Z probe questions

Hoping to find the state of play in the use of inexpensive piezo discs as Z sensors ( [forums.reprap.org] ) I am looking for any problems and areas in need of improvement. To that end, can I ask two simple questions.

1. Have you tried piezo disc Z probe sensing and found it wanting?
2. Do you refuse to consider piezo disc based Z probes?

Reasons why would be cool, but even a simple (but non troll) "it sucks" would at least give me a feeling for how it is going.

Mike

Two things: when probe fails it may destroy your effector and/or nozzle, second is the cost.

Quote
newbob
Two things: when probe fails it may destroy your effector and/or nozzle, second is the cost.

Thanks for the input. I am not sure if piezo discs, assuming that they are correctly set up, are less reliable than anything else - they will even trigger if the nozzle is already resting on the build surface. Having said that, I will try to force a failure and see what I get.

Cost???? Piezo disc can be as little as £0.22 and conditioning circiur maybe £2.00 of components

Mike

Edited 1 time(s). Last edit at 06/09/2017 01:55PM by leadinglights.

Really. I see the 2 complaints as invalid. Maybe I'm a developer, I'd be wise to give my opinion. But I started to unfold with piezo disks precisely because I was dissatisfied with all other types of existing sensors. And it was the best thing I did. The sensor I used on my printer was working every day for 8 months and never failed. It does not even occur to me to use another type of sensor.

The only problem I see in this sensors is the job of adpting them to the various printer designs, or adapting the design to them. Other than that, I do not see any other inconvenience.

Hello again ECJ, I agree with everything you say, but I am trying to get at any complaints in the hopes that if there is anything that developers have overlooked then somebody may be able to point to it. I spent many years in product development and came to see that the developers themselves were usually too close to their own products to see the flaws and pitfalls.

Mike.

I was talking about the cost of only kit that is available ( I'm not criticizing team efforts and I think this is a good way to start but not a way for wide adoption ) but for wider adoption such probe has to be cheaper to end user than Chinese BLTouch clone.

Of course components are cheap but time and effort to put it all together is not trivial. It does not help that currently the only available circuit board design is using components that are hard to solder by hand, and that are not as common as SMDs in 0805, for example.

Once sensor is set up, I think it'll work well without a problem for a long time. However when there are calibration or mounting issues nozzle gets it first. Also, because current circuit is normally open controller cannot detect when probe's circuit is not working (ie not powered on).

Quote
newbob
I was talking about the cost of only kit that is available...

....Of course components are cheap but time and effort to put it all together is not trivial.

That's the reason for the cost, the time and effort the put it together is not trivial

Quote
newbob
It does not help that currently the only available circuit board design is using components that are hard to solder by hand, and that are not as common as SMDs in 0805, for example.

The schematic is open source, if you want a through hole version, make one. As it is you're paying not only for the time to assemble each board but also the time taken to design and develop the PCB (to say nothing of the development and testing of the concept which was largely not my work) . Unfortunately I'm not in a position to donate my time for free, expecting developers to do so is a sure fire way of stifling open source development. 0603 components are pretty simple to solder by hand and the first batch of boards were made with a syringe of solder paste and a heat gun.

Quote
newbob
Once sensor is set up, I think it'll work well without a problem for a long time. However when there are calibration or mounting issues nozzle gets it first. Also, because current circuit is normally open controller cannot detect when probe's circuit is not working (ie not powered on).

The current circuit will work either way as has been discussed in the original development thread.

When you strip away the cost of components, the pcb, the connectors and postage, then take into account the development time and assembly time as well as support I think the price is quite reasonable.

Idris

---

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

I misunderstood the question. Sorry.

Edited 1 time(s). Last edit at 06/09/2017 10:56PM by newbob.

I apologise if my last post came across too harshly, it wasn't intended to. The points that you made appeared to be based on inaccurate or out of date information.

I make the PCBs for the PP20 kit that is available, they're made by hand in batches of 20 or so, assembly and testing is relatively slow and so expensive. Commercial PCB Assembly only becomes cost effective with volumes in the hundreds and at this early stage in production and adoption I cannot risk the capital involved in that. If demand grows and PCBA becomes viable then the cost-per-board should drop significantly. From what I can tell a genuine BLTouch costs $48 at the moment, significantly more than a PP20 kit. I have a particular frustration with certain aspects of how open hardware is treated, if it is not possible to make money from open hardware development then people will eventually stop doing it.

