motor controller / possible power supply problem

Hi everyone. I've finally got my Darwin assembled (minus extruder) and it's all very exciting, but as I was wiring up the axes things started going wrong with the motors.

I'm using the Arduino electronics. I hooked up the Z stage first, and it was working normally in the stepper exerciser, as long as I used the single arduino firmware. I didn't have any problems with the Y either. But by the time I had wired the X axis my stepper exerciser tests weren't working: the Z axis only makes a single step at a time using the buttons for a manual step, and the Y axis responds to the location slider but will only move forward, not backward. The X is now the only axis working okay.

Does anyone know what it means when you can single step, but try to set a location to move to in the stepper exerciser and nothing happens? When I do this I can see the rx light blinking on the comms board but I see nothing from the controller board.

I noticed that my power supply was giving a low 12V voltage so I followed Nophead's advice to put a 10ohm load on the 5V supply. This has still only brought my voltage up to 11V though. Interestingly, when I disconnect power from everything but one controller board and power the Arduino over USB, I can give the Z axis a full 12 volts but then the stepper exerciser doesn't work at all, not even single step. As soon as I bring up the window for the exerciser the leds on the Z controller board show both red and green -- normally they start out both green. I want to suspect the controller data wiring but it worked normally at first!

The heat sinks on all of my motor controllers are getting really hot, even when inactive (I guess for holding torque?). Is this normal?

Help is very much appreciated!

Thanks, Kyle

I have the exact same problem. I found that I can get it working if I use the SNAP firmware, but who wants to use that? If Zach or somebody could help us out, I am sure that Kyle and I would be really thankful.

---

Jay

So I wonder then if the power supply does have anything to do with it. Well, as Jay said, any help debugging this would be awesome.

I don't think that it is the power supply. I just borrowed my friends out of his custom computer, and I don't think that the electronics are drawing more than 800Watts. I didn't test with a mulimeter, but I am assuming that the voltages were fine, considering that is was just running computer electronics before hand.

---

Jay

Hi

I think this a known problem with Arduino electronics. I never could get the stepper exerciser to work properly . The only thing that seems OK is homing the axes and that does not always work.

---

Regards

Ian

Yea, thats exactly what I have noticed. If I were to import an stl file to the host and click build, would the electronics do it?

---

Jay

Hi,

We've just finished to assemble our Darwin too (with extruder, thanks to bitsfrombytes.com) and we got the same problem with the arduino setup:

- Tester for each motor is working;
- When everything is plugged on the arduino, it seems to block when adding a move to the Queue.

I've began to read Zach code, it seems excellent and fairly easy to understand.
I'll dive a more in it during coming days, and let you know what has been found, I can't wait to get my first cup printed

About supply problem: we've burned our first pc-alimentation when all the motors were first connected, any idea why? I wouldn't like to repeat this...

Pc-alimentation as in power supply? That's too bad. But yeah, somehow I get the impression that operating the steppers while this problem occurs is none too good for the electronics... are any of your motor board heat sinks getting very hot when idle? My Z axis seems to get much hotter than the other two, and I wonder if I've already damaged it somehow. Then again, I may very well be conflating the stepper exerciser problem with something that is normal or unrelated. I just think I would have noticed the stepper controller getting that hot before.

Will be interested to hear what you find out looking at the software! I'll take a look also, as soon as I have a little while to work on it.

Kyle

I've corrected a bug with the Arduino firmware library, the bot didn't know when it was atTarget().

This bug could come from enableTimerInterrupt(), jumping values when comparing the axis.current with axis.target, could someone confirm this behavior?

There's still some strange stuff, I'll try to add some code to emulate min/max switchend in software.

Here is the changelog for the axis bug:

[github.com]

You can clone/fork the arduino firmware, and make your own repository:

[github.com]

Have fun,

Alex

yikes!

i'll look into this bug this weekend.

i plan on releasing a new version of the arduino firmware, so i'll get these code upgrades into it as well.

thanks guys!

Ah wonderful, thank you, Alexandre! and thanks Zach, looking forward to trying out that update! -Kyle

Cool, looking for next arduino version.

Last code from [github.com] include software max border on each axis, it allows me to not use the opto-ends.

All it needs now is a way to calibrate the home and the max value automaticly.
The cool stuff I'd like to include is to keep the calibration values and the current axis value ine the Arduino EEPROM memory, so when you stop/start the reprap, you don't need to recalibrate it.
Example of Arduino EEPROM use: [ladyada.net]

I'll keep you in touch as soon as it's in the code, or just follow the github feed [github.com]

Edited 1 time(s). Last edit at 05/23/2008 04:32AM by Alexandre Girard.

unfortunately for us, the gcode stuff is about 13K / 14K max on the arduino. adding more features is going to become a tricky game.

this is why the next step in the long term is to finish the host modifications to support multiple reprap 'drivers' and then start work on the SNAP_V2 firmware. we can then move all the heavy lifting off to the host and be free to do lots of advanced things.

this and actually printing are going to become my top priority.

kylecronan Wrote:
-------------------------------------------------------
> The heat sinks on all of my motor controllers are
> getting really hot, even when inactive (I guess
> for holding torque?). Is this normal?

I believe thats because you always sink the max current from the steppers when stationary because the single arduino config does not have enough free pin outs to control the enable lines, i.e when powered on but not doing anything you cant turn the steppers.

What you could do as a semi hack is take analog 1 which looks to be free and use it to control the enable line of all three steppers, or, if that isnt free take the RX line and add a latch circuit that on data recieved latches for (say) 15sec to the enable line.

The final option is to wire the max min sensors from each axis all in series which gives you an extra 5 free IO lines to use, you just lose the res of which axis limit you hit, although you rectify that in software by moving a couple of stops on each axis in turn to see if the flag clears.

J.

if you use the GCode_Interpreter stuff, it will automatically shut down your steppers to keep this from happening. I'll look into doing this on the SNAP firmwares as well for the next firmware version.

Patches appreciated =)