Precision Piezo Z-probe Now available.

How do I hook um two boards on a single pin ? Is this even possible?

What kind of boards?

Idris

---

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

Sorry, two Precision Piezo boards to a single pin on a Ramps 1.4 (or other boards I guess). In other words: Will shorting two green wires in to a single one work? Or will everything explode?

With genuine Precision Piezo boards from v2.75 and up this is not a problem, just connect the two signal wires together, either board will cause a trigger.

Idris

---

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

Hi folks,

I need some assistance connecting the V2.75 PCB to my RAMPS 1.4. I have the wires coming off the PCB as Red=+, Black=Ground, and Green=Signal. When I connect the other end of the wires to the Z-min pins on RAMPS (Signal near the edge, Ground in the middle, and + near the center of the board), the LCD screen just lights up, the firmware doesn't seem to run, and the PCB gets *very* hot quickly. If I turn the RAMPS end connector around, the printer runs normally, but the PCB doesn't light up.

What am I doing wrong?

edit: nevermind, I figured it out. I wasn't looking closely enough at the pin labels on the PCB. I thought the +, Ground, and Signal lines went across the edge of the board, like the piezo pins do. I didn't notice until now that they instead run vertically. Once I reoriented the plug properly, and rearranged the wires on the other connector, it works properly now.

Edited 1 time(s). Last edit at 05/28/2020 01:46PM by AbuMaia.

---

MakerFarm 8" Prusa i3v
RAMPS 1.4
0.4mm E3D v6 for 1.75mm

Hi Moriquendi

Quote
Moriquendi

You could put the piezo in series but there are some reasons why parallel is probably better; The wiring is simpler, you don't have to run a wire from one piezo to the other, just crimp the two piezos wires together at the connector. I'm going out on a limb here but, a piezo disk is a capacitor, if my understanding is correct the piezoelectric effect is a change in the capacitance of the ceramic due to applied strain (force). Adding capacitors in parallel adds their capacitance, adding them in series reduces the capacitance. If the change in capacitance and therefore the charge flow that we read is proportional to the total capacitance then by putting the disks in series we reduce the sensitivity, while by putting them in parallel we increase it. I'm not 100% confident of my reasoning here but I think I'm right, at least in part.

Idris

Interesting assumption. Have you made any tests that confirm your hypothesis? I would be greatly interesting in the results.

I am currently trying to figure out the optimal setup for my printer. It will use 4 underbed piezos on a triple z-axis (goal is self-leveling) 330x330mm bed. Design constraints have forced me to use a sensor in each corner instead of one piezo next to each of the z-axis. As it has been confirmed numerous times that underbed piezos are too sensitive on heavy beds (and prints!) create false triggers, I currently plan to take my chances with a parallel setup or even mixed setup (2 x 2 in parallel).

Anyhow, if your findings suggest that parallel is inferior, I might change my mind.

Will report back with my results.

Edited 1 time(s). Last edit at 05/31/2020 03:14AM by sgali.

Although I have not tried piezo elements in series, I would think that the voltage is additive at any point in time; even if some of them are giving a negative value. As an example, say after 5 milliseconds the outputs are Piezo1 = 2V, Piezo 2 = 2V, Piezo3 = 1V and Piezo4 = -2V then the sum would be 2V + 2V + 1V - 2V = 3V. In this case, if all else is equal, and assuming the amplifier is a voltage amplifier, the 4 in series should be more sensitive.

In the case of sensors in parallel, if the voltage output from each is the same then the voltage from all the sensors is the same as just from one. If one piezo gives a different voltage then the total voltage is the average of all of the piezos - this is caused by the charge equalising between each of the piezos.

There is a difficulty in this garden of Eden however: the voltage at each sensor at any point in time depends on the pressure on it at that time. With some plate geometries, the time at which the pressure reaches each sensor can be different enough to give thoroughly unexpected results. I have seen a setup where probing of half the bed gives a trigger at 3 grams of pressure within 3 milliseconds of initial contact, the other half of the bed only triggering with 10 grams and one corner not triggering until over 80 grams of contact force.

The good news though is that having 4 piezos on the extreme corners of the bed should avoid this problem. The advisability of supporting a bed on 4 corners though is frequently the subject of holy wars on this forum.

