Tina2 / Tina2S Anatomy & Repair

Moderator: CrazyIvan

User avatar
CrazyIvan
Posts: 245
Joined: Mon Jul 10, 2023 10:25 am
Has thanked: 19 times
Been thanked: 41 times

Re: Tina2 Anatomy & Repair

Post by CrazyIvan »

KenW wrote: Sat Nov 11, 2023 12:44 am Somebody on one of the Facebook Tina2 groups has posted a video of their Tina2 printing all the layers on a slant.

When I requested further information, it turned out the bed levelling calibration only did the first pass (but it presumably had always done that so they didn’t know it was incomplete).

I’ve asked for a video to see exactly what happens during a z offset calibration, but I’m still waiting.

Meanwhile, any comments?
This sounds like a proximity sensor fault.

The proximity sensor is the rod with the rubber tip mounted to the right of the nozzle (see Item 1 in the photos above, noting that the tip is blue in the photos but the original part is orange). The sensor uses a magnetic field to detect the steel plate under the removable magnetic print surface without contacting it, and it is used to calibrate the z (vertical) position of the first print layer. There is a LED on the top end of the sensor (where the cable comes out) which lights when there is steel (or similar) within the detection range of the sensor, and this is coincident with the signal sent to the controller.

The sensor itself can be tested (if the printer is powered up) by bringing something made of steel close to the tip and watching for the LED to come on. (This, and mechanical adjustment, is easier if the guard is removed.) Note that stainless steel (eg cutlery) is not suitable.

The software running in the controller also uses the sensor to take a reading of the bed in three places before every print so that it can automatically adjust the print process for the bed being off-horizontal ("bed levelling" – it doesn't actually level the print plate, it compensates for the plate not being level). It does this by adjusting the z of the nozzle within a layer (when you would expect z to remain constant) according to the x and y of the nozzle and the compensation calculation derived from the bed levelling operation.

I've had two different sensor faults with different consequences, both the result of a broken wire which makes or breaks contact as the cable flexes(the differences being according to which wire out of the three in the sensor cable is intermittent). I've also had an intermittent cable to the heater (see posts above)!
  1. If the +12V wire is broken, the controller believes the sensor has been triggered even when it isn't. The LED will be off.
  2. If the 0V wire is broken, the controller believes the sensor has not been triggered even when it is in proximity of the bed. The LED will be off.
  3. If the signal wire is broken, the controller believes the sensor has not been triggered even when it is in proximity of the bed. The LED will be on or off reflecting the true sense state, but this information does not reach the controller.
The "break" could also be a bad contact in the connector, and note that an intermittent connection might manifest according to where the hot end is positioned (and thus the exact flexure of the cable).

The following assumes the proximity sensor is fitted at the correct height in the hot end assembly, ie it detects the bed before the nozzle makes contact with the print surface, but it is not lower than the nozzle (preferably a little higher – the lower it is, the more likely it is to foul the print). The clamping screws used to fix the sensor in position are of poor quality with rough tips, so they bite on the threaded barrel of the sensor and force it into one notch or another – I found that filing the screw tips flat made exact adjustment much easier.

During Bed Levelling

The bed levelling operation can be performed from the menus, but it is invoked automatically when commencing any print, and as a preliminary to Z offset calibration. The correct sequence of operations is:
  1. The nozzle positions front left, and lowers quickly at first, then slowly for the last 10mm or so until it triggers the proximity sensor, then lifts about 10mm.
  2. The nozzle positions front right (not so far to the right – the sensor has to remain over the plate), then lowers slowly until it triggers the sensor, then lifts about 10mm.
  3. The nozzle positions back right, then lowers slowly until it triggers the sensor, then lifts about 10mm.
  4. The nozzle then remains in this position (menu operation), or remains in this position while it heats up before commencing the print (print operation).
If (due to a fault) the controller thinks the sensor has triggered when it has not, the nozzle does not lower but instead lifts and then moves to the next location. This results in a print commencing in mid air. If the sensor does not trigger, or the controller does not receive a trigger, expect the nozzle to drive into the bed and the z motor will make a graunching sound because it is being driven but can't move (I don't know whether any self-protection kicks in – be prepared to pull out the power plug).

During Z Offset Calibration

Bed levelling is performed automatically as a preliminary to Z offset calibration. Consequently is it very unlikely that the process will get as far as actually calibrating the Z offset if the proximity sensor is faulty.

However, for information: the Z Offset menu operation lowers the nozzle until the sensor detects the bed (this is "zero" point). The exact offset between the sensor trigger height and the position of the nozzle tip can then be dialled out using the rotary knob, in hundredths of a millimetre. The recommended offset is to use a slip of ordinary copy paper as a feeler gauge, and adjust the knob so that the paper is not quite pinched between the nozzle and the bed – it can be moved, but with some resistance. It's easiest if you position the paper under the nozzle before you start the Z Offset operation.

The print nozzle needs to be clean when you do this. If necessary, the nozzle can be cleaned by performing a retract filament operation, and wiping the nozzle with a wad of kitchen paper while it's hot (be careful!).

If (due to a fault) the controller thinks the sensor has triggered when it has not, the Z Offset will leave the nozzle in the air, nowhere near the bed. If the controller thinks the sensor has not triggered, the nozzle will hit the bed and the z motor will make a graunching sound because it is being driven but can't move (I don't know whether any self-protection kicks in – be prepared to pull out the power plug).

Note that if the fault is intermittent, it might not show up during Z Offset.

This video below shows the correct functioning of the Tina2 during a print, which always commences with a bed levelling operation. Any deviation is indicative of a fault.

