Please refer to our team’s proposed BOM (Bill of Materials) for different weight limits: RCJ Lightweight BOM - Google Sheets Reasoning for increasing the LWL weight limit:
- Forced component choices compared to Open League
While Open League robots are restricted by space, LWL robots are restricted by weight. This is a much bigger restriction than space is. Small and minified components are quite easy to find, and quite common, and many things can be made smaller with clever design. Weight can’t really be reduced by designing things better, and this limits our component selection greatly if we want to have moderate performance.
- Example 1: Light, compact, and powerful motors are very hard to come by. We are restricted to very light motors that weigh less than 100g, and the motors that do meet the power demands are only the Joinmax BE-3561 (which is no longer produced or sold), and the Maxon EC-45 flats (which are prohibitively expensive, even with a sponsorship)
Example 2: Solenoid choice for electromagnetic kickers are very restricted. Most kickers are limited by coil inductance, and thus having a bigger solenoid with thicker wire and lower coil density is great for having higher kicking power. This comes with a lot more weight though, with the larger solenoids that many Open League teams use being over 100g. The kicker power circuits and (mainly) the capacitors needed to supply the solenoid also add a fair amount.Example 3: Dribbler motors are quite restricted to only the tiny brushless motors (e.g. A2208), which have quite poor power compared to the larger brushless/brushed motors that most Open League teams use instead. It also draws a lot more current for the same torque and causes a lot of heating. (continuing in next message due to discord character limit)
Google Docs
RCJ Lightweight BOM
[10:25 AM] Froggy111*:*
- Un-negotiable tradeoff between performance, weight and cost
Many Open League robot parts can be made cheaper through DIY and cheaper component selection. This applies to e.g. ESCs, sourcing parts from chinese manufacturers (Aliexpress, Taobao). This is not nearly as much of an option with LWL. Heavier components also tend to be more powerful as previously said, even in the same or smaller form factor. In fact, in most LWL bots, a lot of space goes unused (at least compared to the density of Open League bots). LWL needs a lot of cutting down on metal components, and even decreasing the weight of 3D-printed parts, using less perimeters and infill and making them thinner and overall significantly weaker. The systems that can be put on a LWL bot are also much more limited, with Open League teams usually having more sensors as they have the design freedom to do so.
- Example 1: Usage of machined / laser cut carbon fiber is necessary if wanting to squeeze in more performance. This is a reason that has been previously discussed as well, during the previous weight limit increase. While less of a need now, we still find ourselves needing to discuss this option when deciding on the overall design. This is compared to Open League teams being able to use significantly cheaper laser cut aluminium or stainless steel, or even aluminium PCBs for structural support, which are options orders of magnitude cheaper.
Example 2: Machining of mirrors. Last year, we independently created a mirror geometry solver for undistorted 2D views of the soccer field, and the precision needed demanded machined mirrors instead of heat-moulded mirrors from PVC mirror-coated sheet. While our Open League team could machine them for <10USD each, our LWL team’s mirrors cost >30USD each.Example 3: Fasteners. If all fasteners used are SUS304, that can add up considerably, and it was over 100g in our case. We needed to swap some for aluminium fasteners which strip much more easily to fit in the limit.
[10:27 AM] Froggy111*:*
- It is no longer a defining feature of LWL
- As previously said many times, LWL is much less about the weight limit now, and is more about being an “IR League”, where the IR emitting ball allows teams to more easily get going with their robots. The weight limit seems an unnecessary restriction in this case, adding much more design restriction than Open League’s lower size limit, especially for more experienced and competitive teams. I believe the organising commitee also wants to see more competition and better play in LWL, and that involves upgrading components and adding even more, which are all restricted heavily by the weight limit.
[10:34 AM] Froggy111*:*
One of the main things we want to do is regarding the motors. We have designed (and already made some! this is the first revision though, and we have many planned changes) of a custom FOC driver for 4015 motors that can be found on taobao. This allows us to fit the motor AND wheel, which has been calculated to guarentee providing more than enough torque for even a robot weighing 5kg with very high friction (this motor can easily do 0.3Nm at 24V, and with a coil resistance of ~3 Ohms, should be able to handle 2A RMS without much issue given the duration of a match), including all its control electronics, within a 45mm diameter * 35mm thickness package, giving much more space for other components. This is very useful for Open League, where many teams use Maxon EC-45 flats, and we can achieve similar compactness but at a much lower cost of only around 25-30USD per motor + driver, and 120-140USD for the entire 4 wheel drive-base. That is similar to how much a single EC-45 flat (50W, 42mm, integrated electronics) costs WITH a large discount, and almost half of how much a single EC-45 flat costs without any discount.
