SoliForum - 3D Printing Community — A few easy mods for the SD4

Re: A few easy mods for the SD4

2014-11-20T20:53:07Z

That is correct. High current causes extreme pulsing which is visible as a moire pattern. If you want to see for yourself, bump it back up and do a free air extrusion--you will see the moire as the plastic comes out of the nozzle. You can also feel it if you touch the filament as it goes into the extruder body.

By the way, you don't have to pay too much attention to voltages. Using your eyes and ears will get you much closer. You don't want your motors screaming, but you don't want them growling either. There's a good balance. The extruder is hard to hear, but X and Y are easy to mess with by ear. For E, my approach is to print a long, thin-walled piece and keep turning until the moire disappears.

Granted, there are many things kind of odd about the choices Solidoodle has made for their combination of stepper drivers and motors. Everything is being pushed to nearly maximum capacity, and there is aliasing that you will not eliminate without stepper chip replacement and motors to match. I have a high torque (60 oz-in) motor I want to put on Y, but I don't have big enough bolts to get it on my frame yet. In time...

There is more information on the deficiencies of the Solidoodle stepper choices here. I have not taken the time to check if the motors are the same and if these deficiencies hold up on the SD4, but I'm almost certain they do. It is for this reason that many users find they eventually need more fans on the board--those steppers get hot being pushed like this, and the stock board doesn't even have heatsinks on them which is incredible.
http://www.soliforum.com/topic/459/soli … iciencies/

Re: A few easy mods for the SD4

2014-11-20T20:24:28Z

Now if only I could score a ceramic DMM probe to prevent *it* from shorting things out... ;-)

I did check the Vref on the extruder and it was off. I measured .438. Solidoodle support says it should be 0.250! I dialed it down to 0.250 and the Moire pattern is almost eliminated even with the 0.25 Z height. Yay! - I can go back to easy math. My understanding from the A4988 part sheet is this sets the max current for the DAC. So am I correct to think that the distortion from too high Imax is caused by non-linearity in the stepper response to high current? Seems to make sense - would result in a modulation in the flow.

Re: A few easy mods for the SD4

2014-11-20T17:42:08Z

and for the love of all that his holy use a ceramic screwdriver. i burned up my board ignoring this advice.

Re: A few easy mods for the SD4

2014-11-20T17:15:15Z

It didn't eliminate it, it just covered it up. If you get things properly calibrated you can worry a whole lot less about "I try to make my geometries integer multiples of the bead dimensions". Slice software has come a long ways in the past year so let it do its job and you might be surprised at the results.
Moire is not a real issue as long as you don't mind the appearance. Some like it and some hate it.
With the stock hot end you can come very close to eliminating it with slight adjustment of the VREF. Don't let turning a POT intimidate you. Once you adjust it you'll ask yourself why didn't I do that sooner?
As far as the blobs at the start point go...
You can select random start points in your profile
Or you may want to increase the retract distance and or speed. I have found speed to have the most impact with the stock hot ends.

Re: A few easy mods for the SD4

2014-11-20T17:04:35Z

Yeah it is. :-) Actually, I was referring to the Moire pattern. I don't really understand why going from .25 to .254 eliminated it (seems like an XY thing) but it did. A little annoying since I try to make my geometries integer multiples of the bead dimensions so the math just got a little harder (this is why I switched form .3 to 0.25).

Re: A few easy mods for the SD4

2014-11-20T16:45:11Z

layer height won't help you with that seam, that's just what the stock hotend does. it likes to blob all over your prints because it's a jerk.

Re: A few easy mods for the SD4

2014-11-20T16:12:12Z

..or change to .254 height (a LOT easier :-)). Another observation. If you look closely at the print on the right (no well nut), the horizontal band is gone to the left of the seam (end of the pass). I think this indicates that the bed is just a little slow to descend and has fully settled by the time it completes the circuit. I was confused at first because I knew I don't have Slic3r set to print the outer perimeter first but when I went back to check, I noticed I had made a mistake in my geometry. Instead of having a 1.2mm wall thickness as desired (3 passes), I fat fingered it and entered 0.12mm. So these prints are single bead thickness and the Z settling theory holds.

