Welcome! Log In Create A New Profile

Advanced

Precision Piezo Z-probe Now available.

Posted by DjDemonD 
Re: Precision Piezo Z-probe Now available.
June 18, 2017 01:42PM
I connected the 4 pin probe connector on the Duet right to the Piezo board. 3.3v to Vin, Gnd to Gnd, AD12 to Signal line.

Edited 1 time(s). Last edit at 06/19/2017 08:07AM by ElmoC.
Re: Precision Piezo Z-probe Now available.
June 18, 2017 02:40PM
Quote
Hergonoway
Hi,

since it seems that few people use the piezo on duet, did you directly connect the 5v pin of the piezo to the 3,3 VCC pin on the duet ? or is it related to the INT 5V_EN jumper ?

As ElmoC says its not 5v output from the Piezo 20 module if you supply it with 3.3v.

So on Duet its connected to the z probe connector the first pin (nearest the wifi module/sd card slot) is signal (3.3v), Note the 2nd pin is unused on duet with Piezo probe. next pin (3rd) is ground, the last pin is the 3.3v output from the duet (which powers the piezo 20).

Hope this helps.

Edited 2 time(s). Last edit at 06/19/2017 05:27AM 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.
June 18, 2017 02:43PM
Just to let you know there was a limited release of 5 kits and 5 modules today, they've all gone as I had a list of people who'd emailed to ask when they would be back in stock, and I contacted that list.

However not to worry I have enough PCB's to make another 5 modules and 5 kits which will be put in stock later this week, I will announce that here.


Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions
Re: Precision Piezo Z-probe Now available.
June 18, 2017 05:11PM
Looking fwd to the next batches for my BW.

S
Re: Precision Piezo Z-probe Now available.
June 19, 2017 05:23AM
Quote
DjDemonD
Quote
Hergonoway
Hi,

since it seems that few people use the piezo on duet, did you directly connect the 5v pin of the piezo to the 3,3 VCC pin on the duet ? or is it related to the INT 5V_EN jumper ?

As ElmoC says its not 5v output from the Piezo 20 module if you supply it with 3.3v.

So on Duet its connected to the z probe connector the first pin (nearest the wifi module/sd card slot) is signal (3.3v), the next pin is ground, the last pin is the 3.3v output from the duet (which powers the piezo 20).

Hope this helps.

+ElmoC


Thanks ! I wanted to be sure smiling smiley
Re: Precision Piezo Z-probe Now available.
June 19, 2017 05:28AM
No problem I referenced the duet wifi diagram and made a correction above the 2nd pin from the front (wifi/sd card) is unused with Piezo.

Please can you correct your quote just so no one gets it wrong.

Edited 2 time(s). Last edit at 06/19/2017 05:30AM 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.
June 19, 2017 12:44PM
I installed piezo probe second time, and I was having a lot of trouble getting it to trigger consistently at all (first install had no problems, and it was ultra sensitive).

I took everything apart and found the solder joint on inner side of the disc had broke free. So the disc had no continuity unless there was pressure on the broken solder point. I resoldered it, reassmbled, and it's working again. The solder itself did not break, it was the material on the surface of the disc. Brushing epoxy on the solder joint might help.

The piezo holder might work better with 3 screws to hold down instead of 4. I found it was easy to get one corner semi stuck and way less responsive to trigger than the other three corners. Would probably help to use screws that have a smooth barrel near the head (no threads to interfere with sliding).

Edited 1 time(s). Last edit at 06/19/2017 12:45PM by jonathank.
Re: Precision Piezo Z-probe Now available.
June 19, 2017 01:14PM
The solder joints can be problematic but this is the first module which has suffered, with this design there is the least movement in the leads to the piezo than with most of the designs out there. I'm glad you fixed it.

I'll look into screws with smooth shanks but I've some prototypes where there was no movement at all (the upper and lower parts were both threaded and engaged by the bolts) and they still work, in that case using the piezo more like a microphone and less like a strain gauge. So I don't believe it will make much difference but when supplies of the current screw are low I can look into using that type instead.

You can make it much more sensitive using much less preload ie 3 point fixing but you'll have a wobblier nozzle. You can also use 27mm or 35mm Piezos but unless it's a large delta effector they don't package well into the printer. Plus it's nice to have a small z offset which represents the vertical compliance in the unit.

