Messages

221

Show"N"Tell ( Your Machines) / Re: Denford Cyclone retrofit

« on: March 21, 2014, 05:55:31 AM »

Where are you / they located ?

Wales.

Angel tech, any 8 position turrets on them?
I'll live with my 6 position, but am curious to see an 8 position.

222

Show"N"Tell ( Your Machines) / Re: Denford Cyclone retrofit

« on: March 20, 2014, 07:48:16 PM »

The tool holders are finally bored!

I gave up waiting for Chronos (the excuses have started, and it was meant to be dispatched on Tuesday...), so ordered up a 12mm boring head from RDG which arrived next day, and got the toolholders bored inbetween keeping the Conect running today.
Just need to get them drilled and tapped for the lock screws now.

Then I can get some drills mounted, along with a boring bar and get some more ER collet holders machined up to mount all the drills I'll need.

I can't wait to get it up and running fully, and only have to deal with loading stock, instead of continually changing tools!

And yeah, I'm glad the tool changer won!

223

Show"N"Tell ( Your Machines) / Re: Denford Cyclone retrofit

« on: March 14, 2014, 03:27:16 PM »

Last weekend I managed to get everything pretty well tuned, got the tool holders bolted on, and managed to get them drilled out to 20mm before the spindle ran out of torque, beforing swapping to the boring head.
I got to around 24mm before I had to sharpen the boring, then proceded to rezero things, and command a tool change without moving the Z-axis back to clear the tool holders. All I'll say is the tool changer won, and I now have another paperweight  

Being the last boring bar long enough to get full depth through the holders, I ordered some new ones, however they've still not shown up.
I needed 1/2", and given 99% of the time now I only use insert tooling, I found that Chronos do a 1/2" boring bar set that take inserts, so thought I may aswell give them a go.
However I should of known Chronos are still hopeless at sending things out in a timely manner, so I'm still waiting. I used to use them a few years ago, but got fedup with them taking four days to even bother sending the stuff out, even if I'd paid for express delivery, and it looks like they've managed to downgrade their dispatch times even further.
I'm a bit annoyed because finishing up the tool holders is all that's need to be able to get the lathe making parts.

I may just bite the bullet and order up a boring head that takes 12mm bars, as I've already got a couple 12mm insert bars lying around.

224

CS-Lab / Re: two wire inductive sensor

« on: March 10, 2014, 01:32:11 PM »

They're NAMUR output sensors. I don't know why denford used them, but they basically use a low voltage with the resitance changing when triggered - http://www.tektron.ie/namur1.htm

It's far simpler to bin them and fit NPN/PNP sensors instead, rather than try interfacing to them.

225

General Mach Discussion / Re: Getting error message

« on: March 08, 2014, 10:05:31 AM »

I've had that error before.

From what I remember, I unplugged the SS (making sure it was powered down if you're feeding it from an external 5V supply), and with it still unplugged, started Mach so it produced an error that it couldn't connect to the SS. Then quit Mach, plugged the SS back in, and then reopened Mach.

226

Show"N"Tell ( Your Machines) / Re: Denford Cyclone retrofit

« on: March 05, 2014, 06:34:11 PM »

X axis following error is now sub 10 counts.
After not getting much improvement in Speed mode, I tried Torque mode, but despite trying all sorts of PID values, the results were even worse. So swapped it back to Speed mode, and with a bit filtering and further PID adjusting, managed to get a reasonably good result. 10 counts works out at 0.0025mm of error, so I'm pretty happy with that.

I also had an issue with the spindle VFD, in that it started tripping the RCD while I was in the middle of changing settings in the servo drive. I just turned it off so I could tune the servo. I took the cover of to check for anything obvious, but other than lots of dust it all visually looked good, so gave it a blow out and it's working again. I'm hoping it was just the dust and a bit dampness causing the problem.

Just got the Z-axis to tune, and then I might get to cut some metal with it.

227

Show"N"Tell ( Your Machines) / Re: Denford Cyclone retrofit

« on: March 02, 2014, 02:13:52 PM »

Progress.


Covers on and blocks ready for boring by mc_mtb, on Flickr

I had a bit of an extended run on needing to make bits, so had been suffering from a chicken and egg situation, in that the little lathe takes too much time/manual input to make the bits, so I've not had time to get this up and running, and then when I have had time, things have stalled due to needing some stupid part I should of ordered weeks ago.
The tool holders have now been milled to the correct height, and the other faces tidied up to look a bit neater. Took a couple weeks to get hold of suitable length bolts, as they need to be a specific length, and the supplier sent them to the wrong address.

