I realized that the Trident I’m building needs some parts printed in ASA, which motivated me to actually, you know, use the Zero for the intended purpose.
Toasty Marshmallow work
Given that I still mostly want to switch it over to using a BoxZero styled chassis, I’d rather work more on the final config than poking around with the existing config, but also I’d like to get it fairly sealed up.
The first batch of changes was:
- Sealed the side panels along the edges with foam tape and then also added a strip to help close the gap for the electronics chamber.
- Printed some 1515 clips and then applied some uxcell neoprene insulation sheeting — it’s a recommendation from the Toasty Marshmallow thread; designed for use in cars but seems like it’ll handle the temperature of a printer to seal the top-hat.
- Tuned the rear chassis fan down to 35% because that’s all the cooling it really needs and, at the time, I thought that it was cooling the chamber itself too much.
First roll of ASA
I set it up without all of the mods fully done in front of a window with the fan running and printed a test cube based on what folks on the Voron discord had mentioned for 3DXTech ASA.
Using the Mk I human nose, I can detect a bit of ASA fumes being emitted by the top hat gap but otherwise the room didn’t smell that bad. However the VOC meter shows an elevated amount of VOC’s, which really goes to show that you cannot solely rely on the smell to know if you are safe.
So… as per my longstanding plan, I took my printer for walkies. I moved it down to the back patio where it’s got a lot of air to exchange with. When the sun’s out, it makes for a nicely toasty chamber temperature.
Kinda. I’ll get back to that later.
On the bright side, the settings I gleaned from the discord were about perfect. So I printed a bunch of stuff in ASA, some of which I’ve installed, some of which I haven’t gotten to yet.
The hazards of printing outside… or… you should not trust the toolhead board’s temperature sensor
After a few days of printing outdoors… the hot-end clogged, probably because the build chamber was actually too hot, and also the belts popped loose.
I discovered the hard way that the toolhead board’s temperature sensor is dangerously inaccurate for chamber temperature measurement and should not be trusted. It was a nice touch for the toolhead board to have a temperature sensor but it’s very poorly positioned. It’s right in between the build chamber and the electronics chamber, so it’s not representing either temperature very well.
One way to reveal this with stock hardware is to move the hot-end to around the center of the build chamber and get it heated up. I was finding that the thermistor on the hot-end would be measuring around 50°C and then the chamber temperature would be more around 30°C. So I was worried that the chamber was too cold for ASA when it was in fact quite hot, potentially dangerously so, which really became evident when I got a snap in 1515 thermistor mount printed.
The frustrating part was that I was printing the Dragon Burner toolhead and a ERCF-styled filament motion sensor, along with various structural spares, so this was at a really bad time.
Thankfully, I had already procured a spare heatbreak tube for my Dragon hotend, so it was all able to come back together more or less as it was.
Also, the camera bracket that I printed in PETG has already warped, which goes to show that you really don’t want to be using PETG inside of the chamber.
The dragonburner upgrade
I had a Sherpa Mini that I’d gotten a while back with the intention of having a good “starter” extruder to put on a Dragon Burner. And so I decided since I had to take the works apart, I might as well use the Dragonburner.
I used the Fabreeko Honey Badger 3010 performance fan and the Fabreeko Honey Badger 9200 RPM 4010 P-series blower. If nothing else, the performance fans are noisier?
- The narrative instructions don’t remind you to put the square nut into the slot on the hot-end mount. Also the fastener kit I got just in case there was something weird from West3D includes hex nuts instead of square nuts.
- I was a bit confused as to how to mount the 3010 fan and route the cable. I ended up looking at one of the videos to watch how someone else did it. Another exploded view would be nice there. It also turns out that I’d installed it backwards on the first try, yay.
- The PCB mount doesn’t tell me how to screw the PCB down. From trial and error, a M3x16mm screw with a nut on the other side seems to do the trick. Not singling out this particular design because there are way way way too many projects floating out there that forget to mention these things.
- I had a generic 3950 thermistor bead lying around but the hole is designed for one of the cartridge-style thermistors so I shoved it in there with a bit of thermal glue and then it also turns out that the generic 3950 thermistor bead was giving me bad data.
LDO Sherpa Mini
I went through and looked at what would be a not-overly-expensive new extruder since I’m replacing the existing MiniSB extruder and the LDO Sherpa Mini came up as being small and light and on sale when I was building a Fabreeko cart.
