Robot wars project

Ok guys,
A group of us from school are building a robot to compete in the australian robot wars competition, we are intending to win by raming the competition with two 2500w brushless motors in the 13.5kg weight category!!!!!!! At the moment my questions are these.
1 which picaxe chip runs the fastest while still being able to use the servo command, and have atleast 4 adcs?
2 can anybody think of a clever way of measuring upto 100A at 20v on a picaxe?
3 does anybody know where in australia i can get some lightish (under 400g) dpdt relays that are rated for atleast 100A (this is only is bursts of a few seconds) to use to switch motor direction?
4 I intend to use lead acid cells to keep up with these current bursts, will a 2s2p arangement of 1.8AH cells be able to supply this better than 2p 3.6Ah?

Any comments such as "woah thats a lot of power" are also appreciated

Good luck

Pb technology does not normally use the XsYp terminology. It's only really applicable to LiPo technology which is outlawed for robot wars use for obvious reasons. Also 'p' is not recommended unless absolutely essential.
Size your cells according to whatever voltage you intend to use.

Do you have a link to the rules?
In the US & UK, no voltages above 36v are allowed so that pretty much determines what you can do battery wise.
UK battles last 5 mins. US battles are 3 mins. That makes a major difference when specifying battery capacity.

Don't forget, to be competitive, you must regard the batteries as consumable.
It is unlikely that you will get more than two heats from a set of batteries.

If you design the battery for longevity, (no more than 30% discharge and no more than C/10 current) then you'll be over weight.

The simple equation is to use 18s1p (to use your terminology) at whatever C rating you get within the weight limit.

5kW at 36v (nom.) requires a current of around 140A.
Andrew Cowen is currently building a robot with similar power requirements. I'm sure he'll be able to give some good advice


Measuring current.
Simple. Measure the volt-drop down the 0v line to your speed controller using an op-amp.
No need for drop-resistor and no need for expensive hall-effect sensors.

Fastest PICAXE with servo and 4 ADCs will be the 28X2 (3v) @ 64Mhz

1) PICAXE-28X2 - Though if you want fastest processing and servo control it may be better to have two 28X2's or a 28X2 plus 28X1. The processing 28X2 can run at 40MHz whereas with background servo control it would be limited to 8MHz. The extra cost should be mnimal in comparison to other costs of the project.

sorry wasnt very clear with the motor and battery specs, the motor is rated at 2500w but i intend to run 24v with 100A esc's (electronic speed controlls, then use the relays to swap the output wires around for reverse direction, which is why i want to monitor the current so theat i can drop the throttle it the current draw ever exedes 100A per side. when i said 2s with lead acids i meant two 12v packs is series, if running several in parrallel to keep up with a high discharge rate isnt recomended then so be it ill just use bigger cells. Also many guys in australia are running a123 lipos, but i rulled them out because of the price.

Here are the rules
http://www.robowars.org/info-rules.html

also any suggestions for high rpm counting

oh just thought of another thing, if i use two picaxes with one soley used for servo commands, how do i communicate between then without the serovo pulses stoping? the esc's crack the shits if they loose a signal enen for one or two cycles

The rules don't specifically mention LiPo but I doubt they'd be permitted as a punture WILL cause a fire.

Check rule 4.4.3.
Before you go any further, if you have a PICAXE between your RC Rx and any servo control it will deamed non-standard RC control and you WILL require approval prior to competing. PICAXE is not certified for use in any safety or mission critical circuit so prepared for a refusal.

As for inter-PICAXE communication.
You cannot use any timing critical commands such as serin/out at the same time as servo. (not sure about I2C).

You CAN use ReadADC and pulsin/out with servo. Alternately, you could generate your own frame rate and use other time critical commands.
I'd opt for generating my own frame rate using servo's which can cope with higher rates to give a crisper control.

Don't forget, design for competition is very different for other designs.
It only needs to last for 5 minutes so stretch everything to the limits.
Typical UK robots use 500W 12v Bosch motors driven at 1500W from a 36v supply.

I see you are allowed 48V. I'd suggest using 48v to keep currents lower and save a lot of weight in copper and power loss from I^R issues.

100A ESC's is no where near enough if you want to run those motors close what they could be run at.
Don't hold back. Enter with the intention of winning.

Also, understand that your robot could be a complete write-off within seconds. You will need spares of everything. Motors, ESCs, batteries.
A pick-axed (not the Rev-Ed type) ESC cannot be repaired.

