Reprap calibration, gcode and units

Hey all,
I am at a point where I am trying to get the various motors to run smoothly on my Prusa i3.

In pronterface, if I type in the gcode g0 x10 I am of the understanding that my x-axis ought to move 10mm (or so, pre-calibration) right? I am finding though that it moves around 70mm or so. I think related to this - if I set the rate at 3000min/mm my motors struggle to get anywhere despite putting up their their current. A rate of 100mm/min has them moving at a 'comfortable' speed - (not stated 100mm/min though!)

So, I seem somewhere to be an order of magnitude out.

I thought it may be something to do with these lines in my config:

#define DEFAULT_AXIS_STEPS_PER_UNIT {80, 80,200*8/3,760*1.1}
#define DEFAULT_MAX_FEEDRATE {500, 500, 5, 45} // (mm/sec)
#define DEFAULT_MAX_ACCELERATION {9000,9000,100,10000}

Concentrating on X & Y I have checked the 80s in the steps_per_unit for my setup with [calculator.josefprusa.cz]. 78-80 seems pretty standard for these figures.

So what am I missing??? Thanks in advance anyone who can give me a pointer...

Steps per millimeter is the setting you are looking for.

With the exception of the E axis, there is no "standard" for the steps per millimeter. These values are calculated exactly based on things such as number of teeth and teeth pitch for pulleys and belts, and thread pitch for the z rods. The same would be true for the E axis if it weren't for the variations of the hobbed bolt or whatever you are using to grip the filament. You also need to take into account if you are set up for micro stepping.

What are your parameters? For x and y what kind of belts and pulleys are you using. For z, what is your thread pitch?

Hey thanks,
I'm just concentrating on the X and Y axes for now.

belt and gears: 2mm pitch
gears: 20 teeth
steppers: 1.8 step angle

The electronics hardware I have is basically this set

According to [calculator.josefprusa.cz] this gives me 80 steps per mm - this is what I had in my config (as DEFAULT_AXIS_STEPS_PER_UNIT). Where is a 'unit' defined? Not sure I have seen steps per millimeter in settings in Marlin's configuration header file?

I'd get it if I was 10% or so out, but as I said in my post I'm almost an order of magnitude out? So if I put put in G0 Y1 in the pronterface console I get 15mm over movement.

So then, pro-rating, the rate would be around 80/15 = 5.33. If I stick M92 Y5.33 in the console and try G0 Y10 I get around 10mm of movement. This looks suspiciously like the 5 mm/min value that I get if I put '1 - full step' in as the driver microstepping.

Looking into my drivers, their datasheet says one could change the microstep by changing the voltages across some of driver inputs. So is it possible that my drivers are operating at 'full step mode'? How could I confirm this or otherwise? I had just assumed I would have the default '1/16 step Pololu'

UPDATE - Looking around on the internet I am not sure I quite get the while microstepping thing. Seems to be some debate here and there - is it something I can turn off or on?

UPDATE2 - So I note now that if I get one of my stepper drivers off the RAMPS1.4 I have 3 header there with no jumpers on. If I dig out 3 jumpers and stick them on to get 1/16 step (as per here) , am I going to nuke my driver?

Edited 2 time(s). Last edit at 10/03/2013 03:56PM by aegis1980.

What controller board are you using. If it's a RAMPS, there are jumpers beneath the pololu drivers. If all the jumpers are installed, which is the "default" setting, then you will be using 1/16 microstepping (actually it depends on what driver you are using - I'm assuming the "standard" 4988), so instead of the 200 steps/turn that you achieve with the 1.8 step angle you mention, you are essentially getting 16 times that or 3200 steps per rotation. This difference will throw your calculations way out of whack.

As far as unit go, this calculation itself is unitless until you introduce a unit. You've introduced the pitch as 2mm, so you've basically defined your unit as millimeter.

Jumpers were it!! I had none on RAMPS 1.4 so drivers were working at '1 full step'. Have raided various computers for 12 jumpers and stuck them in. Everything works nicely now on Reprap front...not so good for computers but hey

Now to seemingly underpowered heat bed...

Just had to say thanks aegis, I've just built my own Prusa i3 (boxed) using the same sainsmart electronics and couldn't understand why I had to set my steps per unit to 4.54... like you mine also didn't have any jumpers installed... luckily I've got loads spare from old computers I taken apart over the years

Hopefully if all goes well I might be abled to print this weekend.

I wonder if I could see your configuration.h file?