[wormwood]

Okay, so most of you know to some extent the table on Patriot models before they moved the X axis to the center. I assume the change in location did address this, please correct me if not. Anyway the table twist about the z axis & along the x-y axis was really becoming pronounced when I was trying to rough in a pulley from a piece of 6" rnd. stock. So I spent the next couple of weeks realigning everything, replacing bearings, checking everything & finally noticed the table twist on a fluke when I had the Y axis cranked almost to limits. Surely, I couldn't be the only one who would overlook this, I mean whenever I'm checking alignment or repeatability the first thing I do is send everything to their home position. Well, in short, I'm stubborn & it shows in how I fixed the problem.

JT suggested I machine some extensions of the saddle where the contact the ways & increase the gib length, I told him they was a good idea I think I'll go that route. The next day, I realized I already had an extra ballscrew & nut, so all I needed was some angle bearings & misc. hardware & I could add an extra ballscrew. That was back in Oct-Nov. 2012 & it's been 10 long months coming up with some of the most creative uses for a drillpress in order to machine 5/8" plate steel with relative accuracy to fix, support dual ballscrews. In the end, I can say it can be done with enough patience, time, & a net 30 account with Grainger & MSCDirect. I wouldn't suggest doing this but if your inclined to have a ROCK SOLID table dual ballscrews will certainly provide that. Here's some links to my Dropbox account with photos of the patriot, sorry they're from the iPhone & kinda crappy quality...

Wormwood

https://www.dropbox.com/s/821kahk6oo...%2040%2056.jpg

https://www.dropbox.com/s/lpuio47o9r...%2019%2032.jpg

https://www.dropbox.com/s/wf3zi5p4uu...%20%281%29.jpg

https://www.dropbox.com/s/mvwktip5f5...%2020%2050.jpg

https://www.dropbox.com/s/u7wg1uixus...%20%281%29.jpg

https://www.dropbox.com/s/fmb5sqhbn8...%2047%2020.jpg

https://www.dropbox.com/s/tgn8vjg7ea...%2044%2014.jpg

https://www.dropbox.com/s/1i8m916359...%2014%2036.jpg

https://www.dropbox.com/s/rbmqon4zvw...%2045%2041.jpg

https://www.dropbox.com/s/egyq1pxdt7...%2036%2008.jpg

[bill.anderson]

I went a different route, I disassembled and removed the table and saddle and scraped the ways and dovetails and gibs flat to around 30 pts/square in. I now have .0002in backlash on X and .0003 on Y. Table rock is down to around .0002in. I found that the best test is to CNC mill a 2in circle without backlash compensation on and then cut a 2in circle using a fly cutter on the same center measuring the difference.

bill a.

[countryguy]

Mr. B Anderson. If you are still around, Would you be willing to reopen an possibly dialog about this older post and help guide some of us newbs with a little more info here?? I bought a used machine. Seems in really low-use state. My Son is going into some precision work, has some unique design work going on, and we bought what we could afford. Plain and simple. Your suggestion for table accuracy is very appealing. I'll PM you to see if you might be interested in a little Q&A on this subject? Or anyone for that matter. I simply love to tweak and get exacting performance from my toys. I would love to learn how to do this or at least break down and send over components to someone who can (locally). I did watch some really great Vids from a guy who used to live 10mins from me. youtube: Link: https://www.youtube.com/watch?v=nOJrhrne80s

Just an excited new Shopmaster Patriot owner... Hope the exuberance is not too annoying w/ all my posts going on here. Looking forward to see who's who here and what everyone is doing for "hack a day" improvements on their setups. I was searching for accuracy testing which led me to this post. I'll try your method as well to check out the business end- I'm not sure how you get the machine back to the same exact center on step 2 unless you simply record the DRO readings and then put it back to those exact numbers?

CG.
Best,
JJ / CG.

[bill.anderson]

Sure, The Shoptask machines are fine for manual hobby use since you can compensate for backlash. To get any precision _CNC use you need to
do some accurizing. On my machine I disassembled the cross table (saddle) and scraped in the bearing surfaces. That provided the main improvement.
Scraping is a reemerging skill that can turn one of the cheap Chinese imports into a much more accurate and valuable machine. I attended one of Richard King's seminars and found it to be well worth the time and money I spent.
Richard's class also gave me access to an extremely experienced machine rebuilder. Much of the fun I've had has been "rebuilding" my new Shoptask Patriot
into a better, more capable machine
Bill A.

[TomB]

I'd also like to see a bit more discussion on this thread. I like the way Wormwood attacked the problem, but I'm fascinated by the results Bill got. I'd very much like to do a scraping seminar but they don't seem to hold them in New England anymore. That being the case Wormwood's approach seems most suitable to my skills. He was working with Patriot and I'm looking at a Tri-Power. so I'm not sure connecting the two drive screws at the headstock end would be the way to go, but I can't think of any reason why they could not be connected at the tailstock end. Wormwood put big bearing supports at both ends and I wonder if that is necessary. If the tailstock end of the added screw was bearing supported for rotation and thrust couldn't the headsock end almost float free?