[DFTF] Budgeting for a Competitive Robot...

[Joe Johnson]

This is part of a series of posts called Drinking From The Firehose (http://www.chiefdelphi.com/forums/showthread.php?threadid=97430) on getting Dr Joe back up to speed on All Things FIRST.

Today's topic:
Budgeting for a Competitive Robot...

Okay, we have finally come to the 500 lbs guerrilla in the room.

How much does a competitive FRC robot cost?

This is not an academic question. I have currently signed up for a 2nd Regional (thanks NASA!) but I only have $1,500 left in the account we've labeled “build a robot."

Can I build a competitive robot for $1,500 plus the KOPs?

Assets:


I have some robot raw materials, probably not enough to build a robot but I guess ~80% of the way there (AL stock, some polycarb sheet, some wire, lots of zip ties, all the 10-24 screws & nuts you’d ever want, some #35 chain & master links, … ).
I have a small shop (mini-mill, mini lathe, drill press, bandsaw, drop saw, belt/disk sander, marble block & height gauge, …).
I have student labor (largely untrained at this point, but we will get there)
So… what I have to buy is

sprockets
bearings
wheels
motors
gearboxes
sensors/electronics
pneumatic solenoid valves
pneumatic cylinders
Jaguars/Victors
PWM Cables
Bumpers
2CAN2?
??? 100's of other ticky tack things that add up
Here is what I would like folks to help me figure out:

Give me a base number:


What does a competitive robot cost (all inclusive less KOP)?
Then break it down:

Price for raw materials (AL, polycarb, screws, pop rivets, tie wraps, hold downs, pool noodles, nylon cloth, …)
Price for the main drivetrain (assume use of 2 CIM + CIMpleBox + sensor from KOP to make a 6WD drop center chassis – add in special wheels, chain, sprockets, bearings, etc. that are not in the KOP)
Price per Motor-driven Actuator (MA) include motor, gearbox, chain/sprockets, Jaguar/Victor, PWM/CAN cables, fuse, sensor? Also how many MAs should I plan on? WRT gearbox costs, assume that I am going to have to use stuff from Banebots and/or AndyMark because, while I design robot gearboxes for a living, I am pretty limited on the kind of stuff my kids can make with our shop.
Price to “ante up” for pneumatics (assume KOP stuff pump, tanks, regs, switch,… include costs stuff needed to make the KOP stuff work).
Price per Pnuematic-driver Actuator (PA) over and above ante price above include solenoid valve, cylinder, end fittings, plumbing, sensors …? Also how many PAs should I plan on? If there is a way to cheaply gain cost advantage by ganging valves (for example using a special N valve manifold that works well for FIRST and costs less per channel)
Price for custom electronics (anything above and beyond the KOP stuff. I assume that the KOP angular rate sensor is acceptable, but what else do we need/want?)
Other robot costs not included above (e.g. extra batteries)
Finally, How many MAs and PAs total should I count on (assuming again 4 CIMs are used in the Chassis)?
So…

Don’t spare my feelings here. Tell it to me straight.



Do I need to cancel our plans to go to a second Regional? Maybe even THAT isn’t enough, if I need more, tell me now.



I’ll work to find the funds I need. But before I do that, I need to know what I need to raise...


Joe J.
Team Leader / Mentor
Team #3958
Schrodinger’s Cat
BC High, Boston MA

[EricLeifermann]

If assuming this years kit stays roughly the same as last years kit when it comes to the kit bot. 1114 kitbot on steroids (http://www.simbotics.org/media/videos/2011/presentations/kitbot-presentation-kitbot-steroids) is a great starting point and can get you a competitive bot. The extra parts needed for the kitbot will cost you ~$400 including shipping from Andymark and then that would leave you with ~$1100 to build the rest of the robot and if done properly $1100 can get you a competitive robot. You just have to remember to keep it simple and to build within your means.

[Peter Matteson]

With no old robots to rob for COTS I would say that you need $4000 dollars to build a single competitive robot.

