ATC Slide

[mcphill]

I think you are going to need to stiffen that up. I would add a gusset from the front right corner of your extension bracket to the opposing diagonal face. Maybe you could even make the split line in the bracket be the center of that bent face, and "sandwich" the gusset in between a couple of tabs bent in to the ends?

I would also look for a way to get more support to the front of the bracket/slide. Maybe a square tubing strut down to the column - it could bolt or weld to the right hand face of the extension bracket to give it a bunch of stiffness...

Overall it all looks fantastic. I hope you release the design - I would love to make one!

I think I recall you having a question about controls? It looks like a great use for a PIC, Arduino or Propeller or even a BASIC Stamp processor with a few hooks in to MACH through I/O points... Biggest challenge would be defining what tool slots to use, but it would only take 5 I/O to use a binary coded decimal that would support 16 tool positions... A total of 8 or 10 I/O points should be more than enough. Could get away with much less if you got "tricky" with the programming and didn't need graceful recovery if something goes wrong...

[rwskinner]

2 brackets cut and bent. I had to do the extension bracket in 2 sections because there wasn't enough room on the press.

Tomorrow I'll drill and slot everything and weld the top plate onto the extension bracket and put the steel stiffeners in, if they're needed. I think once I close it in with plate it will be pretty strong. Heck, it doesn't hardly flex now.


It's straight and square. Camera plays a few tricks with it.

Richard

[gd.marsh]

Looking great Richard .. very nice work!

[mcphill]

Ah I missed the top getting boxed - that's not in the PDF

Still think a strut from the front corner to the column may be in order, to keep the whole deal from drooping in the front...

[Starleper1]

Does Alibre have any kind of simulation in it? If it does you can put forces on certain areas where you think it might be weak and it can analyze it for a given material. Solidworks has this, it makes it real nice when your unsure if it will be strong. Just a thought.

Looks great though. Keep up the great work.

[mcphill]

Was thinking about this instead of sleeping last night... The Z will always be at the same height when you change tools, right? How about if you put either some engagement pins, or a ledge, on the moving Z head. The slide would then engage with these pins/ledge when it slides over to the right. This would ensure that the tool changer system is at the exact right Z height every time. Even if the system "sags" over time, it will be lifted up to proper height when the side engages the support.

Just a thought...

[rwskinner]

McPhil, That is actually a great idea. One could easily make a male and a female adapter with a taper that engaged each other. Not only would it align things, but you could make the length adjustable with the stud so the engagement stays the same. If I have problems I may definitely try that idea.

All brackets cut and drilled and I'm working on the slots for the 4" steel mount. Then once every looks like it will work I will cut and weld the top cover on the Extension bracket.

Electronics:
I'm using a NEMA 23 motor at 380 oz/in and a G201 driver. I'm thinking about these inputs so far.

1 Optical home for the Tool Disc.
1 Prox Switch for indicating weather tools are in the slots or not. (Optional)
2 Optical Limit switches to detect when the slide is fully in or out.

1 Output for PDB (Already existing and using)
1 Output for ATC air cylinder extend and retract (Already wired)



I'm still debating using a modbus pic controller I already built or just using the BOB since I have room. I have a feeling using the BOB directly would be much easier but then I'm using up inputs and outputs.

The stepper controller logic would actually be pretty easy since there are 10 evenly spaced tools and the forks have enough slack for some adjustment.

If I use the modbus controller, I could use a register for sequencing and one for reporting errors.


Back to work.... Pics later.

Richard

[rwskinner]

Mounted to the mill. I had only one goof up (so far) and that is I forgot about the round hole and cover where the Z crank used to be. I had to move the entire assembly up another 1.250 in order to get solid metal to drill and tap on the column.

When I remove everything for final welding, blasting and paint I will increase the length of my vertical slots by another inch. It actually clears the top of the tool by about 0.030" but that is too close.

BTW, it I can shove it by hand and it glides over and latches onto the tool holder.

[mcphill]

Lookin' GREAT!

[pete from TN]

Hey man nice work. It is really starting to come together. I hope you have good luck with the build and that it works real well once it is finished. I am sure I am not the only one who is interested to see what you do in the software and electronics to make it all work. That has ALWAYS been the hardest part for me. The mechanicals are a lot more simple in my world anyways. Sure looks good in the pictures. Peace

Pete

[rwskinner]

Okay, Mounted, level, and waiting on cylinder. I'll start wiring electronics this week hopefully before I leave town.

The bracket is boxed in and is pretty darn strong with minimal flex. I milled the slots out more in the 4x4x.250 square tubing. Damn, I forgot how bad it was to mill steel and especially at 3500 rpm and 25 ipm. Hot stuff all over me and in me. I'll be festered up for a month while all the little shards work out from under my skin!

[rwskinner]

Try These instead:

http://www.youtube.com/watch?v=Pu_qkhG48RU]MOV02610 - YouTube

Richard

[mcphill]

Video links not working...

[rwskinner]

http://www.youtube.com/watch?v=-WrjDqbnzTg]MOV02611.MPG - YouTube

Me tired, I'm having a hard time with the videos....

Richard

[rwskinner]

BTW, I ordered another cylinder with cushions inside so that should help a lot. I'll just return the first one I ordered.

If you listen, you can hear the distinct click from the fingers latching the tool holder. This is me pushing it and it gliding over and clicking in. I was too tired to video the PDB opening and the z going up and come back down but it re-engaged the tool with no problem each time I tried out of 3 times.

