Gmoccapy - A new screen for linuxcnc

Hi Norbert,

At the weekend I made a tool setting probe so that I could use the tool measurement in Gmoccapy. It's an excellent addition to my mill, I can wind the knee up or down during the tool change without having to touch off any tools!

I do have a question/bug though. After running the subroutine to measure the tool, the machine continues with my program, but the G code viewer (below the DRO in auto mode) continues to show the tool measurement subroutine. Sometimes it switches back to displaying my program, and sometimes it sticks on the subroutine, it depends which tool I have changed to! I've played around with my tool table but I cant work out the difference between the ones which work and the ones which cause the bug.

Hope that description made sense.

Many thanks, Tom

Tom ,
sound like it could be noise or the pin needs debounce
have you a pullup on the pin to 5v ?, or perhaps fit a transistor as a ttl switch , to make the transition sharper

i also use screened cables , vfd's can make a mess of low voltage signals

Tom,

unfortunately that bug is known! It is caused from the remap procedure, sending a program load/change signal to the hal_sourceview widget and the preview widget.

I have looked at it for quiete a while and I have a solution in mind, but because of leek of time I could not work it out and present is to the other developers.

It is more a work around, let the coresponding hal procedure check the program name and if it begins with "remap_" ignore the change, not sending a signal.

Give me time to work that out.

Norbert

Hi Norbert, thanks for getting back to me.

No problem. Is there a list of known bugs somewhere? I've noticed a couple of other things but they may well have already been reported.

Basher,
I am sensing the input through an input pin on mesa 7i77 hardware. The inputs sense +DC and have an internal resistor, they must be pulled down when disconnected. My setter is 24v DC and earths through the cutter. To get the 7i77 to sense this I rooted it through a spare relay. I bet that's offensive to anyone who knows about electronics, but I'm an ignorant mechanical engineer. Think a circuit with a transistor would have been better..

tom

Hallo Tom

I only know from two "bugs", one with teh preview using remap and the other one is mentioned in the WIKI, "strange numbers in the info frame".

If you have something more, please let me know.

Norbert

Tom, As per mesa tech you could put a 100 ohm resistor from the 7i77-?? input to the + volts on you 7i77 and switch you hal input to -not I did this with some prox switches that switched to ground when active and it worked just fine.

Microsprintbuilder, that's a good suggestion, I'll try that, thank you!

Norbert, I've only noticed minor things.

One is that since the update from 1.2 to 1.3, gmoccapy no longer remembers that I don't want to show the keyboard on any pages (I just uncheck the boxes now whenever I start gmoccapy)

The other is with coordinate systems: If you're making some moves in G54 everything is normal, the DRO shows G54 and it's also listed in the "G-code" part of the screen. If your program then changes to say G55, the machine responds correctly, but the DRO and G-code list still display "G54", it's a bit confusing. This is only the case in Auto mode, if you do the same exercise in MDI the DRO will show the correct coordinate system..

Tom,
a relay is too slow for this , can you give details of the setter your using , an opto or simple transistor will do
and in fact with some setups you may not need anything , the output of the setter should just go to a 7i77 pin , you may need to invert the gpio pin in your hal file
are you using 24v as your field power on the 7i77 ?

For sensing switches with a grounded common with a 7I77 digital input, a 2K 1W resistor to 24V field power will work fine

One is that since the update from 1.2 to 1.3, gmoccapy no longer remembers that I don't want to show the keyboard on any pages (I just uncheck the boxes now whenever I start gmoccapy)

I have not noticed that and as far as I remember I haven't changed that part of the code, I will check on this, but please do me a favor prior to my work.

Please errase the preference file of your machine ( the name is given in the INI file and the file is located in your config folder.)

After a new start of gmoccapy the file will be created with standard settings. Then change all the settings you like and restart gmoccapy. Does it work now?

The other is with coordinate systems: If you're making some moves in G54 everything is normal, the DRO shows G54 and it's also listed in the "G-code" part of the screen. If your program then changes to say G55, the machine responds correctly, but the DRO and G-code list still display "G54", it's a bit confusing. This is only the case in Auto mode, if you do the same exercise in MDI the DRO will show the correct coordinate system..

IMHO it is caused because of the read ahead of the interpreter, and is similar to a program with differnt feed values, vel is shown correct, but the G code shows other values, because the interpreter has read allready a new value.

What happen, if you finish the program, beeing in G55?
OK if the last command is M2 or similar, you should be back in G54, but if you start and finish with % it should stay in G55.

Norbert