$1500-2000
You need to buy transmissions and other power transmission hardware (gears, sprockets, chains) that can easily run this range with reasonable spares and failed concept fallout.

$200-300
Wire, nough said

$600-800
Motor controllers, you only get 4 in the kit of late and this is always the painful number to see.

$250
Fasteners: nuts, bolts, rivets, zipties, master links etc to stock up.

$200
Additional motors, you only get 2 of the CIMS in the kit and you need to buy more if we get to use 4-5 again this year. Banebots motors have had many issues and we never go into battle without spares anymore.

$200
Bearings: we use a lot of bearings and bushings. But we also build lots of spares and failed concepts.

$300
Bumper, fabric wood and pool noodles.

$1000
Shipping: to get the parts to you...

$500
Additional raw materials

$200
Solenoid valves and other pnuematic parts.

$200
Cutters for machine tools.

This is all best guess from memory. I know when you add all this up we easily spend $4000-5000 every year on the supplies to build a single robot. We also have a lot of spares, fallout from concepts that don't make the robot, and shipping costs to bring our robot in under the typical $3500 limit. I have to say with the cost of electronics consuming so much budget when we account our cost I would like to see it go to $4000. We are close every year to that limit, because we are honest to a fault with our accounting, and the electronics is a big factor.

[Clark Pappas]

$1500-2000
You need to buy transmissions and other power transmission hardware (gears, sprockets, chains) that can easily run this range with reasonable spares and failed concept fallout.


Unless I read the original post incorrectly, it appears he wants to run CIMple boxes, in which case he won't be spending this much money on the drive train, even if you buy spares for everything and buy an new set of wheels, say the 6" or 4" plaction.


As for battery costs, I'm going with a 6 battery total on my new team, so with 4 additional batteries it will costs $166 from AM, without shipping costs. However, you will also need additional battery chargers. I don't remember if the KOP includes chargers or not, so you may need to get 2 chargers in addition to the 4 you would get for your new batteries, and even if it came in previous kits, never count on something being in the kit.

[Jared Russell]

As Peter pointed out, shipping costs wind up being a huge part of a team's annual expenses. When you buy from AndyMark or McMaster-Carr, you often end up with capital that you can re-use from year to year (at least for training and prototyping). But when you spend $1000+ on shipping, that is money you will never see again.

There are a couple of things you can do to help mitigate this:

1. Find local suppliers! For bulky items like raw stock, nearby metal suppliers will save you a fortune on shipping in the long run (and if you sweet talk them, you can often work out deals to buy "at cost" or even have the materials donated to you). If you have a McMaster warehouse nearby, you are in luck!

2. Plan ahead. Ground shipping is much cheaper than next day air. The faster you go from "napkin sketch" to BOM, the more slack you will have in selecting less costly shipping methods.

3. Order in bulk. Many sellers (like OnlineMetals.com) offer free or discounted shipping for large orders.

[Chris is me]

Unless I read the original post incorrectly, it appears he wants to run CIMple boxes, in which case he won't be spending this much money on the drive train, even if you buy spares for everything and buy an new set of wheels, say the 6" or 4" plaction.

I don't intend this to be a call out, but there is a little danger in thinking this way. We designed a gearbox preseason last year on the assumption that the KoP gearbox was a Toughbox, and then they switched it out on us.

I would definitely not budget assuming a particular KoP gearbox.

[EricH]

Dr. Joe, there is one thing that you will be very interested in. This started last year, and was pretty successful: the FIRST Choice program. Effectively, some items that maybe there aren't enough of for 2500 teams are put into a "store", first come first served, with all items in the "store" free. This can cut down costs on certain items, depending exactly what's in there. It's hosted at AndyMark's website, or was last year, which makes it fairly easy to place an AM order and have it arrive at about the same time.

IIRC, last year's included such items as some hand tools, an FTC starter kit, old but unused batteries, wire, and speed controllers.

[Peter Matteson]

Unless I read the original post incorrectly, it appears he wants to run CIMple boxes, in which case he won't be spending this much money on the drive train, even if you buy spares for everything and buy an new set of wheels, say the 6" or 4" plaction.