Re: A few easy mods for the SD4

2014-11-20T15:59:55Z

All of that being said, your prints are looking good so keep it up. Now you need an e3d to make those start points less horrendous.

Re: A few easy mods for the SD4

2014-11-20T15:51:14Z

Those "diagonal bands" as you call them are Moire and that is an issue of VREF adjustment and too few steps (1/16 micro stepping) on the part of the stepper driver. On the printer board upgrading to a 1/32 stepper is difficult and requires some serious soldering skills. Other options would be switching your motor to .9 degree motor or a geared extruder like the BulldogXL. There is also the option to upgrade your board to one with removable steppers like the Rumba, Sang or Azteeg-X3.
These upgrades will eliminate the Moire but you can probably remove most of that with a simple adjustment of the VREF.

Re: A few easy mods for the SD4

2014-11-20T15:43:01Z

I stand corrected. You are correct on both counts. I did a quick experiment (two actually) to test the assertions of the well nut and integer step improvements. I created a 40x41.6 cylinder with 1.2mm walls where the lower 20mm was printed at .25mm(~35.433 steps) and the upper 21.6mm was printed at .254mm (exactly 36 steps). I then printed twice. Once with the well nut (on the left) and once without (on the right).

Two observations:
1) The .25mm does show some deformation manifesting in the form of diagonal bands. Harder to see on the print w/o the well nut in the image due to the lighting angle, but it is there on both about equally.
2) There is a improvement in the horizontal defects with the wellnut in place. I was surprised at this (as evidenced in my previous comment) but there it is.

Re: A few easy mods for the SD4

2014-11-19T23:54:30Z

Some of that seems to be rounding error but I do see a tiny bit of backlash, some printers have it worse than others. It just seems to depend on the quality of the rod. I'd throw it on for completeness. It is effortless. And you should also print at 0.2963 or another height that is a multiple of the thread depth. You'll never eliminate rounding error without a metric rod but you'll reduce it by printing with a layer height closer to the thread depth.

Re: A few easy mods for the SD4

2014-11-19T23:22:51Z

Re. the well nut. I've always wondered how people end up with backlash problems on the Z axis. Seems like gravity should pretty much prevent it. I am not seeing any banding to speak of in the Z (see below). I suppose, for $2, why not. But OTOH, if it ain't broke...

layer height is 0.25mm

Re: A few easy mods for the SD4

2014-11-18T00:57:09Z

Liking that plexi top. I'm going to ditch the entire enclosure and go plexi because mine got all bent up from shipping. Neat idea on the filament too. When you buy a thinner, taller roll, you'll find it rolls around on the PVC. I just printed some pieces that fit tightly on the PVC to prevent it from moving around too much.

Edit: while you're modding, get a 5/16-18 wellnut on that Z rod. No more backlash. And an e3d

Love that you ripped that kapton off too.. mine is bare...

A few easy mods for the SD4

2014-11-17T22:07:53Z

I recently made three really simple mods that I thought I'd share.

1) Replace overhead door with clear plexi-glass. Get the hardware store to cut a 13.5" X 14.5" piece of plexiglass (I used 1/4"). While you are there, pick up four M3x10 flat head screws and four nuts. Then is is a simple matter of unscrewing the existing door, laying the plexiglass on top so it is flush with the door and sticks out the back, marking the holes, drilling, and reassembling with the plexiglass. Remove the two stops in the back as well. The plexiglass extending out the back acts as it's own stop. I had access to a mill so I went ahead of machined recesses for the hinges so they would sit flat but I don't think this is really necessary.

2) Leveled the print deck. I noticed that the aluminum plate was curved so that I always had large gaps between it and my glass print bed on the sides. For me it was curved on one axis so I put two blocks of wood on the floor, laid the aluminum print bed across them, and gently stepped on the center (not my full weight), checked for straightness, and repeated slowly increasing the pressure until I no longer saw (much) light between the bed and my straight edge. Worked surprisingly well. Should heat the glass more quickly and uniformly now.

3) I posted this earlier, but I may as well include it. Two quick holes drilled into the PVC pipe supporting the spool, and a length of wire has [edit]dramatically reduced (just had my first tangle since installing - after going through three spools uneventfully)[/edit] my spool tangling problems.