No I/O LED operation in demo mode - bug ??

[beefy]

First of all doing all this in demo mode.

In version 1.2110 I could use MDI for M3/M5, or say M10.1/M11.1 and the associated I/O port/pin LED on the diagnostics tab would turn on/off.

However, trying the same thing in 1.2113 and I get nothing happening on the I/O port/pin LEDs.

Selected the same board / breakout board combo in both cases.

One more question, are there no status LED for M10/M11 or M10.1/M11.1, etc ??

Keith

[dezsoe]

Hello Keith,

You can check the output LEDs, there are no special LEDs for the M10.x outputs.

I made a quick test: AXBB-E demo, M10.1 set to port 3 pin 1. In the diag I see the output changing. How was your setup?

[cncdrive]

Hi Keith,

It works fine at me.

[beefy]

Re-checked again.

This time tried in both Standard and Plasma profiles, same results.

First chose AXBB and everything worked fine with that one. Diag output pins came on, and went off as expected, as well as the Spindle CW LED operating correctly.

2nd chose UC300ETH-UB1 (that's what I used before) and diag LEDs do not light in Standard or Plasma profiles.

The Spindle CW LED came on when it should in response to M3 in the gcode, but the diag o/p LED did not light.

Had Spindle CW / M3 on Port 3, Pin 1. Spindle enabled, no delays.
Had M10.1 on Port 3, Pin 2

If needed here's the two profiles:

Default.pro

(36.54 KiB) Downloaded 655 times

Plasma.pro

(35.16 KiB) Downloaded 661 times

Keith

[beefy]

Another update.

This time went to the workshop for a "real life" test. Using v1.2110 connected to UC300ETH mounted on a UB1 board.

Using plasma screenset and if I press the Torch On button, everything works normal, diag. output LED comes on, and the output on the UB1 activates.

In this condition I type M10.1 into MDI and press enter..........nothing happens.

I click the Torch On button (currently on) to turn it off, and........Torch o/p turns off and M10.1 output turns on.

So it seems the M10.1 command had been acknowledged internally but was blocked by the active M3 / Torch On / Spindle CW signal, and removing the Spindle CW signal allowed the M10.1 to activate.

My custom designed THC is going to make very good use of M10.1 so it's thrown a spanner in the works to find M10.1 not working correctly.

It's interested also, that in demo mode I experienced the problems mentioned in the prior post, but only for the UB1 and not for AXBB.

If you have a UB1 board could you confirm this bug in v1.2110, and of course check that it is, or is not there in v1.2113

I don't want to go upgrading to v1.2113 if this bug is confirmed to also be in there because I have to make screenset changes for my THC every time I upgrade.

Keith

[dezsoe]

Hello Keith,

I've just run some tests. I tested only the 1.2113, because this is the version which will be updated. (Always the last.) The diagnostics page for UB1 in the new screenset is bad. (Most of the LEDs are #14, the magic of copy-paste...) If you switch to the old screenset then everything works fine. I've tried all that you wrote: M10.1 and M3 are independent and are switching the LEDs on the diag screen. I don't have an UB1, so I could test it only in demo mode. In demo mode the with the plasma profile it worked too.

[Robertspark]

you could just change the jumper on the UB1 to use a UC300ETH screenset (which is what I use with the UB1) so that both analog outputs work I think I made that decision.

[dezsoe]

I think it's better to fix the screensets. I've just made it, here you are.
(The Default2019 has some more fixes.)

These screensets are for version 1.2113!

[beefy]

Thanks very much everyone.

I'll try the barebones V1.2113 on my table but without my THC plugin installed, just to make sure that is in no way affecting UCCNC.

Keith

[beefy]

Hi Dezsoe/Balazs,

just an update for you guys. All seems good now.

Got the V1.2113 2019 screenset modified for my THC and installed on the workshop computer.

Even with my THC plugin installed and running, the operation of M3 and M10.1 now seems perfect.
After an extremely tiring/frustrating/confusing bug (simple once I found it) in my THC microcontroller code, it was great to see everything come together.

So I'm completely stumped why my previous install had the problems I mentioned earlier, but who cares now

Thanks again for your help.

Keith.