As for battery costs, I'm going with a 6 battery total on my new team, so with 4 additional batteries it will costs $166 from AM, without shipping costs. However, you will also need additional battery chargers. I don't remember if the KOP includes chargers or not, so you may need to get 2 chargers in addition to the 4 you would get for your new batteries, and even if it came in previous kits, never count on something being in the kit.

I checked my numbers I actually had a 2nd robot in there for drive base.
My assumption was based on 6wd with spares for wheel breakage and no ability to make custom gearboxes for mechanisms:
2 AM Planeteries ~$200
8x 6" performance wheels ~$300
8 drive sprockets ~$100
4 Mechanism sprockets ~$100
10 Hubs ~$150
4 P60 planeteries ~$260
Chain ~$50

For a Total of about $1150.

[plnyyanks]

Shipping: to get the parts to you...

Back in our early days, we used to spend $1,000+ per season on rush shipping for things we need. Now, we've become a little wiser and aren't paying for priority shipping ever. We just try and order in advance and wait the extra day or two. Actually, though, it's not that bad. IIRC, we can get orders from Andymark in 3-4 days in McMaster orders the next day, all without paying for express shipping. It's saved us a ton of money in the life of our team.

And, I think it's probably doable to get one good robot out of $1,500, if you spend your money wisely and think it through.

[Clark Pappas]

I don't intend this to be a call out, but there is a little danger in thinking this way. We designed a gearbox preseason last year on the assumption that the KoP gearbox was a Toughbox, and then they switched it out on us.

I would definitely not budget assuming a particular KoP gearbox.

True, I have been careful to not assume with my new team.

But the point I was trying to make (However badly I communicated it), he plans to use what's in the KOP for drive train, which I don't think would cost that much money to buy additional materials and spares for, even if it is a shifting gearbox (which I doubt will be the case).

[Ninja_Bait]

I believe our BOM came out to around $2500 last year, but I will have to check with last year's captain.

Off the top of my head I can think of a few major price sinks:
7 victors ~ $630
4 mecanum wheels ~ $400
4 Toughbox Nanos ~ $320 (plus we changed the gear ratio and output shaft, don't know the cost of that)
1 extra Digital Sidecar ~ $80
4 BaneBots planetary gearboxes ~ $260

That totals $1700 already, and we poured money into raw materials, spares and dozens of other parts. And a second bot.

Of course, if you manage to not waste an ounce of metal and never bust a part, you could definitely get away with $1500. However, you might want to build prototypes or field elements, too. It's hard to factor in the cost of everything that might come up in the build season, so if you can, try and get a bit more cushion for that budget.

[the man]

Our robots have cost between $1000 - $1500. so I would say your good. I don't have a breakdown list but, they were competitive robots.

[Madison]

Unfortunately, I don't have a detailed break down of our costs available to me at the moment, but last season, we spent ~$7000 to build two "competitive" robots.

A lot of that is shipping; we don't meet often and decisions sometimes come late as a result. We're trying to be better about planning in advance.

I prioritize convenience over price, so we probably buy things that we could make and could save some money there, but I think spending a few hundred dollars more for a few days of extra practice is worth the expense.

We also burned A LOT of money on minibot development last season. Here's hoping that's an idea that never comes around again.

[AllenGregoryIV]

I think this could be a bit easier if people just posted their BOM. We all had to make one.

Here is the BOM from the Discobots 2587 from last year. I think this version is from before our FIRST Regional so it changed a bit by the end of the season but for the most part everything is there.

Discobots BOM 2011 (http://goo.gl/IYBVM)

This only includes the parts that made it to the robot so you probably need to add as much as 50% to this for shipping and parts that weren't used, broken and backups.

[Madison]

I think this could be a bit easier if people just posted their BOM. We all had to make one.

While true, we spent far more money that our BoM would indicate. The BoM shows that we met the rules, but it doesn't show the cost we incur in research, development and breaking stuff.

