Duet setup - instructions in progress

[KAS]

I'm using this mini IR height sensor without any issues with PEI ( tested it on glass, aluminium, tape, glue). It basically shoots light at the surface and bounces back to the IR receiver once the angle the reflection aligns.

https://miscsolutions.wordpress.com/min ... sor-board/" onclick="window.open(this.href);return false;

[Lightpath]

Michael-

I got the board and fsrs from tridprinting. I don't think ultibot had a kit out when I bought them. http://www.tridprinting.com/Electronics/" onclick="window.open(this.href);return false;

I wonder if a piece of felt might work. I have some handy.

[mhackney]

@KAS - that's good to hear, folks have had issues with PEI and had to paint the back with high temp black paint.

Felt should work fine too.

[critical_limit]

Have Davids IR-Probe working with a 3mm PEI Sheet without any problems. No Black Paint, no nothing. Just sanded the PEI with 800grid Paper and clipped onto a 5mm Aluminum Heatspreader.

[geneb]

Isn't 3mm kind of thick for PEI?

Is anyone selling v2-sized circles of it yet?

g.

[mhackney]

I'm not aware of anyone selling affordable disks of PEI.

Some folks are using thicker PEI and clipping it to the bed rather than taping it. I have a sheet of 1/8" ready to try when I have the time. It is much more expensive though. I haven't had any issues with any of my taped PEI but I am concerned about keeping a large disk flat with only edge clips. In the machining world, we would use a vacuum table to attach something flat like this and hold it firmly - but that would be a large expense and quite noisy.

[geneb]

It wouldn't be that bad if you built a venturi vacuum and used an airbrush pump. Might not have the needed volume though.

g.

[mhackney]

On David's link https://miscsolutions.wordpress.com/min ... sor-board/" onclick="window.open(this.href);return false; to the IR sensor board he does mention that PEI is transparent to IR and to paint the underside black.

Can you guys tell if it is reflecting off the surface of the PEI or the back side (aluminum plate or Onyx)? The height offset would tell you that.

[KAS]

My PEI is heavily sanded. It looks almost frosty to the point you cant really make-out any of the shapes/words on the onyx underneath.

That might be why I'm not having issues.

[KAS]

I'm not sure this video will help. I was recording it to determine the cause of the inconsistency and fluttering of the detection.
I ended up having to twist the outer IR LED to correct an alignment issue. Every printing surface seems to detect at different heights.

[youtube]http://www.youtube.com/watch?v=YY5PZOnlXh8[/youtube]

[critical_limit]

My IR-Probe is working perfect. Like KAS said, every surface is showing different probe heights. Think it´s because of better/worse reflection.
But my sanded PEI is giving me repeatable accuracy in 0.01-0.02mm. Same on Glas, which is painted black on the rearside, or on Buildtak.

I have now 3 IR-Probes installed on my Printers and every Probe is working like the other. Never seen this behavior KAS is showing in the video.
Guess it´s a misalignment of the IR-Diodes. But I would ask David for a exchange. Thats not a normal behavior.
As far as I know he changed the Diodes in the latest batch which are working better and giving better results with different surfaces.

cheers
Dirk

[critical_limit]

@KAS:

on the second time I watched your video I recognized that your IR-Board is attached too far away from the surface. Normal probing point is between 3-4mm. But yours is much more.
Maybe one of the IR-Diodes is heavily missaligned. Send this video to David and I´m sure he knows what to do or whats the fault. But I guess the angle between the two Diodes is far out of normal alignment. Please report what David suspects.

[KAS]

@KAS:

on the second time I watched your video I recognized that your IR-Board is attached too far away from the surface. Normal probing point is between 3-4mm. But yours is much more.
Maybe one of the IR-Diodes is heavily missaligned. Send this video to David and I´m sure he knows what to do or whats the fault. But I guess the angle between the two Diodes is far out of normal alignment. Please report what David suspects.

You're absolutely right on the placement, this was an initial setup just to test it out. Everything is working fine now after twisting that outer LED. I'm now using it on a mini kossel I built. The IR sensor would detect the surface about 2-3 inches off the surface with very inconsistent results, but that's all been corrected now.

