You opinion on my plan? Updated: DXF and Pics of parts included

[Darren]

EDIT: I have attached a jpg rendering of my design as well as a DXF file to this post. The plate aluminum for the gantry is 3/8" thick, is that too thin? I don't have 1/2" and had the 3/8" already. Should I put ribs on the outside to add rigidity? The gantry sides are 21" tall, it rides on 1" X 48" supported rails. The Z axis rides on 3/4" X 18" supported rails and final axis is 3/4" X 24" and will also be supported now.

I have all my linear motion bearings and servos with all the aluminum I think I'll be needing. I have one question though. I have two STEEL "C" channel beams that I am considering using to mount my gantry rails to. I'm wondering if it is a good idea to use the steel and mount my rails to it. It seems to be fairly true. Does anyone see any reason not to use that approach. If you have other ideas based on the pics I have provided please let me know!
I have a pic to illustrate my idea for mounting to the steel. I've seen it before:



The two angle aluminum pieces you see will be at the front and back and the steel beams will be mounted above them with the gantry connected underneath where I will connect to the leadscrew nut.

Here are some other shots of the rest of my material:


This is the piece I intend to use to bridge the gantry:


This is a shot of the plate aluminum, gantry bridge piece, and my Y and Z axis linear motion parts:


Extruded aluminum I intend to use to form the table:


The rest of my plate aluminum:

[Darren]

Guess there are no opinions

Another question then that might be interesting. Is anyone using EZ-Lock thread inserts in their aluminum machines to prevent stripping?

http://www.ezlok.com/index2.htm

I am considering using them to assemble most parts but wonder if it is actually necessary.

[ViperTX]

The steel channel is not that perfect (spray them black and use a 2x4x6 machined block with emery paper wrapped around it and pass it over the surface a couple of times and you'll see the surface variation)...it useable if you apply a bed of metal epoxy when you lay the linear rails in the channel.....

[ViperTX]

Threaded inserts are usually not necessary....if you were building an engine...then I would certainly used helicoils in the block. When you thread the end of your extrusions make sure you go at least 1 inch....deeper is better.

[Darren]

Thanks for the opinion ViperTX, I'm a little stuck as to what to use then. I'd prefer to ditch the steel if possible then. I just don't have anything of sufficient gerth to mount the linear rails to. The extruded aluminum pictured below is long enough but I fear it isn't beefy enough. I believe it is 3/16" or so thickness. I'll have to check.

Opinion?

[DennisCNC]

I think with steel mixed in to the machine it might expand with weather and maybe warp the longer ways on the machine.

What is the plate aluminum you have a pic of, cut with? Flow Waterjet?

[Darren]

The plate aluminum was cut with a flow waterjet. I got a lot of those as scrap from a friend.

[ViperTX]

Well for CRS 1018 the coefficient of thermal expansion is 120 x 10 -7 per degree Celsius which basically says for 100 degree shift in Celsius (which is basically 212 degrees Farenheit) the metal would get longer by 120 x 10 -5 inches or .00120. So, I would not worry about it and use the steel channel.

[Darren]

Originally Posted by ViperTX Well for CRS 1018 the coefficient of thermal expansion is 120 x 10 -7 per degree Celsius which basically says for 100 degree shift in Celsius (which is basically 212 degrees Farenheit) the metal would get longer by 120 x 10 -5 inches or .00120. So, I would not worry about it and use the steel channel.

Uh... damn. Went right over my head but I believe ya Anyone think I should worry about the "trueness" of the steel to begin with? I just don't want my rails to warp to any variations in the steel if there are any. I suppose If I don't see any variations as I try to mount it I might be ok.

[DennisCNC]

But if you have the .0012 expantion when coupled to another metal that would create a bi-metal flex effect like a thermostat right.

[Halfnutz]

It looks like your using ball slides and not linear ways. Are you going to support the rails or float them? It looks like you have closed slides so you cant mount the rails except for at the two ends right? If your floating the rails that steel will work great. It probably isnt true enough without milling it to ridgidly bolt the rails to though. But with those closed slides and just the two ends supported with a couple of stand-offs it will work fine I would think.

(The above comments are the authors and do not represent those of CNCZone or its management.)

[Mcgyver]

Originally Posted by ViperTX Well for CRS 1018 the coefficient of thermal expansion is 120 x 10 -7 per degree Celsius which basically says for 100 degree shift in Celsius (which is basically 212 degrees Farenheit) the metal would get longer by 120 x 10 -5 inches or .00120. So, I would not worry about it and use the steel channel.

have you got the lengh in there somewhere... ie total linear expansion depends on the lengh.

i agree its probably not an issue though as aren't the rails also of steel?

My guess is you're right Viper the number will be small, butr it is an interesting point though,with delta between materials' thermal expansion rates it wouldn't take much for parts fastened together to do the banana.