[Nemo]

Here is a shot at some of the costs you are asking about. Here are my assumptions:


Use kit transmission, whatever it ends up being
Use 12T (or whatever) sprockets from kit
Not using 22T (or whatever) wheel sprockets from kit; likely will need to purchase new ones that give the desired gear ratio
Upgrade kit wheels to plaction wheels
Use 4 Jaguars for controlling manipulators
Buy 4 Victors to control 4 CIMs in drive
Use Kit frame with dead axles
Use 1 Fisher-Price motor + gearbox from KoP
Use 1 Denso motor from KoP
Buy 2 RS-550 manipulator motors using Banebot planetary gearboxes
1 pneumatic manipulator: use Festo solenoid from KoP if provided, use Bimba free cylinder order if provided



$156: 6x 4" plaction wheel (6" FIRST wheels can be had for $7 each)
$60: 12x 3/8" ID 1.125" OD bearing
$96: 8x Sprocket
$50: 2x CIM motor
$360: 4x Victor controller
$28: 4x 40 Amp fuse for PD board
$15: 6' of 3/8" sch 80 pipe for axle spacers (is what AndyMark uses)
$129: 2x Banebot P60 64:1 gearbox
$15: 2x RS-550 motor
$24: 2x pool noodle set
$20: 3/4" x 24" x 48" Plywood for bumpers
$54: 4 yards of bumper fabric from Seattle Fabrics

That adds up to not quite $1000. Shipping for that would be something like $50-100 if you ship ground and combine orders. (ground shipping is fast with AndyMark) If you want a spare Jaguar, spare Banebot planetary gearbox, and a couple of spare motors, that is another $175-200.

I'm not brave enough to speculate on the cost of other assorted mechanical materials for actuators, but those costs seem to add up quickly for our team.

I think $1500 is just about as low as you can go for 4 CIM 6WD with any sort solid manipulator.

[theprgramerdude]

We rarely spend more than $1000 a year on the robot. The main reasons being, A. we can salvage a lot of extra electronics and drivetrain items off of past robots or old KoPs that we never used, and, a big one,
B. We get a ton of items donated. Usually, each year we get around $500 + worth of Polycarbonate and aluminum donated from a local manufacturing company. This is mainly what we used to then build the rest of chassis on top of the KoP chassis, as our coach preferred to save money wherever possible.

If you don't think starting off with the KoP chassis is the best idea (I didn't my senior year, but we didn't have the means to not use it), but do have the means (manufacturing technology) to not use it, then I'd certainly try to go out and get a bunch of raw material donated like we did, and use that to make your robot with, rather than purchasing much more expensive COTS items like the pre-made KoP chassis.

[Billfred]

I don't have the breakdown, but I recall 2815 spending around $1,000 on last year's robot. (Despite our success on the field, the robot was average at best; you being you, you would be able to take chicken soup and make it fine dining.)

If you've got a few materials already, you'll be okay. While more is always better, these days a team can have a good season from the kit, the Lowe's hardware aisle, and a couple things from AndyMark or McMaster-Carr.

[Joe Johnson]

I am surprised at how low the numbers are really.

If you think about it, every actuator (whether pneumatic or motor) is going to run close to $200 by the time add in all the associated costs.

I suppose that teams have 5-10 actuators all in (not counting wheels as actuators) that $1K-$2K right there.

How are folks doing full up robots for so little?

Joe J.

[Mk.32]

We were under $1000 this year as a rookie team.
Using just about everything we could from the KOP. We used the kit frame and our lift was just alum sliders and wood. With Brackets for the pneumatic ram (free from bimba) were made out of REM alum we got from the local metal shop, about $2.
It doesn't take much to make a mount for a cylinder or a bracket for a motor.

Though our robot wasn't really that great, it had a good run in off season.
The real cost comes in from prototyping and breaking stuff....

[Alan Anderson]

I suppose that teams have 5-10 actuators all in (not counting wheels as actuators) that $1K-$2K right there.

How are folks doing full up robots for so little?

Motors are in the Kit of Parts. Pneumatic cylinders have been donated by Bimba in recent years. I think the biggest cost to many teams is likely to be speed controllers.

