Bltouch vs Piezo
-
I've got a bltouch on the way I'll test it the same as we test the piezo probes. If anyone wants it after me to repeat the tests for authenticity they're welcome. I cannot imagine a deployable device will achieve the same accuracy. Whether you get 25 microns or 5 microns accuracy probably makes little difference to your prints, but when calibrating a delta more accurate is better.
-
I've used both. The piezo is more precise, and on a delta probing with the nozzle is the only way to go.
I use a bltouch on my corexy. It does well enough for my needs on it for now. I can do full bed (250^2) prints with a great first layer using 3 point autoleveling.
That said, I continue to tinker and try to think of a way to make a piezo work on my corexy as well, but haven't found a reliable enough configuration yet.
-
That said, I continue to tinker and try to think of a way to make a piezo work on my corexy as well, but haven't found a reliable enough configuration yet.
What issues are you having in the CoreXY?
-
I've got a bltouch on the way I'll test it the same as we test the piezo probes. If anyone wants it after me to repeat the tests for authenticity they're welcome. I cannot imagine a deployable device will achieve the same accuracy. Whether you get 25 microns or 5 microns accuracy probably makes little difference to your prints, but when calibrating a delta more accurate is better.
I got a deviation of 0.007 with my bltouch using a repeatability test macro found here somewhere. I remember getting the same result using marlin.
-
I got a deviation of 0.007 with my bltouch using a repeatability test macro found here somewhere. I remember getting the same result using marlin.
Using marlin and ramps i had pretty much similar deviation. Now on my new coreXY i get a deviation of around 0.02 - 0.06 unless i go very slow at around 120mm/min
I am very interested in your scripts (config, deployprobe and retrctprobe). -
Two features in 1.21RC2 may help:
- Separate probe type (P9) for bltouch in the M558 command
- Option to turn heaters off during probing. Some bed heaters are thought to interfere with bltouch.
The multi touch option is not compatible with P9 in M558, but will be in the next RC.
-
Is multitouch documented somewhere David?
-
Two features in 1.21RC2 may help:
- Separate probe type (P9) for bltouch in the M558 command
- Option to turn heaters off during probing. Some bed heaters are thought to interfere with bltouch.
The multi touch option is not compatible with P9 in M558, but will be in the next RC.
Awesome. Will the multi-touch average the results, or does it take just the last one?
I'll test out the P9 setup for the bltouch ASAP.
-
I've used both. The piezo is more precise, and on a delta probing with the nozzle is the only way to go.
I use a bltouch on my corexy. It does well enough for my needs on it for now. I can do full bed (250^2) prints with a great first layer using 3 point autoleveling.
That said, I continue to tinker and try to think of a way to make a piezo work on my corexy as well, but haven't found a reliable enough configuration yet.
On my corexy with titan aero this sensor plate idea https://www.thingiverse.com/thing:2526978 works well for me. Underbed seems problematic as the inertia of the bed moving up to probe seems to trigger the piezos, even when set to very low sensitivity and very low acceleartion. I will try the balanced piezo idea (http://forums.reprap.org/read.php?1,767998,810011#msg-810011) to see if it helps. But assuming it is inertia, not noise (belt driven z axis so not mechanically noisy) then it might not help.
I'd say on a corexy hotend based piezo is the way to go. We're testing some new ideas and my corexy is the test machine.
-
The multi touch does multiple taps up to a configured maximum until two consecutive taps produce height readings within a selectable tolerance of each other.
-
I've used both. The piezo is more precise, and on a delta probing with the nozzle is the only way to go.
I use a bltouch on my corexy. It does well enough for my needs on it for now. I can do full bed (250^2) prints with a great first layer using 3 point autoleveling.
That said, I continue to tinker and try to think of a way to make a piezo work on my corexy as well, but haven't found a reliable enough configuration yet.
On my corexy with titan aero this sensor plate idea https://www.thingiverse.com/thing:2526978 works well for me. Underbed seems problematic as the inertia of the bed moving up to probe seems to trigger the piezos, even when set to very low sensitivity and very low acceleartion. I will try the balanced piezo idea (http://forums.reprap.org/read.php?1,767998,810011#msg-810011) to see if it helps. But assuming it is inertia, not noise (belt driven z axis so not mechanically noisy) then it might not help.
I'd say on a corexy hotend based piezo is the way to go. We're testing some new ideas and my corexy is the test machine.
I'd love to come up with something for the RailCore II. Perhaps a plate like your titan, though we don't have a config for an aero at this point.
I have a spare piezo board & piezos, if there's anything you come up with, I'd love to test it.
My bed is 3 direct driven leadscrews, so underbed is almost definitely not an option.
-
The multi touch does multiple taps up to a configured maximum until two consecutive taps produce height readings within a selectable tolerance of each other.
Gotcha, thanks.
-
How do we use the tolerance setting? What units is it in? Can you give an example of how it would be used?
-
On my delta I use this:
; Z probe and compensation definition M558 P8 H3 F1000 T10000 R0.4 S0.02 A5 ; Z probe is delta effector, Z probe dive height 3mm, probing speed 1000mm/min, travel speed 10000mm/min, max 5 taps, tolerance 0.02mm G31 X0 Y0 Z-0.10 P100 ; Set the zprobe offsets and threshold (put your own values here)However, there is a bug: the recovery time is not applied before the second and subsequent taps. Not a problem if the recovery time is only needed to recover from the horizontal travel move.
-
Thanks that makes sense. I agree the recovery time is probably not relevant on subsequent probing dives.