2500W motors at 24V will draw close to 500A peak when starting. My two 750W motors draw a huge 150A each under acceleration. My robot uses some Albright DPDT motor reversing relays. They are big things designed for 800A max, although I think they are a bit heavy.

I agree with Bb that 100A ESCs won't be enough. You need more like 400A ESCs. Put a load on a robot (eg pushing another bot), and the current will shoot up. Put a load on motors, and they will draw many more watts than stated. They will also start to get hot!

I doubt you can run a 5000W robot in that weight class due to batteries. If running on 24V, you need batteries that can give 200A continuous, and maybe 800A peak.

Looking at lead acid batteries (12v each):
5AH batteries = max 50A
12AH batteries = max 120A
26AH batteries = max 360A (short time, eg engine starting) 9.2kg each
44AH batteries = max 488A (short time, eg engine starting) 13.6kg each

To sum up, you're not going to do this with lead acid batteries unles you have about 25kg of batteries. Even then, they are not going to be happy giving that power for more than a few seconds. It is a very bad to put lead acid batteries in parallel.

Your wires will have to be about 7mm diameter to not melt (heavy). The drivetrain will be another matter. My robot uses 1/2" chain drive - all the sprockets and chain weigh about 2kg. The rigidity of your robot would have to be amazing - it's a lot of power. I use keys and keyways on all the sprockets and wheels in my bot, but with 2500W, I think you would have a very tough time stoping keys shearing.

If geared for speed, expect huge starting current spikes, and lots of stress in the chassis. If geared for torque, good luck stopping it slipping, and try not to make the gearing to heavy!

For measuring current, have a look at Allegros range of hall effect sensors.

High rpm counting - I've done that using slotted optocouplers and striped discs. Will work with any speed depeding how many stripes are on the disks.

A

the picaxes will be working as speed controller- controllers, and there are a few robots in the competition using picaxe based speed controlls so i think they'll be fine, i'm choosing 24v becuse the esc's are only rated at 30v, the picaxes will be working to keep the motors at a reasonable temp and keep an eye on the current, ive had one of these motors in a gokart running 24v at 80A and i pulled me and the chunky pb batteries and frame up hills at 60kph before getting a bit warm and letting the smoke out (and its kinda hard to get the smoke back in), i think main issue is looking after the motors, there'l be plenty of power

andrew replied whie i was writing my last one, brussles motors require very low load to be able to start so i intend to use somsort of clutch in the drivetrain eliminating the start up spike issues, i think the success in the gocart ( using reatively low gauge wire and 100A esc even under the huge load of pulling me around) is a good indicator that the current peak wont be too much of an issue, is i can only pull 50A from low capacity pb's that is an issue however, any other suggestions for power source? i was hoping to spend no more than about $100 because i guessing ill go through a few sets
oh and gocart was running bike chain happily, i dont need to over engineer everything like a commercial product, if it beakes ill just fix it
just thought of yet another thing, at most full throutle will be used for 3-4 seconds, isnt this what pbs are good at?

Sorry, but that's the wrong approach in many aspects.

First off, just because others have done something doesn't mean it's OK.
It just means you all get disqualified.
(sorry Officer, I was only driving at 90mph because everyone else was).

It's a competition. Don't compromise on the ESC. Get the correct one

You don't want anything which might compromise a fast get-away such as a limited ESC due to temperature. It is the responsibility of the driver to keep temperatures down by resting whenever possible. (comes with experience).
A better plan would be to have temperture fed back to the driver.
If you over-heat and have restricted movement you're dead anyway. Better to know you got the most the motor could give before your oponent destroys it. (either way).

Anyway, this a PICAXE forum so I'll try and stick to PICAXE from now on.
There are separate Robot Wars forums for strategy and robot design.

EDIT: after your last post, my last word on the subject, promise.

if it beakes ill just fix it

Click to expand...

Have you ever seen robot wars?

Current up to 100A, maaate, check these out

School project, robot wars, bring on the updates and the photos.
If you want to measure amps upto 100A have a look at these.
Allegro make these, they are small but they work really well if you code them right and if you request a sample they will send you one or two or six.
I actually rang them from oz at about 2am, the lady was shocked at the time here but I was shocked when I got the samples within three days.

I'm not going to tell you the code yet, you have to do some work and testing, remember document everything, it will help you get marks. Do your research and check your "facts"

ACS750-50

Have fun and be nice to your teacher.

those current sensors look pretty cool. given me an idea for a lazy-gadget .
i'll look into it, patent it, then tell you about it

They work well and are easy to use.

But not particually easy to solder to thick wires...

A