[Mk.32]

And gearboxes...
Last year rookie kit you got 4 jaqs + 1 victor. So you only need to add another victor or two.

[IKE]

Budgeting for a Competitive Robot...


Can you define competitive? This can vary the costs considerably.

[theprgramerdude]

Motors are in the Kit of Parts. Pneumatic cylinders have been donated by Bimba in recent years. I think the biggest cost to many teams is likely to be speed controllers.

Plus, most of us have the ability to re-use last years components, so that usually means a few hundred-$1000 in savings per year.

[Kimmeh]

How are folks doing full up robots for so little?


Magic. And donations. But the largest reason is because that's their only option.

I can't give you specific numbers (because I don't remember them) but my team frequently builds a robot well under the perceived "average cost". While we may not have the most competitive robot, that's nothing that a few iterations more wouldn't cure, not more money.

[Billfred]

Motors are in the Kit of Parts. Pneumatic cylinders have been donated by Bimba in recent years. I think the biggest cost to many teams is likely to be speed controllers.

...which we always yank off an older robot we've mothballed because it lacked a control system.

[ayeckley]

If you think about it, every actuator (whether pneumatic or motor) is going to run close to $200 by the time add in all the associated costs.

From a pneumatics perspective, I think that estimate is a bit on the high side. I'm sure you *could* spend that much, but with a modest amount of pre-season planning that number can be brought down dramatically. As a re-rookie some of this won't apply to your situation, but within a year or two your team should be able to reduce your costs by using some of our tactics:

1. Always take Bimba up on the max donation each year, even if you have no plans to use all (or any) of the items. Obviously the strategy is to get the broadest possible range of items to ensure that a *nearly optimum* selection of cylinders is "in stock" for your team down the road.

2. eBay. In the offseason I periodically scrounge parts for bargain-basement prices, and we don't shy away from used items when appropriate. We've bought used cylinders, solenoid valves, etc. and new fittings, etc. for less than 50% of the retail price. Obviously the BOM accounting has to be at full new price, but there's nothing wrong with the robot actually costing less than it does on paper. The loosening of the pneumatics rules in the last couple of years makes this a viable option, and we are careful to purchase items with a long product lifecycle to ensure that we aren't designing-in obsolete products. In our case, we've got dozens of FRC-legal SMC solenoid valves (of all varieties) which will last us for many years and we spent less than $200 acquiring the whole selection.

3. Strategic use of First Choice. For example, IIRC one of the options last year was a pneumatic fitting assortment. There were enough fittings in there to last us for several years.

4. Inventory management. If someone takes charge of the unused KOP items, excess purchased parts, etc. in the offseason to prevent loss then this is a huge cost savings. It almost sounds like some of the posters on this thread are starting with zero parts left over from previous years. Where did all of the extra parts go?

5. Damage prevention. I explain to our students that this stuff is expensive. It's totally OK if it breaks during the heat of competition, but let's not waste our hardware (and our sponsors funding) unnecessarily by horsing around, etc.

[Chris is me]

It almost sounds like some of the posters on this thread are starting with zero parts left over from previous years. Where did all of the extra parts go?

Considering the OP is starting a rookie team, I think assuming that they have nothing is probably a good idea.

[ayeckley]

Considering the OP is starting a rookie team, I think assuming that they have nothing is probably a good idea.

I understand Joe's situation; I'm referring to the veterans that describe spending more to build their robot each year than is intuitive to me. Perhaps I'm expecting everyone to be as thrifty as we are (out of necessity).

[Joe Johnson]

Can you define competitive? This can vary the costs considerably.

This is pretty straightforward for me.

A competitive robot is one that is virtually always playing in the elims and, moreover, is drafted in the first round in more often than not*.


