I have just added the link to the video in the original post as well.
(might as well blonk it here too… 12 Years of 8059Blank. - YouTube )
I will use this post here to talk a bit about the mechanisms.
Regarding the trapdoor - i would say it is about the same concept (it is in 0:48min of the video).
Just that in the video, it was a prototype that we were testing on a 2-section tray. For the actual robot, there will be another about 1/2 section on top - which means the trapdoor will be about 1/3 down from the top of the tray (and not at the end).
This will save us some time without the need to move the cube all the way to the top before dropping off.
We had a descorer attached to the back of our trays in the Singapore Nationals (SingNat).
You can see how effective it is over in this short video that i pulled out…
And after our SingNat, we were exploring combining the descorer with the trapdoor.
Here’s the hastily and badly drawn image of it…

As for the cage, as mentioned earlier, we didn’t even have the opportunity to take a pic or video of it.
So again, relying on my badly drawn image…
It is not a “full” cage, i.e. it does not wrapped around the entire tray or cubes. It is just a frame that covers half of the cubes. So it will be light weight.
2 advantages to this cage:
- To provide a bit more protection against defensive plays… so that the cubes will not fall out so easily.
- To enable 1-second stacking speed.
Some of the things that we notice is that most of the time, it is actually the inertia of the cubes (that still keep moving forward) during stacking that caused the stack to topple. And that’s the reason why so many teams are using PID or some form of motion control to slow down the tilter when it is about to be vertical.
And we also tested and realised that as long as the top cubes are pressing down, the middle section cubes will mostly stay in place (and not topple).
So that’s why our cage is only “half wrapping” the top section cubes.
And with the cage holding the top cubes in place, we don’t even need to slow down the tilter when stacking.
So it is really a case of - (i) drive over to goal, (ii) tilt the stack vertical, (iii) opens up roller, (iv) entire stack drops vertically into goal, (v) opens up cage and (vi) back off.
And essentially, (i) to (v) are all micro-programmed into 1 press button.
As for what happened to SingNat… for the 2nd time in our 12 yrs history, we didn’t clinched the HS tournament champ. And for both times, 8076 beaten us to it.
Many factors came into play… we were a bit complacent to not implement all these ideas or mechanisms that we have during SingNat… of course many unforeseen incidents happened, etc.
but eventually, i must give credit to 8076 that they designed and built their robots to be a lot more consistent than ours. They are worthy winners.
In fact i feel kinda sad for them… finally winning worlds spots but worlds got cancelled.
And they can be assured that 8059 will be hitting back hard in the new season 