CNC Torch-Router Table 2 - Overall Machine - Concept

Yes, I can see the need for a flexible i/o board with wireless communications capabilities. In the short term, linuxcnc could be useful -- though it may well be a step in a irreconcilably different direction.

I could draw a system diagram but all that would really be shown is a mapping to parallel port pins. You connect things to parallel port pins and then delve into HAL to tell the system how it should communicate with those pins.

A LinuxCNC system consists of:

CAD -> gCode Generation via CAM software (no great OS software i'm aware of) -> LinuxCNC -> Stepper encoder (if necessary - depends on step generation capabilities of pc) -> stepper drivers -> stepper motors.

THC is connected through parallel port pins and is configured in the hardware abstraction layer. Gas on/off can be done the same way. Endstops are connected to parallel port pins

There are wizards which pretty much do all the configuration for you included in the LinuxCNC software for specific setups but some manual configuration in a terminal is still required.

I think LinuxCNC is good for a quick n easy setup (relative to the ose case) but, as you say, it requires 1 PC per machine.

Torch Height Controller Concept

Now, an update on my work with the capacitive sensor:

I am getting closer to reading good data out of it. My current focus is the serial data communications between the arduino and the ad7747. I think my method of reading the data and scaling it is a bit whacky. This is due to my lack of experience with i2c and arduino and also to a lack of documentation on reading multiple bytes. I've been getting closer, i think, but more importantly: i'm getting much better at debugging my work. I have been utilizing the serial monitor to act as a debugging platform by "printing" information that shows me where the program is executing and what data i'm getting out of it. Paul sent me some code that he thought would fix this problem but it appears that it does not do what he intended. I have not been able to work for long intervals of time on this project which is another thing I need to investigate.

My goal was originally to spend not much time on this capacitive sensor. Now I think I have a unique opportunity to create a powerful product that can be used in systems outside of our own torch table. I would like to create an open source version of something like this: http://www.agelkom.com.tr/prod02.htm. I would like to spend more of my time working on this project and less time working with the interns. It would also be great if I had help from Gabriel. His knowledge of higher level languages would compliment my understanding of embedded systems and analog/digital electronics. So what I would like to propose is that I be excused from the daily meetings and instead have my own meeting with gabriel in the electronics room and work on that all day until it is ready to be integrated into the torch table. There are some development points that I can foresee: Sensor data calibrated H bridge driver integrate GUI written in python or similar is created by gabriel Lasersaur-controller -> arduino interconnectivity established so first we need to finish the data acquisition process from the chip

then we need to drive a motor (servo would be simplest) for arduino-controlled z height settings

then it would be nice if we could either integrate this THC(torch height controller) into the lasersaur front end or create a separate gui that runs on the beaglebone

I was able to work on this project only 2 real days last week which is not ideal because every time I go do something else I start thinking about other things and lose the train of thought I had. I do keep notes but still it is hard to jump around from thing to thing. In general, i'm in no rush to finish this. That's when I start making mistakes. I think a good, achievable goal for myself is to push forward the development of the capacitive THC module because that has applications in a plasma table for initial height sensing. The open source standard initial height sensing on plasma tables is using an endstop triggered by pushing the tip of the torch into the plate and then retracting a known amount. So a capacitive height sensor would be useful there as well.

What do you think? Are you in support of me spending much of my time on this project and ignoring the myriad other tasks I am well suited for around here? I'm pretty much happy doing whatever but I came here to learn so the things I'm excited about now are this THC module and the velocar workshop.