Calculating required torque for servos for a WALKER!

**Daniel Vernall** · 21st April 2014, 01:32

I'm designing a walker... the 2 times weight allowance is very persuading. But anyway I'd like some help with calculating torque required to push my little friend forward.

Looking at the top end of featherweight I have just over a hefty 27kg (2x times weight allowance) weight to carry.

I know the formula to calculate torque to lift the robot and sustain it, but not the torque required to propel it forward. I'm actually going to be using a pneumatic system to lift it, and servos to propel the body forward once it's in the air.

The servos will drive horizontal bevel gears with a leg bolted/welded to each one providing the movement forward and backwards for each leg.

This is my torque formula to lift the robot:

Torque = (weight/legs on ground)*length of legs*acceleration

N = 28kg/3 = 9.34N
Nm = Force * radius(leg length)
Nm = 9.34N * 0.15m
Nm = 1.4Nm

Account for acceleration of gravity

Nm = 1.4Nm * 9.81 m/s/s
Torque Required = 13.734Nm

Is this also the minimum required torque for my propelling servos? It looks to me like I would not require as powerful servos because the downwards force of the robot is already accounted for by the lifting actuators.

Can someone shed some light please?

Thanks,

Dan

**Eventorizon** · 21st April 2014, 08:40

Its great to see someone having a go at a true walker, but I am just going to point out the first problem I have spotted.

Being able to lift and move your own 27.2kg of robot is one thing but most wheeled feathers can move themselves and push one, two and sometimes three other machines at the same time. If you wanted to actually be able to put up a fight then you would need to be able to move twice your own weight to hold your own in a 3 way match where two opposing machines could be trying to put you down the pit.

If course the maths isn't that simple but I would say you need to be planning to move 75Kg or there-about.

Ambitious design but I can see you either running out of gas, or having no legs left by the end of the fight. But that's just me

**Max** · 21st April 2014, 09:21

Hi,
To calculate the forwards torque I would go about it like this:

When pushing another robot the combined mass is : 27+13= 40kg
So when pushing the kinetetic energy of the system is: 0.5*40*1.5^2= 45J (Assuming 1.5m/s max speed when pushing as walkers are slow)
So assuming you want to get them up to max speed in half a second, power required is: 45/0.5 = 90W
Forward force required is: 90/ average speed when pushing = 90/ 0.75 = 120N forwards force
Sounds like you have 3 legs so 120/3 = 40N per leg so torque per leg = 40*0.15 = 6Nm

Having done all that I realise it is all completely worthless as it doesn't take into account any friction or the driving force of the other robot, I think I've calculated the minimum torque to push the two robots when on an ice rink so not really very helpful.

I will have another go at it later when I can right everything out properly and draw a diagram as my maths teacher always advises me to do, it's quite hard doing it in an iphone.

**Daniel Vernall** · 21st April 2014, 09:57

Ah cool thanks for the responses. So lets use 80kg to account for friction and counter force of opponent.
Therefore:
0.5*80*2.5^2 = 250j (using a slightly faster 2.5m/s)
250/0.5 = 124W
125/0.75 = 166.67N (btw where'd you get 0.75from? Half 1.5m/s)
166.67 / 3 = 55.56N
Torque = 55.56 * 0.15 = 8.34Nm per leg

Sound about right?

**Daniel Vernall** · 21st April 2014, 10:08

I think >8Nm torque servos are a bit pricey for 6. I think the pneumatics may be the way to go.

**Flag Captured** · 21st April 2014, 17:46

I don't know the specifics on the forces at play (maths scares me lol) but what I would mention is that you'll definitely want strong materials for those legs, or protection for them, given how destructive things can get - and make sure you've read up on the rules and actually get your 2 degrees of freedom in there to avoid it being classed as a shuffler, because you don't get the weight bonus for that. I'm sure you already know this, though! I would try to avoid the pneumatics too where possible unless you're confident you could have plenty enough for a full fight and then some because it sounds like there'd be quite a lot of CO2 usage at play...

Other than that, it's awesome to see a potential walker out there - certainly something different, hope it goes well!

**Daniel Vernall** · 21st April 2014, 18:47

Hi thanks for the reply. Indeed! I'm going to be fitting a thick square steel tube cage around the entire body (including the legs) Along with heavy springs in each leg to counter some of those forces.

I should get the 2 degrees of freedom:

Each leg can perform these motions:
1st degree: Lifting/dropping
2nd degree: forward/backwards

At least I think that counts as two degrees of freedom! Please someone correct me if I'm wrong!

**craig_colliass** · 21st April 2014, 21:30

Hi Dan

Naturally you may want to keep your design secret for now but I would cation you a walker that fits the 27.2Kg that can take the punishment of the arena, you will have to pull something awesome out of the roboteering hat! I don't want to stop you in fact I would encourage you to try!

Here are some questions you may want to ask yourself.

1/ How many standard (wheeled) featherweight robots have I made?
2/ How many fights have they had?
3/ How many years have I been running robots in the show environment?
4/ How much money and time do I have for the project?
5/ How much experience and contacts do I have in engineering and design?
6/ How many tools do I have or can use to make the parts for this project?

No one needs to know the answer to these questions, except you. you may not have much experience but be able to make the machine anyway, that's fine.

I don't want to put you off but I would not like you to start and not finish.

In the next few weeks and months and probably years, I will be making a walker too, I have 2 other walkers I use that are not fighting machines Pilgrim and Spiderbot. I am sure you can find them on Utube.

The reason there aren't hundreds of walkers on the robot scene is, they are hard!
the reason I want one is, they are a challenge I would like to do. But even if I manage it (and there is no guarantee) It probably won't win anything!

Craig