Height of printed part is incorrect

8

I have a weird problem with my old 3D printer, it is a Prusa/Mendel type. When I print a 20 mm cube, X and Y are correct, Z is resulting 16 to 17 mm. I have checked the correctness of the movement on Z using the manual controls and there are no issues.

I played a bit with the layer thickness, I have a 0.4 mm nozzle, setting the layer height to 0.12 mm (normally is on 0.16 mm) but no changes in the result. I am printing PLA on a cold bed at 180 °C without any other particular defect.

I would appreciate some direction on how to solve such problem.

FeliceM

Posted 2016-09-30T15:51:15.900

Reputation: 531

This could be related to missing steps – 0scar – 2019-07-05T08:46:53.030

1Is the Z travel consistent across prints? Z travel must be at the root of the problem as the nozzle wouldn't finish the print if there were a 3mm gap towards the end of the print. – emackey – 2016-09-30T17:39:49.727

@emackey there are not other defects on the printed part. I am not sure how to verify such consistency while printing. As i said, I verified it with the manual control and in that modality it is correct. Thanks fornyour opinion about excluding the extruder settings. – FeliceM – 2016-09-30T18:05:08.003

Hi, Felice! I think what @emackey mean about Z-travel is whether the height difference ratio is consistent. So, if you print a box that is supposed to be 50mm high, and it turns out that it is 40mm, you have a 20% error. If the error is consistent, printing a box that is supposed to be 100mm should give you a 80mm result (also 20% error), etc.. – Tormod Haugene – 2016-10-01T07:59:08.710

1@TormodHaugene, thanks. Unfortunately the result is inconsitent. I printe out a 20 mm and it came out 17, the next one was 40 mm and final result is 30. I am starting to consider a problem on the ramp. – FeliceM – 2016-10-01T17:54:05.480

Is this consistent on different Filaments? I once had a bad filament that caused to short prints. – Lars Pötter – 2016-10-02T21:34:42.710

Answers

3

You should check that the steps per mm for your Z-axis are set correctly. This depends on the pitch of the leadscrews/threaded rods driving the axis and parameters of your steppers (microstepping and raw steps/revolution). This Calculator.

Make sure that your layer height is a multiple of a full step of the Z-stepper. The Z-stepper may be disabled intermittently, and when re-enabled it may "snap" to the nearest full step position. If your layer height requires microstepping, you may notice it getting rounded down or up due to this.

For instance, if a full step were 0.08 mm, then 0.16 mm layers would require 2 full steps, printing fine. 0.12 mm layers would require 1 full step and a half microstep. Due to rounding, some layers might be reduced to 0.08 mm instead. This might account for the height discrepancy you're seeing (though 0.08 mm is quite a high, unrealistic amount for a full-step).

Tom van der Zanden

Posted 2016-09-30T15:51:15.900

Reputation: 14 003

Thanks for your nswer. As you said the steps calculation are the result of the mechanical components, so there is no reason to adjust it to compensate a problem coming from somewhere elese. I also tried with 0,16 and 0,2 mm extrusion but the problem is still the same. I am considering the problem coming from the ramp i will try to play a bit with the ramp current. Thanks for your answer, i will check carefully your hints and share the results. – FeliceM – 2016-10-01T17:49:08.827

2

You could check that the stepper driver IC for the z direction is not getting too hot while printing. If this is the case it may be that the current adjustment of the stepper driver is set incorrect causing it to skip steps. however you would expect the same behaviour in manual operation.

Smitje

Posted 2016-09-30T15:51:15.900

Reputation: 61

2

Thanks to your advices I pointed my effort on the root of the possible cause on the Z axis. I swapped the Polou driver between x and z and clearly noted that now the problem is on x. I had to conclude that the problem is with the Polou driver which randomly looses steps. What is surprising me is that on the Z axis I have two motor type Nema 14 which should work at 0,75A therefore 1,5 A total. The Polou driver, old type, is rated at 1A. So it is not a surprise, now that I have learned about this configuration, that the z driver is giving troubles. The problem was there since the beginning and with the ageing of the components it revealed itself. I will replace the actual Polou drivers with the new type rated at 2 A and restart from there. Thanks for the advices.

FeliceM

Posted 2016-09-30T15:51:15.900

Reputation: 531

Maybe adjusting the current (via the potentiometer) to be lower/higher will fix the problem without replacing the stepper drivers. – Tom van der Zanden – 2016-10-03T17:34:14.820

@TomvanderZanden, thanks but I tried. The driver is faulty because gives the same problem also on x where i have only one motor. I even tried to adjust the current but no success. I am replacing it. – FeliceM – 2016-10-05T11:54:20.883

0

A suitable test part for this problem, to check if the bad scaling is linear or the result of missing steps (as was identified in this case) would be a ramp, or sequence of steps. You can check the linearity using a straight edge, and would be able to perform repeated tests to identify of there was a specific Z location where steps were more likely to be lost (for example due to some tightness in the motion system).

Sean Houlihane

Posted 2016-09-30T15:51:15.900

Reputation: 3 622