The Smell of Molten Projects in the Morning

Ed Nisley's Blog: Shop notes, electronics, firmware, machinery, 3D printing, laser cuttery, and curiosities. Contents: 100% human thinking, 0% AI slop.

Tag: Improvements

Making the world a better place, one piece at a time

Makergear M2: Prototype Corner Clips

In the course of normal events around here, the M2 gets tipped to one side or the other. Every time that happens, I rediscover the blindingly obvious fact that there’s nothing holding the glass build plate and the heater to the support spider:

M2 build platform corner

A few minutes with a metric ruler produced some useful dimensions for the ends of the spider’s arms:

M2 Platform Support Spider Pad Dimensions

The Big Box o’ Foamy Things emitted a mouse pad (remember mouse pads?) of exactly the right thickness to bring the corner pads just barely above the level of the glass plate, thus allowing for slight compression:

M2 corner bumpers

That’s a 1/8 inch hole punch, which is close enough to the M3 screw diameter in foam rubber. It worked fine for the balls in the corner support pads, too.

The long-suffering shop scissors produced results about as pretty as one might expect:

img_3157 – M2 platform retaining clips – raw cut

Which is to say, not very.

The material is 6 mil (about 0.15 mm) phosphor bronze, nice and springy. Combined with ripply edges and sharp corners, you get perfectly serviceable serrated knife blades suitable for use in traditional shop ceremonies of ritual scarification of the fingertips.

I stacked the slips, clamped them to the Sherline’s table between sacrificial plastic sheets, used manual CNC to poke a pair of #31 holes (0.120 inch, about the right clearance for M3 screws) at the right spots, and then stacked everything up on the M2:

M2 platform retaining clip oops – in place

The alert reader will notice a third #31 hole at the wrong spot, which was the first one I drilled and partially explains the lack of pictures of the operation.

Sighting across the platform shows that the clip doesn’t lie quite flat on the glass, due to the scissors-cut bending:

M2 platform retaining clip – edge view

However, four of these clips hold the glass firmly to the heat spreader and eliminate the need for the stock bulldog clips, which is what I wanted to find out.

But they’re ugly and I don’t want to explain that extra hole…

2013-07-17
Makergear M2: Platform Support Balls

After adding the insulation below the M2’s build platform heater, I punched a hole in each of the support pads and inserted a 1/8 inch bearing ball:

M2 HBP support balls

The pads measure just slightly less than 1/8 inch thick, so the balls support the aluminum heat spreader plate. Unlike the pads, the balls hold the plate at a constant distance from the spider which shouldn’t vary with mechanical load.

As nearly as I can tell, generic rubber expands by maybe 100 parts per million per degree C, so a 3 mm slab might expand by all of 0.02 mm over a 70 °C range: temperature obviously doesn’t make much difference. However, I’m about to add some hold-down clamps to keep the glass plate firmly in place and that pressure might squish the pads.

Obviously, putting a steel ball between two aluminum plates isn’t something you’d do in a high-stress machine, but the balls must support only the platform and won’t get any shock loading: any shock strong enough to indent the aluminum will probably shatter the glass. I’m pretty sure there won’t be enough motion in the XY plane to produce any wear, either.

Four points do not define a plane, but the spider and the spreader seem close enough to being planar that all four balls make firm contact. The M2 really does have a good mechanical foundation!

2013-07-16
Tour Easy: Yet Another Shifter Pulley

Somehow, I think I’m never going to get around to doing a CNC version of this thing, but at least now I have more pictures…

The overall problem comes from the fact that the Tour Easy frame geometry doesn’t match the expectations of the front shifter: the cable bends over a small finger that, on a diamond frame bike, should simply hold it in position. Here’s the finger, with a very early version of the pulley that just holds the cable slightly higher than the normal position, complete with one snapped wire showing that the pulley wasn’t getting the job done:

Front derailleur cable with broken strand

The obvious solution involves running the cable over a nice, rounded surface that prevents abrupt bending. The most recent version looks like this:

Shifter pulley installed – left view

Yes, the end of the cable sticks out over the chain; I haven’t tucked it in yet.

A bit of lathe work produces a 0.42 inch diameter thin brass disk with a 50 mil half-circle trench around it; in retrospect, the diameter of the trench bottom should be 0.42 inch and the OD should be about 0.45 inch. If you have really good parting-off-fu, you can produce a disk with a finished backside right on the lathe, but I had to drill an off-center hole anyway, so I thinned it on the Sherline:

Shifter pulley – thinning

It looks like this after all the thinning:

Shifter pulley – thinned

One flange is wider than the other: the thin flange faces front and gets a bunch of cutouts, the wide flange faces rearward and must support the bitter end of the cable.

