Intake Help

A lot of intake design choices are specific to the game element each year and generally need to be prototyped. Luckily this isn’t do bad to do.

  1. Retractable 100%. An extended intake can and will get smashed into the wall or another robot and will break.
  2. Polycarb aborbs impacts well and will resist impacts. Aluminum requires impractical amounts of weight to withstand the same levels of impacts.
  3. Generally you want the intake free speed to be over the free speed of the drivetrain. Anywhere between 1.2 and 2x the speed is worth considering
  4. Put the motor in a place where other robots can not mess with your wiring
  5. Depends on the setup, but belts are lighter while chain is less sensitive to imperfect c2c.

No objection to anything folks have said, just adding that you also seem like a team who’d do well to start with an Everybot.


I don’t disagree with your assertion necessarily; however, simply copying an Everybot each year doesn’t really help us learn good practices for starting on our own. We are trying to find good advice for starting points to make intakes each year so we can have an idea of where to start. I understand the “just start prototyping” mentality that people have here; however, with limited resources and machining capabilities we quickly find ourselves spinning our wheels when we “just get something built.”


This is where intake testing started with my team:

They set up a rig with a horizontal bar, spun it up with an impact driver, and just changed the wheels/roller type. A few things happened just after I snapped the photo… I had them back their rig up to a set of bumpers and moved them elsewhere to work on carpet. It is important that the testing happen on realistic surfaces.

Once we are back in our lab again, that intake shape will be assembled out of versaframe and polycarbonate and tested again. We like to power intakes with belts, with the motor placed as close to the intake pivot as possible. I think my team will want an intake that retracts into the robot, but we haven’t figured the geometry out just yet.

On wheels versus roller: I’d say wheels are great if you need to direct the game-piece somewhere (either to singulate or move into a frame/bumper opening) otherwise, rollers will do just fine. I think with retractable intakes, the best ones are usually pneumatic but I expect my team will use a motor because it’ll do the job and we don’t want the added weight/complexity. So we also make choices to minimize the weight of the system (roller, polycarb, belts, heavy stuff near pivot).

I’m also a fan of Everybot and would suggest you’re not doing your team a disservice by looking over build instructions and ideas from outside sources. My team spent lots of time examining climbers from 2013. And we’ll definitely look at the Everybot docs from 2022 (particularly intake) and 2020 (climber). We’ve built Everybot mechanisms both in-season and as off-season projects and it is a great way to improve mechanical skills.


Prototyping intakes is fairly easy. Grab a pair of 2x4’s, stick a hole in each on and run a shaft with wheels through them. Hook a drill up to the shaft and start playing! It’s best to do your testing against a frame with bumpers, so you can figure out any complications there - but once you do that, you’ll be able to figure out exactly where you want it. The rest is just figuring out how to get it out there!

Tons of choices here.

If you are completely lost, I would consider looking at what some of our open alliance teams are doing. 1323 for example, has a pretty nice CAD going with their concept. You may be able to mock up something similar and test it out: 1339 Open Alliance Build Thread - #22 by mrnoble

Bluntness Warning!

You request is conflicting: Give us some advice on how to develop our own original ideas.

It is the end of week 2 of the build season. Have the students, if they are really stumped, build a quick version of Everybot-style intake and ask them how it could be better… how it could meet your needs better… etc. That way they’ve made something that works, AND they get to be critical.

EDIT: my students are also struggling with geometry of intakes! Simple is sort of sloppy, less sloppy gets complicated quickly.


Going through the punch list –

  1. Depends on how squishy the game piece is and the squishiness of your roller, you’re probably safe this year starting testing with an inch or so of compression and adjusting from there
  2. When you need to take the game piece into your robot to do something with it, a lot of teams like to keep the full bumper wrap because this provides some structural benefits for the frame. This means an over-the-bumper intake. Obviously making it wide makes it easier to grab the game piece (to a point). The # of rollers required to get it over the bumper depends on game piece dimensions.
  3. Both work, it’s really the wheel’s/roller’s reaction against the game piece that needs to be tested. A lot of teams use squishy wheels these days because they’re easy to implement (stick a wheel on a piece of hex shaft / churro / thunderhex)
  4. Most games retractable, though there are some games where you may skip it. I’m preferential to pneumatics as the mechanism because it’s simple both mechanically and controls-wise.
  5. Polycarb is great for the most external part because (to an extent) you can run it into things without getting damaged. You can buy a little bit from McMaster-Carr to test it out, or even the local hardware store (check the windows section). However, sometimes some elements of rigidity for the motor, power transmission, keeping the bearings for the rollers/wheels in their holes, etc. Teams will reinforce their intakes with things like sheet metal and long standoffs to get them to behave how they want them to.
  6. A lot of teams try to run the intake with a faster surface speed than the robot can drive, that way you can pick up a game piece on the move. I’d start with 1.5X and adjust from there. With the motors available today, I would use one of these as a starter: 775Pro, Redline, NEO, Falcon 500. You’ll likely get the level of performance you need.
  7. Either is ok, generally on the intake is easier to implement mechanically, but you want to put it in a position where it won’t get damaged
  8. All of these are acceptable solutions depending on your intake geometry. I’m partial to chain and sprockets for most teams because you aren’t locked into a specific power transmission spacing and most teams need to fiddle with things like this on the final robot.

