Backlash Compensation

Is there any provisions for backlash compensation in the SB control software?

No there isn't. It really isn't a good idea either and falls under the category of, "You can't fix hardware problems with software." But that doesn't stop people from trying.

I did a bunch of reading on it last night. What I was able to figure out from a few forums is to get your machine right. Then try and pick up a few more thousands by backlash compensation. But you think is not a great solution either way? It seems to be pretty widely used and accepted. I'm with you though...seem kind of like a giant band-aid. I was just thinking if I could dial it in it might help pick up .003-.005" in tolerance. Just something to play with if it were available. Thanks as always for your hard earned insight.
Erik

This was interesting.
https://machmotion.com/blog/what-is-...-to-correct-it

"backlash compensation" - I don't think so! :D There are some great posts on here discussing keeping things properly tuned and replacing worn parts (IE pinions etc). I had an instance awhile back with some china end mills that were not cutting to my liking and what I discovered was a lot of 'bit flex'. Symptoms were similar to what I would expect from backlash but turned out it was in the bit itself. My machine is over 10 years old and I cut mostly cheap HD ply, so it gets a lot of long run use. The replaceable parts are cheap, quick to change, make my machine a good wood working tool. Russ

Thank you for the feedback guys.

A Shopbot is 'reasonably' tight and backlash free when you measure .003-.007" of play in any one axis. I'd recommend getting a good dial indicator set up and measure everything including gantry AND table deflection when you press down on them with moderate force. There comes a point where you can't get a bolt together machine any tighter and you just have to do the best you can or look at other options. The gearboxes themselves will show .004- 006* in many cases when new.

Slightly off topic....Brady...5 yr old Desktop....Any time frame to replace anti-backlash nut on Z-axis, and spring for spindle support?
What are indications of the need to replace?
LIKE the one piece older Desktops with welded frame, and machined Y-rail area...vs. the new bolted together for my applications.
Sorry for the off topic....and appreciate any info/threads Brady(Thanks in advance).
scott

Scott, What kind of play have you measured on your machine?

I don't have a lot of good things to say about plastic lead nuts when real ballscrew/nut combos can be had for dirt... They are fine if you don't cut over 2 IPS in most cases on these machines.

OK,OK......THIS week I'll buy a dial indicator set Brady:(
Still pretty tight on second passes, BUT Z seems to drop a hair on every RR(5 RR's will eliminate a light pencil mark on wood).
Was just wondering if those years of cutting buttons at low Z had stretched the spring.
I'll get some hard numbers when indicator(and steel plate?) come in.
Sorry I asked the question without any numbers Brady:(
scott

I don't think it's off topic. This can be the anything to do with slop thread. I have seen it mentioned that the gear boxes are able to be tightened up to eliminate some of the slop. Any tutorials for this out there. Would just be interested in learning about the process involved.

There are no tutorials out there for adjusting gearbox lash...this is because you have to rely solely on your mechanical abilities and sneak up on the right amount. If you take too much lash out, you'll wipe the gearbox out (scramble the gears...a la spoons in a blender) after it heats up, expands and binds. Then what?

As I said there's only so much you can do before you have to explore other options...which is what Gary and I have been saying for years. I'm not singling anyone out when I say some will never get it. A word to the wise should be sufficient...Tune up what you have and if that doesn't work for you, sell it and get something that will. Just upgrading the controller and motors with new gearboxes can make a world of difference.

It's SO easy for the snowball effect to occur with these machines and then you're upside down in it and still not where you need to be. I speak to those using CNC to feed their family. Hobbyists can make excuses.

Brady is spot on. You have to decide if this CNC brand will perform as needed to do what you want done. Backlash, flex and slop are things you just can't eliminate from ShopBot machines. The technology is to archaic to accomplish that. I have found ways to work around some of the flex, slop and backlash in my work. Yes, I do use the machine to pay bills and feed my family. Does it compare to what a newer machine can do, NOT EVEN CLOSE. But it's enough for me to do what I do. My machine is run hundreds and hundreds of hours (approaching well over a thousand hours) RUN TIME each year.

One thing I do with my machine is a tune-up after each cutting session. I build this time and material into my shop hourly rate. I check all the nuts and bolts every 3 months. I adjust all the motors and V-rollers, clean up all the tracks and regrease them. This has helped a lot. Before each job I resurface my spoilboard to make sure it is in "sync" with the router bit tip. I use 1/8" sacrificial boards on top of my spoilboard that are surfaced both sides to make sure they are in "sync" with the router bit tip. I replace my v-rollers and pinion gears yearly. Replacing pinions v-rollers is a small cost considering the benefits it brings. As a business, I can build these costs into customer pricing. If I was doing this as a hobby I certainly wouldn't be changing parts regularly. I plan to get a 12 station ATC upgrade at the beginning of next year. I will continue to carry out the same detailed maintenance schedule even after the upgrade. With this approach, I've gotten ten years of decent performance out of my machine.

It's up to each person, and the results they need from their machines, to determine if ShopBot is the way to go or it's time to move on and get a machine that uses the latest technology instead of, at least, decade old technology. I looked heavily at a Thermwood machine and would LOVE to upgrade to one, but I can't afford the price of the machine and the necessary electrical upgrades to my shop. So I'll keep nursing the ShopBot along until I retire. Then I will either attempt to sell it or use it as a hobby machine.

Scott, the thing about the DT tools is that at best they can resolve 0.0002"....on paper. However, the reality is, that number can vary due to hysteresis associated with microstepping. The screw only gives a 2:1 mechanical reduction, which is way too coarse on any CNC router in my experience. Since the SB controller is not capable of outputting steps faster than 6 IPS reliably, that's why the DT maxes out at the speed it does. Less than 17% of a motor's rated torque is available when microstepping...a motor is free to waller around until it gets to a 1/4 step torque detent, which will result in positional loss, if the torque required exceeds that 17%. I've discussed this in the past...as have others.

I liked the DT I had, but it wasn't designed for the kind of precision and features I expected from the tool. Mechanical reduction should have been at least 5:1, the motors should have been more powerful, the screws should have been ballscrews, the spindle should have been 24,000 RPM+ for small diameter tools that require it...the list goes on and on. Those are the main reasons why I sold it. I considered converting to Alpha motors or servos and ballscrews but it was easier to just get something else.

Sounds like I reopened a can of worms. I suppose there is no idea on this forum that probably hasn't been discussed before. I am picking up a 10 year old BT32 the 18th. Part of the attraction is the fact that mechanical parts are cheap and easy to replace. I realize that I won't be getting the state of the art, but turnkey with low investment and good support is huge. I would love a machine I could mill some aluminum to .001-.002" but I don't think there is much out there in "affordable" machines that could do that. I will mostly be doing wood, but hoping with some tuning up the old BT can keep me within .005-.010". On Aluminum...maybe with some practice even better. Originally the idea of the thread was just to try and pick up an easy few thousands. Sounds like the best way on these machines is just to tune it up and like others have said...just accept the fact of what it is.

Thanks Brady!
It does what I need in hardwoods(and just a hobbyist now), but just want to keep on top of maintenance.
scott