ATC cycle in endless loop

I have a old daewoo vmc with a chain style side mounted 24 tool changer and Fanuc OM control. The atc has not worked since I bought the machine but finally had some time to try and troubleshoot. I found some bad solenoids and a bad output card relay. I now got the machine to try a tool change which it does complete but once it finishes the cycle the machine then proceeds to remove the tool it just put in and put it back into the original pocket then back to the spindle and round and round it goes until i press reset. Since it was not working originally I’m not sure if the machine has a memory of the tools and positions that needs to be cleared? I am calling m19 m6 t#. Any ideas why or where to start I have no Fanuc alarms but there is an ATC light that is on sometimes and off sometimes I can’t figure out the rhyme or reason for the light.

I don't have experience with your machine specifically, so you'll have to do some digging to figure out which diagnostic parameters to look at, but this should get you looking in the right direction.

Most likely, the machine is simply confused as to what tool is in the spindle, in the side arm, in the pocket, or some combination of it all.

In my experience with several of our 0m controlled machines, there are diagnostic parameters which store information about what tool the machine has. For more complicated machines that have swing arms, random tool pockets, and sub-pockets for pre-staging tools, the machine has to keep track of which tool it has, and where.


On one of our machines specifically, if you turn the machine off while it is doing stuff with the tool changer, it will go stupid, not have a clue where it is or what tools it has in the spindle or sub arm, and I have had it do similar to what you are experiencing.


What you need to do is look in manuals (if you have any) for tool change information. The information would be stored in a "D" diagnostic parameter on the diagnostics page. It may be in the 400+ range of diagnostic parameters, although it could be lower or higher - it is up to the machine tool builder.

In my mind, if we assume you try to change to tool 1 and this happens, it might think it has T1 in the spindle AND the sub arm AT THE SAME TIME, which confuses it, so it tries to correct the situation. In a case like this, you'd set the spindle tool diagnostic to 1, sub arm/pocket to 2 (or 0 maybe, depends on how it is set). While you are here, you'd also want to make sure that the tool change pot that is facing the tool changer matches the diagnostic bit.

If the tool changer is a "random pocket" type where any tool number could be in any pocket, you might find a whole list of diagnostic D parameters that contain information about which tool is in which pocket, and if that was the case, you'd have to sort it out in a similar manner.



I hope that helps.

It's been a bit since I've played with Fanuc. Is it possible that the M6 macro is stuck in a loop?

Hello tquinn,
First determine if your machine uses a Tool Change Macro. Take a look at parameters 0230 to 0239 inclusive for the registration of the number 6 therein. These parameter numbers correspond to program numbers O9020 to O9029 respectively. If you find the number 6 registered, proceed as follows:

1. Set parameter bit 10.4 to Zero to be able to view the O9000 Series Programs.
2. Find the program corresponding to the parameter were 6 has been registered.
3. Post a listing of the program here for the Forum members to view by either:
a. uploading it to a computer, if connected, then Copy and Paste it to your Post to PM
or
b. copy manually and include it in your Post to PM.

If a Tool Change Macro exists, its unlikely that you will have to execute M19 as well as M6.

How are you trying to execute the Tool Change? Via MDI, or as a Command Block in a Program. If via MDI and if M6 is calling a Tool Change Macro, that may be your issue right there. It depends on how the MTB handles M99 at the end of a Sub, or Macro program when called via MDI. Generally M99 at the end of a Macro Program will return control to the program from whence it was called. When executing the Macro via MDI, you're calling the program directly; there is no Calling Program to return to. In this situation, M99 will return control to the Head of the current program (in this case, your Tool Change Macro) and therefore, repeat the Macro Program infinitum.

Regards,

Bill

Bill,
Finally had a chance to look at parameters 230 to 239 and they are all 0 no 6 is recorded. I believe I am looking in the correct spot. Going to parameters page then page down to show page with 230 to 239. Now what?

Is your toolchanger hydraulic or electric driven? Don't know your Daewoo, but the ones that I did work on had very poor ladder logic programs. As a guess there is a failed sensor for toolchanger home or stop point. The logic is looking for that signal to know when to stop. Since it does not see that signal it just keeps running.

The changer is all hydraulic driven. Since there is no 6 registered in param 230 to 239 that means there is no macro program and the tool change program is in the ladder logic? I have param 10 bit 4 set to 0 and I do not see any 9000 series programs in the program library page. Do I have no 9000 series or just need to look elsewhere?

I can see a proximity sensor that lights up when the chain magazine is at position 1. There is a second prox sensor in the sub pot ( rotates the tool 90deg) that I replaced it also lights up when a tool is in the sub pot. I am guessing that there is a prox sensor or switch in the spindle but I have not looked to in depth there. I will chase the wires to make sure they are ok and the signals are reaching the input.

To answer angelw question we have tried calling the tool change in mdi and in a program with the same result essentially

Just because the light on a prox sensor turns on is not a reliable indicator that it is actually sending a signal to the control. You need to verify the state of the input in the Diagnostics screens.

Got my hands on an electrical diagram with the X addresses on the diagnostic screens and I just went through each of the inputs they all appear to change states when activated. There appears to be some lag on occasion for the diagnostic screen to update. When jogging the magazine fast the input for the prox switch does not change on the screen but the light on the switch itself does. When jogging the magazine slowly the diagnostics page shows each pot position registering which I think means the prox switch is working properly just the diagnostic page is not refreshed fast enough