Pinda Probe Issues: Cannot Keep Z-Offset
Hello All,
I have been having issues since November of last year. I am having the issue described in the manual where once the Live Z-Adjust is made, the printer will not hold this Live Z-Adjust Offset. The manual tells you to make sure the PINDA probe height is not too high. Lowering the PINDA probe, even to the height where the PINDA probe is hitting the filament laid down, does not solve the problem.
A little Backstory: I purchased one of the preorder units, and used the printer all summer. Around august, my thermistor failed, so I purchased an upgrade kit from E3D that included a new heater block design, which allowed me to use the updated thermistor for the V6. If I remember correctly, I started having issues once I got the printer running. I couldn't get a consistent Z-Offset, so I reverted to FW 3.0.3 to see if back-dating could render the printer useful again, but I ended up gouging a huge line in the PEI with the extruder nozzle. Keep in mind that this gouge was not the full depth of the PEI and that there is no visible damage to the MK42 build plate. Also, I believe the gouge was made while the extruder was hot, and since I was printing PET @ 215C (and the glass temp. of PEI is ~217C) I am assuming that it melted the gouge rather than mechanically deforming it. As I was not aware that PR would allow current owners to purchase replacement parts on the site, I bought a replacement PEI sheet on amazon, found at the link below:
The PEI at the link above is 0.8mm in thickness, and when combined with the adhesive, is much thicker than the build surface that comes installed from PR (.175mm + some adhesive, so ~.5mm maximum). I followed the instructions found in the manuals section for replacing the PEI, and when I finally got the new build surface installed, FW 3.0.9 had just been released. With the extended calibration introduced in FW3.0.9, the printer said that my front calibration points were unreachable. I posted an inquiry on this forum and was told to increase the distance from the back of the Y-axis to the Z frame from 100mm to ~101 or 102. This remedied that problem, but then I had too much skew in my x-y axes. Yesterday I verified that the Y-axis distances were within .3mm of each other (101.5 vs. 101.8), and was able to get the skew within a correctable range by loosening all the double-stacked threaded rods and allowing them to get to a "happy place" before retightening them. At this point, I played with the PINDA probe for about 3 hours, following the calibration steps for the kit listed in the manual (for what seemed like the thousandth time), and could not get a consistent first layer.
At this point, the only issue I could determine was that the PEI replacement sheet I bought was in fact too thick, so I contacted PR and purchased a replacement which will arrive sometime this week. I removed the old build surface to prepare for the new one, as well as to inspect the build plate for damage. But thinking about it more last night and this morning, I can't imagine how this would remedy the problem. Over last summer, my the Printer worked well with the PINDA probe at the prescribed location according to the build instructions, which I believe equated to being about 1/2 the height of the nozzle off the print surface. When I was speaking with a gentleman from PR on the customer service chat yesterday, he advised me to check that the PINDA probe responded when the cord was wiggled/to check at all points across the build surface. It passed this test, and he seemed to think the the PEI sheet was too thick as well.
At this point, I am at a loss as to what I should do next if replacing the PEI with the OEM part does not help. I apologized for such a long read, but I was hoping something I mentioned stuck out to one of you. I am not sure what effects having a too-thick PEI sheet would have, as I don't believe it would really impact the magnetic field produced by the PINDA probe, and seeing as it wont even work with the PINDA probe almost dragging on the surface, I wonder if I have a defective MK42 build plate or PINDA probe. From my experience with induction sensors, I believe they are digital, and so the probe is either receiving a signal or not, so I'm assuming that if the XYZ calibration is working, the MK42 and PINDA are not defective. I am not sure what algorithm the coders at PR have developed to develop the G54 global work offset, but since the PINDA probe scans over the induction coil points during XYZ calibration, I assume it records the response to gradually hone in on the center of the point (maybe something like a Newton-Raphson iterative solution).
Regardless I am at a loss as to what I should do next, and would appreciate any input. I had great success with this printer in the first few months I had it, and miss being able to print with such accuracy. Any help is appreciated and thank you for reading my post!
