What are the minimum hardware features for 3D printing

Extruder heater control
Bed heater control
Fan for Extruder
Sensor for bed leveling (I do not think stall detection works for that)
End stops can be replaced by Stall detection
X, Y, Z, E DIR and STEP
Bed temperature probe
Extruder temperature probe

Is Trinamic stall detection reliable or does it need a mechanical switch as a backup? Did I miss anything?

What are you trying to decide/do? There are all sorts of "hardware features" that you haven't listed, and sensor for bed leveling is not needed depending on how you build the machine. Are you talking about a delta or some form of cartesian machine or something else?

---

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

When I wrote my question I had cartesian printer in mind, however I do not see how different it would be from a delta. Primarily I wonder if stall detection is reliable for homing for cartesian and delta printers. I misspoke when wrote 'bed leveling' as I also had Z homing in mind. I understand that Z homing will not work with stall detection if bed has springs or soft build surface.

Expanding on what you said, what are nice to have features?

filament sensor?

I wouldn't call any of those features. With the exception of the stall detection, those are found on every single printer. I would call them fundamentals instead of features.

What you should be looking at is a parts list. "What do I need to get running?"

frame and bed assembly
extruder
hot end
bed heater
linear motion
belts and pulleys
steppers
controller and drivers
power supply
limit switches
fans
physical size

None of your fancy features matter if your machine isn't properly designed and built.

Features are:
multi tool/extruder
enclosure
enclosure heater
filtration
lighting
farm capability
online control
mesh leveling (kind of a crutch to poor build quality)
etc...

Edited 2 time(s). Last edit at 02/25/2020 02:46PM by boredom.is.me.

I should clarify that I'm asking from the angle of what controller would I need.

Multi extruder would require temperature sensor, cooling fan, heater and motor support for each additional extruder. Enclosure would require another mosfet. Mesh leveling would require additional 1,2 or 3 Z motors. Would printer controller control filtration or lighting? I do not think so.

I wonder what is better PT1000 or thermocouple. Personally I think thermocouple because it responds faster and accuracy to few deg is sufficient but it appears PT100/1000 are more popular.

So what you want to know is which controller board you should buy? Are you building a printer? What are you going to put into it?

Any 8 bit controller will run a basic cartesian printer. A Duet 2 or 3 with an expansion board will run just about any printer you want.

The final constraint which is usually the primary constraint, is the cost. How deep are your pockets? If you've only got $25 the answer is easy. If you have $300 there are more options.

---

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

I'm certainly aware of your successful designs. I put together a (cartesian with drill rods as rails so it was a joy to tweak) printer years ago. As I (and my kids with me) learn by doing I want to finish a controller for a printer and have been pondering what to cut out.

So far it seems I can leave out all the end-stops except for Z axis and filament sensor when using Trinamic drivers.
Substitute power mosfet for bed heater for gate driver instead.
Ditch 5V fan power supply and corresponding mosfets
Move away from thermistor sensors to PT1000 and/or thermocouple

Based on boredom.is.me, I think it's worth to have extra mosfets for additional extruder and enclosure heater. Seems as 6 driver controller would be most flexible - XYZEEE - although 7 or 8 would be pretty useful also: XYZZZEEE.

Are you making a controller or a printer?

You should understand something about using multiple Z axis motors. If you use multiple motors, you'll need auto tramming because multiple motors will get out of sync. But that requires extra drivers, cables, and some sort of sensor mounted on the extruder carriage to tell the controller when the bed is trammed, and usually a lot of messing around to configure it, and a lot of trouble when it isn't working right. If you drive the Z axis screws using a single motor connected with a loop belt, the screws won't get out of sync and the bed won't tilt. You'll tram the bed once when you set up the printer and won't have to do it again. It doesn't make impressive, exciting youtube videos, but it's simple, cheap, easy, and reliable.

Endstop switches are simple, cheap, and reliable. If you use some of the little modules that have an LED- either mechanical switch or opto- you can see when the switch is activated without having to mess with the controller.

Thermistors work fine for the bed for any print material, and in the hot-end up to ABS temperature. If you're going to print high temp materials, PT100 or 1000 is good in the hot-end. Thermocouples and PT100/1000 sensors require extra interface boards. Beware cheap PT100/1000 sensors that have teflon insulation on the leads. You won't be able to use those for high temperatures- look for glass fiber insulation (hint- if it costs <$15 it's the teflon type). Thermocouple leads are solid steel wire and won't stand up to repeated flexing. If you cut the steel wires short and attach more flexible wires, you've just created two more junctions of dissimilar metals that will affect temperature readings. You don't really need and can't really get accurate temperature readings. Temperature sensors read temperature where they are located. Move a few mm away and all bets are off. Where is the right place to measure bed temperature? Do you think the entire bed will be at that temperature? Think again. Fortunately, all you have to do is get the temperature in the ballpark. You're going to experiment with the settings to use for your printer and may find that you need to set the bed to 110C for printing ABS, or maybe only 90C. It's OK. Use the setting that works. The same goes for the hot-end temperature settings.

Unless your printer is very small, your enclosure heater will need a lot of power. Bed heaters also need lots of power. Line power switched by SSRs is best.

Edited 1 time(s). Last edit at 02/26/2020 10:37AM by the_digital_dentist.

---

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

From reading issues on the Marlin Github, IMHO sensorless homing on Trinamic drivers is not yet ready for prime time. It's a lot less accurate than a micro switch and very sensitive to any bit of extra friction in the axis.

I can imagine two scenarios where sensoreless homing would fail:

slop ('play') in the carriage that causes deflection which is not repeatable causing accuracy to decrease.
uneven linear motion resistance due to misalignment, bearings, etc which would trigger static threshold or cause to increase threshold to exclude false positives at the cost of accuracy

Reading corresponding forums, it seems people using Duet WIFI appear to have a better results than Marlin users. This could be due to more better hardware and or better tuned software.