If it ain't broke, don't fix it!

Jeez, I had my printer all dialed in and working nicely, then I decided to make it "better".

It had a very heavy transformer to power the 450W 24V bed heater. I wanted to reduce the weight so it would be a moderate 2-man lift instead of a heavy 2-man lift, so I got a 24V 31A power supply, pulled the transformer, and a bunch of other stuff out including the 12V supply that was running the smoothieboard and Z axis and extruder motors. Almost everything runs on 24V now.

The install went fine until I tried printing- suddenly I have blobs at the start of extrusion after retraction. I have tried setting the "recovery" length to negative values, tweaked the extruder motor current, retraction length, played with wiping, coasting, and reran PID tuning for both bed and extruder, etc. and so far I can't get it back to the way it was.

ugh!

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

You have my sympathy, 3D printing often seems more art than engineering. "Improving" something sometimes makes it worse and the obvious answer is often just plain wrong.
Despite being a decently scientific/engineering/critical thinking kind of a guy, lately I have felt a strong need to sacrifice a goat to ensure that my printer does what it logically should.

Mike

p.s., my first thought is that the blobs are caused by the rotor on the extruder motor overshooting try tweaking the current down.

Let us know if the goat sacrifice works, it'll be bad for goats if it does but often this is just the kind of lateral thinking I'm looking for when "improvements" result in nothing of the sort.

My new delta was printing just fine until I decided to try a workaround to get babystepping to work on reprap firmware, resulting in a bad head crash, lots of nice scratches on my new tooling plate bed, and a magnet popped out of my effector - and it was working perfectly before I decided to experiment.

The madness of it is that I'm going to try it again this evening as I'm determined to make it work.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

I think I have the problem figured out. Some test prints exhibit blobbing at the start and end of extrusion just after and before retraction for layer change. The blobs at the end may be fixed by "wiping" in the slicer and the blobs at the start may be due to too high retraction speed/acceleration. I think it may be skipping when it retracts, (30 mm/sec) so it doesn't retract as much as it is supposed to, then when it recovers from retraction at a lower speed (20 mm/sec), it recovers the full amount of the retraction because it doesn't skip. That will produce a small blob at the beginning of print after a retraction.

I'm running a print now, but will do more tweaking of the retraction acceleration and speed when it has finished.

One weird thing is that since the power supply went from 12V to 24V I had to change the extruder steps/mm. Why would changing supply voltage change that? It went from 470 to 506 steps/mm with the power supply change.

Edited 1 time(s). Last edit at 09/20/2016 09:17AM by the_digital_dentist.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

If anything you might expect it to be the other way around, perhaps there was some skipping at 12v but none at 24v and therefore a lower steps/mm value required.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

I think the original steps/mm was found with the hot end off the extruder. The new number was found with the hot-end in place and heated up. Maybe the mechanical load causes some change. Slipping? Not likely...

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

450W at 24V? That is almost 19A! You must have some pretty hefty cables for that. You would need to be using 14 gauge wire minimum. At that level, it might have been a better upgrade to switch to an AC powered unit.

I considered switching to AC power, but the bed moves in the Y axis and I'd prefer not to do that with AC power, though I suppose having 24V at 30+ amps available isn't much different safety-wise than running on AC power. The final decision was based on cost. A new heater would have cost $50-70 and a small 24V supply would have added another $30 or so. I got the big power supply (Meanwell SP-750-24) for $80.

I use a pair of 18 gauge teflon insulated wires for each side of the heater. Since the heater only operates at full power during the initial heat up, about 5 minutes, the slight temperature rise (maybe 5C) in the wires doesn't matter. I could probably use a single 18 gauge wire for each lead without any problems. If I give up a couple watts to heat the wires the bed will still have plenty of heat.

I thought about using the original heater with a lamp dimmer to keep the power into it down to 450W directly off the power line, but if the dimmer failed it might put 117VAC across the 24V heater and that would be very bad.

The CoreXY machine I am building will have an AC powered heater. I have a 24V 10A supply to power the rest of it.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

The moving bed is not an issue if you have the connections properly anchored so it doesn't experience any flexing. I have two AC powered heaters on moving beds and haven't had any issues. Personally, I have experienced nothing be problems with DC powered heaters. I know they work as lots of people have had success with them. I'm just not one of those people I guess.

I found the largest improvement in my retraction (for a direct drive extruder, nema 11 with 5:1 planetary going into an e3d v6) by slowing the retraction speed down to 5 mm/s. Actually, I determined at what speed my extruder could extrude an unlimited length of filament into free air (through the nozzle) without skipping, and I used that as my retraction speed. I use a length of between 0.2-0.4mm, and I get picture perfect retraction. Try it.

That's exactly what I am thinking- I also have 5:1 gearing. I've been running with retraction at 30 mm/sec, but I'm thinking that's too high. I'm not sure why it would suddenly become a problem when I increase power supply voltage, but weirder things have happened. Maybe the higher voltage has shifted the resonance right to 30 mm/sec point and now it skips. What acceleration are you using?

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

I printed six of these last night- it isn't a great test of retraction but it does look like the extrusion is pretty well dialed in. I used 250 um layers for the first 3 mm then switched to 100 um layers. This is the part as it came off the bed. It has a few tiny blobs that will flick off easily.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

For my extruder I'm running 2400 mm/s/s acceleration, with 120 mm/min (2 mm/s) jerk rate. It does look like you have it working pretty well again.

Electricity is weird.

Edited 1 time(s). Last edit at 09/20/2016 02:52PM by n8bot.

After struggling with resonance problems in my Y axis I seem to have finally found the solution. The last time I had it apart to lube the ball screw I measured the bearings. Even though it has a metric pitch (10mm), it turns out that it uses 1/8" balls! I ordered a package of them and removed all the old balls and replaced them with the new ones today and some of the resonance problems seem to have disappeared. It runs smoother and quieter now than before and when it hits resonance it doesn't make the awful shuddering noises it used to make. It also doesn't seem to skip steps when it hits resonance. I'll have to run a bunch of prints to test that.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

It's funny, I mentioned that once my printer was about perfect I started diddling around and had to fix it. Not having an issue is somehow kind of a let down. Anticlimactic. Without the challenge your left with just printing. I decide to finish drawing my new 3 point leveling bed. That's your fault by the way DD.

One more thing... This bit me long ago when I was installing the smoothieboard for the first time... It was driving me nuts with problems that didn't seem to respond to changes I was making to the config file. After screwing around with it for a week someone suggested I try a different uSD card on the board. That fixed everything, instantly.

I decided to try the same thing again- I grabbed a different uSD card, formatted it, copied the firmware and config files to it, plugged it in, and all the problems went away.

I think the problem is that I was editing the config file via USB plugged into the board, then sending an @reset command via pronterface to reboot the board without ejecting the card from Windows first. Doh!

All is good- back up and running and producing great prints again...

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

Glad to see you are up and running again.

Had a similar issue with my printer. I keep my old digital camera at hand and use that to format my SD cards.
It formats faster than SDFormatter and it seems to give them new life, at least 1 out of 2.

Lykle