If you do not wish to buy a kit or a module then the PCBs are available on their own. An earlier version of the PCB can be bought from OSHpark or the CAD files can be taken to any PCB manufacturer. If you do not want to solder 0603 components then the schematic is also available and you can design your own PCB for larger SMD components or through hole. If you don't want to make a PCB then you can use stripboard as I did for prototypes, or a breadboard which I used for months while testing.

I apologise for derailing the thread.

Idris

---

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

I'd like to make the whole thing cheaper. The kit for £20 is about the lowest we can do it now but larger volumes will bring cost reductions. I sold the first 20 modules for £20 at a loss, just to kick things off and get units out there.

I agree with ecj some designs are hard to adapt to. Titan aero for example will need the sensor in the bracket there isn't any other way. I've played with a few designs, nothing I've been quite satisfied with so far. I3 mk2 will also take a bit of adaptation with its integrated and enclosed hotend.

---

Simon Khoury

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

Maybe we need to add a couple more questions. The original two plus two more. As before, the questions are about using inexpensive piezo discs for nozzle contact detection and/or for automatic bed leveling. This is not about FSR sensors, accelerometers or other nozzle contact sensing methods. What I am trying to discover is RepRapers and makers attitudes to these sensors so that we can improve them.

1. Have you tried using piezo discs for nozzle contact sensing and it has not worked?
2. Have you tried using piezo discs for nozzle contact sensing and it has worked but been disappointing?
3. Have you considered piezo disc based Z probes or underbed sensing but decided on something else instead?
4. Do you refuse to consider piezo disc based nozzle contact sensors?

Reasons would be nice but not necessary.

Mike

Hmm! no responses? Am I to assume that piezo disc sensors bestride the field like colossuses (calossi?) shattered bodies of all comers strewn about - capacitative sensors that had lost the capacity to fight, inductive sensors induced to give up, BLTouch sensors that have lost the touch and even IR sensors red with embarrassment, sensitive FSRs that could not resist.

Mike

Edited 2 time(s). Last edit at 06/14/2017 05:08AM by leadinglights.

Was not here in this forum quite a while.

Just now looking into building my next printer.
Until now I never used any sensor for longer than just testing, because they all failed the sooner or later.
For my next build I do consider using under-bed piezo discs (already have a pack of 10 at home )
Will be some time till I get to testing them - have to start frame-construction in the next weeks.

In my Opinion piezeo disks are the most advanced and simplest way of calibration-sensors and I hope they are as reliable as I think of them after reading all the 2 threads on this forum.

Thanks for this great Idea!

Hi Dancer,

Thanks for your positive input, it is always good to feel appreciated. Having said that though, the purpose of this thread is to find any problems that there may be with the use of piezos in this manner.

As Dancer has raised this thread from it's sleep, has anybody - even including nemesises (nemisii??) from past acrimonious discussions anything negative to say about piezo bed leveling?

Mike

The only reliability issues I have ever had are the leads coming unsoldered. Which is less problematic with the nice Murata brand piezos with very thin, flexible leads, and managing the leads well.

Other than that its been smooth sailing, any unit I've setup (and then left alone rather than constantly tinkering) has proven to be reliable. I have put one Piezo20 module prototype through 36800 probes over several days, and it worked exactly the same afterwards (simulating maybe 3 years heavy use) as it did beforehand.

And if you are using one of our PCB's dont turn the potentiometers excessively, they are rated for 20 turns (although they do manage more than this in reality - wd40 helps).

Edited 1 time(s). Last edit at 07/25/2017 06:09AM by DjDemonD.

---

Simon Khoury

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

Hi DJ,

I was shortly thinking of using your pcb, but as the circuit diagram is open source, and I do have the knowledge of printed circuits (and on a first look) do have all the necessary electronic parts at home, I will do the soldering myself.
And as I plan going for under-bed-piezos there is no necessarity for SMD - I can - without any problems - use THT.

I think it will be late autumn or early winter till the new printer might do its first moves - so please do not be angry, if you do not hear anything from me in the next weeks.

Thanks again for the great idea to you, mike, moriquendi and all the others involved in this great but simple idea!

Not angry at all. Sounds great its an open source project, what we love is developing, selling them isn't really the name of the game but a lot of people want to use it. and are not able to make their own circuits etc... so we can make them for them.

Let us know how you get on.

---

Simon Khoury

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