[m0xte]

Note: former EE. I’ve taken it and the provided power supply to bits before I even fired it up.

1. They’re IRLR3103PBF operated within their SOA. They’re fine. Most MOSFETs fail short due to the construction of the MOSFET. I’ve actually got a few scars on my arms from the buggers exploding on me misusing them on purpose as RF PA’s as a ham. I’ve seen them blow up in various bits of expensive US made kit too (tektronix 22xx in particular grr)

2. I’ve had absolutely no problems at all with mine and adhesion.

3. A decent stepper costs as much as a Prusa i3. All the steppers at this price are crap. But they work pretty damn well.

4. These are luxuries really. If it jams the Bowden tube pops out and you clean it up and off you go. I levelled it in 3 mins myself. Skipped step I’m not sure if the scenario. Filament run out doesn’t happen if you pay attention to the Cura filament length and don’t try to wing using up the ass end of a roll.

5. I’m not paying for support. I’ll fix it myself. I’m quite handy with a screwdriver, a compiler and a soldering iron. The thing is just an AVR running open source firmware. Nothing complex.

The main thing is I can get functional chunks of spare parts for this next day via eBay.

As for comparing Chinese support to US support, I’ve had better luck out of the worst Chinese company than Microsoft and Apple.

[2muchcoffeeman]

This sort of attitude ignores the fact that a lot of people would rather spend time designing things and prototyping, rather than fix their printer.

I had the experience of getting and assembling a Prusa MK3 kit for myself, and getting access to a cheap Chinese printer at roughly the same time.

The Prusa cost more than twice as much and I've already made upgrades. Worked correctly right after assembly. I've already gone through several kilos of successful prints. And the frills just make it so much nicer to use: spring steel bed, auto mesh leveling, the Prusa fork of Slic3r just keeps getting better and better.

The Chinese printer required you to attach the gantry to the bed. But after weeks of tweaking. It still doesn't work correctly. It's completed prints. But none of them have been of decent quality. Now, some of the Chinese printers appear to be good. But do yourself a favour and research the specific models if you want to buy one.

Otherwise, if you really care about designing and prototyping, spend the extra cash and spend more time making things. There are a few brands out there. I reckon Prusa has the best kits if you swing that way.

[m0xte]

I think I’ve spent less time debugging it and having software issues than my HP 3636 inkjet though. I care about the result more than the fettling and it does me well.

[BoorishBears]

I have an Ender 3 Pro, it's seen little use and I regret the purchase despite enjoying using our office 3D printers (Taz)

It's just too finicky. Too many odd issues with jams, adhesion. I left it sitting for a while and for some reason when I came back to use it I had to play with the belt adjustment. The magnetic bed doesn't align properly so you need to intentionally line it up in a way that doesn't cleanly overlap the metal base.

And I'll admit recently there's a trend of Chinese companies in the space actually working on conscious iterative design based on feedback, especially Creality... but it's still obvious they're just reacting to whatever pops up, instead of trying to get ahead of what people want (which Prusa excels at)

Creality tried tried to fix the fan issue everyone complained about (fan exposed to dust and filament refuse) in the Ender 3... by mounting the board upside down with the fan mounted on the bottom millimeters away from your work surface. Which made any modifications that needed access to the board a massive headache (I actually didn't bother putting back half the screws, since to access them you need to put the entire machine on it's side... )

-

Long story short, when I want to print something, I want to know, this thing will be ready to go. The hard part should be designing it, if I can't rely on the printer to just work™ when the motivation is there, I just don't bother.

And this is someone who likes to tinker, it's just when I want to tinker on X, I don't want tinkering on Y to get in the way.

[jrockway]

I think he is referring to a step motor controller as a "stepper". People love the fact that Prusa uses Trinamic drivers instead of no-name drivers.

I personally don't understand why step motor control is something that needs to be outsourced. Just stick an H-bridge between the microcontroller and the motor and improve the control algorithm with every release.

[m0xte]

The “no name” drivers are a bit more complicated than an H bridge: https://www.pololu.com/file/0J450/a4988_DMOS_microstepping_d...

The typical H bridge has many failure modes and disadvantages which are handled by the controller ICs. Not to mention a lot less CPU overhead for the AVR running the show.

[jrockway]

Yeah, you can get some feedback from the controllers easily (and then omit physical endstops).

I am waiting for a 3D printer that uses servo motors, though. So many failed prints come from not knowing where the extruder is. But you can just measure exactly where it is, rather than guessing based on how much current the motor used.

[taneq]

Servos don’t actually gain you that much. If you’re meeting enough resistance to skip steps then your print is wrecked anyway.

[imtringued]

Servos will easily triple the cost of the printer.

[IshKebab]

I've done manual h-bridge style control and it just can't match a proper driver. You're going to get a ton more vibration and less power.

I use Leadshine ones normally. They're big but also very reasonably priced and work very well.