The various user manuals for Tina2 & Tina2S (as downloadable at the date of writing) do not agree on the specification for build volume (X x Y x Z in mm):
- Entina Tina2: 100 x 120 x 100
- Entina Tina2S: 100 x 105 x 100
- Weedo* Tina2: 100 x 105 x 100
- Weefun Tina2: 100 x 120 x 100
- Weefun Tina2S: 100 x 105 x 100
And yet, they're all based on the same mechanics, so what's the truth?
In fact, they are all correct in a way, because the specifications should be read as "not less than" – but nonetheless they are also all wrong, because they undersell it.
The build platform is 110x120mm (X x Y), so does that define the limit? No! We can actually achieve a maximum build volume 105 x 120 x 100. How? Read on...
Axis Calibration
Each print axis is driven by a stepper motor, directly under the control of the processor. Stepper motors produce tiny increments or decrements in the position of the print head by "dead reckoning" – it only knows how far to move from the current position, and has no idea where the print head actually is when the power is first turned on.
That problem is solved by "homing". Small PCBs (fitted with microswitches) send signals to the processor when the print head reaches the home point, so the processor then knows where the print head is, and can use dead reckoning thereafter.
The microswitch assemblies are described as "stops", which in electromechanical terms should imply they prevent the motor driving any further by disabling the drive, but in fact they do not – it is left up to the processor firmware to detect the switch actuation and choose what to do. We should instead understand these sensors as setting the datum for the carriage in each respective axis.
There will be a small amount of mechanical travel available after the microswitch has actuated, before the mechanics themselves prevent any further travel. That does not stop the motors being sent signals to move, but they just won't (and will make a clicking noise).
Similar happens at the other end of the travel; the mechanics prevent movement but there's nothing to stop the motors being commanded. There are no sensors in this case though.
X Axis
The X axis motion is datumed at the right hand end (as viewed from the front), which the processor firmware uses to set X = 105.0mm (yes, 105mm – not the advertised 100mm!).
The firmware prevents the axis being driven outside the range 0-105mm, so that's the maximum we can use regardless of the build platform being 110mm, and regardless of any additional travel being available mechanically.
Y Axis
The Y axis motion is datumed at the back (front edge of the platform under the nozzle), which sets Y = 0mm. The firmware allows for Y motion 0-120mm, but the mechanical limits on the actual motion of the platform prevent it reaching the full 120mm by about 3mm (say 117mm) for the Tina2, or even less than that for the Tina2S (see later).
To test this, using the menus:
- Prepare > Auto home
- Prepare > Move axis > Move Y > Move 10mm
- Prepare > Move axis > Move Y > Move 0.1mm
Particularly on the Tina2S, you might have to repeat the above and start incrementing from +105.0, if +115.0 was already beyond the limit of travel. For the sake of continued discussion, I will assume the mechanical limit is 115mm – if your figure is less, substitute it.
Z Axis
The Z axis motion (ie up and down) is datumed at the top. This becomes complicated by auto bed levelling and Z offset, but by the time all that is taken into account, essentially the Z axis travel is limited (in firmware) to the range 0-100mm.
Making Use of the Additional Volume
When printing, the nozzle movements are commanded by the Gcode for the print, which is produced by the slicer software. The actual available build volume is the least of the limits imposed by the firmware, the limits imposed by the mechanics, and the limits imposed by the slicer.
As we have seen above, disregarding the slicer, the limits are:
- X – 105.0mm (limited in firmware)
- Y – 115.0mm (approx, mechanical limitation)
- Z – 100.0mm (limited in firmware)
I use Ultimaker Cura, and the printer definitions supplied as default are:
- Tina2: 100x120x100
- Tina2S: 100x110x100
Why unsafe? Well, even supposing there were no mechanical limitations on achieving 120mm travel in Y, there is no margin for calibration error. If the Y stop is not in exactly the right place, and the removable magnetic mat is not precisely positioned on the plate underneath, trying to print something 120mm (including any raft, skirt, or brim) in the Y dimension, is almost bound to run off edge of the build platform!