Mike.

Thanks Mike for your answer. There is hardly ever an easy answer and to make it worse, my understanding of physics is horrible.

It sounds like there is no way to tell what approach is the most reliable, both in terms of false positivies and negatives. Also, the "fiddliness" of the fine tuning process needs to be taken into account. Unfortunately, as a physics fool, I don't own an oscilloscope and even if, I would not understand it. Hence, I gotta find out the old school way by just playing around once the wiring is done.

Maybe my question boils down to that one: what does your gut-feeling tell you: what setup, series or parallel, is the most promising to start with? (I plan to use un-drilled 20mm piezos).

So I shall enter the holy war to free the path to the Garden of Eden one way or another.

I am having a problem where, the nozzle is not detecting the impact on the glass bed, until it has begun to exert more force than a simple impact. the result is my M851 setting has to be somewhere between 0.18 and has even had to be adjusted as high as 0.76.

I have tried a wide sweep of settings on the VR1 and VR2, the detection seems to only be so sensitive the movement of the gantry down towards the bed (this is on a CR10S) causes a detection, or, the nozzle impact must exert extra force to cause the detection.
I have attempted to "Hack" the effect by speeding up the nozzle Z speed, this way the impulse of energy from the nozzle experiencing the surface of the bed is sharper and causes a larger spike, but also seems to be somewhat less than desirable for an accurate reading. I currently have the first detection Z_PROBE_SPEED_FAST set to 1800 and the Z_PROBE_SPEED_SLOW set to 700.
If memory serves the documentation says to not exceed around 500 as a detection speed. If I set the speed below 700 the detection never fires and the z stepper keeps descending pressing the bed down even more (I'm lucky I'm still using springs on the bed otherwise it would have broken something).

Is there anyway I can modify the detection? It seems I cant make it more sensitive, because the detection is firing when the gantry begins to move, but it needs to have a steeper catch against the touch of the nozzle. So more or less, I need to change the voltage pulse of the impact, while maintaining the detection curve of the expansion card.

Am I making any sense?

I am using the larger 30mm disks, and am using this mount [www.thingiverse.com] I've printed it twice now, once in ABS, and again in PLA, both prints have the same effect (or lack of).

Any help is most appreciated!

Cheers!

You need to reduce the acceleration and jerk for z-probe movements. You have a lot of mass hanging off the piezo sensor so the vibration caused by starting the z-movement is causing a trigger.

You will also need to drastically reduce z-probe speed to get accurate results, I probe at 2mm/s.

Idris

---

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

OK, Hi guys, a newbie to a lot of this stuff, and a complex history behind this question.

For lots of reasons best left out of this post to save boring the hell out of the readership, a number of years ago, I got a lot of hardware to allow us to build a Delta Pi, and the project stalled and never got started.

Now, with Covid changes, and a lot more time on my hands, I'm wanting to get back into moving into 3 D printing. From reading over the time since we started, there have been a lot of changes since I got the Pi hardware, and I want to save myself some grief with getting it working so I can print usable output, but as far as I know, the basic Delta PI hardware works well when correctly set up and configured, so I don't want to waste what we got back in the day.

The original plan was a Arduino Mega, with a Ramps 1.4 and display unit, and the Hot end is an E3D-V5. The rest of it will be standard Delta Pi, and as far as I know, I have the relevant bits, like the heated bed, the glass sheet, and a Server Power supply that could double as a welder supply, if pushed, it's capable of 120 Amps, so I should be able to assemble a reasonable piece of hardware.

So, my reason for posting here in this thread,
Can I use the Piezo Z probe with the E3d V5 Hot end that I have, or am I better to get a more recent hot end to save some pain?
As the Delta Pi was originally set up for Mega / Ramps 1.4, should I continue with that hardware, or look to upgrade to a more capable 32 bit board, and if so, any recommendations given I want to use the Piezo Z probe as an upgrade?

Thanks

---

Shore, if twas easy, we'd all be doin it

Irish Steve

Quote
sgali
...
Maybe my question boils down to that one: what does your gut-feeling tell you: what setup, series or parallel, is the most promising to start with? (I plan to use un-drilled 20mm piezos).

So I shall enter the holy war to free the path to the Garden of Eden one way or another.