After all, LDO is the “premium” brand as far as people in the community are concerned.
I am glad that I did not assume that just because LDO was selling it, it was the premium version. I spent a lot of time fighting with it which was made a lot harder because it was very obvious that LDO had reduced the spec on the stepper motor and LDO provided no manual whatsoever whereas the Sherpa manual was very incomplete.
And, dono, I’m OK with the Sherpa manual being incomplete, because I gave them exactly $0.00 for their design. On the other hand, when the BOM specifies a LDO-36STH17-1004AHG6 or LDO-36STH17-1004AHG stepper but I got a LDO-36STH17-0354AH(G8T) on mine. I was not able to find any datasheet via LDO’s site or github repos but I eventually found some random PDF on the web. Which is, of course, dangerous because I could totally generate my own fake datasheet and nobody would have any way of knowing.
I found a datasheet for that extruder, but not on LDO’s site, so maybe it’s not accurate? I do know that my Formbot kit came with basically no manauals but at least it came with the datasheet for the stepper motor, so assuming that the random datasheet I found is correct, I think the correct run_current is 0.16A (0.35A spec × 0.707 × 65%).
Given that, as best I can tell, I got delivered a weaker motor, I do wonder what else has been changed from the open-source release. And, sure I could take it apart and measure everything but then I might as well have just saved myself the time and got the parts and printed a Sherpa kit, right?
I don’t know if I got the RIDGA gear set or the regular gear set that has problems with runout. I did see the Bondtech logo on the box so I assume it was potentially genuine? Assuming that they didn’t tweak anything else, this would presumably mean that the gear ratio is 50:8.
Neither manual specifies what sort of tension I’d want (one of the people on the Discord told me to try pulling the filament, tighten the knob until it doesn’t move, then turn it 2-3 turns after that) nor how much PTFE tube I needed.
Everybody treats the LDO products as “premium” but I wasted quite a few hours trying to figure out why nothing was making sense and it was made worse because I didn’t have a reference point for what “normal” was. For example, the Voron manual states “Typical rotation_distance values should be around 22.6789511 for Afterburner, Stealthburner and Mobius (update gear_ratio to 50:10 for Stealthburner with Clockwork 2 or 80:20 for Mobius).” And in the end, a few folks on the Voron discord who I don’t think are LDO employees helped me resolve some of the confusion.
I got the Sherpa from Fabreeko and it’s no longer on their site anymore, so I don’t need to warn you not to buy it.
More work
- I installed the Purge bucket so I could get a bit more printable room without dealing with skirt issues. And then I looked at the provided macro and my start macro and the accurate chamber temperatures I was seeing and did some major refactoring to my start macros. It’s actually a lot closer to the better print_start macro.
- I turned up the fan again for the electronics chamber since it wasn’t hurting the chamber temperature after all.
Further efforts
- I need to revisit how I’m routing my filament yet again. Routing through the side seems to be causing trouble with the Dragon Burner. I think I want to route it through the back of the printer so it’s probably going to be a semi-custom variant on the URBI.
- I have the tophat filtered exhaust ready to go, but I want to print a fresh panel for it.
- I have the Kirigami Bed Brace and Fan printed. Driftrotor was nice enough to print a custom version of this. I haven’t felt up to dismantling the bed just yet.
- I also have the Voron V0.2 deck plate cover for Kirigami mode that I printed out of PET-GF before deciding that I didn’t want to deal with it creeping printed, again waiting for the dismantling the bed energy.
- I have the Massive nuts mod printed as well. I might hold off on it until I’ve got the accent color ASA open?
- I also have some titanium bolts for the bed mount.
- There doesn’t seem to be any camera mounts that mount outside of the printer so I don’t fry the camera so I’m going to have to make one myself.
- I still haven’t done the resonance tests with an accelerometer so I can start to get this thing going at maximum speed.
- I probably want to either print another zerofilter or maybe a Nevermore v6.
Conclusion
This weekend, I’m annoyed because I was hoping to spend more time this weekend working on both of the albatrosses in my living room (the Trident and a secret project) instead of doing some maintenance because the toolhead umbilical board thermistor misled me and took longer because my LDO Mini Sherpa sucks.
My prints are probably not yet PIF-grade because there’s still tuning to do. But I’m pretty happy with the print quality. I found a few print-in-place bits that I knew never quite printed right on my Ender 3v2 and tried them out and they printed pretty well.