Welcome! Log In Create A New Profile

Advanced

Precision Piezo Z-probe Now available.

Posted by DjDemonD 
Re: Precision Piezo Z-probe Now available.
January 12, 2020 08:19AM
Quote
Dancer
I do not know.

But there are 4 Piezos... (Standard 220 x 220 Heatbed, 4 Holes on the Edges)
Most important is to wait 1 second before each probing, because the piezos trigger from the Y-movement.

Pictures of the Printer to come (SD-Reader not working in PC...)

I see, FOUR piezos, one at each bed corner, compensation if the two Z are out of synch, the way you attach the belt to the X carriage and path ....

As this has already been shot down a few times, no point discussing further !

Edited 2 time(s). Last edit at 01/12/2020 08:32AM by MKSA.


"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)
Re: Precision Piezo Z-probe Now available.
January 12, 2020 08:48AM
I am not so sure that four-point mounting deserves the bad rep that it has got. Certainly, four-point mounting is overconstrained and invites the uneducated to bend the bed by straining two pairs of opposite corners against one another. I don't see that the method deserves to be ignored because people neither read the manual nor think about the implications as they twiddle the adjusting screws. Equally certainly though, having supports at each of the four corners does make for a more rigid structure and it can be made lighter.

As far as piezos go though,(Edit: As four mounts on the corners has) having no external mass to contribute to the dynamic instability will at least go a little way to making for more predictable behaviour.

For myself, I will keep on with three-point mounting as I have become a great fan of exact constraints - but I won't conduct a holy war against others who work in a different way.

Mike

Edited 1 time(s). Last edit at 01/12/2020 02:25PM by leadinglights.
Re: Precision Piezo Z-probe Now available.
January 12, 2020 05:21PM
thumbs up

I just did it that way, because I had this heatbed on my shelf. And as I do not bend anything on this bed, because it is just put on top without screwing, there will not be any problem. - As you say: You gotta know what you do. spinning smiley sticking its tongue out
I know, that the mounting of the X-belt can be improved, but with my measurement-methods I can't see any problem arising in my prints.

There is one thing on this printer that is much worse (not really a big thing) - it has Z-Wobble in Y-direction about 0.2 to 0.3mm.
Interestingly not in X-direction.

As far as I have seen, it is because the bushings for Z-movement have some play in Y. Not in X because the cable tie puts them straight.
Seems I'll have to rework the whole X-Axis.

@MKSA:
I do not know what your problem is.
Could you show me, how you get two Z-Screws in absolute 100% height-sync? - Forever???
Sorry this is not possible. Because the steppers have 200 full steps per rev (or 400 full steps per rev) and if you shut down the steppers, they go from their microsstep to the neares full step. This does make some 0,.. mm - with my 1,8° Steppers one full step with M8 Leadscrew is 0.00625mm - that is much more than my Standard Deviation. And that can easily be compensated with bed leveling with the piezos.

You can't even get them in 100% height-sync when driven by one stepper, because you would have to mount them, the pulleys that drive them, and the nuts at 100% exactly the same height and angle.
This error is normally compensated by tilting the heatbed accordingly with the adjusting screws.

Sorry MKSA. I think that you are just repeating what you heared from others instead of thinking yourself. hot smiley
Over and Out.
Re: Precision Piezo Z-probe Now available.
January 13, 2020 04:28AM
Quote
Dancer
...
Sorry MKSA. I think that you are just repeating what you heared from others instead of thinking yourself.
Over and Out.

Can an admin prompt this guy to remove this personal attack ?

Of course, the rest of the post can stay as it is so entertaining.


"A comical prototype doesn't mean a dumb idea is possible" (Thunderf00t)
Re: Precision Piezo Z-probe Now available.
January 13, 2020 07:41AM
Quote
MKSA
Quote
Dancer
...
Sorry MKSA. I think that you are just repeating what you heared from others instead of thinking yourself.
Over and Out.

Can an admin prompt this guy to remove this personal attack ?

Of course, the rest of the post can stay as it is so entertaining.

Maybe not the friendliest thing to say, but I have also seen you being very direct. It's definitely not a personal attack. Heck, even I myself can be direct and not the nicest. I hope you manage to let it slide of your shoulders. Let's focus on positivity.