Neither series, nor parallel, simply put none, or one in the carriage/hotend.

Do you know any working system using 4 under bed piezo ?
DO you know any working system using 3 leadscrews plus motors to level the bed ?

Many "projects" well acclaimed but so far, no feedback, gone.

Fact is, this kind of "solution" although could be made, brings here more problems than it solves in addition to being more complex than solving the initial problem in the first place.

Edited 1 time(s). Last edit at 06/02/2020 04:04AM by MKSA.

---

"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)

@Irish_Steve

If you're looking to upgrade your controller board then I would suggest you get a Duet2 (either wifi or ethernet). It has lots of advantages but one of the biggest for me and the reason I will never go back to a RAMPS is that configuration is so much easier. Rather than having to upload a new firmware to update settings you just edit the config files. The duet also allows for Precision Piezo probes to work in analoge mode making tuning much easier.

If the v5 uses a standard groovemount then it will be compatible with the Precision Piezo Orion, alternatively under bed sensing is ideal for a delta printer.

@MKSA

Yes I have customers who have used 4 under bed piezos with great success.

Yes I know of many printers using 3 leadscrews to level the bed, the Railcore 2 does this.

Thousands of people are using my products with great satisfaction.

This thread is for announcements and support of Precision Piezo products. If you have questions or comments regarding PrecisionPiezo products specifically please post them, if you wish to comment on piezo sensing in general, this is not the thread to do it in.

Idris

---

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

I have to agree with MKSA to the extent that many interesting projects never result in feedback to the forum. In trying out new ideas the failures are as important as the successes but only if the details are also published.

As far as my gut feeling goes, I would go with parallel, but if the leads on each piezo are extended then both can be tried.

On the 4 underbed piezos, mounting a bed on 4 corners has a major problem in that it is what is called "over-constrained" where the act of mounting or adjusting can pull the bed out of flat. There is however a minor advantage in that bed can be lighter if the mountings have the right amount of compliance - although getting the compliance right is not an easy trick. If you should decide to try this 4 underbed piezos I would like to hear how it works out.

On the 3 leadscrews with motors, once again I am in MKSA's camp. Two motors for the Z axis on many Cartesian printers does work after a fashion - but is is still a kludge, 3 motors is too much of a kludge.

My attitude differs from that of many on this forum in that I think that trying out ideas to see if they solve problems is always worthwhile. To re-phrase something I have seen somewhere:

Quote
A comical prototype that works proves that an idea is possible and not so dumb - although it may still not be practical

Mike

@Moriquendi

He asked for an opinion, I gave it. I didn't advise NOT to use your product, but one, in the hotend/carriage. It does the job and is difficult enough.

The example you gave proved my point, too complex, problematic compared to building a proper machine, use of a simpler approach.

leadinglights made quite good prototypes, analyzed the issues extensively and published the results. Self explanatory.

Edited 1 time(s). Last edit at 06/02/2020 10:56AM by MKSA.

---

"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)

As an owner of one precision piezo early version, one Orion first version and one Orion last version my advice will be to use the last Orion version if possible. Just make a good mount suitable for your hotend and you will forget about leveling problems. And I strongly advise about using the Orion in analog mode, it's unbeatable. I didn't try underbed setup, but from what I read it's more complicated and hard to predict, setup and control.

Hi!

I am a (very happy) user of a Precision Piezo Orion kit (Module Mounting Option: Groovemount to your printer) since May 2018 (Order No. 10716 ).

I am now planning to redesign my carriage, so I'd love to understand if

• has anyone managed to find a reliable carriage mount solution?
• can I convert my kit to that, in case? I suppose the anser is "no" cause I have the PCB and not a bare murata sensor
• are there some design specs of the PCB so I can attempt a design without dismantle my current setup?

For the records, I have an Hypercube corexy, with 8mm rods on the Y, E3D V6 original direct with trianglelabs "titan aero", Marlin on an ultratronics and I print on glass.
Many thanks in advance, comments & insults are welcome!

---

Disclaimer: io ce la metto tutta a darti consigli sensati, sta a te non incendiare nulla :-)
CoreXY HyperCube. [www.thingiverse.com] | Cerchi test? Make: test set: [www.thingiverse.com]

You guys wouldn't happen to have the PCB and disc Step or Solidworks files for the universal Z probe kit?
Thanks!

