A newb's quest. Understanding the firmware (marlin & printrboard rev D)

More homework while I wait for pieces parts. As background, I'd note that I'm neither new to electronics, programming, or PID loops. I am, however, naive when it comes to this platform and the arduino concept and would like to reduce the learning curve as much as possible prior to firing up the machine (prusa mendel variant using the printrboard rev D, marlin current version)

I've been going over the marlin code, and I'd be grateful if those familiar with the info could answer a few questions about the firmware configuration.

1) TEMP_HYSTERESIS: What is the practical limit of precision for this constant? I'd prefer to operate with no more than a 1 degree variance in either the hotend or bed temp control loop, but I really don't know if these chips are capable of providing better than the default 3 degree spread in the first place, without causing some undesirable side-effect..

2) PID_INTEGRAL_DRIVE_MAX: What is the practical limit on this constant, for these processors? I see it's default is 255, but I have no idea where the real boundaries are for the 90USB1286.

3) Is there a known JHead default PID triplet floating around that hasn't quite been incorporated into the firmware dist, or do I just have to figure out my own? I see that there are preset starting PID triplets (Kp, Ki, Kd) for Ultimaker, Makergear, and Mendel parts V9, but I'm not sure if any of them represent my board.

4) EXTRUDER_MTR_STPS: The default value is set to 260, but to be honest, I'm not quite sure.why. Shouldn't that value be just like any other 1.6 deg/step motor @ 200?

5) BOLT_DIAMETER: Is this measurement the measured diameter of the bolt shaft or the minimum diameter of the hobbed section?

6) DEFAULT_MAX_ACCELERATION: Why are the default X and Y values so different? They use the same gear & motors, they should have the same value to maintain cartesian integrity. Is this due a difference in intertital loading between the extruder head assembly transport (X) and the platen transport (Y) or is there some other practical reason that might help me "grok"?

Thanks,

x

Edited 1 time(s). Last edit at 09/04/2012 02:43PM by xiando.

Good questions! Sitting in on this one...

Not exactly a newbie to printing (Started printing about 8 months ago) but the firmware still confuses me a lot.

The only thing I know about the PID max is whats stated in the firmware "too big, and heater will lag after changing temperature, too small and it might not compensate enough for long-term errors." Also I've never seen anyone set it higher than 255 or lower than 80, but it's not like I read peoples config files everyday :-P

If I'm correct your E steps will vary because different people use different hardware. most people using a Prusa/Mendel variant will have stepper motors, but you'll still find programs like Skeinforge defaulting for DC motor. Also don't most stepper motors use 1.8 deg stepping?

Finally I think your correct about the Max Acceleration varying as an inertial compensator. The Y plane moves a lot more mass or at least a lot more surface area than the X plane and then you have to worry about base-layer adhesion. This is all just an educated guess of course.

Hopefully one day we will truly "grok" this madness.

1) hmm... so the 255 may be a result of the 8-bit controller...that's what I suspected, but I really don't know

2) I understand about the steps varying a bit, to tailor the response, just odd that what I would have figured to be a 200 step motor is set to 260 by default (presuming that's the one you were responding to). And yes, afaik, most steppers recommended and used in the reprap initiative are or appear to be bipolar, 1.8 deg/step motors

3) Thanks for the agreement regarding the deviation between X and Y intertial loading. after a few moments of head scratching, I thought of that as a potential source for the disparity. At least I don't feel totally blind in guessing my way around the dark here.

finally...oh I will grok it...one way or another. Hopefully with as little pain and as time/cost-effectively as possible, but either way. If I get answers outside of here, I'll back post.