Caveats:
FIRST isn't about building robots. I know. But I also know that Inspiration comes in a lot of flavors and I, for one, believe it is a lot easier to inspire my kids when we set "playing after lunch on Saturday" as a worthy team goal.
There is more to making a competitive team than building a competitive robot, but there is a minimum level of robot performance that will meet the above standard.
Of course there are competitive robots that fall through the cracks. But I am talking about a statistical idea, a sort of thought experiment: imagine that the same regional were run 100 times in 100 parallel universes, a competitive robot would be drafted or drafting 95+% of the time and, if not a drafting team, drafted in the top 8 50+% of the time.Joe J.

*assuming the associated team is not full of jerks, the drivers are reasonably competent, ... It is easier than you'd think to get yourself crossed off a draft list. I advise my teams, "Be the team you'd want to partner with."

[Peter Matteson]

The rookie number the number I came up with was so high because I assumed the following: No spare motors, full legal amount of CIMS, no pnuematics lying around the shop, no motor controllers, wire, electrical connectors, fasteners, not sure of the raw material stock...

I understand Joe's situation; I'm referring to the veterans that describe spending more to build their robot each year than is intuitive to me. Perhaps I'm expecting everyone to be as thrifty as we are (out of necessity).

I get what your saying, but I see the opposite side of things. There are many crip crap items that will chew a budget quick if you don't have a stockpiled inventory of small items that people seem to forget you need to buy. These things aren't necessarily in the robot BOM because that rules don't require it. That's why I say a $3500 budget robot really cost about $4000 to build. As I stated earlier add shipping costs to get the parts and that bill goes higher.

Anyone who claims they can build a robot for under $2000 is either re-using a ot of COTS, motor controllers, and pnuematics from old robots or has drastically under estimated the value of items that they have scrounged as well as consumables that don't factor into the robots BOM cost.

When I start to add up costs or mill bits, grinding wheels, argon, tungston, weld wire, emory cloth, drill bits, carbide inserts, bandsaw blades, cutting oil, tap magic, taps, loctite, sander belts, etc. The costs become high very quickly. I feel like a lot people here have overlooked these types of items, granted not everyone uses all of these, but they are REAL COSTS in factoring the TRUE PRICE to build a robot. I'm not even touching upon buying capital equipment like tools and machines that you can use for years,

I'm just trying to add up all those little items that on many teams just show up in the shop because people go and buy them on their own. Speaking from experience myself and 3 of the other engineers on my team in some of our leaner years have spent upwards of $500 each on these types of items, prototyping materials, and field parts.

[Ninja_Bait]

This is pretty straightforward for me.

A competitive robot is one that is virtually always playing in the elims and, moreover, is drafted in the first round in more often than not*.

This sounds kind of weird to me; when a team is only going to one or two regionals a year, having their robot "always playing in the elims" is a silly kind of statement. That sounds more like a competitive team to me, one that has played enough years to justify the above trend.

I think it should be more along the lines of "usually wins their matches and is a likely pick for elims". That leads to the same result as what you said, but is a little bit clearer from a statistical point of view. It will be easier to gauge the trend off of 2 dozen matches than off of two alliance selection rounds.

[BrendanB]

This is pretty straightforward for me.

A competitive robot is one that is virtually always playing in the elims and, moreover, is drafted in the first round in more often than not*.


If this is your definition of a competitive robot, then more importantly budget your time. Having the time to debug your robot is very key. We ran our robot for about 2 hours before our first regional which was enough time to find out that: our supershifters stopped shifting due to an issue when we assembled them, our arm needed a balance of surgical tubing to lift the tubes, and some code malfunctions. On Thursday at our regional we never had a practice match in which everything worked and we found out more: pnuematics on the deployment were an issue, a piston came loose, more code issues, and another supershifter malfunction. Come friday morning our robot never broke down for the rest of the season including champs, we averaged around 30 points all weekend doing one low logo and a first place minibot, and were the first pick of the 5th seed in a very deep week one regional.