Going to sit in the hot tub for a while and have a whiskey drink. Tired and sore.

Later,
Richard

[rwskinner]

Maybe, it's easy enough to add if I have an issue. I'd use a Link with right and left handed threads so the height could easily be adjusted. Being all Aluminum except for the 5" wide Extension bracket there isn't much weight there. Currently, it's very easy to adjust and level out.

I'm very sure there will have to be minor modifications made but we will tackle them as we run across them. One by one.

Richard

Still think a strut from the front corner to the column may be in order, to keep the whole deal from drooping in the front...
__________________

[rowbare]

Air Cylinder. It's a double acting 1-1/16" bore x 12" stroke cylinder that will be controlled with a 12 volt 5 way spring return solenoid valve. I did not see any cylinders like I wanted with internal cushions so I will have to devise a way for it not to slam open and closed. Needle valve is my first thought.

Richard,

I was at my favorite used equipment dealer yesterday. They had just gotten a bunch of stuff from a prototype shop and one item was an air cylinder with a small gas shock mounted on top of it. A couple of aluminum plates held the two base ends and the two rod ends together. I haven't put any air to it to test it but I am sure this was done to damp the movement.

bob

[rwskinner]

That also sounds like a good idea. You don't happen to have a picture you can share do you?

Richard

[rwskinner]

Wow, I was too tired, I was responding to spambots !!

Everything was torn down, sandblasted and a few coats of paint. Nasty weather here so it will take a few days to set. I won't be back in town until Friday evening anyways.

Stepper motor is wired up, axis tuned, and I set Mach up for 1000 steps per inch which is really 1000 steps between tools. So I do a G0C5 and that is tool 5, G0C10 is tool 10. Works like a champ and seems to be pretty accurate with nothing more than a line scribed for now.

In the M6Start the code will really be calling C0 thru C9. Got to rig up a homing switch. Hadn't decided if I'm going to use a Optical like my others or just put a microswitch in there.


I wrote some code for my modbus board and had it driving the Gecko but it was too slow. My digital outputs on that one large board is done with a IO Expander and by the time I toggle bits on and off via I2C it has too much of a delay. I could use a different pic but I think for now I'll use Mach since I have enough IO for the 5 axis and a few more inputs.

Later,
Richard

[rwskinner]

Here is my M6Start Macro. It's been testing in simulation mode.
I have messages in there and long dwells mainly for testing.

'Richards Toolchanger Macro

Sub Main()
OldTool = GetOEMDRO (1200)

'Get axis positions
x = GetToolChangeStart( 0 )
y = GetToolChangeStart( 1 )
z = GetToolChangeStart( 2 )
a = GetToolChangeStart( 3 )
b = GetToolChangeStart( 4 )
c = GetToolChangeStart( 5 )

Tool = GetSelectedTool()
NewTool = Tool


MaxTools = 10 'Tool Holder Slots
ToolDown = 2.0 'Height for Spindle to sit down on Tool
ToolUp = 0.5 'Z Height to clear tools during tool change.
MyFeedRate = 5 'FeedRate for bringing Z down to Tool
SlideDwell = 10 'Large Dwell for testing
PDBDwell = 5

'Test here - If Not Axis are not Homed then Exit


If NewTool = OldTool Then 'If tools are the same then exit
Exit Sub
End If


If NewTool > MaxTools Or NewTool < 1 Then
Message("Invalid Tool Requested")
Exit Sub
End If


Code "F" & MyFeedRate 'Set a slow feed rate for testing


Message("Move ATC to Old Slot")
Code "G00 G53 C" & OldTool 'Go to ATC Slot for Old Tool
While IsMoving()
Wend


Message("Lower Head to Unload Tool")

Code "G01 G53 Z" & ToolDown 'The Z-axis goes to ATC height
While IsMoving()
Wend


Message("Extend ATC to Get Tool")

ActivateSignal(Output2) 'Extend Slide ATC to Tool
Code "G4 P" & SlideDwell 'Wait for Slide
While IsMoving()
Wend

Message("Release DrawBar")

ActivateSignal(Output1) 'Send signal to release PDB.
Code "G4 P" & PDBDwell 'Wait for PDB
While IsMoving()
Wend

Message("Move Head up out of way")

Code "G00 G53 Z" & ToolUp 'Put Z Axis up out of Way
While IsMoving()
Wend

Message("Move ATC to New Tool")

Code "G00 G53 C" & NewTool 'Go to ATC Slot for New Tool
While IsMoving()
Wend

Message("Bring head down on new tool")

Code "G01 G53 Z" & ToolDown 'Z-axis moves down to where it touches tool
While IsMoving()
Wend

Message("PDB Grip Tool")

DeActivateSignal(Output1) 'Send Signal to PDB to Grip the Tool.
Code "G4 P" & PDBDwell 'Wait on PDB.
While IsMoving()
Wend


Message("Retract ATC Slide")

DeActivateSignal(Output2) 'Retract ATC Slide
Code "G4 P" & SlideDwell 'Wait on Slide.
While IsMoving()
Wend


Message("Bring Axis Up - Not Needed")

Code "G00 G53 Z" & ToolUp 'The Z-axis up
While IsMoving()
Wend


Call SetUserDRO (1200,NewTool)
SetCurrentTool(NewTool)

Code "G90" 'Make sure we in Absolute Mode
Code "G00 X" & x & " Y" & y 'Move back to we were before tool change


End Sub