PDA

View Full Version : Help Please With Auto Tool Height Mach 3 Problem



howser37
10-06-2014, 05:49 PM
HI, have been reading these forums for quite a while and after eventually deciding i didnt have the skill or knowledge to build my own machine at the moment, have recently purchased a heiz S1400 CNC second hand, with very little mileage.

It came with a computer with mach 3 already set up and so far i have been using it as is, without changing any settings

The CNC came with an Auto Height setting puck that plugs into a terminal on the z axis gantry and i am suing mach3 to set the auto tool height.

At first it was a little intermittent, ie sometimes it would work fine and sometimes it would not and i would have to gog the machine to a slightly different postition, however now it has stopped working completely, so that when the z should back off when touching the plate, it carries on going and i have to hit the emergency stop button, this is happening every time now.

I have done some googling but have not found anything that i have either found to help or that i understand lol

In Mach3 whilst looking at the diagnostics page the digitize led is on when the probe is not plugged in and goes off when i plug the probe in, the led stays off when the bit is touching the plate.

Not sure if that info is any help, but might be a starting point

Anyone have any ideas?

howser37
10-06-2014, 06:30 PM
oops, sorry about duplicate thread, i did not think it had posted

Lee Roberts
10-06-2014, 07:50 PM
oops, sorry about duplicate thread, i did not think it had posted

No problem, I've removed the other one for you.

When you plug-in the LED turns off - that’s how it should be as far as I know. The led should come on when you touch the puck with the probe, so I think there is a connection problem, the probing systems I know of are all basic in how they work, as in, all they really do is close a connection.

I would have a good look at the wiring in the probe/puck and see if you can see anything adrift (lose connection), if the "probe" part of the system is simply the cutter, see if you can find out where/how the other side of the circuit is connected to the spindle.

.Me

howser37
10-06-2014, 08:20 PM
Hi Lee, i will take a look inside the puck again and see if i can see any problems, it is the kind with a plastic button that presses down in the middle with what looks like 3 arms sat on ball bearings inside, and i assume that when one of these arms is lifted off a ball bearing the circuit is broken and the digitize led should register, but it does nothing.

So i assume it cant be related to the bit/spindle as there is no electrical contact there

regards
Adam

howser37
10-06-2014, 08:51 PM
cant see any problems inside the puck, but for some reason it is not turning the digitize led on when i break the contact either by hand or using the bit in the spindle grrrrr

JAZZCNC
10-06-2014, 11:28 PM
Best to start at the source and it's Possibly a lose connection on the breakout board anyway. Trace the wire back into the control box and see if you can identify the input it's using on the breakout board.
Then to test the breakout board hasn't fried it's input take a short wire and place between the (+) input and Gnd(-) watching the digitise input in mach to see if toggles on/off.

If it does then you know it's either the wire or puck. If not then you probably have a blown Input on the Breakout board.

howser37
10-06-2014, 11:55 PM
Thanks JAZZ, i will take a look at that tomorrow, it sounds plausible to me as it was working intermittently before it stopped working alltogether.

howser37
11-06-2014, 03:32 PM
I have checked everything i can think of connection wise, it all checks out ok from the touch plate through the control box and to the computer.

Is there some way of telling on the diagnostic screen of mach 3 if a signal is being received when the puck is pressed.

Mach 3 senses the puck being plugged in and sets the input to pin 15 port 1 in the auto settings but detects nothing from the puck being pressed.

None of the input status leds change when the puck is pressed

Clive S
11-06-2014, 07:01 PM
Have you tried what Dean suggested here
:- Then to test the breakout board hasn't fried it's input take a short wire and place between the (+) input and Gnd(-) watching the digitise input in mach to see if toggles on/off. You can always set up a different input pin on the BOB and then of course in Mach ..Clive

howser37
11-06-2014, 08:06 PM
12554

I am a bit clueless regards checking the input guys.
This is the inside of my control box.

The signal i assume comes in on the ribbon cable via the ST port on the right of the picture and terminates next to the parallel port.

Can anyone talk me through testing it so as i don't fry myself and the control box

JAZZCNC
11-06-2014, 08:57 PM
This will 99.9% be because you have the wrong Input number in Mach3. You can't trust Mach3 auto detect.

Try changing the Probe input to any of these. 10, 11, 12, 13 or 15. Getting it wrong shouldn't hurt anything, if the input is shared by another switch ie: limit or E-stop all that will happen is when you short the probe the limit will trip.
If you go to Diagnostics tab in mach3 on the right you'll see a bank of Led lights. When you touch the probe the Digitize Led should light up. If it's on already without touching then go back into inputs and change the active Low state.

If any of these input numbers don't work then you either have some wiring issue or the board input is not working. If you have limit switches then you could swap the probe with this input to check your wiring, you'll need to set the probe input to same input number as limit.



Do you have a manual or give more info about the model of machine.

howser37
11-06-2014, 09:46 PM
Hi JAZZ, I set the probe to the z home/limit switch pin and sure enough when hitting the limit the digitize led changes, however when pressing the probe plate it does nothing.

The probe plugs into this machine via what looks like 2 core speaker cable and it plugs into a small contol box mounted on the side of the gantry.

I traced this back into the control box via a cable labeled as ST

I just did a continuity test to make sure both sides of the wire were making it through the control box and to the computer, this is about as far as my electrical skill goes lol.