[Lightpath]

Felt appears to have worked well. I no longer have random triggers on the FSRs.

Now on to trying to get extrusion working. I've installed a new bondtech QR, but it doesn't seem to be working. I have checked the wiring very carefully.

Based on the diagram in the manual, when looking at the front of the stepper with the connector pointed down, the wiring is A BC D. That corresponds to Red Yellow Grey Green. AC are phase 1 BD are phase two.

So using your wiring scheme:
Motor Due
A Red -> yellow
C Grey -> green
B Yellow -> Blue
D Green -> Red

When I power up the printer it moves one step then stops, and the only thing that will convince it to show any sign of life again is a power cycle. Once power cycled, if I try to extrude it moves exactly one step and then freezes again.

I've set my steps per mm to 494, and the amperage of the stepper to E1200:

M906 X1000 Y1000 Z1000 E1200 ; Set motor currents (mA)

The stock extruder appears to function normally.

I've spent a fair bit of time troubleshooting this, but I'm really scratching my head on this one.

[mhackney]

Lightpath, in thinking about it, felt might be too compressible and require some adjustment in the triggering height. It should be consistent once dialed in. Just keep your eye on it.

Did you read my blog post on wiring the QR (or steppers in general)? http://sublimelayers.blogspot.com/2016/ ... g-for.html" onclick="window.open(this.href);return false;

[critical_limit]

@ lightpath:

assuming your wiring is correct, try to lower acceleration for the extruder massivly.

I had an 5:1 extruder motor which only worked well when I lowered acceleration from 2000 down to 700. above it was lossing steps and sometimes stalling.

[Lightpath]

I did read it, but I think I confused A/B and 1/2. I have everything wired correctly now and it works.

I hear you on the felt perhaps being a little too squishy. The calibration looks like it works beautifully. I'll keep an eye on it though.

Edit:

I've still got some issues, but I am at least printing!

I need to do a better job of profiling the filament I'm working with, I think. I'm under extruding. That's a different problem, however- At least now everything is working mechanically.

Leveling works brilliantly, however I have to run the autoleveling routine every time. If I don't it starts printing a few millimeters above the bed, and extruding into air. All in all there's more to be dialed in, but at least I am getting close. I will probably grab a few of the neoprene washers you use from home depot, since I think the felt may not be optimal. I'm also going to print off different fsr mounts.

[Lightpath]

I do have something else going on that concerns me...

I use a 24v power supply with an SSR to heat the bed, and the Duet is wired in parallel. When I turn on the heated bed the SSR light comes on solid, my 24v peak fan slows to a crawl. The hot end takes forever to warm up with a 24v heater cartridge.

Typically when using the stock controller the SSR led would flash, indicating it was using pwm to heat the bed. Is there a setting to configure pwm on the bed heater output? On the plus side the bed heats up crazy fast. On the down side I'm betting letting the heat bed gobble up 750 watts of power at 24v is probably going to end badly. Although... It hasn't blown the fuse. I need to check on that.

At least now I know why I didn't bake my 12v fan/heater cartridge... This only started once I installed a 24v fan/heater cartridge.

[mhackney]

Yes there is a setting for PWM on the bed heater - read M301 here: http://reprap.org/wiki/G-code" onclick="window.open(this.href);return false; I use S0.8 to scale on a similar setup.

[geneb]

I got a refund on a Kickstarter project I'd contributed to, so I ordered two Duet boards. Blue MAX and Red Sonja will be my scratch monkeys.

g.

[Rjaallen]

Many thanks mhackney,

just got the duet up and running with the fsr autocal and latest dc42 firmware and it's working great,

your guides have been extremely helpful in this process!

Rob

[mhackney]

Thanks Rjaallen, great to hear.

[Eaglezsoar]

I got a refund on a Kickstarter project I'd contributed to, so I ordered two Duet boards. Blue MAX and Red Sonja will be my scratch monkeys.

g.

Darn, I was hoping the Orange Sonja would be the scratch monkey.

[geneb]

That would be Orange Menace, Eagle.

g.

[Xenocrates]

So here's the question Gene, how many Max's do you have, if two are scratch monkey's? And what have you named them all? I'm more than slightly curious.