Precision Piezo Z-probe Now available.

Quote
MKSA

Quote
DjDemonD
Good idea definitely worth considering. Depends if we go to a microprocessor or stick with analogue. It's worth thinking about as most issues are reverse polarity rather than anything more complicated.

The other advantage of a software solution is that you can better protect the design from the thieves from over there.

This is the difficult part. The only way we can be open source and yet keep the clone problem to a minimum is if we develop things at pace, release source files only after we have product on the shelves (not 12 months later like Ultimaker) and provide a community/support that encourages people to either build their own system with our full support or buy from us (or our distributors).

---

Simon Khoury

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

Bltouch tested (finally).

Massive thanks to Chowa for sending me one to try out. Installed on duetwifi equipped machine, using the instructions in the duet wiki, without any major issues (except you need to comment out any M307 commands in either config.g or config_override.g which relate to Heater 3 (H3) as these override the configuration of heater channel 3 as a servo signal).

I ran 120 probes in 4 batches of 30.

The first run achieved

range 15 microns
std dev 3 microns

the next three runs got the same results:

range 20 microns
std dev 4, 4, 5 microns so average over 90 probes 4.33 microns

So it is quite reproducible and fairly accurate. Reports of it achieving 50 microns accuracy are wrong, its more like 20 microns. Which is good enough for 3D printing.

However it still has an offset so whilst it's much smaller than an inductive/capacitative sensor, and you can mount it closer to your nozzle you still cannot probe anywhere you can print, unless your axes are much larger than your bed (i.e. a printer with relatively inefficient use of space). It is larger than an IR probe which can be mounted much closer to the nozzle, so reducing the XY offset, and the effect of effector tilt, if being used on a delta printer. But unlike an IR probe it is surface-neutral and can probe on any surface. It still needs to be deployed and stowed, which although isn't particularly difficult using duet, might be a small challenge to setup in some firmwares. You do not lose any z-height using it provided you can fit it under your carriage/effector, which should be easy enough with v6 hotends, but might be a challenge with something smaller like deltaprintr mini hotend.

If I compare it to piezo (and I am trying to be fair) I'd say piezo has no offsets so you really can probe anywhere you can print, is equally unfussy about surfaces, does lose you some z height in some common configurations, but can achieve marginally in most cases - higher accuracy and lower standard deviation when well tuned, and equipped in an accurate machine.

---

Simon Khoury

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

The BLTouch works quite well when it's been mounted solid and all the offsets have been configured. It isn't great for smaller printers though since the places it can't probe becomes quite significant. Also, not having to constantly clean an oozing hot nozzle during levelling, which I need to with the piezo, is a great bonus.

I use different print surfaces all the time so I naturally gravitate towards the physical touch-type probes.

Simon,
Happy you got it working. I've had limited success with it. The piezo is the best choice for me. There's more to a product no matter how good its manufactures think it is. It's called support for the product and you and your team won the gold

Here's a little video I did of the bad vs. good polarity.


Then UBL probing.

Quote
Moriquendi
Not if you make it open source you can't, that's kinda the point.

I'll do some investigation into the rectifier, however, there are developments coming that might make it unnecessary anyway.

Idris

*sh1t*, I just ordered the current universal version....
I had hoped the teething problems were already solved. But if the only new feature will be polarity protection, I can live without...

If you see my order in the mailbox, I can wait for the new version. Hope there is no big price difference?

Edited 1 time(s). Last edit at 02/26/2018 03:12AM by o_lampe.

Thanks for the videos Jim. And the kind words Stef.

---

Simon Khoury

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

Quote
o_lampe

Quote
Moriquendi
Not if you make it open source you can't, that's kinda the point.

I'll do some investigation into the rectifier, however, there are developments coming that might make it unnecessary anyway.

Idris

*sh1t*, I just ordered the current universal version....
I had hoped the teething problems were already solved. But if the only new feature will be polarity protection, I can live without...

There's always a minor tweak around the corner so if you need a probe buy one. The very first boards we used a year ago probe just as well as the latest one, they just weren't so easy to setup.

---

Simon Khoury

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

Quote
o_lampe
*sh1t*, I just ordered the current universal version....
I had hoped the teething problems were already solved. But if the only new feature will be polarity protection, I can live without...

The new features aren't really applicable to the universal board. We have two families of products and they're aimed at different types of customers;

The universal board maximises flexibility at the cost of ease of use. It is the most adaptable to different sizes and numbers of piezo disks but with more variables, tuning is more complex.

The PP20 board and module is designed to be easy to use. The first boards worked in the same way as the universal boards but the latest revision makes tuning much less complex, though with less adaptability. We can do this because we're designing for one specific application, one piezo size and one mechanical arrangement so we know much more about how to tune the circuit before we start.

The developments I'm talking about are another step down this road, it may filter across to the universal board but if it does it will take quite a while. As Simon has said, nothing really fundamental has changed, the new boards are no more accurate than the original prototypes eighteen months ago but they are hopefully easier and more convenient to use.