Benjamin
Re: Pinda Probe Issues: Cannot Keep Z-Offset
Benjamin,
Not 100% it will fix your problem. I had a similar issue. It drove me crazy. I believe I have it fixed now.
I started by redoing the X and Y belts - the tension was to high on the X and low in the Y.
I then did a hard reset of the firmware settings.
Next I re-flashed with the latest 3.10 firmware and ran the simple V2 calibration routine to get close. I used the procedure offered by Jeff Jordan.
During the issue I had my Z-offset up to -4 mm
I have successfully run the printer four times with a Z-offset of -0.650.
Hopefully this works for you.
Best,
Chris
Re: Pinda Probe Issues: Cannot Keep Z-Offset
Benjamin,
Not 100% it will fix your problem. I had a similar issue. It drove me crazy. I believe I have it fixed now.
I started by redoing the X and Y belts - the tension was to high on the X and low in the Y.
I then did a hard reset of the firmware settings.
Next I re-flashed with the latest 3.10 firmware and ran the simple V2 calibration routine to get close. I used the procedure offered by Jeff Jordan.
During the issue I had my Z-offset up to -4 mm
I have successfully run the printer four times with a Z-offset of -0.650.
Hopefully this works for you.
Best,
Chris
Chris,
Thanks for the tip. I will be sure to try that once my replacement PEI arrives. I had intermittent layer shift issues previously, so it is possible that could be part of the issue.
Benjamin
Re: Pinda Probe Issues: Cannot Keep Z-Offset
Short answer:
I tend to think it's a too thick sheet as well
Longer answer:
The key thing that comes to my mind is the following
The PEI at the link above is 0.8mm in thickness, and when combined with the adhesive, is much thicker than the build surface that comes installed from PR (.175mm + some adhesive, so ~.5mm maximum)
The PINDA is supposed to have at best a sensing distance of around 1 mm. Supposedly, the actually usable distance might vary from probe to probe. If your PEI sheet already is 0,8mm thick plus the adhesive on top of that and lets assume you have a slightly "deaf" PINDA you might be exceeding the actual max distance for your probe and then all bets are off.
So, I wouldn't do anything more at this moment. If its still flakey with the new PEI, i'd try to lower the PINDA a tiny bit (few fractions of a turn) and then see where we're at.
Re: Pinda Probe Issues: Cannot Keep Z-Offset
The PINDA is supposed to have at best a sensing distance of around 1 mm. Supposedly, the actually usable distance might vary from probe to probe. If your PEI sheet already is 0,8mm thick plus the adhesive on top of that and lets assume you have a slightly "deaf" PINDA you might be exceeding the actual max distance for your probe and then all bets are off.
So, I wouldn't do anything more at this moment. If its still flakey with the new PEI, i'd try to lower the PINDA a tiny bit (few fractions of a turn) and then see where we're at.
Stephan,
Thanks for the input. Originally I was thinking, "Well if it's too thick, why wouldn't it have just jammed itself into the build plate?", which I now realize is because I had it so low in the first place. And then I remembered that during the XYZ calibration it slowly lowers itself, and I was very attentive to make sure it didn't contact the build surface using the paper trick described in the manual. I'll have to bring the PINDA back up anyways, as it was dislodging deposited filament during my last test, so I think I'll put it at the prescribed height and then go from there.
Benjamin
Re: Pinda Probe Issues: Cannot Keep Z-Offset
Hello All,
I order a replacement build surface from PR, but also bought a 5 mil replacement piece from CS Hyde. The PR item hasn't shipped yet, but the CS Hyde item arrived yesterday and I installed it shortly thereafter. This looks much closer to the OEM build surface compared to the Gizmodorks item i purchased on Amazon. I returned the PINDA probe to the position shown in the assembly manual, and ran the XYZ Calibration. The printer still wouldn't hold the Z-Offset. I repeated the same process each time, lowering the PINDA probe each time, with no improvement in result. I have factory reset the printer several times, checked all electrical connections, and I believe all my belts are tight enough (Although I'm really not sure how this would impact the print). At first, I thought it may have been a glitch where the Z-Offset is saved but not applied to the next print when the original print is cancelled, but I manually adjusted the offset to get a successful print and let it finish.
