Preventing prt alpha x gantry strut deflection

[Chuck Keysor]

Hello. I am slowly reassembling a PRT Alpha 144 x 60. I have some questions, that require some set-up, so please bear with me. Thanks.

From old posts, I understand, that some people with PRT Alphas have had problems with the aluminum struts of the X gantry deflecting. I didn't find any detailed information on this, but while I am reassembling my unit, I would like to head off this problem, IF I have it. But of course I also don't want to fix a problem that I may not have.

The fix that I had read about involved placing one threaded rod down the center of each strut, and then putting those under tension. This then serves to stiffen the strut and eliminate the deflection, as I understood the post.

I have attached a photo of my machine so that hopefully the type of PRT alpha I have can be recognized by the experts who read this forum. My unit was manufactured in the fall of 2004.

I used a digital hieght gauge and recorded the distance from the table top to the center of the bottom edge of the long aluminum strut, with the spindle pushed off to the edge as far as possible. Then I rolled the spindle as close as possible to the center of the x gantry, without hitting my height gage, and recorded the change in height to the table top. Then I repeated this for the second strut. On one strut, the deflection was only .003 inches, and on the other it was .007 and these seemed like good readings.

Question 1: Did I correctly understand the problem with PRT Alphas described in the old posts? Since my readings seemed so good, I am concerned that I might have misunderstood the problem, and thusly did not correctly test for the posted problem.

Question 2: If I did correctly understand the problem, and measure correctly, can someone verify that my small measured deflections are in fact OK? And if so, are there any other things that I should consider to improve my x gantry as I reassemble my unit?

Question 3: If I did not correctly understand the problem, I have attached a detailed image of where the strut attaches to the motor car. There are some labels there and notes to hopefully aid in any replies that may be generated.

Thank you, Chuck

[gerryv]

Chuck, you're off to a very good start in your analysis and approach. I too am considering the truss rod; in my case as a just-in-case improvement.

In principle, I agree with the approach of using the threaded rods to stiffen the struts. However, I'm not so much going to challenge the conventional wisdom on this as just add a footnote for debate.

Where I think this approach needs clarification is IF the threaded rod is of a smaller cross-section (in this case, diameter) than the open space within the strut.

My point is that if there is any existing degree of bow in the strut and the truss rod is well anchored at each of its ends, then tightening it will apply a further bending force to the strut rather than straighten it. I do agree that if there is little or no clearance, then the effect would be to straighten as well as further stiffen it.

This is theoretical of course because things other than basic physics comes into play: The rod may stretch first if its too thin or made of low-grade steel, the areas of the strut where the tightening washers/nuts are seated may collapse inward marginally before the strut bends further, the threads may give way, etc. I'm simply assuming that you've chosen a good heavy, high quality steel rod suitable for the loading you'll apply so it would not stretch first, thereby negating the benefit it meant to provide.

Assuming this reasoning to be correct, I believe there is a simple way around a situation where there is significant lateral space or gap between the rod and the inside walls of the strut. That would be to mount a number of custom shaped and sized washers along the rod such that they would keep the rod "centred" through the length of the strut. I suppose a tiny undersize would be needed to allow the rod and washers assembly to be slid in.

This point may have been identified or assumed already by others of course and I've missed it (or I'm wrong so no need) but I offer it as a consideration that may need to be discussed because I think that the truss-rod approach offers a low-cost, easily undertaken improvement and I salute those who conceived of it.

Hopefully, some of the engineering oriented members will address this - either way.

[Brady Watson]

Chuck,
A few observations and comments...

The 'rod trick' mainly pertains to 4G PRT gantry machines only, which do not have the 3G (what you have there) 3X3 box sitting on 3X3 angle 'bulkheads' - Instead they have CNC bent components that bolt to the aluminum extrusions under the Y rails. There is no 'gap' between the Y rail supports and the bulkheads. I see how the recommended B7 rod trick doesn't seem very applicable in your case.

Two possible scenarios here pursuant to the history of your machine. One, this was an 'in between' low volume variant/ missing link between a 3G and 4Gen tool (unlikely) or you bought this used and someone ditched the original cad plated utility strut that went under the Y rails and replaced it with a piece of extrusion. I would assume the latter since the gold utility strut appears to still be intact on the Y car. These were AL extrusion on a 4G, which really didn't roll on the scene until around 2005.

Before I bought a PRT Alpha gantry, I beefed up my original gantry on my 5X16' PRT. This was after I got a return on my investment and wasn't sheepish about modifying the tool to make it better. I removed the strut under the Y rails (where you have the extrusion now) and replaced with 1/8" wall 1.5 X 2.5" box steel that I welded to the 3X3 bulkheads. I drilled and tapped holes for the Y rails to bolt to. I also squared up the gantry (which was a bear on those models) and laid a few welds where the rails met the bulkheads as well - making sure that it would never go out of square again.

Your deflection numbers are very good. The rod trick is for tools that are 'rollin heavy' with 2 spindles or a bunch of stuff on the Z axis, causing deflection. ShopBot advertises 0.015" in any axis, so you are well within that spec with the numbers you report. Keep in mind materials expand and contract that much thru the course of the day, and that level of precision is not required for 99% of what the tool is capable of doing.

Hope that helps!
-B

[beacon14]

It might be worth assembling and testing the machine before deciding whether it's worth the time and trouble to beef up the gantry. One of the first things you will do with your newly assembled machine is to surface the table/spoilboard. When you do that the table will now be parallel to the gantry, including any slight sag in the center. So when you machine your workpieces, the gantry will stay parallel to the surfaced table and workpiece even though it may be sagging very slightly compared to a perfectly flat surface.

[Chuck Keysor]

Thank you Gerald, Brady and David for your replies.

Brady, I neglected to say that my machine had originally been built as a PRT, and maybe a year after it was made, it was upgraded to an Alpha model by the original owner. I bought this unit from the second owner who only had it for a few months. But in addition to all the documentation for the PRT unit, came a full set of factory instructions for making the Alpha upgrade. So quite possibly that would account for my machine's oddities which you noted in your post.

With the information supplied, it is clear that my best course of action will be to continue on with the reassembly and learning how to run my machine.

Thank you again for your much appreciated advice! Chuck