Also... your order is already in the post...

Idris

---

{Precision Piezo} Accurate, repeatable, versatile z-probe plus piezo discs, endstop cables, pt100, 50w heaters.

Quote

Also... your order is already in the post...

Good to hear I won't miss new features. That's always bugging me, buying new hardware and seeing updated stuff one week later...

BTW: The order confirmation and the shipment note got filtered by my spam filter...

Q: Has anyone used the piezo pcb to probe CNC parts? They often use multidirectional probes. Top/down and sideways. With the right probe head, the piezo would trigger on vertical and also on horizontal touch.

Edited 2 time(s). Last edit at 02/26/2018 08:36AM by o_lampe.

Quote
o_lampe
Q: Has anyone used the piezo pcb to probe CNC parts? They often use multidirectional probes. Top/down and sideways. With the right probe head, the piezo would trigger on vertical and also on horizontal touch.

You might not need anything more complicated than one disk with a long bolt through the centre, lateral force on the bolt would flex the disk, as would vertical though I'm not sure whether one side of the disk would cancel out the other.

Idris

---

{Precision Piezo} Accurate, repeatable, versatile z-probe plus piezo discs, endstop cables, pt100, 50w heaters.

I imagine designing a mount with one disc perpendicular to the other and attached to a probe of some sort would be possible. But I think Idris is right one disc is likely to be able to manage it witha very simple mount.

Be interested to see what you come up with.

---

Simon Khoury

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

The probe of a Coordinate-measuring machine works with 3 sensors and they are on the XY plane. If one sensor disconnects, the probe triggers.

With a single drilled piezo, It should work too.

but I would do is differently.

3 piezo at 60° on the XZ plane.

It should increase sensitivity for probing in the XY plane.

Edited 1 time(s). Last edit at 02/26/2018 10:28AM by tech-raton.

Don't most contact probes use a Renishaw probe? edit, you edited while I was replying, we're talking about the same thing

My concern with a single piezo would be that a lateral force on the bolt would bend the disk into an S shape. The ceramic on one side of the disk experiences strain in one direction while the ceramic on the other side is strained in the other direction. The charge generated by each side of the disk would then cancel out and the result would be significantly lower sensitivity to lateral force than to vertical force. How much of a problem this would be I do not know.

Idris

Edited 1 time(s). Last edit at 02/26/2018 10:38AM by Moriquendi.

---

{Precision Piezo} Accurate, repeatable, versatile z-probe plus piezo discs, endstop cables, pt100, 50w heaters.

I dunno...

I have used some old tools at school (10 years ago).

Some tools had even some CCCP markings...

I'm with o_lampe on the idea of buying new hardware, then seeing a new version before the one you ordered even arrives. It's happened to me several times, sometimes where the new feature or revision is something that I'd really like.

I got the universal kit, too. I had been hoping for polarity protection, but given the voltages involved, I may just make my own little bridge rectifier for the sensors. I have a box full of components, I'm sure that I can find a bunch of little diodes.

I am really looking forward to the parts getting here.

---

MBot3D Printer
MakerBot clone Kit from Amazon
Added heated bed.

Leadscrew self-built printer (in progress)
Duet Wifi, Precision Piezo parts

Its not so much polarity protection, nothing gets damaged if the piezo is plugged in the wrong way round, it just doesn't work as well. The clamping diodes protect the circuit from excessive voltage spikes. The advantage of the bridge rectifier would be that it wouldn't matter which way round you plug in the piezos. If you're aware of the importance of piezo polarity it's not a problem, the issue is that unless you're aware of it you can get frustrated trying to work out what's wrong.

Idris

---

{Precision Piezo} Accurate, repeatable, versatile z-probe plus piezo discs, endstop cables, pt100, 50w heaters.

Yes it a fairly minor change and one that mainly reduces support questions for us to answer. Once the polarity is checked and correct you'll never need a rectifier or to worry about it again.

As for old tools, my Dad has a big box of spanners and wrenches use originally for world war two aircraft maintenance.

Edited 1 time(s). Last edit at 02/27/2018 03:58AM by DjDemonD.

---

Simon Khoury

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

Hi. I have asked this on the duet forum but thought i migt want some feedback from the users. Ordered pcb with 20 and 27mm discs. My setup is rigid with no printed parts that hold the extruder. I have totan aeto. Couple of photos:







Which way would you suggest mounting it? I could mill the aluminium bracket I have to accomodate piezo but dont know how deep as i was told thicknesses varies.

@agniusm
The piezo needs a plunger to trigger. IMHO, you could mill the piezo groove deep enough for all piezos and vary the length of the plunger, right?

Re: CNC probe
I don't like the idea of drilling holes through sensors, so I'd make a probe with a "dogbone" shaped head. The tilting point would be closely below the bone head.
When it got touched radially, it tilts and one side of the bone head would trigger the piezo, while the other doesn't deform it to an S-shape.
Might make a sketch later to illustrate...