At the 0mm end of Y travel, the nozzle will print near the front edge of the build platform. Where this is, relative to the actual edge, depends on the precise location of the Y stop, and cannot be changed (except by how the magnetic mat is positioned on the plate). Thus the Y stop calibration should be a mm or two in from the edge, and the definition in Cura set to 115mm (or less for the Tina2S, if your measured figure is less), to provide ±2.5mm margin for error so the nozzle won't run off the plate.
As for the X definition, that can safely be increased to 105mm. As noted above, the build platform is 110mm in X, which already provides a ±2.5mm margin for the 105mm travel.
Changing the definitions in Cura is straightforward. Click the printer drop-down near top left, and click "manage printers" then "machine settings". On the "printer" tab, the left "printer settings" column shows the build plate dimensions, which can be altered.
We've gained 5mm in X, and approximately 5mm in Y on a Tina2S (depending on your measurements), but decreased the over-optimistic Y setting for the Tina2 by 5mm.
Can we do better? Yes we can, but that requires mechanical tuning.
There is a "gotcha" though. Despite the Cura definition now being 105x115x100, it will only slice if the model is slightly smaller than this. The difference appears to be 2 line widths, as 104.1x114x1x100 will slice, but 104.2x114.2x100 won't (assuming 0.4mm lines).
Tina2S Y-Axis Fouling
I investigated why the Y axis on my Tina2S would not reach the full 120mm travel, and found three reasons:
- The cable to the bed heater was secured in a way which caused the platform to foul the printer chassis when near its maximum forward travel;
- The anchor connecting the Y drive belt to the platform fouls the chassis when near its maximum rearward travel;
- The position of the Y stop.
If your Y travel reaches the full 115mm target, there is no need to do anything about the cable fouling. If not, or if you are aiming for 120mm, you might want to correct the bed heater cable fitting (this only applies to Tina2S, there is no bed heater on Tina2), and the anchor fouling.
By peering under the right hand edge of the build platform, you might be able to see a zip tie securing the cable to the bed heater, which then does not clear the plastic chassis when the platform is fully forward.
The photos below show stages of disassembly to get to it. If you are not confident of your abilities... don't even begin dismantling – the point of no return is almost immediate, and limited Y travel is preferable to making your printer inoperable!
First, from the underside of the printer, release the Y belt from the carriage bracket (you might need to slacken the belt, see 'Y Belt Replacement' HERE). Top-side, remove the magnetic mat to reveal eight screws through the plate. These secure to housings for the linear bearings on the Y guide rods, remove them all to release the plate.
...and disassembled:
Look closely, and you'll see the zip tie fitted in such a way that the "block" sits up (thus requiring greater clearance), and located further forward than is entirely necessary. This could be an assembly error on my particular unit, which perhaps does not affect some or all other units:
I removed the zip tie and fitted another in a different orientation, which brings the "block" away from the front edge of the plate and gains a crucial extra few mm of travel:
Meanwhile, the anchor should have about 2mm filed off in the critical area, as shown below. I advise removing the anchor from the plate to do this – you don't want to risk damaging the heater.
Re-assembly is tricky, but do-able. Just be methodical. Place the upper halves of the linear bearing housings over the bearings, then aligning everything place the plate on top, refit the 8 screws, then reconnect the Y belt.
The extent of travel is now only limited by the placement of the Y stop, so what do we do about that?
Y Axis Datum Calibration
As explained above, the existing Y stop microswitch is positioned conservatively, in other words the datum is a couple of mm inside the front edge of the build platform. We don't really want the nozzle exceeding the edge of the platform, but at the same time the Tina2/2S firmware permits (correctly or not) 120mm travel.
The (now modified) Y axis motion also allows at least 120mm travel, except for the Y stop getting in the way. To take advantage of the extra travel, we need a means to adjust the position of the Y stop.
The Y stop is located on the left at the rear of the chassis, identified here:
...and is secured by screws to two of the pillars more clearly shown here:
I thought long and hard about how to make the the Y axis datum adjustable, and for a number of reasons the best way seemed to be a replacement Y stop assembly, with a custom PCB and fitted with the requisite connector and microswitch, which can slide on its screws and be clamped at the required position.
