Homing not backing off

[johnswenson1]

I'm in the setup phase and can jog the axis just fine, all fine work great, but when I try and do a home x operation it moves left until the limit switch clicks, it stops, but doesn't back off, the limit switch stays on.

If I try and jog to the right I get an error popup:
Attempt to Move outside safe limits (X - 598.000)

Y does the same thing. (except the number is a little smaller)

I tried slowing it down to 500, that didn't make any difference.

I tried starting at different locations - no difference.

The limit switches all work when manually pressed.

The XYZ current controls are all at halfway except for A which is at 20%.

I'm using the latest software.

I've gone over all the limit switch wiring, it looks good, resistance checks are all very good.

I'm not sure what to do next.

Thanks,
John S.

[JuKu]

There are a couple of settings on TinyG that affect this.

Have you clicked Basic Setup - Control Board Settings - Reset to defaults?

Something is wrong with the control board settings. The easiest way to correct might be to take a screenshot of your current Basic Setup page, and click Control Board Settings - Reset to Defaults. (You might have not done this earlier?) This will reset all TinyG settings to conservative defaults, so you can use your screenshot to restore those that you have already trimmed.

If this doesn't cure it:

You can also compare your TinyG settings to a known good set. Sending $$ to TinyG will dump all your settings to the log window in human readable form. To change a setting, the format is

$xlb=8

...for an example. (My guess is that the xlb and xzb parameters are the culprits here (and same for y), but since these are off, some others may be as well.) The example parameter dump is here: https://www.liteplacer.com/Downloads/Goodsettings.txt.

I'll be happy to have a look as well; do the dump and send it to me (ctrl+A to select all, copy and paste to a text document, send to juha at liteplacer.com).

[johnswenson1]

I tried the board reset, no difference. I dumped the tinyG, compared it with the settings in your file and xzb, xlb and xlv were the same. I tried changing xlb to 20.0 and it made no difference.

I tried swapping the xmin and xmax switches, that didn't make any difference, no matter which is which it stops when the switch trips and doesn't backoff at all. That doesn't seem right.

I went through and did continuity tests on all the switches from the contacts on the switches to the screws on the tinyG board, they all seem to be behaving correctly: all the switch inputs are tied to 3.3V until one switch is activated, at which point one and only one wire gets connected to gnd.

I'm at a loss right now. I'll email my settings to you.

Thanks for the help,

John S.

[JuKu]

I got your settings; nothing is wrong as far as I can see. I experimented on my machine, and this is the expected behaviour:
-When homing hits the correct switch, it backs up slowly the latch backoff amount (normally 8mm on X and Y). Then it backs off faster zero backoff amount (normally 2mm)
-If a wrong switch is hit, nothing happens, the machine just stops. When the wrong switch is released, you get an error message and TinyG needs a reset to get out of the error situation.

Maybe you do hit a wrong switch, but you don't see the last point, because it is not easy to release the switch, as the motors are still engaged. (We found a bug in TinyG firmware: If a limit switch is hit, motors should be powered off, even when homing. At normal operation, limit switch disables the motors as expected). Please try this:

Send command $mt=1 to TinyG. This causes the motors to power off one second after a move is completed. Try homing one axis. You hit the switch, the machine stops. The motors now power off in a second. Manually move the machine and clear the switch. If you now get an error message, this means that the switch that triggered is not the switch the homing operation was waiting for. (Homing Y, moving the bat towards you, should trigger the switch on left corner pillar. This is Ymin. Homing X moves the gantry to left. The switch that triggers is Xmin.)

The motor timeout needs to be bigger, as the absolute position is lost at motor power down. The command to set it back to the default 5 minutes is $mt=300

[johnswenson1]

Hi Juha,
thanks for looking into this, I got it fixed! It now homes properly.

It turns out I had TWO wiring errors which were making it tough to track down.

1) The wires from the jacks on the control box to the TinyG had min and max transposed for all the axis.

2) The labels on the control box end of the cables had zmin and xmin transposed.

When I got both those straightened out it all worked fine. I don't know what in the world was going on when I wired this thing up.

Now I can actually do the calibrations and start using this wonderful tool.

Thanks for the help.

John S.