Machsupport Forum
Mach Discussion => Brains Development => Topic started by: drobertson123 on July 01, 2019, 04:34:46 PM
-
I am attempting to use a brain to multiplex a touch plate and Auto Z Axis probe.
The touch plate is normally open and the probe is normally closed. This limits my ability to directly wire them together in one port.
My attempts to do this with a Brain have been very frustrating. I set up Input #1 and Input #2 as the inputs to my two different probes. I then created the Brain so it would us an OR to trigger the Probe Input if either of the Inputs got triggered.
This actually works fine. I can test it easily and both the touch plate and z-axis probe will cause the Digitize/Probe LEDs to go on and off in my main screen. The problem is that the G31 probe command doesn't seem to respond to this artificially create Probe input. Even though the Digitize LED lights on the screen the G31 command is still moving the probe down.
I did some independent tests with the touch plate and Z-Axis probe independently wired to the Probe signal. Both work perfectly fine. It only fails when the signal is generated by the Brain.
I attempted to disable the port config for the probe just in case that was interfering but then any attempt at probing resulted in a “No Probe Active” message and nothing happens.
Here are some combinations of settings I tried for the Probe Port settings;
Enabled: True Port: 1 Pin: 13 Emulated: False -> Works from touch plate but not from Brain triggered by Z Probe (Port:1 Pin: 13 is the correct setting for the touch plate)
Enabled: False Port: 1 Pin: 13 Emulated: False -> Probe is not triggered by brain, G31 returns No Probe Active
Enabled: True Port: 1 Pin: 13 Emulated: True -> Probe/Digitize is triggered from any of the inputs from the Brain but it does not stop the G31 and get a sample for Z height
The main problem seems to be that the brain triggered Probe output won’t cause the G31 to react if the probe is enabled but the G31 command returns “No Probe Active” if the probe port isn’t enabled.
Just in case it is relevant, I am running version R3.043.066 of Mach3.
-
Hi,
just as a matter of interest Mach4 has four probe channels.
Craig
-
what Controller are you using? or just PP and BOB?
-
He's using an ESS.
Just in case it is relevant, I am running version R3.043.066 of Mach3.
Actually, it is relevant. Warp9 recommends 3.043.062.
-
Thanks Gerry
Yes, I am using an ESS from Warp9 and a C11G BOB from CNC4PC.
I actually was running 3.043.062 when I first had this problem. All the same failures existed. I tried upgrading to the .066 version as an attempt to fix the problem.
If there is an issue I can roll back to .062 but so far I have seen no difference in the performance of the two versions.
-
so if you are using an external Motion Controller (ESS) i am pretty sure (i know definitely from CSMIO) that
the "real" probing (G31) is handled on the Motion Controller.
IMHO you have no Chance (i had no Chance on the CSMIO) to "manupilate" probe Input via brain.
the only way, in my case, was to modify my probes by external Hardware to be NO contact's to wire them
parallel.
-
I was worried about that.
There are some LM339 quad comparators in my parts kit. It shouldn't be too hard to make a basic multiplexer from that. If that fails it will be on to Mach4 or something else. Having only one probe input is extremely frustrating.
As an alternate idea. Is there any way to switch the active Port and Pin settings for the probe using a macro? I could just do that with a button on the screen if I needed to.
-
As an alternate idea. Is there any way to switch the active Port and Pin settings for the probe using a macro? I could just do that with a button on the screen if I needed to.
i was looking for a possibility to modify pot&pin Setting via VBScript, but had no succes, witch does not mean there
is None.
-
The NO and NC switch contacts could be ORed into one signal with an NPN transistor and a two resistors.
The NC input is inverted so the output is NO for either input.
The resistor values are not critical and there are many small signal NPN transistors that would work in the circuit.
-
Wow, thanks for that. You design is simpler than mine and I have the parts right here in my project box.
I was using an LM339 Comparator for the same concept. It would have worked but your circuit is simpler and easier to do.
One quick clarification of your design. I assume that the other side of the NO and NC switches are going to V+?
Thanks
-
The other side of both switches go to ground.
The NC switch normally shunts the current from R1 to ground. When the switch operates R1 is released to turn on the transistor, pulling the output to ground.
The NO switch directly pulls the output to ground.