrill in stable phenolic), we measured optically the positions to within +/- .0025" maximum error. (Our measuring machine is calibrated to NTSB standards every 6 months, to +/- .0002" )

Note that I am describing XY 2 dimensional accuracy here. The yaxis carriage on our machine has not be modifyied yet. It is as delivered, angle iron with electrical uchannel supports. The increased weight of the Colombo auto tool change spindle and the extra stepper motor and the aluminum trolley increase the weight and resultant sag as the trolley rides in the yaxis across the rails. At the center, we measure about .010" deflection in the Zaxis. This may be a concern if you are not cutting planar (flexible) materials like plywood that will be held under vacuum against the table top. Because the top is machined true after mounted in position, it will also exhibit the same sag in it's surface. Then, flexible materials will conform to the table top, and will machine 'true'. This point is expressed elsewhere in other threads.

Also important is your daily squaring technique, which is also discussed in previous threads.

Good luck with your PRT!

[gerald_d]

Jim, if you are still considering an upgrade of your y-rails, here is a proposal:


4713

The u-strutted rails mostly sag because they twist due to the V-roller pressing down off the centre of the strut. (Ever twist a length of u-strut by hand and realise that it has close to zero torsional strength?)

Do not try to make plug welds with a stick welder unless the hole diam is 4x the plate thickness - and this would be far too big for this application. That is why the MIG (GMAW) welding is proposed. (Probably 250 amps 1.0 mm wire) Test this on some scrap first.