![]() |
#211
|
|||
|
|||
In order to keep the 'sparky stuff' simple for us (and future noobs), although your right, i'd like to add it's not necessary to add the 7408 to the schematic (which makes it more complex and daunting).
The 4N27 with the simple led and resistor schematic I posted will work (I tested this). Disclaimer: with Eding Cpu5b, but I see no problem why it should not work with Mach3 either ? |
#212
|
|||
|
|||
Fox,
I agree that your schematic will work, but I also know that it will not work forever. Thirty-nine years of experience building process control computers lets me know that attaching multiple outputs to a single input will fail - eventually. A 7408 chip cost about $0.25. It solves the problem that attaching multiple inputs to a single input creates. I'm certainly not saying that you are wrong and that I am right, but experience is a valuable teacher. If the goal is to build a robust system that will run reliably without operator intervention, then "tried and true" electrical engineering principles should be considered. To me, there are few things in life as annoying as troubleshooting a circuit that worked yesterday but doesn't work today. Using a 7408 chip follows approved practices. I have built an interface board using a 7408 chip, but I replaced that board with a much more functional board that uses a microprocessor. The important point being that attaching an output to its own input follows standard industry practice. |
#213
|
|||
|
|||
Mike, your posted image is rather small and hard to read. You do not have the full size uploaded somewhere with a link to it so we can see it better do you?
|
#214
|
|||
|
|||
I am not using a single input.
|
#215
|
|||
|
|||
Pete,
I'm sorry about the quality of the image. I use Eagle, which exports the image. As far as I know, there are no controls to darken the image. The resistors on the diode side should be sized to allow about 10mA of current to pass through the diode (1.2k for 12V and 2.4k for 24 volts, or 1k and 2.2k). The resistors on the transistor side simply "pull up" the voltage. Those resistors can be 4.7k to 10k. Fox, Sorry about the mix-up. I saw the schematic that showed all outputs going to the same input and my old mind played tricks on me. I thought that it was your schematic simply because you posted another schematic earlier that was somewhat similar. Yes, your schematic shows one output going to one input. If your break-out board has enough inputs, your design would work perfectly. However, I would suggest that you add a pull-up resistor (4.7k to 10k) to each input just so that the input is pulled HIGH until the transistor of the 4N27 pulls it LOW. For many years the military required all inputs to be pulled high through a 2.2k resistor before giving the design a military spec rating. Many microcontrollers have a built-in 100k pull-up resistor on all data lines. I still add an external pull-up just in case I forget to turn on the internal resistors through the program. Floating inputs are hard to diagnose when the input randomly seems to switch states. Some break-out boards have a SIP resistor strip that handles that chore, but others do not. |
#216
|
|||
|
|||
Thanks for the advice, mine has indeed enough inputs and internal pullups ... but yes, basically if it does not work add a lot pull up resistor. Should also be noted in your manual ( that info is in my manual)
|
#217
|
|||
|
|||
Steve Stallings at PMDX designs break-out boards properly. He uses a 10k pull-up resistor on critical inputs. His 10k plus an external pull-up is equal to 5k, which is very close to the 4.7k value found in many "how-to manuals". Arduino manuals typically suggest using a 10k resistor.
Other break-out boards don't include pull-up resistors. Because many CNC builders struggle with electronics, adding the external resistor will help if their BOB doesn't use a pull-up resistor and it will do no harm if their BOB does have the resistor. With the proper tools, it's easy to see how signals work, but I would guess that few CNC builders have an oscilloscope and the other tools required, so when something goes wrong, people get frustrated, especially if the circuit works sometimes and fails other times. |
#218
|
|||
|
|||
I'm back, the princess dragged me away from my 'Machine" to go on vacation. Yellowstone and Jackson Hole. Worth the 2400 mile drive.
Connected my motors to the Gecko outputs and the inputs to the PMDX-122. After a day of playing with addresses and jumpers was finally able to get the motors to jog from the keyboard. Finished cutting and installed the spindle mount, installed the spindle. |
#219
|
|||
|
|||
Added an internal ground wire to the spindle and connected the power cable and water cooling.
Connected the VFD. Manual for the first run. Ran the spindle as Fox suggested, 20%, 50%, 20%, 50% and 100%. 25 minutes each. Coolant temperature rise (input to output) was 2 degrees F with 8.8 GPH flowing. Very quiet. Much quieter than the smallest handheld router I have. |
#220
|
|||
|
|||
Looks great. Makes me want to get on the ball and get a spindle going.
Mark |
#221
|
|||
|
|||
Nice !
I made a mount for my coolant reservoir under the table today... where's yours going to be ? |
#222
|
|||
|
|||
Right now I have it in a plastic box on the floor. Going to hang it under the table as soon as I find the right container.
|
#223
|
|||
|
|||
Nice, well done - neat looking panel, now lets see some dust...
![]() |
#224
|
|||
|
|||
Found some mini boxes to cover my motor terminal blocks.