Edited 1 time(s). Last edit at 01/13/2020 07:42AM by Ohmarinus.


http://www.marinusdebeer.nl/
Re: Precision Piezo Z-probe Now available.
January 17, 2020 05:40PM
Z-Wobble gone.
X-Belt was to tight. loosened it up a bit, now prints nearly perfect - just some finetuning in slic3r left
Re: Precision Piezo Z-probe Now available.
January 23, 2020 06:19PM
I've made an e3d effector mount for all you Rostock Max owners out there, fits the Piezo20 v1.4 PCB
An assembly guide is included
Enjoy!
[www.thingiverse.com]
Re: Precision Piezo Z-probe Now available.
January 28, 2020 07:19AM
Hi everyone,


I'm thinking to upgrade my Rostock Max using one of the piezo probes DjDemonD and company provides, but I'm having doubts about the best way to do so. Let me briefly explain my setup:
My Rostock MAX v2 glass bed got broken a few years ago and since then I replaced it with a cheap glass. Since then I haven't been able to print to the full width or consistently for that matter. Lately I'm thinking of purchasing a spring steel bed Energetic beds on aliexpress and place the magnetic sticker directly into the heated bed so as to obtain better thermal conductivity and hopefully a flatter surface. My fear with this is that probing might damage the heated bed if something goes wrong with the probing like a faulty piezo, too little sensitivity in certain probe points or anything else. I hope I can hard code a safety min Z to prevent that (I'm trying Klipper at the moment). I also rather install 3 piezos on the bed rather than using the Orion because it's kind of annoying to rewire a Rostock and also because I want to keep the hotend as fixed as possible (I'm using a genuine E3D Volcano so the installation of the Orion should actually be pretty straightforward). However, since the Rostock has a large bed I wonder if it would be more precise to use the Orion instead of a bed setup. I'm also wondering why I mostly find FSR setups but little on piezos for the Rostock (except CodeBusters proposal the other day). Is it because of practical or popular reasons?

So, to sum it up:

  • Orion or Universal with 3 piezos on the bed? Is there a recommended set up for the Rostock MAX v2 or any big delta for that matter?
  • Is the use of piezos recommended if I'm going to probe with no glass (but spring steel sheet)?
  • Will I'll be able to configure the piezos on Klipper or should I change my firmware?
  • FSR vs Piezo on the Rostock user community? Practical or popular?
Re: Precision Piezo Z-probe Now available.
January 28, 2020 10:55AM
I have been doing some research recently on problems which people have had on underbed piezos. Although many people - including myself - have had no problems with underbed piezos, there is a dynamic problem which is critically dependent on layout and this has plagued other users. The ultimate result of this dynamic problem is there is a significant chance that you may have bad spots where the sensitivity is not great. There are some details on [reprap.org]

From my own observations (please note that I am not a God or a Guru but I am a Geek) For larger beds, the piezo in a probe is likely to be more trouble-free. It is better to have the piezos on an under-bed setup beyond the periphery of the bed and it is better to get rid of mechanical noise and set the piezo as sensitive as possible.

I would point out that the problems I found on under-bed piezos will be as bad or worse on under-bed FSRs. The reason for this is that the sum voltage of several piezos will equal the total contact pressure (lots of caveats such as piezos all being equal). FSRs do not have this advantage and can suffer badly from contacts on parts of the bed beyond the triangle of the piezo sensors.

Mike
Re: Precision Piezo Z-probe Now available.
January 29, 2020 03:20AM
Quote
CaboGuataca
Hi everyone,


I'm thinking to upgrade my Rostock Max using one of the piezo probes DjDemonD and company provides, but I'm having doubts about the best way to do so. Let me briefly explain my setup:
My Rostock MAX v2 glass bed got broken a few years ago and since then I replaced it with a cheap glass. Since then I haven't been able to print to the full width or consistently for that matter. Lately I'm thinking of purchasing a spring steel bed Energetic beds on aliexpress and place the magnetic sticker directly into the heated bed so as to obtain better thermal conductivity and hopefully a flatter surface. My fear with this is that probing might damage the heated bed if something goes wrong with the probing like a faulty piezo, too little sensitivity in certain probe points or anything else. I hope I can hard code a safety min Z to prevent that (I'm trying Klipper at the moment). I also rather install 3 piezos on the bed rather than using the Orion because it's kind of annoying to rewire a Rostock and also because I want to keep the hotend as fixed as possible (I'm using a genuine E3D Volcano so the installation of the Orion should actually be pretty straightforward). However, since the Rostock has a large bed I wonder if it would be more precise to use the Orion instead of a bed setup. I'm also wondering why I mostly find FSR setups but little on piezos for the Rostock (except CodeBusters proposal the other day). Is it because of practical or popular reasons?

