Deflection in Z carriage

[carl]

I think I have a way to solve the deflection in the Z carriage from heavy hard cuts. It goes back to to old drafting boards on how they used to keep the straight edge even across the board. Here is a picture to show you what I am talking about. Naturally the wheels would be mounted on top of each other on both sides, a couple of turn buckles to adjust the squarness. I would like to get someone's input on this before I try this on my machine. I think this would be a cheap way to solve the problem other than spending a thousand dollars on a stepper motor and a driver.
Carl
5455

[carl]

Update from previous post: I know that everyone has had this problem, about the racking on the Y axis in doing heavy cuts. It seems to leave chatter marks and not very acurate cuts because the bit is always trying to play catch up. The only way to really solve this is to spend a thousand dollars to upgrade your alpha with another Y stepper motor. I did a little upgrade to mine, spending only twenty dollars from the local Home Depot. It seems to work really well. It goes under the same principle as the old drafting tables with the straight edge that you can push from one side or the other and it will always be square. Here are a few photos to show what I did. - it still needs to be painted
5456

[carl]

5457
The cables are going around the wheels, and to the opposite corner, so if you were to pull the cables tight it will create an X. I just wanted to show everyone how you can upgrade your machines too if you would like. Carl

[dvanr]

Hi Carl,

Nice work! I'm squinting at your photos trying to make out how you set up the second pullies ( the first ones being the closest to the bottom of your sketch).

I can see the first one in the photo and I can see a cable that looks like it is heading for the second pulley , but not sure where the second pulley is.

Some details on the hardware be appreciated. Are the pulleys ball bearing? What size cable did you use? How did you determine when the cables were tight enough?

PS noticed the Z bracket in the bottom photo. That make a big difference in chatter too?

[jhicks]

Carl, Looks like a great upgrade idea. Unfortunately I can't see clearly enough either in these photos. Could could send a photo I can enlarge to see the details and the parts as suggested by Dick?
Please e-mail me directly when you have time. Thanks

[GD (Unregistered Guest)]

Carl, I am sorry to tell you that your scheme is not going to do what you are hoping it will do. the carriage will still be able to twist because you have the carriage attached to each wire with freely rotating rollers. The parallel rule of a drawing board is different - the rule is fixed to the cable at 2 points (one continuous cable only) and the rollers are fitted at the far corners of the board

[carl]

Thanks Dick and Jerry!
I will take more photos and send them to you sometime this weekend.
To explain a little more in detail, the second pulley is the same as the first. The pulley's are stacked on top of one another and mounted to the bracket. The cable starts from one corner, goes up to the pulley, around into the second pulley, around that pulley into the far opposite corner, and the other cable starts on the opposite side and does the same thing in reverse. By tightening the right cable, it will pull the squareness one way, and by tightening the opposite cable, it will pull the squareness the opposite way. The cable is a 3/32" cable (which is probable a little over-kill, but it is better safe than sorry) and the pulley's are doorwall roller wheels. It is a good idea to use a square when tightening the cables.
I hope this helps, Carl

[carl]

GD - I guess you never used a drafting table

[GD (Unregistered Guest)]

I used them, designed them, and built about 200 for our local university in 1970.

[carl]

I appreciate your feedback, but I already have it set up and it has been working for me.
Carl

[bleeth]

A similar system is used for the Accuride "anti-rack" retracting door system and has worked for many years. Their latest improvement has been to put the tensioner adjuster in such a position that it can be readily accessed from the front of the carcass when the doors are open. It does work with the two ends attached to a fixed position and I think your adaptation is a great idea. Have you worked it with some "heavy hogging" yet? I too would like some better pics if you would.

Dave

[carl]

Hey Dave,
I have done some 3/4" thick ply at 3. per second. I haven't had the need to do anything more. I appreciate your info. The next thing is to take it apart and paint it. SB only allows a small picture so I can email more photo's to you directly if you would like.
Thanks, Carl

[GD (Unregistered Guest)]

After racking my brain why that system wouldn't work for a drawing rule, it was because it bent the plastic rule up. The tension in the cables under the board bowed the ruler upwards. However, there is little chance that the cables will curve the carriage of the SB upwards, so yeah, it should work.

http://www.mymall.ws/Affiliates.d/Woodworking.d/Tools.d/rockler.d/articles.d/accuride.HD.php

[bleeth]

Yes I would Carl, my e-mail is in profile. 3"/sec. through 3/4 ply with a Prt sounds quite good. Were these straight runs or shapes and curves? After just retuning my car and carriage for the umpteenth time I tested by running some symmettrical arcs at various points and check them for equivilency. Keeping the bot accurate for this is important as I do template work for some other shops.

Dave

[jhicks]

Carl, thanks for the enlarged photos. The design looks simple and effective. Is Dave correct that you are running a PRT at 3"/ sec?
When I've spoken to other folks, including shopbot, the general concensus was over 2.5" one could experience accuracy issues with PRT's. Especially with such a heavy cut. I see the newer carriage in your photos looks like what I've seen on Alphas but not sure if this heavier rail system was ever used on PRT's.

[gerald_d]

The "classic" PRT's are quietly back on the market. They use the latest rails and carriages, but the pre-Alpha motors and controller. Link (http://www.shopbottools.com/prtspecial.htm)

[bleeth]

I just asked about that the other day and the price quoted for the new carriage is $1500. I believe some angle brackets running from the struts to the rails could stiffen up the "classic" considerably and prevent wracking. One would just have to make sure it is dead square to start with. If they are mounted low enough on the strut then they will not interfere with the hold down bearings.

