Can we make extruder return home after every layer to check for missing steps?

I think this is a simple way to check for missing step, no need for closed looped stepper motors. Perhaps a small program that reads the G-code, looks for when Z increased, then inserts "G28 X Y ;". This will allow the printer to move correctly every time, even after a missed step.

However, I'm not sure if the stepper motor can keep up with the transition. Maybe it needs G1 code with X Y coordinate first, then the next line. Another problem is oozing, It may need G10 and G11 to stop some of the oozes.

I don't know how should I write the program, but I think I got the algorithm:
-check where Z increase then insert
-G10;
-G28 X Y;
-G1 X*** Y*** Z***; (the *** is replaced by the coordinate of the first move in)
-G11;

I have no idea if this will work or how well it would work, what do you guys think?

How are you going to determine if a step was missed? How much time are you willing to waste? If your printer is missing steps, you should be looking for the cause.

The limit switches used for homing are usually not especially precise. Besides slowing the print, you'd probably end up adding Z wobble to you print problems.

If your printer is missing steps, as the other poster said, find and fix the problem.

---

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

Fyi, I have owned a 3D printer for a year or so, I know how loose belt result skiped step or not enough current on the stepper motor result in motor moves to the nearest step, or too high jerk results in quick changes in movement, I know all the problem.
This is just a concept to make 3D printers a bit easier to use, perhaps exchange the speed for reliability so that it can be more consumer friendly printers. As for the end stop, It can be changed into an optical one for better precision. The Z axis wobble can be fixed by using a ball bearing and held the axis in place.

Edited 1 time(s). Last edit at 06/30/2017 01:28PM by Shizen.

You misunderstood my comment. If the home position switches are not precise, the extruder carriage will stop at a slightly different location each time you send the carriage home. Then, when you resume the print, the extruder carriage will miss the target point by that variable error amount, layer after layer and the result will be layer misregistration that looks a lot like Z wobble in the print. Then there's the filament that drools out of the extruder while you're going back and forth between the print and home. Then there's the hairs that may form on the print as you go back to home position at the end of each layer.

All these problems that you're creating have solutions, but why create them in the first place?

Edited 1 time(s). Last edit at 06/30/2017 03:15PM by the_digital_dentist.

---

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

Well, I have no idea how much difference moving to home each time would make. I have aware of oozing each time moving to home as I mentioned in my topic.

Like I said before, I'm just proposing a new concept, I haven't tested it myself so I have no idea how efficient, effective or practical it may be. You can't tell if it's a problem until you try it yourself (I can't do it yet since I don't know how to write such program).

If any steps have been skipped, it will be a multiple of 4 full steps. So what you could do is a modified XY homing move that stops when endstop switches are triggered, but doesn't reset the axis positions. Then the firmware can adjust the axis positions by a multiple of 4 full steps if necessary to get them as close as possible to the home positions. That way, a small amount of jitter (i.e. less than one full step) in the homing switch trigger points wouldn't cause Z wobble.

I'll give some thought to implementing this option in RepRapFirmware. Most slicers let you set up layer change scripts, in which you could call this function, along with any retraction etc. that you want to do.

Edited 1 time(s). Last edit at 06/30/2017 05:52PM by dc42.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Just a question, linking to the threads on 1.19 beta and step errors, for example [www.duet3d.com] how does the m122 report calculate the step errors?

---

Simon Khoury

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

The step error count in RRF is just a check in the firmware that the calculations have given a sensible result, to guard against rounding error and firmware bugs. Nothing to do with missed steps due to the mechanics snagging.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

I could see having a "fail-safe mode" where the head moves to what it expects to be home at the end of a layer (maybe every tenth layer or something) to check and see if the home sensors trigger when expected. If they trigger much too early, or don't trigger at all, then it stops the print and alerts the user with a text message or something.
Not to rely on the home sensor for layer alignment, but just to keep the printer from wasting extra time on a garbage print.

That's a great idea.

I wonder if there is any real data on endstop repeatability? Maybe a test routine could be devised for RRF which would enable testing of endstops. That way we will quickly clear up how precise and repeatable they are. If this testing done by Russ Gries on duet forum [www.duet3d.com] is anything to go by micro-switches are very accurate with some getting down to 15 microns accuracy and repeatability (brand name switch with no levers).

But I'd like to compare them to optical, and we even have a piezo endstop design to test.

---

Simon Khoury

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

Virtually every type of switch is repeatable <0.001" over short time periods. Limit switches, inductive switches, and photointerruptor/optical sensors are frequently used in CNC builds.

The problems arise if you need a repeatable position over long time periods. If you show up a year later with a different temperature, humidity, atmospheric pressure, and supply voltage; it is very difficult to judge how inductive/magnetic/optical sensors will react.

Almost all of the designated repeatable homing switches I've seen are mechanical plunger types. Omron and similar run $120-300 a piece. Similar parts from misumi would be [us.misumi-ec.com] and [us.misumi-ec.com] which repeat to 5um and 0.5um respectively.

That's useful information.

But we are talking about repeatability during one single print, if we are thinking about an occasional re-home to check for missed steps (probably unecessary with a correctly setup printer) or just a quick check to ensure the machine has not got out of shape and save an unattended machine from carrying on regardless that will suffice, yes?

---

Simon Khoury

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

I just can see any good reason to home my delta on every layer. The home position is 290mm from the build plate. Endstop repeatability depends on the speed at which they're triggered, so you can't just rapid till the endstop triggers.

If my printer missed a step and the print failed, I'd rather correct the problem and reprint. It will consume far less time than rehoming every few seconds.

I think Feign's idea of it being an optional periodic check every x layers might have some merit. But I do wonder if this is a solution looking for a problem, a bit like using closed loop motor systems which aren't worthwhile with respect to current reprap 3d printers, unless it's to use cheap dc motors to make a cheap mass market printer.

Edited 1 time(s). Last edit at 07/11/2017 06:09AM by DjDemonD.

---

Simon Khoury

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

If you had an encoder on each stepper but did not try to use this to make a true closed loop it but purely to warn of missed steps this would serve the same purpose as homing without taking any extra time. If this worked it could be expanded to gently tweak things for minor errors or announce if slippage had exceeded some limit - the complexities of a true stepper closed loop would be avoided just by looking at it from another point of view.

Mike