At this point I have no other theories of what could be causing this problem. The only thing I can assume is that either the PINDA probe or MK42 build plate is defective. I originally bought the printer last summer, and there were some times where the printer malfunctioned and pushed the nozzle into the build plate during Z-Calibration. Recently, the printer has not been able to get that low, even with the PINDA probe basically at the build surface. However, the build plate does not appear to be damaged or warped from any of those instances, and there is no indication of any malfunctioning components other than the issue described. Any further troubleshooting tips would be appreciated and if you have any questions at all, I will attempt to answer them to the best of my knowledge.
Benjamin
Re: Pinda Probe Issues: Cannot Keep Z-Offset
Im not quite sure what is actually happening. So let me ask some questions;
Does it reliably detect all 9 points during a calibration?
When you say: "it won't hold the Z-offset", what do you mean by that?
Does it reset your offset? Does it never save the offset or do you need a different offset every time you try to print something?
Re: Pinda Probe Issues: Cannot Keep Z-Offset
Im not quite sure what is actually happening. So let me ask some questions;
Does it reliably detect all 9 points during a calibration?
When you say: "it won't hold the Z-offset", what do you mean by that?
Does it reset your offset? Does it never save the offset or do you need a different offset every time you try to print something?
It has reliably found all the points, except for once yesterday (Right Top I believe). I reran Calibrate-Z and it found it. First time it has ever done that.
I can give a clearer example regarding it "Not holding an offset":
I did this process several times over the past few days. After changing the height of the PINDA probe, I do a hard reset and go through the calibration sequence (Factory Reset, Reflash 3.0.10, Self Test, XYZ, Z). Then, I print the Prusa.stl that comes included with the PR software. I use the flat base of this part to tune the "Live Z-Adjust". To do this, I bring the feedrate down to 50-60% (so I have time to crank the offset down before I start dragging a big ball of filament around). I have a decent eye for getting good squish/adherence, so I just eyeball it and let it complete the first layer before stopping the print to ensure the individual lines of the shell don't break apart. I then restart the Prusa.stl print, and the issue I'm having is that even though the Z-Offset hasn't changed in the machine UI (eg. I set the Z-Offset at -0.565 for a good squish in the first print, and it reads -.0565 when the second print is underway), the nozzle is several millimeters above the build plate. Is that a clearer explanation?
Re: Pinda Probe Issues: Cannot Keep Z-Offset
Thanks for the explanation. I am, however, at loss what would cause this.
I think by the fact that all points are detected we can assume that the PINDA is working. Also means the inserts in the bed should be fine.
I am grasping at straws here.. unfortunately.
There have been numerous cases on the forum where z-offset varies per print. Usually the problem is avoided by raising the z-axis before preheat / start of of a print. But I've never seen it vary by that much, not "several millimeters".
You think it could be a problem with z-axis motion perhaps? Some clicking from the motors indicating lost steps?
- I myself had a problem where the wire bundle got caught on the rambo case and the printer lost steps during the homing process.
- Another problem I am aware of with the Z-axis has been the black dustcovers sitting on to of the motors being too tight adding alot of friction.
I've had problems after cancelling a print using the X button on the front panel - apparently, that's a No-No (i wasn't aware of).
(when this happens and you rerun calibrate-z, the next print is ok?)
I think we need somebody with more knowledge than I got to help.. Sorry..
i'll go stare at my MK2 maybe something else pops into my mind.
Re: Pinda Probe Issues: Cannot Keep Z-Offset
My Live Z adjust will does not commit to memory either. I printed Jeff's test square and finally got adhesion for my PLA. I dialed in about .450 and was getting great results until I went to print something else and the .450 was still in the LCD but the print was like .000 but adding another .450 up to .900 would get good results. This process continued up and up. Another issue was the speed of the first layer. Since the Live Z was not saved I had to adjust on the fly and do a quick math problem to add .450 to whatever was showing in display. I slowed the print speed down from 100 to 24, dialed the live Z up to whatever (+450) and let the first layer go down good and slow. After the base was finished I speed up the print and was able to walk away while Pug Buddy printed flawlessly. I have since printed several other items with no problem. Not what I was expecting, but its working and I am happy for now. Also scuffing bed with dish sponge and cleaning with 90% ISP ALC does the trick for me.
