Welcome! Log In Create A New Profile

Advanced

Precision Piezo Z-probe Now available.

Posted by DjDemonD 
Re: Precision Piezo Z-probe Now available.
September 12, 2017 07:22AM
Vasan - I thought about putting it between titan and motor but there is not much space due to the motor having the flange on its mating surface. If you intend to mount it via a bracket try our titan bracket design or see what you come up with. We're giving away a free precision Piezo kit to the first 10 people to come up with and share a mount /piezo sensor design that works to 0.01mm accuracy and 0.007mm std deviation. See precision Piezo Bounty in this forum.

Edited 1 time(s). Last edit at 09/12/2017 07:24AM by DjDemonD.


Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions
Re: Precision Piezo Z-probe Now available.
September 12, 2017 08:13AM
Quote
DjDemonD
Quote
klcjr89
I am not sure what to change in my start script.

[Try G30 Z0 instead of G28 Z if it's not working with G28.

Also consider moving to bed centre, then M400 to pause (I presume that's a 400ms pause?) then G28 Z or G30 Z0. This allows mechanical noise to settle before probing.

I can home fine with G28, but issuing a G30 or G30 Z0 does nothing at all and just gives a vague message:

SENDING:G30 Z0
ZProbe not triggered
Re: Precision Piezo Z-probe Now available.
September 12, 2017 08:39AM
Have you tried defining a spare pin as z probe and just connecting the signal output from piezo board to both that pin and your z endstop pin? If g28 works then I suspect it's the z probe and endstop not sharing a pin issue. Silly really there's no reason they can't.


Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions
Re: Precision Piezo Z-probe Now available.
September 12, 2017 09:07AM
Quote
DjDemonD
Have you tried defining a spare pin as z probe and just connecting the signal output from piezo board to both that pin and your z endstop pin? If g28 works then I suspect it's the z probe and endstop not sharing a pin issue. Silly really there's no reason they can't.

I did this and posted it on the previous page unless you mean something different to try: [forums.reprap.org]
Re: Precision Piezo Z-probe Now available.
September 12, 2017 09:12AM
Yes that was it must have missed that sorry.


Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions
Re: Precision Piezo Z-probe Now available.
September 12, 2017 09:20AM
At this point I'm disappointed, not at the piezo, but no reply from Arthur. They actually make it hard to get support. Things really shouldn't be this difficult to setup.
Re: Precision Piezo Z-probe Now available.
September 12, 2017 09:55AM
I wouldn't take it too hard 3d printing is difficult if it was easy it wouldn't be so satisfying when it works. It would just be a 2d printer for £40 from the supermarket.


Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions
Re: Precision Piezo Z-probe Now available.
September 12, 2017 10:02AM
Digging through google, would this be related to my problem perhaps? [groups.google.com]
Re: Precision Piezo Z-probe Now available.
September 12, 2017 10:30AM
Maybe because inductive sensors float between triggered and not triggered as they're design to switch relays to stop door closing motors or similar. Try a small load if you like in series with the signal like an led or low value resistor.


Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions
Re: Precision Piezo Z-probe Now available.
September 12, 2017 10:40AM
If you turn the left hand pot till the LED is permanently lit then do M119 what is the result?

Idris


{Precision Piezo} Accurate, repeatable, versatile z-probe plus piezo discs, endstop cables, pt100, 50w heaters.
Re: Precision Piezo Z-probe Now available.
September 12, 2017 03:27PM
If the LED stays blue the result is that it's triggered.

Edited 1 time(s). Last edit at 09/12/2017 03:27PM by klcjr89.
Re: Precision Piezo Z-probe Now available.
September 12, 2017 05:07PM
If I use the X axis endstop pin for testing, it does the same thing so it can't be a wiring issue.

Edited 1 time(s). Last edit at 09/12/2017 05:07PM by klcjr89.
Re: Precision Piezo Z-probe Now available.
September 13, 2017 04:19AM
I have made some progress, albeit in the hacker spirit (nice thing about open source). So inserting print statements into the Zprobe.cpp file and also cleaning up the config .txt file, I can now probe with G30 (haven't tried G32 yet).

Over 10 probes starting at a position 299mm away from the bed, the max/min deviation between the lowest value of the 10 probes and the highest value of the the 10 probes is: 0.0213mm (21.3 micron).

The standard deviation over the 10 samples is: 0.006590262mm (6.6 microns)

Would love to work on getting the max/min differential lower but are my results are what to be expected of a tuned Piezo20? Thanks!



Edited 2 time(s). Last edit at 09/13/2017 04:30AM by klcjr89.
Re: Precision Piezo Z-probe Now available.
September 13, 2017 04:33AM
The deviation is about expected the range is usually 10 microns, so maybe lower probing speed, or slightly tweak the vr2 (left) pot to be closer to the threshold between led on and off. But 20 microns out of a 200-300 micron first layer will work.

Good work on the code changes.

Edited 1 time(s). Last edit at 09/13/2017 04:34AM by DjDemonD.


Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions
Re: Precision Piezo Z-probe Now available.
September 13, 2017 04:43AM
Probing over 299mm you'll also be affected by the accuracy of your z-axis drive.

Idris


{Precision Piezo} Accurate, repeatable, versatile z-probe plus piezo discs, endstop cables, pt100, 50w heaters.
Re: Precision Piezo Z-probe Now available.
September 13, 2017 04:43AM
Quote
DjDemonD
The deviation is about expected the range is usually 10 microns, so maybe lower probing speed, or slightly tweak the vr2 (left) pot to be closer to the threshold between led on and off. But 20 microns out of a 200-300 micron first layer will work.

Good work on the code changes.

