Rods "Spaceship" Scratch built Plasma Cutter build

Robert, that was not addressed at you, i thanked you in the above post before i replied as i did agree with all of what you said, save for the little bit that Linux is hard, it was hard, now it is much much easier than any windows for normal users.

tommylight wrote: Robert, that was not addressed at you, i thanked you in the above post before i replied as i did agree with all of what you said, save for the little bit that Linux is hard, it was hard, now it is much much easier than any windows for normal users.

Tommy. sorry if I offended you. I hold you in the utmost respect and you have given me a lot of guidance on this journey of mine. I know you have had great success with your proma config and others have too.

This build has become an experiment in the use of the external offsets branch and guided by a desire to build a THC within linuxcnc rather than rely on an external THC controller. I have an external controller here to fall back on but at this stage I'm not done with external offsets yet.

Since I've started this, there are at least 3 others who have built working systems using PID based control within Linuxcnc that are giving good performance. Eventually, I am sure a working solution will emerge and I truly hope it can be locked down into a standard part of linuxcnc. The reason why I want to do this is that it has been said that the best THC's are built into the motion controller (eg. like the Hypertherm controller mentioned a few posts ago). Because Linuxcnc is_the_motion_controller and we have a means to accurately read the torch voltage into linuxcnc, I can't see why this can't be done.

rodw wrote: Tommy. sorry if I offended you. I hold you in the utmost respect and you have given me a lot of guidance on this journey of mine. I know you have had great success with your proma config and others have too.

Because Linuxcnc is_the_motion_controller and we have a means to accurately read the torch voltage into linuxcnc, I can't see why this can't be done.

Rod, you have never offended me. Even if someone did offend me i would not bother to reply.
On the other hand, i get really p/o/ when some individuals insist on what is absolutely not true about Linuxcnc.
Linuxcnc is extremely flexible and can be made to do almost anything as far as CNC goes, so i have no doubt you and others can make it work, and Icelander has a working solution already.
Moving on.

Tommy. I think with the benefit of hindsight a plasma machine is not recommended for a starter CNC project. A simple milling machine should be able to be configured and working in a day using stepconf or pncconf if a new user takes the time to read the introductory docs. And I agree the current Linux distros are a no brainer to install and much nicer than Windows today.

Anyway back on topic! I mucked about replacing the faulty proximity sensor with one laying around that initially tested as working only to find it was faulty once installed (which explains why it was laying around no doubt). So I'm waiting on parts again.

This looks like its the second time this same sensor has failed. I'm wondering if it was becasue it was an NPN sensor and required a pull up resistor to work with my Mesa card. The new one is a PNP one so I should be able to remove the resistor. Maybe it will last longer.....

Well, the last round of testing came to a sudden stop when the proximity sensor on the Z axis home switch failed. This is the second time its failed. Originally, I bought a swag of NPN connectors and I added some pullup resistors.

So this time I ordered some PNP sensors and thought it would be easy enough to unhook the relevant resistor. You can see it in the foreground with one leg waving in the air.



But I must have created a short when unsoldering the resistor and I let a lot of smoke out of the sensor. As per usual what should have been a 10 minute job took hours so I did not get any more tests done this weekend.

So I have a question for the brains trust. Should proximity sensors be protected with a current limiting resistor?

rodw wrote:
So I have a question for the brains trust. Should proximity sensors be protected with a current limiting resistor?

Yes... But like everything.... It depends.... On your breakout board circuit.

Am I reading the resistance correctly.... 11k ohm?

Thought that would be a bit high... What voltage are you running the limit switch circuit at?

Please bear in mind that proximity sensors have a leakage current.... They are not quite digital... If the resistors too high there will be a voltage across the resistor when it's off and it may be high enough to imply that the sensor is on. This is more of a problem if you have electrical noise or if you run the sensors in parallel....

www.google.co.uk/url?sa=t&source=web&rct...28UbibFqqS8tDHGudUFr

www.google.co.uk/url?sa=t&source=web&rct...a-5OKbX1-rnzI6FOHKTf

Some good info , just word search "leakage"

Also the recommended wiring arrangements for sensors (more an issue if you have multiple sensors on one input)

Another one

www.google.co.uk/url?sa=t&source=web&rct...hhJrQw376pMyRQOsmpAt

Hi Rod,

You have no luck. Why not try IFM sensors 24vdc in combination with Finder relais 24Vdc? Then you are also more protected to your Mesa equipment.

Thanks guys, I think its all fine. I did notice in Robert's second link that they used a 100 ohm resistor with a couple of diodes so I might try adding that to the new PNP sensor and worry about the others when they fail. Fortunately, I've left a fair bit of room for extra components on my distribution power board on top of the gantry.

Really, so many of my issues has been a lack of knowledge of electronics, crappy soldering skills and failing eyesight!

I have not really had any mechanical issues, its all been electronics.