I was going to risk just relying on the grub screws to hold the X servo pulley in place, however after I realised just how much effort it takes to get all the slideway covers off/on, and that I really want to minimise knocking any of the tooling of line by anything slipping, I ordered up a suitable broach set, and took the pulley back of.
Everything is now back on how it should be.
I'll take the Z servo pulley of and broach it at some point, but I'm in no rush as it's far easier to get into and not as critical if it does slip.

I've got a dedicated computer setup now, so have been working on the final coding, during which I discovered an issue with the PLC ladder causing the toolchanger to go into an endless spin if an invalid position was requested. That's now sorted, so that it shouldn't do anything if an invalid position is requested, and if it does somehow go into an endless spin, it should stop after 10 seconds (I could probably reduce this, as the maximum change time should be sub 3 seconds).
Thanks to Hood, I've also added some extra lines to the tool change macro so any program should halt with a warning message if an invalid position is requested during a program.

Main things I have left to do are bore the tool holders, and properly tune the servos. I done a quick tune earlier, and have the X-axis holding sub 50 pulses, and the z about 14 pulses, but I think X should be better.

I also need to make a proper stand, but that won't stop it working!

228

General Mach Discussion / Re: ATC and selecting tool issue

« on: February 26, 2014, 04:31:45 PM »

Thanks for that.

It's probably bad habit on my part, as I just type T0 via the MDI on the small lathe prior to moving everything to zero with a G0 prior to shutting it down.
Now I think about it, I can't remember trying anything above T0, as all the other tool changes are done in the g-code files.


I'll add that code into my M6 macro as a safety net, but I still need to sort out the PLC. If a T0 is commanded, the changer starts spinning endlessly even when the tool change allowed signal is removed, and the only way to stop it is to stop the PLC, or send a valid tool number and cycle the change allowed signal which has to be done directly via the KFlop Kmotion manual controls. I thought I had a sanity check in the ladder, but I''ve obviously missed something :-/

229

General Mach Discussion / ATC and selecting tool issue

« on: February 25, 2014, 06:29:58 PM »

I'm finally getting around to finishing of my cyclone, but during testing various things, I've discovered an issue with selecting tools.

If I issue a full tool change command i.e.T0101, things work fine, however if I command an abreviated version i.e. T01, then it requests tool 0 which is invalid (it has also highlighted an issue with my PLC logic, but that's another issue).
Is this how Mach turn usually works?
Or have I missed something?

Here's my toolchange macro-

Code: [Select]

oldtool = GetUserDRO(1202) 'Get the current tool position (we could use GetCurrentTool, but UserDRO1202 is updated directly from KFLOP via the macropump)
newtool = GetSelectedTool() 'Get the new tool position
if oldtool = newtool Then 'if we've already got the requested tool loaded
End 'then we can quit this macro without doing anything
End If
ActivateSignal(Output10) 'this activates the tool change allowed signal from the KFlop to PLC (KFlop bit 144 - Port2Pin16)
SetOEMDRO(1006,newtool) 'transfer the new tool position to the OEMDRO (transfers to KFlop persist, where it gets transferred via MODbus to the PLC)
NotifyPlugins(19006)
While Isactive(Input1) 'wait for tool changer OK signal to turn of (this is so we know the new tool position has been transfered)
Wend
While not Isactive(Input1) 'then wait for tool change ok signal to turn back on - KFlop bit 141/port2pin13
Wend
DeactivateSignal(Output10) 'turn of tool change allowed signal
SetCurrentTool(newtool)
End
I'm using a KFlop, which communicates with a PLC via modbus to control the toolchanger.
I've checked and the KFlop is being sent a request for position 0, so I'm sure this is something Mach is doing.


One other quick question while I'm here.
If I add code into the Macro to check for a valid position, what code is needed to halt the current program if an invalid position is requested?
i.e.
If newtool < 1 or newtool > 6   Then
   What goes here to bring everything to a halt?
End If

230

General Mach Discussion / Re: Modbus / MPG help...

« on: February 15, 2014, 05:21:25 PM »

I'd guess it's using the serial modbus.

Got a link to one?