tom.w
Posts: 6
Joined: Fri Nov 11, 2016 4:03 pm
Contact: Send private message

faulty PINDA probe(?!?)

Tue Jan 10, 2017 11:49 pm

Hi all,

I've just had what looks like either a faulty PINDA probe or a callibration memory loss issue.

After some fairly heavy use since I bought it I tried a print first thing of the morning and got a message saying:

Bed leveling failed. Sensor didn’t trigger. Debris on nozzle? Waiting for reset.


I gave Z calibration a try and not only had the two Z motors become misaligned but after moving the hotend to the very top the motors tried to drive it straight into the heated bed, resulting in the same error message.


This surprised me a bit as I understood the point of going to the very top to be as there was a known distance to descend. :?

Furthermore there is supposed to be a limiter which stops the hotend from descending more than 1mm down from the calibrated point - does this not apply during calibration? Makes sense that it doesn't but...


I disconnected and reconnected the probe connection from the board and thought I'd try a full recalibration - XYZ (with finger hovering over reset in case it tried to puncture the heated bed of course).

This started off great! It found the correct Z level and the first couple of calibration points were found just fine. So fine in fact that I made a cup of tea only to be horrified by the sound of the hot end gouging out a good square inch of the heated bed. (pic attached).


So, my question you all is...

:?: Any idea what's going on? I'd guess faulty PINDA probe - unlikely though it seems - but maybe the board is fried (it has seen some pretty constant use in the last few weeks).

:?: Why did the 1mm-under-the-calibrated-height failsafe not work though? (I did also test this by starting a new, supervised print before the calibration effort with the same result - it was trying to push way more than 1mm below sanity)

:?: Anyone seen anything like this before?

All comments gratefully received!

david.b14
Posts: 539
Joined: Mon Oct 31, 2016 12:42 pm
Contact: Send private message

Re: faulty PINDA probe(?!?)

Wed Jan 11, 2017 6:22 pm

Just to see if the pinda is faulty, what happens if you do an autohome and when the Z is being found, tapping the bottom of the pinda with a metal coin?

Does it trigger?

tom.w
Posts: 6
Joined: Fri Nov 11, 2016 4:03 pm
Contact: Send private message

Re: faulty PINDA probe(?!?)

Wed Jan 11, 2017 7:13 pm

Thanks for the reply...

I just tried a galvanised screw and a steel (non-stainless) screwdriver - neither worked and there is no light on the probe.

Does this mean we are in probe / RAMBO fail territory?

EDIT:

I also tried again with XYZ calibration:

* Finding the bed was fine (presumably this is pre-programmed?)
* Calibration points 1, 2 and 3 worked fine - or so it seemed. Light was on the probe the whole time.
* When it came to point 4 the light went off...

Can it just be a faulty connection? That is twisted to break contact when it is too far in one direction?

david.b14
Posts: 539
Joined: Mon Oct 31, 2016 12:42 pm
Contact: Send private message

Re: faulty PINDA probe(?!?)

Wed Jan 11, 2017 7:28 pm

My guess is an issue with the probe or maybe the connections.

Here is a pinda on ebay which is not the same one used on the MK2:
-http://www.ebay.com/itm/LJ12A3-4-Z-BY-Inductive-Proximity-Sensor-Detection-Switch-PNP-DC6-36V-/140907575134?pt=LH_DefaultDomain_0&hash=item20cebef75e

Some specs:
-Inductive Proximity Sensor Detection Switch PNP DC6-36V LJ12A3-4-Z/BY
-Output Type
-PNP NO(Normal Open)

You may want to do a continuity test for each of the wires that go from the rambo to the probe.

Also, you may want to put the probe lower to see if it will trigger afterwhich you will need to redo the z calibrations. Thingiverse has a pinda probe level tool that may be helpful.

Also check that all the parts holding down the bed to the x-frame part and bearings are not loose.

As a test, you could tape a dime to the calibration point that is not being found.

david.b14
Posts: 539
Joined: Mon Oct 31, 2016 12:42 pm
Contact: Send private message

Re: faulty PINDA probe(?!?)

Wed Jan 11, 2017 7:48 pm

Also give this a read, make sure the holder has not shifted:
http://www.thingiverse.com/thing:1958083

User avatar
PJR
Posts: 3237
Joined: Mon Jun 01, 2015 6:37 am
Location: Kamnik, Slovenija
Contact: Send private message Website

Re: faulty PINDA probe(?!?)

Wed Jan 11, 2017 8:30 pm

Most likely a break in the probe cable. Best bet, right behind the extruder body when the cable bends 90 degrees to enter the cable bundle.

Intermittent cable break so it works when the nozzle is in one position but not in another.

Peter
Please note: I do not have any affiliation with Prusa Research. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage or loss.

tom.w
Posts: 6
Joined: Fri Nov 11, 2016 4:03 pm
Contact: Send private message

Re: faulty PINDA probe(?!?)

Wed Jan 11, 2017 8:40 pm

I didn't do a continuity check but a highly scientific "wiggle of the leads" showed that there is, indeed, a shonky connection going on at the probe end of the wiring.

When the printer is reaching 0 in the X axis it is breaking the connection, but it reconnects at x > 100mm or so.

Cause unknown for now - it wasn't moved since it last worked well - must just have been a manufacturing defect / being built in a way that wore the connections too quickly.

Ah well.

Still the solution is simple enough - I will need a new probe and a new heated bed since the nozzle tore up the last one so badly... right on the embedded metal calibration point...

Do you happen to know if Prusa sell the heated beds and probes separately? I can't find them on the store... don't worry if not I've a ticket raised with them, I'd rather get exactly the same parts as come with the kit as a security blanket if nothing else.

Thanks again for the help!

tom.w
Posts: 6
Joined: Fri Nov 11, 2016 4:03 pm
Contact: Send private message

Re: faulty PINDA probe(?!?)

Wed Jan 11, 2017 8:42 pm

PJR wrote:Most likely a break in the probe cable. Best bet, right behind the extruder body when the cable bends 90 degrees to enter the cable bundle.

Intermittent cable break so it works when the nozzle is in one position but not in another.

Peter


That's what I figure.

How fixable / avoidable in future do you think?

Lump of solder and a roll of electrical tape job?

User avatar
PJR
Posts: 3237
Joined: Mon Jun 01, 2015 6:37 am
Location: Kamnik, Slovenija
Contact: Send private message Website

Re: faulty PINDA probe(?!?)

Wed Jan 11, 2017 8:52 pm

It is totally fixable, once you have identified where the break is. If you remove about an inch of cable around the bend I mentioned and solder splice / heat shrink you should be OK.

The parts are available, but you will need to get access to them - they are only available to users who have purchased the Mk2. You need to speak to support via "Live Chat" to get access and also to purchase the parts (they are usually kept as "out of stock").

Peter
Please note: I do not have any affiliation with Prusa Research. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage or loss.

tom.w
Posts: 6
Joined: Fri Nov 11, 2016 4:03 pm
Contact: Send private message

Re: faulty PINDA probe(?!?)

Wed Jan 11, 2017 9:19 pm

Awesome, thanks for your help Peter, I'll take a look when I'm a bit fresher.

I'll probably get new parts in anyway - I'll certainly need a new heated bed - but would be nice to have a go at the fixing anyway.

What's the prospect of avoiding future damage? Is there a way of strengthening the cable? Another length of filament maybe?

Return to “Prusa i3 kit - Building, calibrating, first print, maintenance, ...”

Who is online

Users browsing this forum: ales.f and 3 guests