I would determine 1, 2, & 3 via prototype testing.
In most games the frame perimeter rules will likely drive you towards a retractable intake.
We used mostly 1/4" polycarb last year, but it did get broken pretty badly at one competition. So, this year, aluminum tube.
As others have said, you don’t need a particularly power motor, but you do want the surface speed of your intake > max forward speed of robot. Motors generally get attached wherever they can fit w/o interfering with something else.
We used chain and sprocket last year, it was kind of a hot mess, I think we will do belts and pulleys this year.

1 Like

I think what you’re hearing several of us here say is that the Everybot is literally itself an excellent prototyping platform. It (and other successful designs) is a fantastic source of best practices and rules of thumb. In fact, my students almost always find that they learn much more about good engineering and new design by starting from existing robots themselves, rather than starting from generalized rules of thumb I provide them like your OP requests.

It’s not at all that the rules of thumb in this thread are wrong; it’s just that the act of iterating from an actual baseline design is often far more useful and educational (and a lot more like my day job) than asking others for the rules of thumb we’ve developed by doing exactly that ourselves. It’s not at all “simply copying”. It’s a cornerstone of professional engineering. And it makes it much easier and more useful so you really can in fact “just start prototyping” next year.


This statement is totally correct. This is why the people assembling the products I and my coworkers design generally don’t understand how we arrived at the design.

The proper way to learn from building an Everybot is to study the design as you build it. Studying the documentation explaining the thought processes behind the design decisions leading to the final design will teach you and your team how to think about the problems in the future.

As others have suggested, take the Everybot as a starting point and experiment to learn what does and does not improve it. You stated the one intake system you and your team build was not effective. There are many possible solutions. Many of them are bad. Some of them are really bad. A few of them are good. When a solution is really bad, it is difficult to learn from it because there are often so many different factors that make it bad and discovering and fixing just a few of them will not make that solution good. Starting from a solution that is already good, one can experiment with one factor at a time and learn from that experience.

Many (most) engineers and programmers start their careers by being assigned to add a feature or fix a flaw in an existing product. That requires studying the existing design to understand how the new feature can be integrated and/or to find and fix the flaw. The existing design serves as an example for the new engineer or programmer to learn what does work. I don’t know of any company that would assign a new engineer or programmer to design something totally from scratch. The young engineers or programmers who insist they can take on such a task are exhibiting hubris and I have never seen one succeed.


do you wanna share that CAD :eyes:



It is a little buried in their build thread.

We prototyped an intake using some 2x4s and screws. When we did it we set our wheels at about 2" of compression on the ball. When the ball ran through, of course, the intake moved up. (Not really sure how much it moved, but my guess would be around 1.5-2"). We are planning to use pneumatics to raise and lower our lift, and are about to send our CAD to be CNCd, but I have some concerns about the amount of compression on the intake. If we use pneumatics I know it will allow some give, but probably not 2 inches worth. Should we move the mounting point of our intake higher (less compression) to account for the pneumatics or should we leave it as is with 2" of compression? What amount of compression are others running for the intake?

what type of wheel where you using compliant wheels are awesome on intakes for this very reason

We’re using 1.5” of compression on the front roller of the intake, and 1” for the upper rollers, with ThriftyBot compliant wheels. Our prototypes still had some problems with the ball pushing the intake upwards, so we designed our linkage to mitigate that by pulling the roller back as it moves up. If that doesn’t quite make sense, feel free to check out our CAD for more details, as well as our Open Alliance build thread.
Our Google Photos Album also has several videos of the intake in action.
Hope this helps!

We are using 2" stealth wheels

Looks like a good design. What thickness poly are you using for your intake arms? Also, would you recommend we go with less compression since we are using pneumatics and stealth wheels?

We’re using 1/4" polycarbonate for impact resistance. I’d say 1" to 1.25" of compression should be fine for stealth wheels, and I don’t think pneumatics make much of a difference in terms of compression.

How much have you run this intake? How many cargo through it? See this thread.

We have already run over 500 game pieces through our intakes and what we have seen is that there is severe degradation of the intake and handoffs over a few short cycles. The ThriftyBot Squish Wheels are great - until they aren’t with this game piece.

Also, if you are feeding the balls (not off of the official carpet), it will give you false data. Get the robot on the carpet and attack the cargo. Don’t start by chasing cargo - start with cargo in stasis. That will give you a good benchmark. Then chase the cargo and see how it operates.

It is amazing how different your intake will work in the first few cycles vs. cycles after running for 10-100 times. 4607 has been running ‘driver practice’ for over a week now and we have seen some very unhappy results with many different wheels, compression, diameters, spacing, etc. And depending on the wheels, there has been some siginificant degradation/damage to the cargo. I would expect that Inspectors will be on the prowl for teams that are damaging the game piece. We ‘think’ we have found some of the culprits, but we have not fully negated any one wheel type (but mecanums with too much compression seem to be culprits).

1 Like