Auto adjusting first layer gap idea.

Im sure we all know how critical getting the first layer squished down is.
My old Mendel has a fairly flexible bed and a powerful geared extruder which all means that plastic will always extrude even if the gap is zero. It sort of makes its own gap even if I have adjusted it too close.
My new DIY printer is very different. The thick glass bed is very inflexible and the direct drive extruder will give up with any resistance. As the bed is so large I usually watch the first layer and adjust, as required, the Z offset by hand. This is possible as my 2 Z lead screws are linked by a belt. If I adjust it too close I can hear the extruder stepper skipping steps so I adjust just above that.

It occurred to me that this resistance should be detectable is some way, perhaps by stepper current or strain gauge and the gap adjusted accordingly. Im sure some code guru could ‘map the contours’ of the bed under the print (if not perfectly planar and level) and adjust the gap accordingly. The contouring could be blended back to level over the next few layers.

No doubt this is already being done somewhere?

Yes its been done, its called auto bed leveling. Very handy.
This one uses a servo: [www.youtube.com]
This one doesn't: [www.youtube.com]

Not sure if thats exactly what you wanted, but it gets your z height set perfectly which then means your first layer is whatever you set it to be.

The first one looks like a simple 3 point z probe which is ok if your bed is flat (not level)
The second one looks like it takes multiple points which is much better, however I was really thinking about using the extrusion pressure during the first layer skirt to sense the gap.

@Vince,

Monitoring the pressure inside the barrel is on my bucket list.
Barrel pressure monitoring would allow for a robust, quantitative part manufacturing.

Monitoring barrel pressure would allow for:
Increase object precision,
Auto adjust for filament diameter,
Detect temperature swings,
Detect slipping extruder,
Detect increasing resistance,
Detect blockage,
Optimize retraction,
Control oozing,
etc.

I think a redesign of the hot end would be required, an insert placed
between the thermal barrier, or a bespoke nozzle/die to mount a sensor.

What kind of sensor works best, how/where will you mount it?

A2, do you think stepper torque measurement is a reasonable starting point, if adjusted for temperature?

@Vince,

A combination of about 7 things would be "factory" , for example:

1) Airtripper Extruder Filament Force Sensor – Design & 3D Print
[airtripper.com]

2) Closed-loop stepper controller progress
[forums.reprap.org]

3) Barrel pressure sensor.
4) Barrel temperature.
5) Die temperature.
6) Extrudate temperature.
7) Extrudate force sensor.

This is over kill as a solution to the topic of this forum, but it will allow for the most control of your process.

Quote
Vince
My old Mendel has a fairly flexible bed and a powerful geared extruder which all means that plastic
will always extrude even if the gap is zero. It sort of makes its own gap even if I have adjusted it too close.

It occurred to me that this resistance should be detectable is some way,
perhaps by stepper current or strain gauge and the gap adjusted accordingly.

An extrudate force sensor might work if the sensor range is paired with the extrudate force against the substrate.

What I like about an extrudate (not extruder) force sensor is that it gives you a sort of an ensemble averaging of the whole system.
Off the top of my head I don't have a suggestion of how to separate the extruder push back force and the extrudate push back force.
I would have to reevaluate airtrippers rig to see what if any mods would be required.
A bench mark could be constructed with various parameters anchored/fixed.
This is a cool idea to test.

Note:
Airtripper is measuring the force generated at the proximal end of the extruder.
You want to measure the force (push back) of the extrudate against the substrate (distal end).

I never understand these discussions.. For now I have two printers running, both using an aluminium print surface with kapton on it and they almost never need to be calibrated on the Z-axis!

Both platforms are put together in the first try without making a prototype and they both work flawlessly from the beginning on. On the ABS printer (Air 2) I haven't set the Z-height since June 4 and on the PLA printer (MendelMax 1.6) I have just finished it, but since I re-calibrated yesterday the whole printer hasn't changed a thing in the Z-height. I might have to design new Y-axis clamps for the MendelMax, but besides that I really think making it so over-complicated is just generating more variables for failure.

However, I quite enjoy over-complicated solutions to a simple problem and would not mind the least seeing you guys develop this further. I think that as with each new technique you first need to go all the way and beyond before you can simplify and take a few steps back.

One question, would the cost of a hotend with this pressure sensor in the print head be very expensive? I can imagine the currently available pressure sensors cannot be mounted directly inside the print head because of heat problems, but I can imagine a membrane that would transfer the pressure onto a sensor or the likes?

Edited 1 time(s). Last edit at 01/03/2014 06:27AM by Ohmarinus.