Re: Pinda Probe Issues: Cannot Keep Z-Offset
Hi Guys,
I find that I get different first layer results if the extruder is very close to the build plate during heating, rather than further away from the build plate...
initially I tried to 'Adjust' this problem out,
which lead to me going round in circles...
what I do now, is make sure the extruder is about 100mm above the build plate during initial heating, and this seems to have resolved the issue...
so it looks like the pinda is temperature sensitive.
the Live Z adjustment needs a specific Gcode to save it to memory after calibration, this G code is in 'V2calibratioan and JeffJordan's improved calibration option.
Good luck,
regards Joan
I try to make safe suggestions,You should understand the context and ensure you are happy that they are safe before attempting to apply my suggestions, what you do, is YOUR responsibility. Location Halifax UK
RE: Pinda Probe Issues: Cannot Keep Z-Offset
I'm having exactly the same problem that benjamin-b2 was having. Right down to ordering a too-thick PEI sheet and installing it because I didn't know I had to be logged in to PRUSA to find the PEI sheet.
Any solution?
RE: Pinda Probe Issues: Cannot Keep Z-Offset
I can't state this as a fact, but Live-Z isn't stored until the basic Layer 1 calibration has been performed. That zig-zag pattern that is basically useless, but; unless it is run the Live-Z number is forgotten when power is turned off.
So start with running the on-board calibration. Adjust the value to something close enough to working, let it finish. Then, print a part and do a fine adjust. Record the number you ended up on, exit the Tune-Live-Z menu, stop the print. Then power down. Clean the bed, and power up again, check the Live-Z number. It should now be what you left it at.
ps: the Prusa PINDA has a detection height of 2.0 mm +/- a few percent. Normally, you adjust it 1.2 mm above the nozzle tip (some set it to 0.8 mm above). Live-Z then adjusts for material thickness.
RE: Pinda Probe Issues: Cannot Keep Z-Offset
@tim-m30 THANK YOU! Now I understand. Unless you do the built-in first-layer calibration, live-z won't work correctly. I'll give it a try...
I skipped the built-in because I run a .25 nozzle and read somewhere it's calibrated for a .40. But I guess I'll have to run the built-in and then, additionally, do a few live-z.
Hope it works!
RE: Pinda Probe Issues: Cannot Keep Z-Offset
@tim-m30 THANK YOU! Now I understand. Unless you do the built-in first-layer calibration, live-z won't work correctly. I'll give it a try...
I skipped the built-in because I run a .25 nozzle and read somewhere it's calibrated for a .40. But I guess I'll have to run the built-in and then, additionally, do a few live-z.
Hope it works!
I hope this works for you - I had a similar experience with 3.7; and read a few others who reported similar. I'll keep my fingers crossed.
The other half of the issue is the PINDA is temperature sensitive and trip points have rather large differences as it warms up. Enough that between a Live-Z set at 25c and a print started when the PINDA is at 35c you'll swear you didn't even run a Live-Z, ever. There is a Temp calibration that is supposed to help, but it also has a few known issues. So adding a defined "acceptable" PINDA temp, and letting it warm up to it, is one way to minimize the problem.
RE: Pinda Probe Issues: Cannot Keep Z-Offset
I'm on a MK2S so I don't think the PINDA has temp compensation. But I've never really has a problem with first layer Z. Must be newer firmware. Hopefully I'll have some time this week to get back up and running.
RE: Pinda Probe Issues: Cannot Keep Z-Offset
I'm on a MK2S so I don't think the PINDA has temp compensation. But I've never really has a problem with first layer Z. Must be newer firmware. Hopefully I'll have some time this week to get back up and running.
Oops - I keep repsonding to MK2 issues with MK3 info - I am sorry. This new forum is very unhelpful that way: I'm in the Mk3/S forum, hit recent, and I get results from MK1-MMU and SLA1 posts.