Alarm 167 Zero Ret Margin Too Small

[lovell110]

I need some help! I have been dealing with this alarm for quite some time. I don't do super precise work so it has not really been a thorn. I have some down time right now so I would like to clear it up. I have checked the limit switches and the trip dog and everything is secure. I tested the limit switch and it works fine. My manual is original with the parameter list included. I am the third owner so Im sure there were some changes over time. The Z grid shift number presently does not match with the original settings. How do I change the grid shift settings? I have tried and the readout says "function locked". If I change the grid shift will that also affect the tool change position? Any help would be greatly appreciated.

[Geof]

I don't think it affects the tool change position but maybe you should make a measurement first from the table to the spindle nose so you can check.

The 'Function Locked' refers to Setting 7; Parameter Lock which you have to turn OFF to change a parameter. You also have to push E-stop.

[lovell110]

Geof,

Thanks for the reply. I was having a brain fart with the #7 parameter. I keep forgetting about that. I did play with the Z grid numbers and got the alarm to go away. I did see something I did not like on the distance to go readout. Everytime I zeroed the Z axis the Z number would not repeat. It changes to a different number each time I do this. The difference is not drastic mind you but Im still concerned. I'm trying to figure out why this is happening. I can only think it has something to do with the encoder. I really don't have a lot of experience with these things and Im really hesitant to call my local HFO. I've learned more from guys like you on this forum than any tech I've had. I understan about job security but I basically can't afford the rates the HFO demands. I'm a weekend warrior just trying to make some extra cash. I hope you might have some ideas for me. Thanks again to all of you for this forum and your willingness to share.

[Geof]

Originally Posted by lovell110 ....I did see something I did not like on the distance to go readout. Everytime I zeroed the Z axis the Z number would not repeat. It changes to a different number each time I do this. The difference is not drastic mind you but Im still concerned. I'm trying to figure out why this is happening.....

I have never looked at that display so I have no idea whether it stays constant or not.

You did say you had looked at the limit switch, did you clean it to make sure it is not operating slowly? That, I think, could make this display vary.

[Donkey Hotey]

I could be wrong but it seems that resetting any axis limit each time a machine started wouldn't be very good for repeatability.

I would bet that the machine simply jogs to its last known limit position to verify that it still sees the limit switches and knows everything is safe. I'm not surprised that it doesn't repeat the same number every time it finds the sensor. There's probably a tolerance band that is acceptable for it.

For it to repeat every time, the switch/sensor would have to be repeatable to 0.0002". I could be totally off-base but that's my educated guess.

[Geof]

The machine does not use the switch to zero, it uses the Z channel pulse from the endcoder. Encoders give innumerable counts per revolution on the A and B channels, that is their job, but they have an index pulse that occurs once every revolution...very precisely. That is the Z channel (and I hope I have the correct terminology)

The way the machine homes is it moves a little distance past the zero position, and this triggers the microswitch or proximity detector.

Then it reverses direction and the switch resets or the proximity detector loses it signal. As soon as the controller detects this it starts counting encoder pulses and looks for the Z channel pulse. Home is when it sees the Z channel.

If it counts too few or too many pulse between the switch closing and the Z pulse it gives the zero return margin too small error.

[lovell110]

Now I'm really concerned about this! I am going to work tomorrow and I'm going to fire up every Haas mill we have and try to see if they have the same thing. I have a nice age gap between the three (VF-2,VFOE,Mini). Hopefully I'll get some kind of idea of whats going on. Keep up the ideas fellas! I do my best to keep you updated!

[Geof]

Actually I thought a bit more, and came to the conclusion you should see the distance to go reading vary. Here is my reasoning:

The controller counts the encoder pulses between the switch closing and the Z channel pulse.

The Z channel pulse always occurs at the same spot.

But the switch is not going to close within 0.0001" of the same position every time; microswitches or proximity detectors are just not that precise.

So you should see a variation in the readout because the switch is not precise.

But....the variation should go up and down; there has to be a theoretically constant position for the switch to close, but sometimes it will be a bit early and sometimes a bit late.

You need to sit down at the machine, turn it on, do power up restart and then record the readout.

Then repeat that a lot of times.

If the readout varies up and down, don't worry.

If it drifts constantly in one direction, worry.

And I am sorry, if it is the last one I don't have any suggestions.

[lovell110]

Ok

I came into work and tried the zero return on all my machine and there is a little drift like you had mentioned. There was a larger drift on the machine with the limit switches versus the proximity switches. Now that I see all these machines do this I rest a little better. I still have one question! Does the home position have a relation to the accuracy of the parts being made? Example a milled slot depth. I was having thoughts of repeatability problems from day to day. I know checking your parts is the name of the game but getting rid of as many variables as I can makes me a happy camper.

P.S. My machine distance to go number changes both ways! Thanks for the help Geof!

[Geof]

Originally Posted by lovell110 .... I still have one question! Does the home position have a relation to the accuracy of the parts being made? Example a milled slot depth.....

Changing the home position changes the position of your Work Coordinate system, G54, or whetever you are using. So if the only operation you were doing on a part was cutting a slot then the slot would be moved by the same amount, but if you where machining all over the part it would not make any difference.

[miljnor]

I dont know how the distance to go works on your newer machines but on the older machines it has to have a reference.....Ie somewhere its supposed to be.

So if you tell it to go 6" then it has something to look for....when telling it to go home it has no reference and is just basically counting pulses and giving you some random distance to go....so variations can be had since its determining DTG from a random source (maybe not so random but definitely not defined by you)

And as always ignore if you already have an answer to your problem

[Geof]

Originally Posted by miljnor I dont know how the distance to go works on your newer machines but on the older machines it has to have a reference.....Ie somewhere its supposed to be....

It does seem to have some function during homing. We had the Z axis servo replaced on our HL1 and when he had finished the technician adjusted the grid offset and checked the value in the distance to go after zeroing the Z axis.