Did you see where I was 299mm away from the bed, and returned back to Z299 between each probe? That might be more a testament of the machine's leadscrew and stepper but doing the same test over again 20mm away from the bed yielded 5.8 micron standard deviation.
Re: Precision Piezo Z-probe Now available.
September 13, 2017 04:45AM
At these sort of numbers I agree with dc42 we are probing as accurately and repeatably as 3d printing requires and the machines are capable of.


Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions
Re: Precision Piezo Z-probe Now available.
September 13, 2017 04:48AM
Next up: electrically probing and using plasma to clean the nozzle smileys with beer

Am I crazy?

Edit: G32 is working wonderfully now.

Edited 3 time(s). Last edit at 09/13/2017 05:05AM by klcjr89.
Re: Precision Piezo Z-probe Now available.
September 13, 2017 05:29AM
Electrically probing is the way to go. If you know where you can get conductive PEI for cheap then I have a great product idea....

This thread was a good test [www.duet3d.com]

the control was nozzle to aluminium and it yielded the most accurate result, but don't want to print onto aluminium, do we?


Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions
Re: Precision Piezo Z-probe Now available.
September 13, 2017 05:38AM
Would electricity (plasma) also be a method of cleaning the nozzle? Or since it takes the path of least resistance, it might only clean a portion of the nozzle? sad smiley
Re: Precision Piezo Z-probe Now available.
September 13, 2017 05:59AM
Giving it a good vigorous pass through a wire brush with nozzle at 150 should do it.


Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions
Re: Precision Piezo Z-probe Now available.
September 13, 2017 06:08AM
Quote
DjDemonD
Giving it a good vigorous pass through a wire brush with nozzle at 150 should do it.

Do you know what this setting is for? The doc description doesn't make sense to me:

And if you do know, what should I set it to for the Piezo20?

Edit: this setting doesn't look like it's used for the rectangular grid strategy, only the 3 point strategy.

leveling-strategy.three-point-leveling.tolerance     0.03           # The probe tolerance in millimetres, anything less that this will be
                                                                    # ignored, default is 0.03mm

Edited 3 time(s). Last edit at 09/13/2017 06:14AM by klcjr89.
Re: Precision Piezo Z-probe Now available.
September 13, 2017 07:45AM
So I have a rectangular strategy with 19x19 points setup but when I do a bed level test print the results aren't as good as they should be sad smiley

Is a 19x19 grid too much for a 32 bit board? It is not listed in the docs anywhere; I'd like to know the biggest grid possible.

Edit: I see the duet can handle a 21x21 grid max, so what would the azteeg (LPC1769) handle?

Result of 19x19:



Edited 6 time(s). Last edit at 09/13/2017 08:16AM by klcjr89.
Re: Precision Piezo Z-probe Now available.
September 13, 2017 08:05AM
RepRapFirmware on Duet WiFi/ Ethernet supports a grid of up to 21x21 points.



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].
Re: Precision Piezo Z-probe Now available.
September 13, 2017 08:16AM
Quote
dc42
RepRapFirmware on Duet WiFi/ Ethernet supports a grid of up to 21x21 points.

As you posted that I edited my post because I found the info on google, what I need to know is the max grid for the LPC1769 based boards.
Re: Precision Piezo Z-probe Now available.
September 13, 2017 11:38AM
Sounds like a question for Smoothieware support. There is a forum section here for Smoothieware, but the Google group at [groups.google.com] is more active.



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].
Re: Precision Piezo Z-probe Now available.
September 14, 2017 08:08AM
I just experimented with the nozzle as an electrically based probe and I got a standard deviation of 1.3 microns homing at 1mm/s on the second approach after the first touch at 25mm/s. Nozzle was wired as GND and bed was wired at signal level.



Edited 4 time(s). Last edit at 09/14/2017 08:15AM by klcjr89.
Re: Precision Piezo Z-probe Now available.
September 14, 2017 08:50AM
That's a great measure of your machine's ability. However, frustratingly, getting really close to this level with a piezo or strain gauge is very difficult, as by their nature they are much more compliant than nozzle to bed via electrical contact. But it's a great control test. Subtract that from your piezo results and in theory you're measuring just the piezo not the whole system.

If a suitably adhesive build plate could be made conductive for a reasonable cost, then it would offer incredible probing accuracy and yet great simplicity, with the proviso you had a fairly clean nozzle and a little heat to flatten any residual plastic.

That being said as fun as this is its moving from engineering to physics as we already have enough precision to make the printer perform to its best.


Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions
Re: Precision Piezo Z-probe Now available.
September 21, 2017 11:09AM
I'm looking for feedback and ideas for version 2 of the piezo circuit, initially this will be implemented in the next batch of kit PCBs but may then be added to the PP20 as well.

So far I'm looking at the following;

Mounting holes, 2x M3 holes on a removable tab, if you don't need/want them, cut them off.

Power/tuning LED, this will allow you to be sure that the sensor circuit has power before probing

Longer output pulse, each trigger event will produce a 0.5-1s output making it much easier to check whether the probe is working and the firmware is reading it properly.

Any thought's or suggestions would be gratefully received.

Idris


{Precision Piezo} Accurate, repeatable, versatile z-probe plus piezo discs, endstop cables, pt100, 50w heaters.
Re: Precision Piezo Z-probe Now available.
September 21, 2017 02:40PM
Q1 assuming this is for mounting electronics I'm using the T nut and m3 screw
Q2 Yes, very useful as the blue led is normally off
Q3 Yes,yes,yes!! Definitely yes. 500ms adjustable for now

A free bee -a bandpass filter using that spare op amp section and spare ground pin for the shielded sensor piezo cable. Since we are basically using audio freq it may be useful to limit the band width.

Steve
Sorry, only registered users may post in this forum.

Click here to login