So you really not going to make any more of the 20 boards?
A little design consideration from my project:



The 20 board would have made a perfect fit on top of a 40 mm fan.
The new board is obviously to long, but it would fit with the mounting hole cut and the connector facing inwards. Then again my mounting possibilities are rather difficult.
Furthermore i would only require one piezo connection...

2 ¢

Okay so I'm thinking of biting the bullet. My Delta printer is now just doing nothing because indeed, as expected the Force Sense Resistor solution is not good enough.

I have two questions:
- Considering these 'corona times', are there v2.85 modules shipping at the moment?
- What is the best method of mounting? (see image)

---

http://www.marinusdebeer.nl/

For most purposes, you should use the piezo disks in radial mode even though it introduces some unwanted compliance into the system. The additional compliance is usually so little that it can be safely ignored. Compression mode does give a very rigid structure but needs greater care to get it working well. Bending in one axis can give problems with some piezos that where there can be a substantial difference in output depending on which direction it is being bent in - you may have to rotate it until you find an acceptable sensitivity.

A note on compression mode: When a piezo disk is compressed the output polarity is in a consistent direction but if it is very lightly loaded and there are irregularities in the gap between the parts sandwiching the piezo there can be some very strange outputs. In the drawing below a piezo is shown between a pressure block and another surface: Either or both of these may, and probably will, have surface irregularities in the micron region. The initial response of the piezo will be in a sense determined by which direction it is being bent in before the piezo is seated and the pressure response predominates - this may result in a signal changing its polarity. Having a preload of greater than 600 grams seems to eliminate this problem.



After trying many different sensors for Z probing I am fairly firmly convinced that the common piezo disk gives the best performance and the best bang for the buck. Having said that, there are several problems that they suffer from and it needs a bit of understanding to avoid these problems. Until a commercial company brings out a purpose-made piezo sensor as robust as these disks at a reasonable cost you will have to put up with study and tinkering to get it right.

Mike

Thanks for your explanation Mike, so it seems that I will have to design a nice bracket for the heated bed to support a radially mounted piezo disc. Even though the bed then might be a bit loose on it's feet this will give me the best between wear and sensitivity.

I'll wait for the seller to respond so I know if the piezo discs are currently being shipped. Had some bad luck with some orders the past few months so I hope to be able to just go ahead and order the parts. Right now things are taking a huge leap forward with the machines I have. Soon I hope to have finished every machine I currently have sitting in maintenance or unfinished build status. The Delta only really needs a better system for bed leveling so this seems like a no-brainer to go this route with the Delta. Plus I can reuse the inductive sensor I had on it for my custom Prusa or the Replicator 2x I'm converting to an open source HW machine.

Edited 1 time(s). Last edit at 07/03/2020 03:47PM by Ohmarinus.

---

http://www.marinusdebeer.nl/

The compliance is not enough to be called loose. There is an old posting at [reprap.org] where I measured the deflection with a 500 gram load and got less than 40 microns. That is for a single cell and includes a little compliance for the mounting. As you will be sharing your load between 3 piezos with a simpler mounting I think the figure at the piezos will be closer to 20 microns per kilogram on the bed.

Mike

Quote
leadinglights
The compliance is not enough to be called loose. There is an old posting at [reprap.org] where I measured the deflection with a 500 gram load and got less than 40 microns. That is for a single cell and includes a little compliance for the mounting. As you will be sharing your load between 3 piezos with a simpler mounting I think the figure at the piezos will be closer to 20 microns per kilogram on the bed.

Mike

Yes I was thinking to use a spring-loaded system where the bed is held down by six springs. Each piezo element being surrounded by two springs on each side. So with loose I mean that the bed just isn't fixed to the frame by bolts but by indexing shapes and by springs. I can try to draw up a design today and I'll email the v2.85 producer to ask if they currently deliver.

---

http://www.marinusdebeer.nl/

I'm really not making any more of the Piezo20 boards, not unless you want to special order a large number. The Piezo20 PCB was designed for a single product using a technique that I can no-longer support, drilling holes in piezo disks. The Universal PCB does everything you need to run a system like the Piezo20 but is also easy to use for underbed sensing, unlike the Piezo20 PCB. The mounting hole is designed to be easily removable, or, you can mount board a little further away.

