PDA

View Full Version : ball screws VS. acme w/backlash comp (manual mill)



DaHui
06-04-2008, 12:40 PM
I'm wondering which will give me more reliable positioning performance: ballscrews (presumably 0 backlash) or and acme screw and DRO with backlash comensation...on a Bridgeport mill...assuming all other factors are not considered.

Okay, I know you aren't "supposed" to use ball screws on a manual machine. However, I worked in a shop for a while that had a bridgport that was converted to CNC and then converted back to manual...except they didn't remove the ballscrews.

I fell in love with it, the best bridgeport I ever used. Under normal conditions I didn't experience any wander caused by the driving effect of the ball screws and it was geat not to deal with paying attention to my feed direction during my initial positioning. It was spot on every time.

So here we are today, thinking about getting my own bridgeport. I've been looking at the new accurite DRO's that have backlash comp built into them. However, I've also found a great deal on a "nearly new" set of NSK ball screws for the brideport. Coincidence?

So my choice is a less fancy DRO (cheaper than $1,300 for the Accurite 200S) and the ball screws or get the bells and whistles DRO and keep the acme screws. Any advice?

lazlo
06-04-2008, 01:12 PM
Ballscrews have backlash too - the ballnuts are usually mounted in pairs and pre-loaded to take up the backlash. It's just that a ground ballscrew and ball bearing (in the ballnut) can have a smaller clearance, and virtually no wear, so the backlash will be less.


I've been looking at the new accurite DRO's that have backlash comp built into them.

How does that work? Why would the DRO care about the backlash of the table screws? The DRO tells you the absolute position of the table, which in itself eliminates any backlash issues (the DRO doesn't change until you wind the leadscrew past the backlash).

In any event, I used to believe the common advice that a manual machine with ballscrews would self-feed under cutting pressure, but a friend in San Antonio has a Bridgeport 1J with ballscrews, and it feeds like glass.

DaHui
06-04-2008, 01:44 PM
Okay you got me, it's not zero, but I can deal with a couple of thou in backlash for 90% of the work I do...but not as much as an acme will give...especially if it's an old worn out one.

The new Accurite is proramable for backlash...so you would measure it and then enter the value. I know CNC software like mach2/3 (and I'm sure others) have the same feature but I've never used it so don't know how reliable it is. I suppose in theory it should be quite effective.

I guess I could get the ball screws AND the Accurite and just be in debt for a lot longer but have a super duper accurate bridgeport.

rantbot
06-04-2008, 02:11 PM
I guess I could get the ball screws AND the Accurite and just be in debt for a lot longer but have a super duper accurate bridgeport.How would it be super duper accurate? It looks to me like it would have exactly the same accuracy as a plain ol' Bridgeport used by someone who knows how to handle blacklash.

toastydeath
06-04-2008, 02:33 PM
How would it be super duper accurate? It looks to me like it would have exactly the same accuracy as a plain ol' Bridgeport used by someone who knows how to handle blacklash.

Signed.

Get the DRO. Ballscrews are nice but are not something I'd put on a machine without dropping a control on it as well. At least in my mind, the zero backlash "issue" isn't much of an issue for manual work. At least in my case, whenever I do something manually where backlash is an issue, I'm doing it wrong.

lazlo
06-04-2008, 03:07 PM
How would it be super duper accurate? It looks to me like it would have exactly the same accuracy as a plain ol' Bridgeport used by someone who knows how to handle blacklash.

My point exactly. The "backlash compensation" feature seems pretty hokie -- you just turn the handwheel until the DRO starts changing :)

SGW
06-04-2008, 03:25 PM
Are you sure you're not talking about the DRO's linear error compensation feature? That has nothing to do with backlash. DROs don't need to worry about backlash. They are totally independent of the leascrew.

Mark Hockett
06-04-2008, 03:43 PM
Ballscrews have too coarse of a pitch for fine work on a manual mill. I have both a lathe and a mill that are CNC but can be used manually, they both suck to use manually because of that.

pcarpenter
06-04-2008, 05:51 PM
I will second or third what has already been written...a DRO needs no backlash compensation. Its not driven off the leadscrew or dials. It measures only the movement between the table and saddle or saddle and knee which inherently means that it never sees how much backlash your mill has in its feed screws.

It is certainly true that a DRO alleviates the need to pay attention to removing backlash in reading the dials...but with a DRO, you no longer use or need the dials. This is true of all of the DRO's with which I am familiar and not some "special feature". As SGW points out, the linear error compensation is a different issue. It allows you to compensate for error found in the scales themselves. If you have some other more precise standard for length measurement, you use it to determine some travel while monitoring how much travel the DRO says you had, so you can establish the amount of error per inch. This can then be preset in the DRO which then adds or subtracts to or from the value the scale actually registered as it displays distance. Since most seem to actually handle the measurement in metric calculations, my guess is there might be some rounding error for which this is handy too.

I have yet to establish just what I am going to consider more precise than my DRO scales to use as a standard for this measurement. I suppose some fixed stops and some Jo blocks would do the trick.

The shop downstairs has a manual BP with ballscrews. Its OK but I found it tends to coast if you get in a hurry with the table feed. There's not as much resistance to travel with that moving 300# or so of table mass. I never take my hands off the feed handle while cutting and have only used the mill for facing stuff, so have not dealt with self feeding issues or problems with the table wondering while say boring a hole etc..

Paul

mochinist
06-04-2008, 11:51 PM
Go to products and then retrofit kits and choose bridgeport or clone. Never used one, but I hear they are nice.

http://www.rockfordballscrew.com/

macona
06-05-2008, 01:38 AM
Mark, my mill has 5 tpi ball screws just like a normal machine.

But.... The problem is the super high efficiency of ball screws allows the screws to be backdriven. Supposedly it has pulled the handles out of operators hands while they are running them.

Acme threads will not back drive, they basically lock in place from friction.

Get a DRO unless you plan on going CNC, and if you plan on doing that get a old CNC mill with a dead control and retrofit that.

Mark Hockett
06-05-2008, 01:46 AM
Paul, macona,
I just checked my CNC Milltronics knee mill and it is a 5 TPI ballscrew, the same TPI as the acme screw on Manual knee. But I still like the manual mill for manual work much better as it just seems to have a better feel for hand work. It is my CNC lathe that has a really coarse TPI which makes it almost unusable as a manual lathe.

pcarpenter
06-05-2008, 04:33 PM
The guy in charge of the shop downstairs is on vacation, so I went and cleaned the goop off the way surfaces and lubed up the ways on the Bridgeport that has ball screws. Probably not a good way to go in my opinion. Outside of feel, if you stop with the handle in other than the 6 'clock position, it wants to fall down and move the table for you. This means in positioning that you would have to get it in place and hold the "sloppy loose feeling" handle carefully without moving it while operating the table lock with the other hand. Probably not real conducive to accuraccy really.

I would be especially concerned while using a boring head. You have a relatively slow moving cutter on one side only trying to shove the table first one direction and then the other. Forget to lock the table and you would get a nice egg shaped bore.

Paul