I haven’t been here for a while, and missed this explosive conversation. I’ll try and sum up a few things that I’ve noted, as well as put in a few of my own views (instead of studying for exams)
- Holonomic or Not
In my opinion, it’s never ben necessary to have a holonomic drive, and with v5, it has become increasingly popular to create holonomic drives (especially in turning point), because the limited field space and resistance to defense, as well as integrated velocity PID on the motors made creating drives with mechanums much more viable, however, for TT, you really need consider whether or not you need the extra mobility from side to side - the same motion can quickly be achieved with a good driver on a tank drive setup, and would only require a little bit more space, watch 574C or 929U lining up for shots in TP.
- Power of the motors in pushing each other
Yes, v5 motors are more powerful than 393 motors, but there is still a limit to how much each motor can push - you won’t have as bad of a situation of let’s say turbo drives stalling against HS ones, but you will still lose a pushing battle against a stronger drive, but then I beg the question: how often are you going to push head to head, when there’s so much open space. If you think that you’ll be doing it a lot, go for a stronger drive, otherwise, you can use less motors or gear down. If you think that your driver has the ability to control the robot well, and use physics to their advantage (pushing from a side / on an angle), a 2m drive will be just as effective as a 4m drive, it goes down to which driver can get the upper hand in positioning.
More motors does mean that you can get more power down - this means that you’ll be able to accelerate quicker, reach top speed quicker (if at all), and keep travelling at that speed all the time. With a 2m drive, and a heavy robot, especially with quick direction changes and defense, I can quickly imagine that it’ll become difficult to control. Think about it this way, if you have a bowling ball rolling down a track, it’ll be a lot easier to stop it with your arm, rather than with your finger, because your finger can only do so much work. Same with the motors: both your finger and your arm can start moving the bowling ball, and both can stop it, but which one will take longer - do you need that time? or will you be able to make up for it by building a quicker scoring mechanism elsewhere.
Do also note that with the increase in power, you can gear up your drive - go past 200rpm to 250, 280, 330, or even 360 rpm on 4" or 3.25" omni wheels, which will allow you to move around the field quicker than the rest of the other robots. Obviously, on 2 motors, this would mean you take a while to get up to speed, and on 4 motors you would still be at a disadvantage compared to a 200rpm ‘base’ drive, but then you think about how many small motions you make and how many bigger motions you’ll make, and you’ll have to make a decision to see whether you want faster acceleration to top speed or a higher top speed - for this season, you’re going to want some decent acceleration to get around defense, but i can imagine that there’s going to be a fair amount of straight up driving between towers, scoring zones, and rushing to stray cubes.
- Other scoring mechanisms
Generally, the debate seems to focus a lot around how fast you can “cycle”, or how long it takes you to score x amounts of points. If you think you need more than 4 motors to create a super fast stacking thing, and then only 2 motors on your drive to get it there, that’s all the better for you, but generally, you can make things work with 4 motors. Even in turning point, where there were so many different objects to manipulate, teams managed to create wonderful ratcheted systems with even just 2 motors to do all their shooting, intaking, scraping, and cap lifting. with that in mind, given 4 motors, you should be able to complete at least 4 individual tasks, and looking at TT, you need to consider if you need more than those 4 tasks, and if you want more than 1 motor on each task - for example, you might want 1 motor to intake the cubes, 1 motor to stack them, 1 motor to out-take the stack, and 1 motor acting as an intake lift so that you can get the towers.
- How to build the drivetrain
This is something that has been covered extensively throughout this post, where people have been talking about the advantages and disadvantages of chains and gearing. 2381C and 2381X used chains and gears to transfer power throughout the entire season, and we can confidently say that chain will work just as well as gears, you can look at our reveal if you want to see what we did. Chain will allow more versatile mounting, and it can snap, but it rarely did. The only time 2381X had chain snap was when 2381C literally broke the chain (they snapped an actual link in half by hitting us from the side, catching the chain, and pulling us around the field via the chain). Chain will have a little bit more slop on the drive, but you can use tracking wheels to easily counteract this - think pilons (5225a) in itz. Gears are more direct, but will also have a certain amount of slop - in fact, even direct driving has some slop - it’s just how much you think is acceptable, given the number of field aligns you do, and the motions that you need to make. In driver control, the slop is basically not noticeable. I know that a well built chain or geared drive will have similar amounts of friction, so don’t worry too much about that. I think that, for you guys, if you want to explore, there is another option for the drive. Usually this type of drive isn’t done, because historically it has introduced a LOT of friction into the system, but as build quality generally gets better, it’s something that you can start to explore. This is a 3m drive, where 2m power the wheels, and 1m powers a shifter to switch between high torque and high speed settings, allowing you to have good pushing power under defense, or high speed on an open field. The actual gearing ratios might be a little bit weird, but generally 5:3 and 3:5 back makes sense, on let’s say a 200rpm drive. If you want to explore this option, be prepared for a lot of pushback from the community - it’s difficult to do well, and nobody has done it well in VRC in quite a while. This type of drive would probably put one wheel as direct driven by gears, and the other wheel chained to the first wheel, similar to what 2381X for TP. In the same vein of that idea, let’s say you only needed 3 motors for your scoring mechanism, then you could either create a 5 motor H drive, or a 5th motor as a transmission, to allow your robot to go between 3:5 and 5:3 on 200rpm, or perhaps 5:3 and 7:1 on 600rpm.