The functionality of the Piezo20, a compact, hotend mounted probe, has been replaced by the Orion.

Yes, everything that's sold in my shop is in stock and ships within 2-3 days. Shipping can be affected a little by the pandemic but not badly, depending destination only a few extra days.

I currently have no plans to release a new version of the Universal PCB

Idris

---

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

Quote
Moriquendi
I'm really not making any more of the Piezo20 boards, not unless you want to special order a large number. The Piezo20 PCB was designed for a single product using a technique that I can no-longer support, drilling holes in piezo disks. The Universal PCB does everything you need to run a system like the Piezo20 but is also easy to use for underbed sensing, unlike the Piezo20 PCB. The mounting hole is designed to be easily removable, or, you can mount board a little further away.

The functionality of the Piezo20, a compact, hotend mounted probe, has been replaced by the Orion.

Yes, everything that's sold in my shop is in stock and ships within 2-3 days. Shipping can be affected a little by the pandemic but not badly, depending destination only a few extra days.

I currently have no plans to release a new version of the Universal PCB

Idris

Hi thanks, meanwhile I ordered a v2.85 board with three 27mm piezo's. Really happy to start playing with it soon. Tomorrow I'll fire up Fusion360 and do a first sketch! Isolated from heat, mounted with springs for pre-tension and more. Basically already had designed mounting feet for the heated bed in such a way that I could add piezo functionality as an option later on. Ahhhh, good new stuff

---

http://www.marinusdebeer.nl/

Quote
Moriquendi
I'm really not making any more of the Piezo20 boards ...

If my models are correct the 20 board is only 2 mm wider, but 5 mm shorter, plus it provides you with two full grown mounting holes. Thats the board only. The unfortunate placement and orientation of the connector on the universal PCB adds another whoping 9 mm in length. Depth is about the same.
Aside from the support of 3 piezos (which wouldnt matter for a hotend kit), i dont see how the new form factor is an advantage.
At the hotend mounting space matters.

Now the Orion is a unnecessary large fellow as well that leaves no choice in mounting options.
I get your point, but I still think the 20 setup was ideal for the hotend. It is suffering a premature death in favor of multi (bed) piezo capability, which probably wont use as many.

Not talking about you drilling piezos, China happily jumps in there too.

Demand for the Piezo20 reduced drastically with the introduction of the Orion, I do not wish to maintain stock of a product that does not sell well, not to mention updating it every time a new iteration arrives. The Piezo20 did not die prematurely, it was replaced by a superior product.

The Universal PCB can support all the applications that the Piezo20 could, perhaps not as neatly but it can. The Universal PCB is not a new formfactor of the Piezo20, it is a different product designed with different criteria. The Universal sells well, the Piezo20 did not once the Orion was available. The Orion is as small as it can be, the overall size of an Orion module is barely any larger than a Piezo20 and has a number of advantages.

I'm sorry that the Universal PCB does not fit your requirements as well as the Piezo20 would but there simply isn't enough demand for me to keep manufacturing it.

Idris

---

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

Quote
Moriquendi
I'm really not making any more of the Piezo20 boards, not unless you want to special order a large number. The Piezo20 PCB was designed for a single product using a technique that I can no-longer support, drilling holes in piezo disks. The Universal PCB does everything you need to run a system like the Piezo20 but is also easy to use for underbed sensing, unlike the Piezo20 PCB. The mounting hole is designed to be easily removable, or, you can mount board a little further away.

The functionality of the Piezo20, a compact, hotend mounted probe, has been replaced by the Orion.

Yes, everything that's sold in my shop is in stock and ships within 2-3 days. Shipping can be affected a little by the pandemic but not badly, depending destination only a few extra days.

I currently have no plans to release a new version of the Universal PCB

Idris

Do you no longer support the drilling technique due to inconsistent quality, and DIY-ers asking for support when they might not have drilled it properly?

It's inconsistent how ever you do it and a has a fairly high failure rate. More importantly, the piezo ceramic is lead based, it's inert and until you disturb it by drilling it. Drilling it produces dust which is toxic and I don't want to have to deal with it. I have alternatives to drilled disks for almost all applications, eg the Andromeda pcb.

Idris

---

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