Tina2 Malfunction – Bed Levelling [SOLVED]

[CrazyIvan]

I have some connectors on their way, and a new sensor (but I hope to keep that as a spare). I plan to print a new fixture to replace the existing plastic anchor seen fitted to the left screw on the top of the head, but spanning both screws and providing a mounting point for an intermediate connector to the sensor.

The problems with replacing the sensor at the moment are that:

• Cable ties have to be cut to release the existing sensor, and replaced after fitting a new sensor;
• Unless you buy a replacement part from the manufacturer, a generic sensor is bare-wire so you need to source a suitable connector (JST XH series? - not sure yet) / salvage the existing crimp connector (not easy) / splice onto the old cable (easiest);
• The cable is still subject to flexure as the head moves, and therefore still subject to failure.

My plan will address those points, but I need a working printer to print the fixture! There's a hole in my bucket, dear Liza dear Liza...

So I'll bodge it to start with.

[CrazyIvan]

...and we're back!

No, I didn't bodge it. What I actually decided to do was install the new sensor as per the original (more or less), but made some improvements to fitting. With a fully-fledged working printer, I can prepare the parts for an upgrade and hold that in abeyance until needed or the whim takes me.

The trickiest part was trying to crimp the sensor leads into a new JST XH 3-way connector, as I didn't want to spend £20 or more on a crimp tool (but I might have to in the end). I found the only way I could make a reliable joint (without the crimp tool) was to solder them, just squeezing with pliers was not accurate enough.

The improvements I made were to bring the cable sheath into the anchor instead of it stopping short, bringing the sensor cable through the anchor and making a swan-neck in it, and providing the leads to the heater block a bit more slack.

Something I don't like is the grub screws gripping onto the sensor body – I wish there was thread at each end of its mounting so that the locking nuts supplied with it can be used. The grub screws grip onto the thread, which makes the adjustment notchy rather than continuous. If I can, I'll slip some shim in so the screws press on that, and then the adjustment should be smoother. (Update: I discovered the problem is that the screw tips were not flat. Filed them off, problem sorted.)

[slybunda]

hello, registered to say i got a similar problem to you with the proximity sensor cable, here is a video of its issue:


and here is a video of where i have pinpointed the problem itself:


seems to be a break in the cable, im going to go a different route in fixing this and change the sensors wire itself and use silicone wire which are more flexible.
i believe the connectors on the board are jst xh 2.54.

[CrazyIvan]

Welcome to the forum. Thanks for posting up video.

seems to be a break in the cable, im going to go a different route in fixing this and change the sensors wire itself and use silicone wire which are more flexible.
i believe the connectors on the board are jst xh 2.54.

Yes, it's definitely a break in one wire, I suspect the 0V connection. If the break was in the signal wire the light would still come on, and if the break were in the 12V wire it would appear to the controller that the bed had been sensed early. In the latter case the behaviour will be different. It also very much depends where in the travel the break occurs as to what actually happens; in the above video the connection is intact for the first sense position but opens on the way to the second sense position.

I thought about a length of flexible wires through the umbilical, but for the moment simply replaced the sensor complete with cable and made sure there are no kink points to concentrate the strain (see photo above). OK so far! If you look, you might find the break is immediately where a cable tie is.

What I had in mind was a printed cable guide to fit on the hot-end, but I needed a working printer for that so...

Note that silicone is only the insulation on a wire; unless the conductor itself is suitable constructed, it will still fracture under repetitive flexing.

Just to note:

The connectors appear to be JST XH type (fan 2-way, sensor 3-way, heater block + thermocouple 4-way). The JST XH spec is 2.50mm pitch, but several suppliers of supposedly "JST XH" connectors say 2.54mm pitch. However, the difference between 2.50 and 2.54 over 4 pins is too minor to make a difference, and I have found they fit regardless.

[slybunda]

I think main cause is the ziptie clamping too hard on the cable will have a good look at the printer over the weekend to see if i can use a clamp on a ball joint to put less strain on it like cars have on some hoses.
Also i contacted weefun support who are sending me a replacement proximity sensor under warranty for no cost so thats the backup plan sorted.

[slybunda]

looks like the sensor is still under warranty. weefun support is sending me a replacement one. i will still be looking to mod the gantry to reduce cable stress or look into converting to the use of a flex ribbon cable.

[slybunda]

Replacement sensor arrived, incase anyone wants to know its got part number: LJ12A3-4-Z/BX NPN 4mm, 6-36v DC

[slybunda]

All my wiring finally arrived so will get to work sorting this out. Worked out a few changes i can make to improve on the wiring to hopefully stop it from getting damaged again.