I lined it up in the shifter, filed a notch to fit around the shifter finger, scribed the hole location, clamped it down, and drilled the hole:

Shifter pulley – center drilling

I think the hole could be on-center with the larger disk; now that I’m keeping better notes, I’ll try that next time. If so, then I can drill it on the lathe, part it off to the correct width, and hand-file the backside flat. The general idea is to have the cable pass over the finger, which almost happens with the smaller diameter.

Some tedious hand-filing produces notches that index over the finger and clear some protuberances on the shifter arm. This is the front face of the pulley that sits against the shifter arm, with a 5 mm socket head cap screw for scale:

Shifter pulley with bolt – front face

The rear face has one side of the trench filed away to get the cable out of the trench and around the bolt:

Shifter pulley with bolt – rear face

Then it looks like this from the right side of the bike:

Shifter pulley installed – right view

A pleasant morning with some Quality Shop Time…

2013-07-14
Makergear M2: Platform Lighting

Adding a strip of white LEDs under the X stage helps shed some light on events atop the M2’s build platform; this was very nearly the first improvement after getting the printer, but somehow I’ve never written down where that nice white glow comes from.

This view shows the strip from below, looking up from the -Y direction in front of the stage:

White LED strip under X axis frame

I originally screwed the wires into the terminals from the hulking 12 V Dell laptop brick for the platform heater, but then I had to unscrew the wires whenever I moved the M2 and I didn’t like sharing the connectors with those huge conductors. Now the LEDs are in parallel with the extruder fan (which runs continuously), sharing the FAN1 screw terminals inside the electronics case.

The M2 firmware uses PWM to cut the 19.5 V supply from a much smaller laptop brick down to roughly 12 V RMS for the fans, but that isn’t such a Good Thing for LEDs. The strip has 120 Ω resistors that drop about 2.4 V at 20 mA from a 12 V supply, leaving 9.6 V for the LEDs (at about 3.2 V each). Running from 19.5 V means the resistors will see about 9.5 V and pass nearly 80 mA, four times the nominal rating, during each PWM pulse.

Based on those measurements, the light output doesn’t go up by nearly a factor of four during each pulse.

I plan to add a 12 V supply to the LinuxCNC box, probably by recycling the 12 V brick from the M2, which will get the LED current back down to a reasonable level. With any luck, they’ll survive this mistreatment and not carry a grudge.

You could, of course, just power the LEDs from a separate 12 V wall wart, but that adds Yet Another Thing when I carry the M2 to demos.

2013-07-12
Makergear M2: Heated Build Platform Insulation

Although I don’t have any data to support the idea, it seems that there’s far too much heat loss from the bottom of the HBP. Admittedly, air is a great insulator, so most of the energy should go into the aluminum plate, but having air blow over the bottom can’t be a Good Thing. There’s a very thin space between the bottom of the silicone heater element and the black aluminum spider supporting the corners, so I added a thin cardboard sheet:

HBP insulation – cardboard base

The curiously shaped cutout clears the heater power wires, the thermistor in its lug, and the thermistor wires.

Atop that goes a pair of very thin cotton cloth sheets (again, not much to focus on, so it’s a bit blurry):

HBP insulation – cotton sheet

And then the plate fits atop the corner support pads as usual. I suppose the heater duty cycle should be lower at any given temperature, but I don’t have any records to compare against.

2013-07-08
Makergear M2: HBP Connector Stabilization

Given my experience with the TOM’s platform heater connector, I was bothered by having the M2’s heated build platform connection flex as the Y stage travels back and forth. After replacing the Z axis motor, I added a strut to the Y axis stage to stabilize the HBP connector.

This overview shows the aluminum strut sticking out to the rear (Y+) end of the platform support spider:

HBP connector support strut – overview

A closeup shows a quartet of 4-40 holes drilled and tapped along the strut’s midline:

HBP connector support strut – mounting detail

Admittedly, that’s a bit of a kludge, but I didn’t want to drill holes in that nice steel bracket… particularly since I’d have to dismantle the whole stage to get to it. The four screws wedge the strut firmly in position and have jam nuts on the bottom so they don’t loosen.

I extracted more wire from the braided sheath and moved the cable a bit further out at the cable tie holding it to the Y axis stage, then cable-tied the HBP connector to the strut.

With the stage all the way to the rear:

HBP connector support strut – at Y min

And to the front:

HBP connector support strut – at Y max

The wires may break, but now the HBP connector and heating pad joints should survive!

2013-07-05
Makergear M2: Grippier Z-min Switch Mount

The printed bracket for the M2’s Z axis home switch doesn’t get a good grip on the oiled steel rod, so it can slide around just a little bit when nudged. That doesn’t happen often, but when it does, all your careful alignment Goes Away.

A single wrap of silicone tape solves that problem:

Z min switch on silicone tape

While I was in there, I replaced the socket-head cap screw I’d been using with a longer hex bolt and swapped the nylock nut for a plain nut that’s easier to adjust. I should file the raised markings off the top of the bolt head so it presents a smooth surface to the switch.

That was easy…

2013-07-04