Our robot was extremely basic, we bought a kitbot on steroids from AM (kit chassis, 6wd 6in plactions, and supershifters *no longer sold by AM but it will cost a couple hundred more to buy the components themselves), some 1x1x1/8in aluminum tubing for the base, kit drawer slides and piston for the deployment, and some more tubing and a small piston for the claw driven by two window motors and tada you have our robot. https://s-hphotos-sjc1.fbcdn.net/hphotos-snc6/206214_178539498878492_178534675545641_463051_6986 653_n.jpg

I don't know what our BOM was however most of the cost was starting costs as a rookie with the metals, cims, controls system, wiring, etc. It is also really hard to budget what a competitive robot is. I can look at a other teams who did worse than us at our regional but had more on their robot in materials, electronics, etc. You can throw as much money as you want at a robot but that won't make it a winner.

Only you know what your team is capable of putting out (ours was about 9 students with a majority freshman). We stuck with what was simple and wound up falling back from the second row to the bottom and we were thrilled with the results. In short: keep it simple, get it done!

[Joe Johnson]

This sounds kind of weird to me; when a team is only going to one or two regionals a year, having their robot "always playing in the elims" is a silly kind of statement.

First of all, I didn't say "always" I said, "virtually always" and then I went on to explain in my caveats, the most appropriate one I quote below:

Of course there are competitive robots that fall through the cracks. But I am talking about a statistical idea, a sort of thought experiment: imagine that the same regional were run 100 times in 100 parallel universes, a competitive robot would be drafted or drafting 95+% of the time and, if not a drafting team, drafted in the top 8 50+% of the time.
This comes from the idea that a competitive robot should be competing to do something. What? Well, for me that means to win the competition. How do you do that? You have to be playing after lunch on Saturday. If you don't make it to the elims, your robot was likely not competitive (or you were unlucky, or your team had a reputation for being jerks, or ... which I cover in another caveat).



As to whether two drafts will tell me more about the competitiveness of at team (and their robot) than W-L-T record with 24 matches, of course it does!

Karthik (http://www.chiefdelphi.com/forums/member.php?u=1777) and kin, back me up here!

So... ...among the goals I have set for my rookie team is to build a competitive robot. Yes, there are larger goals and we are working on those as well. But I do think that there is a minimum capability robot that meets my definition.

Great drivers can do a lot, great planning can do a lot, great scouting can do a lot... ...but a box-o-rox is just not going to get make it to the Elims better than 95% of the times they compete (again, in a thought experiment sense not literally).

Bottom line: I am trying to gauge what I should budget for this activity.

Joe J.

[Ninja_Bait]

I understand what you're saying, but I still think that success in a match environment is more indicative of robot quality than success in the competition structure. Let's agree to disagree before this turns into an nasty semantics-slinging fight.

What BrendanB said was very wise - inventiveness can trump money any day. In 2009, the very competitive "number 7"/"roller-dumper" bots were probably possible with a $500 budget (the KOP, 2x4s, PVC, lexan and belting was basically all it took). In 2011, a similarly cheap but competitive robot was probably possible, though I can't think off of the top of my head what you'd have to do to get it - the game is also a factor in what kind of money you'll need.

$1500 is an admirable and probably achievable goal; if only every team could pull it off, FIRST would be all the richer. But this question is ultimately unanswerable. Experience can tell you so much, but any curveball will totally flip experience upside down. I seriously doubt that $1500 is enough to cushion the team from huge mistakes or massive changes in the game, and this is coming from a team that has never spent less than its $8000 engineering budget (of which the robot was probably only half our expenditures). If you can get the extra money, do it - if you don't use it you lose nothing, but if you don't have when you need it, you'll lose 6 weeks of hard work.

[Jon Stratis]

Pulled straight from our budget sheet:
2011 - $1900 spent "for the robot" (includes spare parts and some new tooling, but not a lot).

Is it possible to build a robot for less? Yes, absolutely. If you have a firm budget, however, I suggest you start with some significant planning the first week or two of build season.

Figure out what you want to build, then figure out how much it will cost to build that. Start taking out capabilities until you're within your budget.