Feel free to experiment you've all the parts we supply with a kit, let us know what you come up with.

Edited 2 time(s). Last edit at 06/19/2017 01:16PM 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.
June 22, 2017 06:50AM
So I have released some new stock:

3 modules
5 kits

Are now available to buy on the webshop
[www.precisionpiezo.co.uk]


Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions
Re: Precision Piezo Z-probe Now available.
June 23, 2017 03:24PM
Quote
ElmoC
I connected the 4 pin probe connector on the Duet right to the Piezo board. 3.3v to Vin, Gnd to Gnd, AD12 to Signal line.

Hi I've just switched to a duet, do you have an updated version of your configuration files since the last time?
Re: Precision Piezo Z-probe Now available.
June 23, 2017 03:30PM
Quote
Hergonoway
Quote
ElmoC
I connected the 4 pin probe connector on the Duet right to the Piezo board. 3.3v to Vin, Gnd to Gnd, AD12 to Signal line.

Hi I've just switched to a duet, do you have an updated version of your configuration files since the last time?

The config file I posted was for use with the piezo probe. At least I think it was. I was having issues getting it to work consistently on my Cartesian printer, but it is working great on my delta. Here are the parts of the config file in regards to the probe.

M574 X2 Y2 Z2 S0			; set endstop configuration (all endstops at high end, active high)

M558 P1 X0 Y0 Z0 F120 T3000 H2 R0.2	; probe is used for homing any axes
G31 X0 Y0 Z0 P10			; Set the zprobe height and threshold (put your own values here)

Re: Precision Piezo Z-probe Now available.
June 24, 2017 04:27PM
Quote
ElmoC
Quote
Hergonoway
Quote
ElmoC
I connected the 4 pin probe connector on the Duet right to the Piezo board. 3.3v to Vin, Gnd to Gnd, AD12 to Signal line.

Hi I've just switched to a duet, do you have an updated version of your configuration files since the last time?

The config file I posted was for use with the piezo probe. At least I think it was. I was having issues getting it to work consistently on my Cartesian printer, but it is working great on my delta. Here are the parts of the config file in regards to the probe.

M574 X2 Y2 Z2 S0			; set endstop configuration (all endstops at high end, active high)

M558 P1 X0 Y0 Z0 F120 T3000 H2 R0.2	; probe is used for homing any axes
G31 X0 Y0 Z0 P10			; Set the zprobe height and threshold (put your own values here)


Thanks, what's the R 0.2 for on M558? I mean the purpose smiling smiley

Edited 1 time(s). Last edit at 06/24/2017 04:40PM by Hergonoway.
Re: Precision Piezo Z-probe Now available.
June 24, 2017 04:36PM
R is recovery time, I believe it waits for the specified time after a move to a probe point before beginning the probe. This cuts down on the effect of mechanical noise which seems more problematic on cartesian machines than deltas.


Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions
Re: Precision Piezo Z-probe Now available.
June 24, 2017 04:43PM
So it could technically limit the "wobble-to-falseTrigger" effect when it start to probe? Practical knowledge!
Re: Precision Piezo Z-probe Now available.
June 24, 2017 04:44PM
I believe it's useful definitely worth trying.


Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions
Re: Precision Piezo Z-probe Now available.
June 24, 2017 04:58PM
Quote
DjDemonD
R is recovery time, I believe it waits for the specified time after a move to a probe point before beginning the probe. This cuts down on the effect of mechanical noise which seems more problematic on cartesian machines than deltas.

That 0.2s definitely made a difference on my Delta.
Re: Precision Piezo Z-probe Now available.
June 24, 2017 05:00PM
The other thing that's very useful is lowering acceleration, motor current and jerk before probing, then restoring it afterwards.


Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions
Re: Precision Piezo Z-probe Now available.
June 24, 2017 05:06PM
Quote

The other thing that's very useful is lowering acceleration, motor current and jerk before probing, then restoring it afterwards.

If you are using RepRap firmware, like Duet, the jerk is not used during probing.

Also, the M906 command to set the motor current allows the use of the H parameter to set current during probing so you will only need to adjust the acceleration.
Re: Precision Piezo Z-probe Now available.
June 24, 2017 05:18PM
I was trying to decipher the duet wiki about this. How are you using it?


Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions
Re: Precision Piezo Z-probe Now available.
June 24, 2017 05:55PM
Quote
DjDemonD
I was trying to decipher the duet wiki about this. How are you using it?

You just use M906 twice. Once as normal with the regular currents. Second with the H1 parameter for the probing currents.

M906 X1000 Y1000 Z1200 E600		; Set motor currents (ma)
M906 X500 Y500 Z500 H1			; Set motor currents for probing

These are just made up numbers right now. I don't have the probe on my Cartesian right now so don't have the real numbers I was using. I haven't gone back and updated my Delta yet.
Re: Precision Piezo Z-probe Now available.
June 24, 2017 05:57PM
Superb this will cut down on config file complexity. Maybe we need to convince David to do a similar thing with acceleration.


Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions
Re: Precision Piezo Z-probe Now available.
June 24, 2017 05:59PM
Quote
DjDemonD
The other thing that's very useful is lowering acceleration, motor current and jerk before probing, then restoring it afterwards.

Once you have configured recovery time, reducing jerk should have no effect, because jerk is not used in Z probing moves. I wouldn't expect reducing motor current to be necessary either, unless you use low microstepping and the resulting vibration triggers the probe, whereas reducing motor current will make positioning less accurate. As for acceleration, the only one that should matter is Z acceleration, because a high acceleration at the start of the probing move could cause false triggering. Cartesian machines generally have low Z acceleration anyway. Delta machines can manage high Z acceleration, but it isn't necessary because Z acceleration doesn't affect printing speed the way X and Y acceleration do. So you may as well reduce Z acceleration generally if need be, not just during probing.

In summary, for a nozzle contact sensor I think it should be sufficient to configure enough probe recovery time for the sensor to recover after the travel move, and make sure that Z acceleration isn't too high especially on a delta or other machine with a belt-driven Z axis. That's also what we have found with the prototype Duet3D PCB effector.

If for any reason you do want to reduce motor current temporarily, use M913 in your bed.g file or in the macro that you use to do G29 bed probing. The M906 H parameter was only implemented in another user's fork of RRF.

Edited 1 time(s). Last edit at 06/24/2017 06:00PM 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].
Re: Precision Piezo Z-probe Now available.
June 24, 2017 05:59PM
I did talk to him about it, but he didn't seem too enthused about it. His last comment didn't make a lot of sense to me. Sounded almost like he was saying slower acceleration would be good, but then at then end says it wouldn't help.
Re: Precision Piezo Z-probe Now available.
June 24, 2017 06:00PM
[github.com]

Forgot to post this link.
Re: Precision Piezo Z-probe Now available.
June 24, 2017 06:11PM
Quote
dc42
Quote
DjDemonD
The other thing that's very useful is lowering acceleration, motor current and jerk before probing, then restoring it afterwards.

Once you have configured recovery time, reducing jerk should have no effect, because jerk is not used in Z probing moves. I wouldn't expect reducing motor current to be necessary either, unless you use low microstepping and the resulting vibration triggers the probe, whereas reducing motor current will make positioning less accurate. As for acceleration, the only one that should matter is Z acceleration, because a high acceleration at the start of the probing move could cause false triggering. Cartesian machines generally have low Z acceleration anyway. Delta machines can manage high Z acceleration, but it isn't necessary because Z acceleration doesn't affect printing speed the way X and Y acceleration do. So you may as well reduce Z acceleration generally if need be, not just during probing.

In summary, for a nozzle contact sensor I think it should be sufficient to configure enough probe recovery time for the sensor to recover after the travel move, and make sure that Z acceleration isn't too high especially on a delta or other machine with a belt-driven Z axis. That's also what we have found with the prototype Duet3D PCB effector.

If for any reason you do want to reduce motor current temporarily, use M913 in your bed.g file or in the macro that you use to do G29 bed probing. The M906 H parameter was only implemented in another user's fork of RRF.

Okay. Now your last comment on the github issue makes more sense.