@agniusm - it is a tricky concept as we may presume that 100% solid is best, but there has to be some level of compliance in the assembly for piezo to work, we are only talking about 0.1mm and this is not an amount which will cause problems when printing, as it is purely in the vertical/Z direction, not in the lateral XY direction.

You can mill the bracket in some way, but please print one and try it first. The piezo will need to be insulated in some way as its metal surfaces are its electrical conductors.

If the piezo sits in a groove near the uppermost end of the bracket where the force will be concentrated when the nozzle taps the bed, and it is squeezed or bent by pressing into the centre from one side whilst being supported on its periphery on the other side then this will work. Make the recess match the piezo you have. It might also be worth in your setup to simply bond the piezo onto the side of the titan extruder, and see if you can get it to work as a microphone. The tuning window will be much narrower but not impossible to tune.

Edited 1 time(s). Last edit at 02/27/2018 04:28AM by DjDemonD.

---

Simon Khoury

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

I will insulate it with kapton. I cant print the bracket as my bed would not reach then and i will loose some. Probably not the major but still..
I wander if i could make couple of washers out of something to allow for some movement between my bracket and rail block

I'd consider adapting my motor plate design and putting it between bracket and carriage, testing it, and if it works for you then you might design something firmer.

[www.thingiverse.com]

---

Simon Khoury

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

I was trying to calculate my new z offset, but I must be doing something wrong. When I try the G30 S-1 command, I get 0.04. Could it really be that small?

Yes, mine is -0.03

Your hotend is probing, so the zoffset is minimum.

The 0.04 is your flexibility and the displacement to trigger.
It can be very small.

Quote
SpectreGadget
I was trying to calculate my new z offset, but I must be doing something wrong. When I try the G30 S-1 command, I get 0.04. Could it really be that small?

Yes it can be that small but unless you are planning on a 0.05mm first layer you might as well set it to 0.

---

Simon Khoury

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

Just wanted to report on, for those who use Duet/RRF that the latest 1.21RC firmwares now have an option to do multi-touch probing. You need one of the 1.21 RC (current is RC3) installed.

Then in your probe definition, you add

M558 Ax Sy ...........

Where x is max number of probe touches, and y is tolerance required to move on to the next point in mm.

So I now on one machine as an example, I use:

M558 P5 I1 F300 X0 Y0 Z0 R1.0 H15 T3000 A5 S0.03

So I am asking for up to 5 touches, and it moves on when they are within 0.03mm of one another. With an accurate printer (the one in the example above is not very accurate - its made from a fishtank) and a precision piezo probe using a tolerance value of 0.02, perhaps even 0.01 should work well.

What this really means is no more accuracy since the piezo probe system is very accurate inherently, but it does mean that one aberrant probe point, caused by nearby vibration or a false trigger, does not result in the resultant calibration being junk, potentially (if automated) leading to head crashes or a junk print.

Big thanks to DC42 and Duet3D for introducing this excellent feature.

Edited 1 time(s). Last edit at 02/28/2018 03:08PM by DjDemonD.

---

Simon Khoury

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

Quote
tech-raton
Yes, mine is -0.03

Your hotend is probing, so the zoffset is minimum.

The 0.04 is your flexibility and the displacement to trigger.
It can be very small.

Quote
DjDemonD
Yes it can be that small but unless you are planning on a 0.05mm first layer you might as well set it to 0.

Thank you, I appreciate the validation.

Hey Idris...

First test completed, and I haven't tuned it yet...

Recv: Finished!
Recv: Mean: -0.010381 Min: -0.016 Max: -0.006 Range: 0.010
Recv: Standard Deviation: 0.004904

Second test... tensioned the hotend...???

Recv: Finished!
Recv: Mean: -0.036376 Min: -0.036 Max: -0.036 Range: 0.000
Recv: Standard Deviation: 0.000000

Edited 1 time(s). Last edit at 03/04/2018 01:51PM by Nitram.

Hi guys,

I finally received my piezo kit and now I need a little help. I have Tevo Tornado currently running Marlin 1.1.6 with BL Touch and original E3D V6 installed, configured and working properly. I'm relatively new to 3d printing, but installing the BL Touch watching the instructions in a YouTube video was easy. I don't want to make some fatal mistake, please help with wiring and firmware changes to replace the BL Touch with Piezo. I have the kit with V2 board (with the two mounting holes that can be snapped off). I designed my own mount to accommodate the piezo, V6 and the nimble sidewinder that I'm expecting in the next 2-3 weeks, so mounting and mechanical installation (and I believe the tuning) is not an issue. I did read the manual and watched all the YouTube movies, and I did almost read this hole topic, but the information is kind of scattered and there is no easy to follow instructions especially for the firmware changes. I will appreciate any help I can get.

Thanks in advance,

Valentin