Just so you know i was only using the auto detect in mach 3 as a means of testing, the digitize is setup on pin 15 and was when i bought the machine, the auto height worked when it was demonstrated to me and it has worked intermittently since and recently nothing.

Unless i have changed some other setting which could affect it.

I tried changing to low active on the pi as well, this makes the digitize come on when the probe is plugged in, so i assume it would make it go off if the probe was touched, but as usual there was nothing at all.

PS. also with low active set the auto tool height acted very strangely by seeming to come down one step at a time.

howser37
11-06-2014, 09:51 PM
I forgot to say i also tested an edge finding probe that plugs in the same wayand the result was the same, no reaction to the touch off

howser37
11-06-2014, 09:59 PM
1255512556

This is what i have re manuals and wiring, they are for a different model but listing but basically the same

howser37
11-06-2014, 10:11 PM
12557

Just found this also

thanks for any help

Ger21
11-06-2014, 11:12 PM
If both devices plug into the same place and neither works, then you have a wiring problem between the plug and the PC. If it was intermittent and kept getting worse, it may be a broken wire.

JAZZCNC
11-06-2014, 11:21 PM
12557

Just found this also

thanks for any help

Ok well this shows it's using Limit C for both probe and limit which is pin13. But like Gerry says because you tried it on another input that works then you have wiring fault. (provided you changed the Probe to same number as limit you tried)

To be honest I'm not 100% sure that link is the same controller you have because it says uses a second breakout board and i don't see one in your box unless it's hidden.? I see a ribbon cable disappearing away underneath the base plate where's that going.?

howser37
11-06-2014, 11:48 PM
Thanks JAZZ and Ger21, no they arent exactly the same, but its as close as i can find. Heiz dont seem much use when it comes to manuals to be honest. and i cant find one online for my exact setup.

Mine is the S1400 model, its almost same as the smaller models but with the addition of hiwin linear rails

Mine has no second breakout board, just the one control box.

I dont know much about wiring or what happens in the control box, but is it still possible to have a wire problem when continuity tests out ok?

I assumed as long as the signal from both puck wires is making it to the PC then it was ok, but sure im probably wrong as i dont really understand the technicalities

PS, my control box looks more like the one in the first manual i posted up. Basically the photograph i posted first shows everything in the box

JAZZCNC
12-06-2014, 12:10 AM
Bed time now but this is a very simple concept. When you touch the probe it just completes the circuit from Gnd to Input +

This is why if your input is working and setup correctly in Mach3 then it must be a wiring or probe problem. To test this and bypass the wire/probe then just put a short wire across the pins going to the probe wires.

howser37
12-06-2014, 12:28 AM
Ok thanks JAZZ, i will try that

Clive S
12-06-2014, 08:56 AM
It is the wrong manual but the way I am reading it is. The probe is connected to pin 1 on the 9 pin control plug which goes to pin 15 on the D25 plug to the PP. Have you checked the continuity on the cable pin 15 to pin 15 each end of the cable going from the machine control box to the PP on the computer. Or from pin 1 on the control plug through to pin 15 at the pc end of the cable PP. Some more pictures might help. ..Clive

howser37
12-06-2014, 05:05 PM
Thanks Clive, yeah that's how i read it as well. I have done some more testing just now by jumping the probe pin and the ground pin, first checked just pc parallel port with mach 3 on diagnostic screen (digitize goes out), then end of parallel cable that plugs into output of contol box (digitise goes out), then to 9 pin input of control box ( digitise goes out) all good so far. then traced the st cable back to the what looks similar to an audio jack on the side of the gantry where the probe plugs in, i plugged in probe to the port digitise led goes out(which i assume is correct as it is now making a circuit waiting for the circuit to be broken by the touch off. however pressing probe does nothing as usual, then tried completely different edge finding probe and again this does nothing, (digitise doesn't change on diagnostic screen.

I then gained access to the box containing the port that the probes plug into and jumped both pins with the wire attached to the top of this port (digitise changes as it should) so my thinking is now that there is something wrong with this jack internally?

Here is a picture

12560

JAZZCNC
12-06-2014, 05:14 PM
Looks like a nice rats nest you have there.!! . . . . But you have found the problem so that is good. Personally I wouldn't use any Jackplug has they are too unreliable and you need instant contact with touch probe with no delay thru dodgy contacts. Can't you wire permanently with a coil wire and leave attached to machine.?
what do the probes look like.?

howser37
12-06-2014, 05:22 PM
lol JAZZ, it wasnt a rats nest until i yanked it all out to get to the jack, what i don't understand is if it was a problem with the internals of the jack why would plugging in the probe trip the digitize light every time reliably?

will post some probe pictures later

thanks

howser37
12-06-2014, 05:46 PM
Ok just did a test with an old set of in ear earphones, i plugged them into the jack, and the digitise light went off, in my mind this should not happen as the circuit is not complete, so i am assuming that the circuit is being completed inside the jack and therefore a probe touch cannot break this circuit as it is not just being completed in the probe but in the jack as well

howser37
12-06-2014, 07:31 PM
thanks guys, got it fixed now, it was a loose terminal on the jack

Clive S
12-06-2014, 07:55 PM
thanks guys, got it fixed now, it was a loose terminal on the jack

Very pleased you got it sorted. Happy milling.:applouse: ..Clive