Alternatively, come up with an "incremental" design - you can add on assemblies/capabilities in stages, and stop when the money runs out. We did this with Overdrive to alleviate worry that we wouldn't have time to do everything we wanted. First, focused on getting something that could drive around the track and score points that way (only two balls per alliance = 1 team won't be interacting much with a ball). Next, build our mechanism for handling a ball - essentially sucking it onto the top of the robot, and spitting it out. If needed, that can be bolted right onto the drive base and we're done. Finally, build the elevator that's going to let us hurdle the ball. This way we gradually increased capabilities without requiring that everything is finished to have a functioning robot.

[Bob Steele]

I don't think that win-loss record is a very good indicator of competitiveness.
It certainly says something about a robot.
I do think that being picked for eliminations is a much better indicator.

My reasoning here is that matches are completely random and a team can (and often does) win matches not based on their own ability.

Teams are picked (and note that I say picked and not seeded) because other teams think they are competitive.

The issue here is that what about the seeded teams? Most would be considered competitive.

I also think that a robot needs to be reliable...

We could consider these three factors to be Robot Seed + Robot "attractiveness as a partner" + Robot Reliability

I would propose that a value for competitiveness (for a robot) be a synthesis of something like this:

[(some constant) X Number times seeded/ number of regionals] + [(some constant) X number times picked for eliminations / number of regionals] + [(some constant) X number of matches that the robot is functional on the field / number of matches scheduled for robot]

I would have to do some numerical evaluation with teams that we consider to be competitive to come up with values for the constants. I would think that my initial reaction is that the weight (or constant) for picking should be higher than the other two.

Just my ideas.
A competitive robot does not mean a great team... and we are talking robots here.

[IndySam]

We get a $3000 budget from our school for robot parts and tools. We always spend every penny + This year we have already spent a big chunk on new cRio chassis.

But we also build at least an extra drivetrain and enough extra stuff to have a effective practice robot (we use every bit of our withholding allowance we can to help reduce coats.)

[IKE]

...
Can I build a competitive robot for $1,500 plus the KOPs?
..

Yes, (assuming you have the assets you listed), but it will be difficult. If at all possible, try to double that. Doubling it will allow for the possibility of 2 different successful strategies.
1. Build 2 simple yet relatively effective robots and use the one for practice, development, and more practice.

2. Build a more complicated robot that uses a lot of good COTS parts (that aren't cheap) and thus free up development time on game piece manipulators, collectors, or whatever other custom stuff is needed. This will also allow you to "waste" some funds on things you don't "need". "Waste" would be defined as ordering the $50-250 dollars worth of stuff that you think you need, but by the end of build find you are not using. This buffer reduces the budget stresses that can drive a team lead crazy.

********************************
This season I had the pleasure of watching the OCCRA (Oakland county Competitive Robotics Association) Championship. OCCRA has pretty strict student build and tool usage limitations that require teams to keep it relatively simple to construct. Robots are about 2ft x 3ft x 3ft, so just a bit smaller than FIRST. Of the 20 teams, there are only a couple that I would estiamte would have broken $1500. Because the teams get to keep working on their robot, the "path" they follow tends to be more like path #1 from above. The game was a lot like FRC2005 (Triple Play), but used balls instead of Tetras. By the end of the season. Almost every team could score, and over 1/2 of teh field could score 4+ game pieces. In a pick and place game, having this capability will put you in the top 25% or better which should make you a captain or first round pick at most regionals. The OCCRA teams get there by having more development time and more practice driving than FRC teams get. Similar behaviours occur in VEX where students have more time to test/tune/tweak. The practice bot, or doing multiple events does this in FRC.

************************************************** *****
If you want to make it to Einstein, then I would listen to Peter from 177. They seem to have the approach down.

[Brandon Holley]

Joe-

I would post our numbers, but they would more or less be duplicates of Pete's.

As I'm sure you remember from a few years ago, there are always little do-dads that add up over the course of a season. Tools, spare motors, etc. it tends to add up.

As far as your goals for robot/team performance- ours are more or less the same. My experience is telling me that $1500 may be able to get you to a competitive robot, but if you want to build a "rock star" robot, I think you will need to be closer to double that number at a minimum.

Bottom line: I think you should give yourself a larger budget to get the robot you are looking for.

-Brando