To discourage outright copying, I'm not illustrating my replacement part, but I am offering it for purchase through my shop (click) as part of a calibration kit. I had considered supplying the bare PCB for users to transplant their own connectors and switches, but (a) some revisions of the OEM unit have surface-mount (not through-hole) connectors, and (b) the microswitch is difficult to get off (even for me). Consequently, I will supply complete, assembled & tested, replacement Y stops, ready to go. Just fit and calibrate.
To calibrate the Y axis datum (not using spacers):
- Download this Gcode file (sliced for PLA/PLA+, 0.4mm nozzle, Tina2/2S only), and copy onto a microSD card for your printer (or prepare your own).
- Fit the Y stop, and lightly tighten one screw* (leaving the other loose) with the Y stop about mid way along its travel, temporarily.
- Make sure the magnetic mat is positioned on the build plate accurately.
- Run the print, preferably in a white or light-coloured filament so that it contrasts with the magnetic mat.
- The print should produce squares in each corner of the mat, inset from the edge symmetrically by the time the calibration is correct. Measure the inset from the front edge and compare that with the inset from the back edge. Back off the screw slightly to adjust the Y stop with resistance. If the inset from the front is greater than the inset from the back, move the Y stop towards the rear (or vice versa), and re-clamp. The distance to move it is half the difference of the insets (a digital calliper is a great help, if you have one).
- Clean the magnetic mat and repeat from step 3 until the margins are symmetrical within 0.5mm (the best that can be reasonably done). Do not destroy the final print (see "X Axis Calibration" below).
- Tighten the remaining screw.
Note that if the Y stop is too far to the rear, the travel limit will prevent it being actuated and the printer will shut down reporting an error. Unless you have made the anchor modification above, you will probably find the Y stop cannot be moved far enough back to achieve symmetry without the error. The easiest way around this is to settle for the best you can do, and displace the magnetic mat to compensate (without which, printing cannot use the full 120mm Y dimension without running off the build plate).
To be honest, this is all a bit fiddly, especially with the difficult access to the Y stop. I came up with a much less hit-and-miss alternative process, involving printing a special "spacer":
- Print a "zero offset" spacer;
- Fit the Y stop using the zero offset spacer;
- Run a calibration print, measure the offset to be dialled out;
- Print a spacer with that amount of offset;
- Fit the spacer.
With the Y axis datum calibrated, Cura can be told to use the full 105x120x100mm. There will not be a danger of the print going off the build platform (presuming the magnetic mat is appropriately positioned), because slicing accounts for the line thickness when positioning the nozzle, so that 120mm is the overall size of the print and not the actual travel of the nozzle (which will be 120mm minus one line width).
To achieve proper calibration (or a maximum-sized print), it is critical that the magnetic mat is positioned accurately – otherwise the calibration measurements will be invalid. To that end I found it helpful to fit the base plate with fences, which locate the magnetic mat in the same place every time. The Tina2S mat is accurately 110x120mm, but the rubber mat for the Tina2 is slightly oversize which makes this more awkward... but being oversize the calibration is not co critical anyway.
However, as noted above, Cura artificially limits the size of an object even within those figures and we have to settle for 104x119x100 – I am investigating a solution, but my current thoughts are this is an oversight (bug!) in Cura. Increasing the definition in Cura to (say) 106x121x100, to compensate, is not an option.
X Axis Datum Calibration
It is satisfying (but not essential) to calibrate the X axis datum so that the rectangle test also prints symmetrically side-to-side. This can be achieved by packing the mounting of the X stop with washers, on the screws between the PCB and the mounting pillars.
Notice I have turned the X stop upside down – not essential, but it just seems better!
The calibration kit (see HERE) includes suitable (0.5mm thick) nylon washers, and a zip tie in case you don't manage to re-use the existing one.
To calibrate the X axis datum:
- From the Y axis calibration print, measure the difference between the left and right insets. If the right inset is greater than the left inset, there is nothing more that can be done (not easily, anyway).
- Presuming the left inset is greater than the right inset, round the difference to the nearest mm.
- Remove the X stop PCB, and refit with that number of washers on each screw, inserted between the PCB and the mounting pillars.
The proof of the pudding is in the eating, and here is my final calibration print:
...and a 104x119x100 test print (12½ hour print run!):