Some thoughts on what you're mulling:
The heated bed on the v2 is not perfectly flat. Its shape changes with temperature. I checked this with machining parallels.
The only other person I've seen use a flex bed on Max v2 was Keegan on the SMC User's Group. His stack was the melamine snowflake, the Onyx bed, a machined aluminum heat spreader, then the magnetic sticker, bottom to top.

Precision of the least squares calibration at the furthest extents of the bed relies on two things according to dc42 :
1. Including probe points near the edge of the bed
2. Allowing 7 factor (delta arm length is adjusted) calibration
I get .05mm deviation across the bed. That's pretty friggin good, considering Duet/SE300 hot end users get .03mm deviation as a matter of course.

FSR is common because that's one of the two options Rostock Max v3's came with for the brief period SeeMeCNC manufactured them, prior to making up the v3.2
It is also common that both the FSR and HE280 (hot end accelerometer probe) had bad results.

I think you should go with some form of the hot end probe. Orion is good. My design is sort of a mashup of an e3d effector I liked and the Orion. It actually isn't that hard to "rewire" the Rostock hotend. In fact, all I did was run 3 extra wires up the side of the extrusion. The kapton-taped connections are still there. Check out my guide.
Re: Precision Piezo Z-probe Now available.
January 29, 2020 10:52AM
Quote
CodeBuster
Precision of the least squares calibration at the furthest extents of the bed relies on two things according to dc42 :
1. Including probe points near the edge of the bed
2. Allowing 7 factor (delta arm length is adjusted) calibration
I get .05mm deviation across the bed. That's pretty friggin good, considering Duet/SE300 hot end users get .03mm deviation as a matter of course.

My recommendation is not to use 7- or 9-factor calibration, because (a) that can only give good results if you probe well outside the normal printing area between the towers, and (b) if you use well-made rods (e.g. the magnetic ones from Haydn Huntley) then you will know precisely how long they are.

I use 8-factor calibration on my delta (endstop heights + delta radius + tower position corrections + bed tilt) and the calibration deviation using 16 points is usually below 0.025mm.

Under-bed piezos/FSRs can give good delta calibration results if and only if the trigger height is consistent across all the probe points.

Edited 2 time(s). Last edit at 01/29/2020 10:56AM by dc42.



Large delta printer [miscsolutions.wordpress.com], 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.
January 30, 2020 03:44AM
How far out? On escher3d.com delta wizard it recommends only "a very long way out opposite each of the towers" but not specifically. The max radius of the v2 extends probably ~15-20mm beyond the delta tower triangle. Perhaps even 40-50 beyond the virtual tower triangle.
I, and probably Cabo, are both using well made rods, injection molded ball-cup arms.

Edited 1 time(s). Last edit at 01/30/2020 10:29AM by CodeBuster.
Re: Precision Piezo Z-probe Now available.
February 03, 2020 05:10AM
Thank you all for the feedback! Yes CodeBuster, one simple option is to pull the cables not through the inner hole of the extrusion profile but from one side, I might do that indeed. And thanks for the bed deflection point, I will go with an aluminum spreader as well then, will look for the post on the seemecnc forum to get more details on that setup.

Yes, my Rostock uses injection molded ball-cup arms, so length should be consistent across them all.
Re: Precision Piezo Z-probe Now available.
February 16, 2020 01:04PM
Quote
han45
Hi guys, I'm using the Piezo Orion on a E3D V6 setup with a bowden extruder and it works really well.

Now I ordered an E3D Hemera and I want to use the Orion with that. Is there any chance to make it work? Maybe use the Orion vertically between the Hemera and it's mounting plate to detect the touch on the build plate?
Any pointers on what I can do are welcomed.

I second that. Been using the Orion for over a year, and finally decided to upgrade my cheap MK8 extruder for a Hemera in direct, but would have to go back to a bltouch if I can't find a mounting solution for it, which I really don't wanna do. Anything in the pipeline for an Hemera mount?
Re: Precision Piezo Z-probe Now available.
February 24, 2020 07:46AM
Yes, I'm working on a hermera mount but not for the Orion...


{Precision Piezo} Accurate, repeatable, versatile z-probe plus piezo discs, endstop cables, pt100, 50w heaters.
Sorry, only registered users may post in this forum.

Click here to login