ROBO1 robots?

assuming

How's about 17 servos, with 10 in the legs, for just over $1645.

And a 600 mAh battery, guaranteed to run it for maybe .... 10 minutes?

Only $1645? Heck, for that price, where is the buytwo.html page?

-- Gordon

So, I'm assuming, when one thinks of "budget robotics", this isn't really the first thing that comes to mind :):). Actually, this is the big problem with building any servo-powered walker. Most of the $$ goes into buying the servos, even cheap ones.

dan wrote: [snip] : So, I'm assuming, when one thinks of "budget robotics", this isn't : really the first thing that comes to mind :):). Actually, this is the : big problem with building any servo-powered walker. Most of the $$ goes : into buying the servos, even cheap ones.

I figured as much. I've recently seen some video of a Japanese competition and thought it was the coolest thing I'd seen in a while. Time to start dropping quarters into the piggy bank to pay for one of these things - Then play with the programming, which sounds even cooler.

fun, DLC

The Robo1 uses digital servos for the torque, but in my (admittedly limited) testing I found it possible to increase torque on a standard servo by running it at a 200-300 pps frame rate. The servo got hot, but with some intelligent control I think it might be possible to dynamically change the frame rate for those times when torque demands are lower.

Robo1 is storing speed and acceleration settings in the servo, which is something digital servos can do, but you'd have to do this with a dedicated microcontroller (the $5 kind). One MCU could handle several servos.

You still need a ball bearing servo of decent quality...something along the lines of $25 at discount (rather than $50-80 for digital). That's $425 in servos. Figure maybe $100 for the controller, and $50 more for other electronics. If you have access to a metal brake and basic metal-working tools you'd be able to make the metal frame pieces.

In all you're still looking at about $600 to build your own from scratch, not to mention a lot of work.

-- Gordon

I have a milling machine to cut frame parts.

If someone wants to come up with dimensions for them, I'll cut them and see if we can get a biped put together.

Rich

I'll tested std servos with update periods of 15 and 10 msec, and they ran fine, and "seemed" to have a lot more torque, but I didn't try it long enuf to notice heating. Alternately, I've used 40-50 msec update periods, and the servos went really wimpy. I'm not 100% certain whether the different effects are due to more/less torque per se, or just to different responsiveness in the internal feedback loops.

With the longer update periods, you can turn the horns a long way before the torgue builds up. I actually thought this might be a software way of creating a "servo saver" - ie, a way to add in some compliance to leg joints, eg, during footfall, etc .... might not work too well for a biped, however ... tilt, tilting, over and down ....

Try 4 msec periods. That equates to 250 Hz. Some digital servos run at

300+ Hz.

If you don't mind the hack, you can always cut away some of the plastic housing of the servo and stick a small heat sink onto the side of the motor.

It has to do with the period of time power is applied to the motor. In an RC servo the motor inside is powered only when the servo receives a pulse. Between pulses it receives no power. This is why if you remove pulses from a servo (we're talking analog here, not digital) they simply stop.

Given a maximum 2 msec pulse every 20 msec means the motor is really only receiving current 1/10th of the time. Going with a higher frame rate may affect the PI loop in the servo, but different brands may behave uniquely. I only tried it with some GWS servos, and there was some overshoot (expected) in moving from one location to another. Given a more intelligent controller, you could build acceleration ramps to tackle most of this. This is what goes on inside a digital servo.

When the torque isn't needed by the motor you could revert to a 50 Hz frame. That would conceivably prolong motor life, and for sure extend battery life. Digital servos consume a lot of amps.

-- Gordon

Well, several of us have CNC routers and mills and such, and guys like Pololu, Lynxmotion, and Bruce Filener provide reasonably-priced laser cutting by the minute. What we need are those dimensions! The design is the trick.

-- Gordon

So does anyone else believe, perhaps wrongly, that when robots do finally crack the home market you'll be able to pick up the same $30 servos off of Ebay for 3 for a $1?

I can dream can't I? :)

BTW, the Robix USB UsBor controller is a really nice way to roll your own servo cluster type robot. It's about $200 for the controller alone. It comes with a Java client-server architecture that makes it easy decouple the motor control from the physical PC location and chain controller's together. In addition, it gives you "motor strain" values for each motor.

I'm just a customer. I assembled a 6 DOF arm using the UsBor and also did some minimal work controlling it with a P5 virtual reality glove.

If you're interested:

They sell these these by the gazillions - at least the analog type - for use in model cars + planes - so doubtful we'll see much price drop.

Now, this is really interesting. I've never dug into the servos, but given this info, I can see why they get wimpy with 40-50 msec update. I can easily set my servo controller to shorter update periods, so I'll play around some more and see what happens. I suspect this could be useful with a servo-powered mini-sumo when you need some xtra oomph during the tussle [don't tell Dennis, his sumos win too often already].

- dan ============

I think it would be possible to provide your own electronics and loop closure for standard RC servo mechanics. Use the servo's pot and motor and an external microcontroller with a multichannel A/D to read the pots and perform PWM Generation for many servos. A number of microcontrollers have multiple timers or better yet multichannel PWM generators. One amp H bridge IC's should provide adequate drive to the motors.

I am familiar with the Freescale dsp56800 family and I think that you could run at least 6 servo's on their controller with a single PWM generator (their PWM Generators output 6 channels each). The dual PWM controller, dual A/D chips like the DSP56807 could probably run 12 servos. I am currently building a 4 channel controller for a different system with the DSP56807 and I think that it the cpu horsepower, RAM and Flash space to do this.

Bob

Si,

The '807 eats servos for breakfast. Don't forget that the internal electronics include a bridge. A bit of copper, a simple loop, and you can even make a springy feeling servo...

Check out

not RoboOne, but similar and is a completely open design with all support documents.

Scott

Servo used in Robo1 shown on the soz site =3D KRS-784 ICS Digital Servo KON784 $89.99

17 servos*$89.99 =3D $1529.83. So, $1645 for the entire Robo1 bot including 2 servo controllers sounds like a real bargain. [Gordon can now buy 2].

8=2E7 kg/cm is only 120 oz-in, if I calc'ed right, so these aren't especially large servos, but at Speed: 60=B0 @ 0.17 sec, good speed for the size.

You will also notice the lynxmotion walkers are up in the $500-700 range. Most of the $$ there is also in the servos. My octopod with chintzy analog servos cost $200 for servos, and about $20 for everything else [not inc controller].

[snip]

Dan,

Most of the servos I've seen have been in the $30-$40 range. Is there something special about these $90 servos? If so, do you know what is it?

Thanks, Robert

Looking at a Tower Hobbies ad, all 6 Futaba digital servos shown are in the $59-99 range.

Hitec digital servos, $35-90 range.