So far so good!

As I write this I am 2hrs into a 4hr+ print, and the first print of a part of my own design. This magnitude of a print is a first for me with my personal 3D printer. Little Fisher is sailing along which is great after several years of not so much success (early printer, before people used PLA).

Mine is a Beta, absolutely standard except not running top panel, only 'replaced' part is the BuildTak which got messy during original learning curve with calibration. Original arms, haven't even soldered the endstop wires.

I'd leave it as-is but the extruder motor does get a bit warm so may do the new geared extruder just for that.

Turned it on and was printing as soon as the file uploaded through the web interface. Really Cool!

Edit: The most useful mod would be a way to hold a 1Kg reel of filament....

Edited 1 time(s). Last edit at 10/19/2015 04:43PM by Theolodian.

Success!

Are people still having problems with dimensional accuracy? This was my first print where dimensional accuracy mattered or could be checked at all really. The overall size was 77x40x60 lwh and it is accurate to within 0.25mm everywhere, including straightness. Very impressive overall.

If you haven been following my previous issues I had been running an arm length of 162.8. This was back to 160 for this print, but my radius is still 82.4. There is a hint of a strongly curved pattern in the surface finish, maybe slack in the system? Very odd visual effect but doesn't affect the part too much.

We've seen patterns in large, flat, vertical, areas of prints. We think it's more to do with the stepping of the motors, as they try to draw a straight line from what, effectively, is a series of curves. If you can post a picture, it may help to determine if this is the same thing.

Ian
RepRapPro tech support

Ian,

Thanks, that describes it exactly. I like the honesty of it, signature of being made on a delta I guess.

I published some photos showing the difference that changing to 0.9deg/step motors on my delta made at [miscsolutions.wordpress.com]. But the pattern was barely noticeable even with standard 1.8deg/step motors.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

How many microsteps is the Duet running? I also could make my belts a bit tighter.

We supply 1.8 degree (200 full steps per revolution) NEMA 17 stepper motors. The Allegro A4982 offers full, half, quarter or sixteenth stepping, and the Duet is set to use sixteenth. Generally, there's no point going higher than 1/16 micostepping (or, accoding to the following, more than 1/10 microstepping), as the motor isn't accurate enough to take account of more steps; from: [www.geckodrive.com]
"A step motor is a mechanical device that is manufactured to a certain tolerance. Typically a standard motor has a tolerance of +/- 5% non accumulative error regarding the location of any given step. This means that any step on a typical 200 step per revolution motor will be within a 0.18-degree error range. Stated otherwise, the motor can accurately resolve 2000 radial locations. Coincidentally this is the resolution of a 10 microstep drive."

There are other stepper drivers that can give more microsteps (1/32 up to 1/256), but it's arguable if that will make any difference to accuracy or resolution. So using a motor with a finer angle, as dc42 has done (nice write up!), is the better way of increasing resolution and accuracy, at the cost of speed (because there is a finite limit on the number of steps you can send per second, and more microsteps means you have to send more steps to move 1mm). Another option would be to reduce the gearing, to get more steps per mm, by using a smaller pulley (ie with fewer teeth) on the motor.

Ian
RepRapPro tech support

Great answer. Many thanks.

I would add the following to Ian's reply:

1. The torque per microstep drops off as you increase microstepping. When the torque per microstep is too low to overcome the friction of the system, the effector won't respond to individual microsteps but will only move on each e.g. 2nd or 3rd microstep.

2. Not all microsteps are the same size even if the stepper motor is perfectly constructed and the motor follows each microstep. This is because the torque due to current in the windings has to compete with the detent torque.

3. Increasing motor current will help with both (1) and (2). I can't remember what current the tower motors on the Fisher are normally run at, but if it is 800mA then increasing it to 1A might possibly improve print quality. It will also increase heat dissipation in the board. Cooling of the board may already be poor in the Fisher because it is mounted horizontally under the bed, so if you increase motor current then it might be wise to add a cooling fan to blow air across both sides of the board along the row of stepper drivers, as I do in my delta.

4. 0.9 deg/step motors provide not only double the resolution, but also almost double the incremental torque.

5. If you change to 0.9deg/step motors, the maximum speed may be reduced for two reasons:

(a) At the same speed, the motor generates twice the back emf. I had to increase my printer supply voltage from 12V to 24V to overcome this.

(b) The electronics has to generate step pulses at twice the rate. I did a lot of work on my firmware fork a few versions ago to achieve 300mm/sec on my printer using 0.9deg/step motors (160 microsteps/mm). I think that is fast enough.

Edited 1 time(s). Last edit at 10/22/2015 12:13PM by dc42.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Thanks again. I am running 150mm/s rapids on my Fisher, any more than that seems abusive to the hardware. It is only acrylic after all.

So you're pulsing at 50KHz per axis. How many lines of Gcode can the Duet handle per second?