[jhicks]

Gerald, I see it and I get it. Thanks.
Nice to know but $1,500.00 sounds a bit rich to me. I still don't know if Dave has an Alpha or a PRT though. I like 3"/sec on a PRT regardless. If its achieveable with repeatability and consistency I want to go there.
What say you Dave? Are you running a PRT with new rail & carriage or an Alpha?

[bleeth]

I'm running a classic PRT and so stiffening the rails and car is a little different from what Carl did to achieve his 3 ips. He says he is getting it on straight and angle cuts and has eliminated chatter. $1500 seems a bit rich to me too, but eliminating chatter would be almost priceless. SB makes a large case of comparing it's alpha to the "big boys" in terms of speed and accuracy but those machines produce a smooth edge finish with one pass. I have yet to figure out a way to do it with my bot without a clean up pass. Hence the interest in stiffening the carriage and car.

[gerald_d]

I may have caused some confusion with the term "classic".......

In the context of car stiffness, the PRT moved from:
- angle iron and bolted steel utility strut y-car
to:
- folded profiled plate with bolted aluminium extrusion y-car
when the Alpha series was introduced a year ago.

The extruded alu is now standard for all models, whether they have Alpha motors or not. Dave, I presume your "classic" still has the bolted angle iron and utility strut car construction?

[bleeth]

Yep-Gerald, I recall a similar discussion about y alignment and stiffness in which you discussed your solution of a spot weld on one of the bolts to maintain rail distance accuracy. I'm wondering if the angle bracket solution wouldn't kill four birds with the same stone; distance accuracy, anti-wrack effect, constant squareness, and chatter. Since we are dealing with a big "erector set" some basic bridge/span type gussetting should improve the overall charecteristics. Eliminating chatter and hence vibration should also result in bolts throughout staying tighter and reduce maintainence cycles.
One would need to make sure the y rails are properly squared and then make sure the folded profile (strut) is also aligned. I have thrown a dead straightedge on the side of the strut and seen that it is NOT perfectly straight and so some slight shimming of the angle bracket may be necessary.

[gerald_d]

Dave, are you trying to stiffen your...

- y-car (the small one that rolls on the gantry in the y-direction) and that is often blamed for cut quality?

or:

- the gantry (the big animal that rolls across the table in the x-direction) and that is often blamed for out_of_squareness? (and what I tacked together)

Carl is talking of ways to hold the little y-car square to the gantry. Other guys have used a second y-motor for this, but they say they don't need a second y-motor when they stiffen up the little car. Welding stiffener gussets to this car is a very simple thing to do (if you can weld) and you can easily make a stiffer car than the Alpha.

For the big gantry, some brackets can be added to make it as stiff as the Alpha, but this will only help with squareness and not really with cut quality.

[stevem]

It cost me approximately $180.00 to fabricate a welded steel X gantry. It shouldn’t cost more than $75.00 to fabricate a Y carriage, including paying a professional welder. I am currently experimenting with cutting wood (normally I cut plastic at 150 IPM) at 350 IPM on a PRT machine. Results are good so far with no chatter marks.
5458

[gerald_d]

Steve, I still think that is the best gantry design I've seen on a SB!


What material/size are the two main beams?

And, if it is not asking too much, could you mail me a better photo please? gdpost(at)gmail.com

[stevem]

Gerald, You're making me blush. The tubing is 3 X 2 X .1875" on the drive side and 3 X 3 X .1875" on the other side.

[bleeth]

Gerald: My last post on this thread addresses the gantry. SB's system has been to bolt a set of holddown brackets under the strut to the rail and from what I understand the alpha uses a slightly different shaped version of the same bracket cutting down on the amount of spacing necessary. I think angle brackets or gussetts laterally fixed would do a better job. To stiffen the car and hold it square on the "classic" I see a similar system mounted at the four inside corners from the car struts to the carriage. It seems to me step one is to make sure the gantry rails stays parallel and don't wrack. It's shifting while pushing through plywood all day will result in the holdowns getting pushed out of place and then here comes the bouncing car and chatter!! (As well as parts that are supposed to be the same size being different). I have found that the gantry being kept square is a more often occuring malady. In heavy use though, any retweaking of the alignment that needs to happen on a regular basis when no extraordinary shocks to the system have occured is more than there should be.

[bleeth]

Steve-6ips with a PRT? FANTASTIC. If you pull that off driving a bit through full width 3/4 ply with complicated shapes you have made a believer of me. Please send me some more pics or a sketch

[Brady Watson]

Dave,
Stevie is leaving out a few details...such as his Mach2 control software conversion. Using this setup with twice the voltage on the power supply, you can effectively double or triple your move speed. The higher voltage pushes the torque band higher in the RPM range. You will however, need new stepper drivers such as Geckos to make use of Mach2, since the Allegro drivers in the SB box are both soldered in and limited to about 44v or so.

-Brady

[bleeth]

Brady: You beat me to it-I had a chat with Steve today and he filled me in on all of modifications. As he put it: "There isn't much shopbot left and soon there will be less." He has done a lot of what I was thinking about vis a vis the frame components but before I go quite that far (It's not the cost of the steel and welding, it's the overall time involved) I want to try some of the simpler things to stiffen it all up.