[10:44 AM] Froggy111*:*
Another idea we want to attempt is the use of Lithium-Ion cells instead of the traditional Lithium-Polymer batteries. Lithium-Polymer batteries are great, but their form factor isnt, usually being a “big component” that decides the placement of many other things. Instead, we want to try using high discharge 18650 Li-Ion cells, which can handle 12C/30A continous discharge with decent thermal management. These cells can be plugged into a custom PCB at the bottom of the robot, and the cells can be compactly stored using vertical space that is otherwise wasted, with very convenient replacement by simply flipping the robot over, removing the bottom cover and swapping out the cells. An Open League robot can fit 6 cells using only space in the sides and the back of the bottom side, with more if stacked, and a LWL robot can fit 9, only using space that will not otherwise be utilised fully. Fully charged with 9 cells, a LWL robot can sustain over 1.1kW continous power, at 37.8V, which is a bus voltage rarely seen for soccer robots, and usually only in the kicker. It also will have an energy capacity of almost 300kJ fully charged. Doing so also gives us the design freedom to implement safety measures by measuring every cell’s voltage seperately, monitor temperature and perform current limiting to the rest of the system. These cells are heavy though, at around 50 grams each, 9 cells would be 450 grams. Adding the motor+wheel’s around 170 grams each, thats already 1.1kg, making it impossible to actually make a robot.
[10:46 AM] Froggy111*:*
While many teams seem to stay under 1.4kg, including Open League teams, I believe this is due to following the status-quo on component choice, and we should not limit innovation because stagnant design choices do not challenge the limits of the rules. I believe an increase in weight limit will bring stronger competition and better robots to LWL. A lot more things are noted in our planned BOMs for different weight limits. The approximate cost tallys can also be seen and compared. In summary, a 1.4kg robot might cost ~1500USD, while 2.2kg and unlimited weight limit robots might cost ~500USD, a third of that price, with much more accessible and replaceable components. We propose to at least increase the weight limit to 2.2kg, which is quite reasonable, considering Open League teams mostly stay within that, to let teams innovate and save costs, or potentially make it unlimited. We hope the community and the comittee can support this change and make LWL more competitive and exciting.
[10:46 AM] Froggy111*:*
RCJ Lightweight BOM - Google Sheets
Google Docs
RCJ Lightweight BOM
[11:41 AM] jakubgal*:*
I agree, lets increase it. But I would be even for calnceling it. Less rules better.
[1:58 PM] JX*:*
i agree! The weight limit makes innovation much harder and contributes far more to increasing robot cost than to new interesting designs. I think currently LWL serves as a lower entry into soccer (historically it has been, given the previous rules limiting LWL participation to only two years. Although whether or not LWL is easier is debatable, I think it often treated as such) The larger space and presence of IR on the ball simplifies the construction of a working robot as one does not have to work around space constraints and computer vision, but the weight limit forces many design decisions that make designing the robot much more difficult Having to design around a tight weight limit doesnt really seem to inspire innovation in making more lightweight bots (this is often an inherent limitation of materials, where the more expensive ones i.e. alu 7071, titanium - are stronger and more lightweight, or components such as motors, where expensive motors are often more power dense i.e. maxons), and instead forces sacrifices of innovation in LWL (ive seen a few LWL teams that mention wanting to integrate a kicker or dribbler - including some teams that already designed and manufactured it, but being unable to do so because they will exceed the weight limit Overall, I think the costs (both to innovation and actual component costs) of the weight limit are much more than the benefits, and at the moment I think this cripples new designs in LWL in such a way that its actually far more difficult to make a good LWL bot than an open bot (especially when combined with other nerfs such as the inability to use LiDARs). Even if the goal is for LWL to be different but not easier, I think the current rules lead LWL to be far more restrictive in design choices and makes it difficult to do something new and innovative