Post 1 in this thread provides details of disassembly and replacement (viewtopic.php?t=560)


User avatar
CrazyIvan
Posts: 245
Joined: Mon Jul 10, 2023 10:25 am
Has thanked: 19 times
Been thanked: 41 times

Re: Tina2 Anatomy & Repair

Post by CrazyIvan »

I have revised the above. Playing around, I realised the z offset starts with a bed levelling, so it is impossible to run them separately.

slybunda
Posts: 66
Joined: Wed Nov 29, 2023 9:12 pm
Been thanked: 10 times

Re: Tina2 Anatomy & Repair

Post by slybunda »

Some good info on this thread, will help me lots when it comes to fixing my tina2

slybunda
Posts: 66
Joined: Wed Nov 29, 2023 9:12 pm
Been thanked: 10 times

Re: Tina2 Anatomy & Repair

Post by slybunda »

got printer up and running today. fixed proximity sensor wire and also changed the noisy fan for a new one. new fan is 15mm thick compared to old fan which was 10mm thick. new fan got more airflow and is of better quality so should last longer. also made fan extension cable so if ever need to change fan no need to undo the full wiring loom, jst connector on the end of the extension cable allows fan to plug in to the wiring right next to the fan itself. same wiring mod done for proximity sensor too.
Image
Image

slybunda
Posts: 66
Joined: Wed Nov 29, 2023 9:12 pm
Been thanked: 10 times

Re: Tina2 Anatomy & Repair

Post by slybunda »

from the looks of it there seems to be 20mm thick fans that are direct replacement fitment too. will get one of those when this ones time is up.

User avatar
CrazyIvan
Posts: 245
Joined: Mon Jul 10, 2023 10:25 am
Has thanked: 19 times
Been thanked: 41 times

Re: Tina2 Anatomy & Repair

Post by CrazyIvan »

I'll be surprised if the existing screws are long enough. In fact, I'm surprised they were long enough to fit 15mm.

Bear in mind also you are affecting the centre of mass for the hot end, which is cantilevered off the X carriage, so this might alter the dynamics a little. Not that I think that necessarily matters, just watch out for print defects which might be the result of too much acceleration/deceleration in X.

slybunda
Posts: 66
Joined: Wed Nov 29, 2023 9:12 pm
Been thanked: 10 times

Re: Tina2 Anatomy & Repair

Post by slybunda »

Hi, i used 5mm longer screws to do the job. Also iv added tye Z screw stabiliser mod that you print from thingyverse and add in a bearing to cut down considerably on gantry tilting.
Also the fan specs of the one I used:

Nidec U30R12NS1Z5-51
Dimensions: 30x30x15 MM
Voltage: DC 12V
Operating Voltage: DC 4.5 ~ 13.8V
Current: 0.05A
Input Power: 0.6W
Speed: 7500 RPM
Air Flow: 4.4 CFM, (0.125 m³ / min)
Noise: 20 dBA

And here is the Z screw mod i used:
https://www.thingiverse.com/thing:6050289

So far printed few items and all is good, printed a calibration cube which came out fantastic much better than the one it printed when printer was fully stock.
Will get more pics up of the printer with fan mod on it once i copy the pics to the computer.

slybunda
Posts: 66
Joined: Wed Nov 29, 2023 9:12 pm
Been thanked: 10 times

Re: Tina2 Anatomy & Repair

Post by slybunda »

Here is video of what my original fan sounded like:

slybunda
Posts: 66
Joined: Wed Nov 29, 2023 9:12 pm
Been thanked: 10 times

Re: Tina2 Anatomy & Repair

Post by slybunda »

this is the main mod everyone with this printer should do. the 2 zip tie method on the cables so they are not pinched down horizontally onto the gantry but are held upright placing considerably less flexing on them:

Image
Image

User avatar
CrazyIvan
Posts: 245
Joined: Mon Jul 10, 2023 10:25 am
Has thanked: 19 times
Been thanked: 41 times

Re: Tina2 Anatomy & Repair

Post by CrazyIvan »

slybunda wrote: Sat Dec 16, 2023 11:08 pm this is the main mod everyone with this printer should do. the 2 zip tie method on the cables so they are not pinched down horizontally onto the gantry but are held upright placing considerably less flexing on them:
A "before" photo would have been useful. The umbilical on my Tina2 is not "pinched down horizontally" in the way you describe, although there is a hole in the metalwork for the cable clamp shown in your photo. I'm sure the umbilical would not have enough free movement for the X motion if it were clamped, but I did buy it second-hand.

slybunda wrote: Sat Dec 16, 2023 5:11 pm Also iv added tye Z screw stabiliser mod that you print from thingyverse and add in a bearing to cut down considerably on gantry tilting...

https://www.thingiverse.com/thing:6050289
I am concerned about this. So far as I can see, the Z axis lead screw needs to be floating, so that any inaccuracies in the guide rails, or any slight bend in the lead screw itself, do not cause the lead screw to bind. As you see, there is a residual socket for the top end of the lead screw moulded into the top of the case, and the lead screw has been made too short to reach into that socket – I presume deliberately.

Fixing the top of the lead screw should not be necessary, or appropriate. The Z motion is constrained by the guide rails at each end of the X gantry, and by the linear bearings which run on them. The guide rails are kept parallel by their top and bottom fixings, and they have to be dead straight. If you've gotten away with fixing the top of the lead screw, then (1) your lead screw must (luckily) be exactly parallel with the guide rail, (2) your lead screw must be perfectly straight, and (3) your printed stand-off must be exactly the right size.

How did you manage to remove the top panel to fit the stand-off? So far as I can see, there are so many cables and such like, it's a major job.

Post Reply