HEVO / HPCE / PrecisionPIezo universal piezo Z probe adapter

by Pumpal Feb 21, 2018
Download All Files

Thing Apps Enabled

Please Login to Comment

I'm struggeling with this. Tapping with finger or object triggers it. But if the plate touches it it doesn't it only triggers when the plate goes down again .... Where should i look for the problem

Hi - first of all, while setting up sensitivity of the Piezo PCB, use your print removing tool (spatula?) to apply knocking/force upwards, just as when the bed reaches the nozzle. Start from here - ensure a little move in upwards direction or a slight knock in up direction reliably triggers the probe.
From what you explain, I am only guessing, but here are my thoughts:

  • maybe the two screws on the sides of the piezo are too tight and so the piezo does not have any space left to bend when probe touches , try to unscrew them a bit - they should not be tighten completely!
  • Play with the PCB sensitivity - up it until the blue led stays on all the time, then lower just a little so it just goes off. Start from there
  • up the home speed - set it to 5.
  • remove the Z homing retest speed reduction. I am using Repetier, and so will show that setting here to serve as guideline point:

ENDSTOP_Z_RETEST_REDUCTION_FACTOR 1 - (that makes the second move up when homing, the same speed as the initial move)

Other related settings (Repetier)
HOMING_FEEDRATE_Z 5 (note that this one should also be checked in EEPROM, as it may already be saved and then it overrides the Configuration.h value! )

HOMING_ORDER HOME_ORDER_XYTZ (home X and Y, heat the probe, go to the defined X and Y coordinates (see bellow), and home Z)
ZHOME_X_POS 30 (only works with XYTZ home order, see above),
ZHOME_Y_POS 30 (only works with XYTZ home order, see above)
ZHOME_MIN_TEMPERATURE 170 (heat the head to that temperature before homing Z - any fillament ooze becomes soft, but not new oozing occurs usually)
ZHOME_HEAT_ALL 1 (related to above)

ZHOME_HEAT_HEIGHT 3 (important as default is set to 20 and that screws auto-level and distortion probing)
Z_PROBE_DELAY 200 - wait that long after x/y move before probing Z (important in autolevel and distortion probing G32/G33)

Hope you manage to deal it in.

Thanks, i reprinted the parts and that fixed it :-), its working now i'm tuning in my hypercube... Think the botom hinge was not ok and had to much play causing it to move upward and not into the piezo.

Hmm, that's interesting. I always have to file a bit the hinch so the parts can fit together.. maybe underextruding your old printer? Also you do not drill the horizontal holes, right? The screws should cut a thread inside in both parts.. Glad that it worked though ..

Love your design and trying to remix it to my mgn15 on my hevo,
can you please share any "base" file version, i have Fusion360 / Solidworks, so probably any export you'll do will be great.
Much Appreciated!

As I said in the description, I'm a lame designer :) and use Tinkercad.
Here is the project - it is public: https://www.tinkercad.com/things/j58bZaWPPw4 . Please let me know if you have problems opening it.

What I did is took the files from the original project that this thing is a remix from and the HEVO X-Carriage and used the Tinkercad tools to blend things together. If you can redraw it in a real CAD program, I am afraid it should be from scratch .. anyway, keep me posted on your progress please ..


Yep, l8er i saw the tinkercad link, but unfortunately the only export possibility is stl/obj
so will have to design my own, will post and link (remix) to your design.

I printed this and tried installing it on my HEVO but I am a bit confused...

I added a brass insert into the base and 4 m3 dowel pins into the actuator but when I came to assemble the mount on my printer, it seems to me that the single bolt used to fix the hot-end mount needs to be longer so it goes through the actuator and the base and then by tightening it will pull the hot-end mount onto the actuator so that there would be no movement?

I'm not sure if I have explained that very well. Would it be possible for you to produce a "exploded" drawing of how the whole thing, including a hot-end mount, should be assembled?

I'll try to later today, but if I'm getting you correctly you are confusing where the x mount tightens to. The actuator part has 4 pins to connect with the actual hotend assembly and then from the back goes a screw into the hotend assembly brass insert in the middle. base then attaches to that with the two horizontal screws and the adjustment ones on the sides of the piezo. all that goes to the x carriage as normally

Ahh! I understand. I use a short screw to attach the hot-end assembly to the actuator and then I attach the actuator to the base. :-)

Sorry for being stupid.

Nah, you are not. I did a bad job making appropriate pictures and writing the description.
How did it work for you by the way ? I am very happy with the results so far - once dealed, only pressing print. Attaching a photo of a first layer from today //

So I have recently found the time to test this and I am getting some odd results. I wonder if you have any thoughts... I have a custom Titan Aero mount made to the standard HEVO mounting patter that is plugged & bolted into the universal base as shown.

