Random X homing movements

Hi guys, i'm quite new to the rep rap world.
My DIYtechshop printer, when I ask it to home (Either G28 X0 or in repetier host) it homes at random places!
I'll open up the printer, making sure that the X axis is at the endstop (which is at Xmax around 230mm), I'll then make it move to 0 manualy, then i'll ask it to home. But strangely it stops about 10mm further.
The worst thing is that repetier thinks it is at X230 even if it moved around 10mm.

My X axis stepper is calibrated to the belt and pulley that I have (20teeth, nema 17 motor, on ramps).

I have changed this:
#define X_MAX_LENGTH 300
#define Y_MAX_LENGTH 300
#define Z_MAX_LENGTH 165

because i thought that this:

#define X_MAX_LENGTH (X_MAX_POS - X_MIN_POS)
#define Y_MAX_LENGTH (Y_MAX_POS - Y_MIN_POS)
#define Z_MAX_LENGTH (Z_MAX_POS - Z_MIN_POS)

might be the problem, but either way it does not change anything.

I have attached my configuration.h so you can have a look at it.

Thanks in adavance!

You may not be calibrated correctly. Or you may be skipping steps. I would check your ref voltage on the drivers. You can check to see if its skipping by putting slight pressure against the carriage and see if it has problems moving. If so bump your ref voltage up a bit.

Thanks for the reply!
I've tried to tweak with the voltage but if I put it any higher the movements sound too difficult whereas where they are now (voltage reduced by 30 degres) everything is smoother. I also tried adding more into it but they either just skip, or dont move at all. So I don't think the pots are the problem.

Also for the fact that it might skip steps, it doesn't physicaly skip any if that's what you're asking because I would hear the belt. I don't know if it might skip some in the firmware itself though. It might be an explaination because sometimes i'll manualy move 50mm and it will move 7.53, 23.56 or some random number. It most often moves exactly to 50mm when I move it manualy but it never homes corectly.

try turning your homing acceleration down

Yeah, I just tried that now and it doesn't change anything. It's weird, but what I found out is that the X carriage only likes to move in one direction! The X- direction. Due to the fact that my endstop is located at Xmax it doesn't home properly. When I say "it doesn't like to" it moves, but only by about .01mm or 1.12mm or .56mm, just random numbers!

I droped my acceleration to 500 mm/s², and my retract acc to the same. My max acc is at 3000mm/s².
I don't know if this is still too high but i'm guessing it's not because the belt moves and moved fine towards X- even before I lowered it.

Thanks for again for the replies!

My config.h is in the attachements.

Anyone have an idea what's going on???

Edited 2 time(s). Last edit at 03/29/2014 12:46AM by samsam481.

Your configuration looks pretty ok, but here are some thoughts as I compare it to my setup.... You can probably increase your homing feed rate to 3600 and you should be fine there. If your end-stops are normally closed (most standard end-stops are) you can leave the *_MIN_ENDSTOP_INVERTING set to false. If they are normally open, set those to true. I'm using a RAMPS 1.4 board and I do have the ENDSTOPPULLUPS define enabled, which turns on the 4.7K endstop pullup resistors.You might try that. Not related, but wrong, is your line 18, which automatically fills in the build date for you. You can change the date you inserted back to a space. You should probably leave DISABLE_Z set to false. There's no special advantage to letting the z motors sleep, and they might lose their step position. If your Z axis is screw-driven then the DEFAULT_MAX_FEEDRATE should be a bit smaller, something more like 2-3 (150-225 RPM) instead of 30 (2250 RPM).

Edited 1 time(s). Last edit at 03/29/2014 04:06AM by Thinkyhead.

---

|
| Lead Developer of Marlin Firmware
| Help support my work at Patreon and Elsewhere.
|

Wow, this is realy frustrating it's like it's trolling me or something.

First of all the pullups are disabled because it just screws up everything, and because I don't think I have resistors as all I have to connect to the endstops are a blue and a red wire, no black (ground I suppose).
When I turn the pullup option on or off (taking off the // begining of > //#define ENDSTOPPULLUPS < ), it just turns the endstops useless, (always open, or always Triggered).

What I've found that "works" is I put X_MAX, Y_MAX, and Z_MAX as true and the rest to false. I have literaly tried every possible combination, Pullups disabled, enabled, endstop inversion true false, everything!
This is the closest to "working" as it has gotten in a while yet it doesn't stop when It hits the endstops because theres a weird delay. It moves fine in both X+ and X- directions, but when I trigger the endstop and I see what it does with M119 it takes 1second before it is triggered!

It would work if it wasn't for the delay, and to make it even more frustrating, now the Z axis also has the same problem, it continues when it hits the endstop, whereas it didn't yesterday.

By the way my endstops are NC, ill attach a photo (google).

Alright, turns out it's interference. Very good video explaining it (subed): [www.youtube.com]
I haven't fixed it yet but close, I can move in X+ when the interference has stoped (hence when the X max is no longer triggered).
I'll M199 a couple times until the X_Max is open then I can home.
It's obviously not totaly fixed but I get what's going on!

Edited 3 time(s). Last edit at 03/29/2014 03:07PM by samsam481.

Interference, I hadn't considered that. I found no RF interference issues with my setup so I assumed it was rare.