Also did an air cut of a 10" circle .1 in deep worked fine. Then I performed the calibration of my travel in Mach3 and after Mach3 it picked the new number of steps the motors got really loud. I'll follow up with some screen shots of my Port Setup and Axis Selection and the Motor Tuning and Setup. |
#225
|
|||
|
|||
I like your method of connecting the stepper motors better than using the connectors I used. If I find myself swapping out my gear steppers for belt drives, I think I will do what you have done.
|
#226
|
|||
|
|||
I use a standard size electrical junction box. thats the only box that I can find off-the shelf.
|
#227
|
|||
|
|||
Thanks DB, I figured if I used a connector it would need to be mounted some how anyway. The terminal block seemed like a good compromise until Ken said something about a finished look. These boxes caught my eye at the electronic supply store.
|
#228
|
|||
|
|||
Clean look;
what I still don't really understand why you not just run them to your e-stop boxes ? I end mine with XLR connector into the side of the e-stop box, but you could also just use a strainrelief ? |
#229
|
|||
|
|||
Fox,
The E-Stop boxes I got from Mike were set up with 3 switches. E-Stop, Start and Stop. There was hardly enough room for the proximity wires. Also it didn't seem right to run noisy un shielded motor wires in with the control wires. My motor wires are also 16 awg (surplus wire). More of an evolution taking place rather than forward thinking design. Based on earlier discussions I decided to run 4 separate proximity wires back to the 122. I have thrown out the start and stop buttons in favor of more room in the E-Stop box. Don't see any reason for them at the present time. In hindsight the E-Stop boxes would be more accessible if they were on the ends of the main beams. Chasing an e-stop button for that matter a stop button that is on a moving or runaway gantry seems odd to me now. That's how it ended up where it is. Kind of a drifting design. John |
#230
|
|||
|
|||
Its a good idea to have the motor wires shielded as well.
|
#231
|
|||
|
|||
John,
I (would) use shielded motor wires indeed. You wires are mixing anyway in you e chain. I also played with the idea of adding 4 estops on each end onto the downwards pointing 45 degree section of the x beams ( saves them from bumping into them by accident, still easily slapped in an emergency) but instead i added only 1 physical e-stop with loose extension wire on top of my pc cart that can move with me. The estop on the gantry makes sense when you think some more about it, though. It's the part that can hit you, cut your hand or a cut a cable etc etc. It's convenient to be able to resume the action from the same box when you stopped it, but not necessary indeed. I would advise you to put an e stop on the z, funny things can happen when changing a bit, holding material etc, and you want an estop in reach then! I use bigger boxes for the e stop so you can add pcb boards, buttons and the wires more easy. Last edited by Fox; Tue 15 July 2014 at 15:58.. |
#232
|
|||
|
|||
Fox,
My motor cables are shielded. You have a good point about an E-stop near the bit change area. I view an "E-Stop" as a button that removes all power from everything and it doesn't care about Mach 3 conditions. A real Emergency! Then there is a 'Stop' button that tells Mach 3that you want to stop but also will continue. More of a Pause Button. The E-Stop is easy to do. It goes through the latch on my contactor. Where do I connect the Stop/Pause Button? I saw something written on this but don't remember where. |
#233
|
|||
|
|||
You can read about setup buttons on this thread: http://www.mechmate.com/forums/showthread.php?t=281
Your machine looks almost complete. I started cutting with mine very plain and I'm still doing improvements. Only miss vacuum hold-down... Did you use the 4N27 Opto Isolators? |
#234
|
|||
|
|||
At 3~6A and over 60Vdc, the stepper motor power cable is the biggest source of noise. Shielding it is to keep the noise in rather then out.
You should worry about it affecting others rather then worry about it being contaminated. ![]() |
#235
|
|||
|
|||
Thanks Paulo! That's the one I remember seeing. I received the 4N27s but decided it was easier to run the 4 separate wires to the PMDX122 since the ports were available. Maybe in the future after I get the machine running.
All of my cables are shielded. |
#236
|
|||
|
|||
Remember that PMDX122 has only 4 inputs you can use: 11,12,13 and 15. I'm using 11 for Z- zero, 12 for pause/start (toggle), 13 is damaged and only just 15 for limits.
Last edited by parrulho; Wed 16 July 2014 at 14:38.. |
#237
|
|||
|
|||
I used Brains for the feedhold button. Not that hard to set up.
|
#238
|
|||
|
|||
Good to know. Thanks Pete, I'll look for information about it.
|
#239
|
|||
|
|||
Quote:
![]() Cheers Bruce |
#240
|
|||
|
|||
Sorry for such a long wait for a response. I watched the Brain instruction videos linked to from Mach website and then just played with the Brain editor to figure it out. Basically you add the input you want to use by clicking the plus + on the toolbar, after clicking Ok click back on the input you just created to highlight it green (sometimes Brain is not friendly and will not highlight, just click anywhere on the screen and click on it again and it will turn green. Now add lobe logic (no operation (pass through) or Invert), click the plus + button again and when done click Ok, finally click on the lobe logic box so it is green and then click on the upside-down green T on the toolbar (terminator), for feedhold, click on ButtonPress and select the button press action you wish to do, in this case feedhold. Save the Brain and the enable it from the Mach Brain Control. You can use View to watch the operation of the brain logic.
|
![]() |
|
|