If I barely touch the Titan Aero anywhere the blue led flashes on the PP20 board :-) Which is a good thing.

But when I probe, the bed seems to have to push the nozzle/head assembly quite a long way before the probe triggers and sometimes it doesn't trigger at all. I have made the Z travel faster to try and create more of a "hit" than a gentle push and I think it improved slightly but it is not reliable.

Do you have any suggestions or comments?

Thanks in advance

@Pumpal, please ignore previous comment. I took the assembly apart and re-built it and now it is triggering reliably :-) I ran 10 BED_TILT_CALIBRATE commands in Klipper on my HEVO. Here are the results (I do not know if they are good or bad):


I had the same exact problem with the original mount (the one with the drilled piezo). It was easy to activate by hand but not from the home procedure. I got tired of trying and thus was the reason for making this thing. It has been working great since then..

If you get something flat, like a knife for example and try to push the nozzle in vertical direction (push, not knock), do you get a strong signal?
How was the piezo fit inside the nest ? If you try to lower the tension by unscrewing a bit the two screws on the sides of the piezo is there any change in the results you are getting? Or if you try to tighten a bit more ? Anyway, see if applying pressure in vertical direction with something flat triggers the probe reliably for starts..

It was the way I had assembled it I think. Anyway it is working much better now. I have adjusted down the sensitivity some more I am getting quite consistent results:



EDIT: Second results pasted are much more like it - good job!

Looking at the first pasted results I think they were not perfectly consistent. With such, I would try tightening the two screws around the piezo some more and if it doesn't help, reduce a tiny bit the sensitivity... We are looking at not much then +-0.01 inconsistency in readings.
Also oozing filament does not make my probe trigger earlier if I probe at working temp .. that will be a problem with a real print as it is unavoidable to have some oozing when the extruder gets to temp... my probe successfully smashes it without triggering. I also added to my startup G code 2 additional Z homes with a little move in between so any filament gets cleaned and so the Z home is more likely to be reliable.

Thanks - I have assembled it again, but not had time to test it for a variety of reasons. It's probably going to be a while yet too :-(

  • I am using Klipper firmware and haven't worked out how to configure the probe and a Z end stop yet
  • I need to take it apart again as I have just received a new board from PP (I was going to use the Universal board, but now have one of the very new PP20 boards with only one potentiometer on them.
  • I need to do some work on by heatbed and glass to make it more stable
  • I have just received a cloned Titan Aero from China that I would like to design a mount for

Lots to do and no time!!!

When I open and print teh V1.1 Base, the bottom two holes for the dowel pins are not holes. I see them in the thingview here, but not in simplify3D or 3DBuilder, or FreeCad. I can drill them out, but just an FYI.

Thanks for letting me know - will upload a new version tomorrow immediately.

Should be fine now - fixed!

Cool, Also, I'm thinking something like this for the hingeline:

I might wait till I can sneak one in on a mcmaster order here at work. But I'd drill the hole out a touch for the 4mm shoulder.

I'd say the move is so little that it really doesn't matter what you will put there. It is just to keep it together. Ordinary screws seem to work just fine.

I see, so I need a second set of the roll pins, and my complete fan and hot end mount setup can all go on with this?

I'm going to start printing as well, I have the precision piezo kit in a box somewhere, I'll take a look at what I might need to make this happen. I'm super excited, I've been using BLTouch, but I haven't been super pleased with its consistency over time, this will be a welcome alternative.

Exactly - a second set of the dowel pins and a brass insert .. Keep us posted. Thanks.

Comments deleted.

This sounds almost perfect for my needs. I have been looking for a piezo mount that wasn't "tied" to the hot-end choice - I haven't really made that choice yet ;-) (I'll use a clone E3D v6 for now but possibly go to a Titan Aero or a Nimble later.

I will try to print this over the next few days and let you know how it works. Thanks!

I am using it for few weeks now with results that are definitely superior to before when I was using the standard probe with the drilled piezo. The key is to tighten it well enough to get repeat-ability and still a sensitive probe with clean triggering.

When I was designing this, it was intended for a Titan Aero and that was the first I was using it with. But it is too heavy for the default 8mm X rods that are sagging already with the bowden setup, not to speak if you put another 300gramm on top. The sag I measured to up to ~0.5mm in the middle, whereas it is between 0.1~0.2 with the Nimble. If you intend to mount something heavier as X, you should consider 10mm rods or one of the linear rail mods (haven't seen one completely convincing myself as of yet)...

I hope it works good for you.
Definitely would appreciate your feedback.

Thanks. Will let you know as soon as I have tried it.