Boxford 125 lathe Tool Changer Comp FERROR (SOLVED)

Another thing I find a bit strange are your stepgen settings:

setp stepgen.2.steplen 1
setp stepgen.2.stepspace 0
setp stepgen.2.dirhold 35000
setp stepgen.2.dirsetup 35000

that means you have a step length of 1 nanosecond and a step space of 0?

This is what mine look like:
# these are in nanoseconds
DIRSETUP = 2500
DIRHOLD = 1000
STEPLEN = 1000
STEPSPACE = 1000

Aciera wrote:

Another thing I find a bit strange are your stepgen settings:

setp stepgen.2.steplen 1
setp stepgen.2.stepspace 0
setp stepgen.2.dirhold 35000
setp stepgen.2.dirsetup 35000

that means you have a step length of 1 nanosecond and a step space of 0?

This is what mine look like:
# these are in nanoseconds
DIRSETUP = 2500
DIRHOLD = 1000
STEPLEN = 1000
STEPSPACE = 1000

I think the stepgens are correct 1& 0 are normal and mean something to Lcnc this is from stepconf my-mill
setp stepgen.0.steplen 1
setp stepgen.0.stepspace 0
setp stepgen.0.dirhold 35000
setp stepgen.0.dirsetup 35000

Re the Adir
T2 False
T3 True
T4 F
T5 T
T6 T
T7 T
T8 F
T7 F went forward correct
T8 T
T6 F went reverse
T7 T forward
T8 F
T1 T
T3 T
T8 T
T6 F went reverse
T3 F went forward correct
T1 T
T3 F went forward correct

The tool numbers get mixed up to the actual tools ie. they end up out of sync

Thanks for your patience.
My files should open ok on any 2.8 version and don't need a machine connected as it a P/port. The above was done on the real machine

So the Adir pin has no correlation to the actual rotation direction of the stepper?

[edit]
Ok found this in the docs. Still not sure you got this setup right though,

Aciera wrote:

So the Adir pin has no correlation to the actual rotation direction of the stepper?

[edit]
Ok found this in the docs. Still not sure you got this setup right though,

(float) stepgen.<chan>.steplen - Length of a step pulse (step type 0 and 1) or minimum time in a given state (step types 2-14), in nano-seconds.

(float) stepgen.<chan>.stepspace - Minimum spacing between two step pulses (step types 0 and 1 only), in nano-seconds. Set to 0 to enable the stepgen doublefreq function. To use doublefreq the parport reset function must be enabled.

Ok for the avoidance of doubt I will hard code them in.

I set them to 20000 in the hal file. Unfortunately It made no difference

When it goes the wrong way I don't think it does the reverse move probably because it already going that way.

I guess I was barking up the wrong tree, sorry about that.
Anyway i installed it on my buster 2.8 and I am playing around with it.
I noticed that the "adir" signal sometimes gets reset to 0 at the end of a tool change but mostly not.

The tool numbers get mixed up to the actual tools ie. they end up out of sync

You mean that "toolchanger.currenttoolnumber" in halshow does not correspond to the commanded toolnumber or that the physical toolchanger does not move to the correct position?

Aciera wrote:

I guess I was barking up the wrong tree, sorry about that.
Anyway i installed it on my buster 2.8 and I am playing around with it.
I noticed that the "adir" signal sometimes gets reset to 0 at the end of a tool change but mostly not.

The tool numbers get mixed up to the actual tools ie. they end up out of sync

You mean that "toolchanger.currenttoolnumber" in halshow does not correspond to the commanded toolnumber or that the physical toolchanger does not move to the correct position?

Yes. the physical turret get out of sync when it goes backwards. Then pressing T1 does not go back to its correct position.

From what I understand I think the toolchanger is controlled by the comp .the readme seems to say that it hijacks the stepgens.

I do not understand C or Python . I notice in the files the guy that did the comp left hhis email there but its an old file.

I suppose I could set up a stepper motor with a cardboard disc on it to act like the turret !!

I notice that the toolchanger.position-cmd for a specific tool shifts as I change tools. I would have expected that the respective tools position remains the same.

[edit]

I wonder if this has something to do with this "fudge_factor"

Aciera wrote:

I notice that the toolchanger.position-cmd for a specific tool shifts as I change tools. I would have expected that the respective tools position remains the same.

[edit]

I wonder if this has something to do with this "fudge_factor"

{
                        position_req -= ((moves / divisor) + fudge_factor);
                        if(position_req < 0)
                            position_req += 360;
                        position_cmd = position_req;
                        progress_level = 3;
                        }

I notice that in the side panel there is a jog control according to the docs it is only used to set the reverse to lock the turret by adjusting and then putting the setting numbers into the comp. The jog buttons for/back don't move the turret to the next tool it is only for getting the numbers in the comp correct.

Ah, I was wondering what those Jogging buttons were for. The back jogging didn't seem to make any sense to me at all.

Aciera wrote:

Ah, I was wondering what those Jogging buttons were for. The back jogging didn't seem to make any sense to me at all.

There are no sensors on the turret so it has to move past the ratchet then reverse until the motor stalls.

I gather you have seen that the odd numbers degrees are different than the even numbers.

Ie they are not at 45'