Quote

2RepRapFirmware, DC42 version, cmm sub-version supports "H1"

This confused me as it sounds like it was in the main code. Wasn't sure 'cmm sub-version' meant.
Re: Precision Piezo Z-probe Now available.
June 24, 2017 06:12PM
I think I've mainly been reducing motor current during prototyping the sensors, to minimise the effect of head crashes. Perhaps I'll try just the R parameter and set low z acceleration by default in config.g, even on the delta.


Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions
Re: Precision Piezo Z-probe Now available.
June 25, 2017 04:32AM
The settings for RRF I am currently using on my delta based on these ideas are:

M558 P5 I0 R0.2 F300 X0 Y0 Z0 ;digital piezo sensor, output rise on contact, recovery time, probing speed, not used to home axes
G31 X0 Y0 Z-0.14 P10 ;sensor is nozzle,z offset and debounce value.

I've changed my Z acceleration overall in config.g to 200.

In bed.g now I'm NO LONGER using - any lowering motor currents, any higher microstepping, or any slow down (speed/accel/jerk).

And its calibrating to 0.008 as it was before just faster.

So to summarise the R parameter in M558 is important 0.2 seems to help but set longer if you have a lot of false triggers.
Set Z acceleration low such as 200.

Thanks DC42 for the input its very valuable.

If you aren't using RRF/Duet then setting Z acceleration low is still useful.

Edited 3 time(s). Last edit at 06/25/2017 05:17AM 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.
June 25, 2017 08:50AM
What are your M906, M201, M203, and M566 values you are using?
Re: Precision Piezo Z-probe Now available.
June 25, 2017 12:42PM
I've got two machines with Duetwifi and piezo sensors.

1st machine - Kossel XL delta, 27mm Lykle design piezo effector, v1.20 board (early kit board).
Motor current M906 X1100 Y1100 Z1100 E800:800 I60
Accelerations M201 X3000 Y3000 Z200 E200
Max feedrates M203 X15000 Y15000 Z15000 E3600
Jerk M566 X1200 Y1200 Z1200 E100

M558 P5 I0 R0.2 F300 X0 Y0 Z0 ;digital piezo sensor, output rises on contact, probing speed, not used to home axes
G31 X0 Y0 Z-0.14 P10 ;sensor is nozzle and debounce value.

2nd Machine Custom corexy 20mm piezo 20 module groove-mounted into a titan extruder, running off v1.21 kit board.
Also testing Titan Aero mounting bracket with built in 20mm piezo, working very nicely too. Watch this space for them on the site.
M906 X800 Y800 Z800 E600 ; Set motor currents (mA)extruder current just enough will click not strip
M201 X2000 Y2000 Z200 E1000 ; Accelerations (mm/s^2)
M203 X15000 Y15000 Z3000 E3600 ; Maximum speeds (mm/min)
M566 X600 Y600 Z30 E20 ; Maximum jerk speeds mm/minute

M558 P5 I0 R0.2 F400 X0 Y0 Z0 ;digital piezo sensor, output rises on contact, probing speed, not used to home axes
G31 X0 Y0 Z-0.1 P10 ;sensor is nozzle and debounce value.

Hope this helps.

Edited 3 time(s). Last edit at 06/26/2017 05:54AM 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.
June 26, 2017 06:54AM
So we've been testing a bracket for use with titan aero hotends/extruders (and titan extruders if you wish). Its based on the official bracket by e3d. The design was by Moriquendi.



TitanMountBracket4_mouse_ears.

Also on Thingiverse.

See it working on YouTube.

There is a recess in the base into which you bond (superglue/cyanoacrylate) a 20mm undrilled piezo disc, with the brass only side facing the bracket. The bracket must be used in the orientation as shown in the stl above, it won't work if used sideways/90 degrees rotated, or upside down.

The piezo is attached to one of our boards and then tuned in the usual way.

When the nozzle contacts the build surface the bracket flexes and this flex goes right through the piezo generating the signal.

If anyone is interested in helping us to test it and has a piezo kit already (we will try to get more made up and on the site ASAP), please print it in ABS preferably, alter the mounting holes to suit your setup, I've put "mouse ear" pads on the corners to help adhesion just cut them off after printing.

Edited 2 time(s). Last edit at 06/26/2017 06:57AM by DjDemonD.


Simon Khoury

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

Sorry, only registered users may post in this forum.

Click here to login