Know of any medium range drive systems for robot ?

Well, I'd say that $160 is still cheap. It is really a problem finding
powerfull motors that are accessible pricewise. If your application allows
for scavenged parts, you may want to create one or two search agents on
ebay. It is hard, but I found one 12V DC motor with gearbox and controller
(about 40rpm final) for $10 shipped. I was amazed that it worked when I got
it.

Cheers

Padu

You might want to look at the children's electric riding toys for
parts. You might find some of these at the goodwill, salvation
army thrift stores, yard sales and such. Cheap variable speed
battery powered drills from places like harbor freight might be
of interest if you need the wheels to be independently operated.

1) cheap
2) on time
3) meets specifications

pick any two

Know of an affordable controller that can power cordless drill motors ?
I have actually already taken apart a harbor freight drill (had one). I
am not really a electronics man (may be by the time my robot is done).
I have thought of using a servo to control the variable speed, but im
not sure if that would be exact enough.

It does come with HRFZ44N (transistor ?) on a heafty heatsink.
I looked it up and its a :
49A, 55V, 0.022 Ohm, N-Channel UltraFET Power MOSFET
Anyone know why its in there, and if/how i can use it ?

I would think you could actually use one of the small controllers
to drive the power MOSFET in the cordless drill.

Do you know how ? I dont know much about electronics. The mosfet has
three wires. Cant be too difficult ? Is it just a power amplifier ? Im
going to do some searches.

tsdave@gmail.com scribbled:


In this case it isn't really an amp.  It is a simple switch that allows
for higher power passthru than a generic transister.  a signal on the gate
allows current to pass from the collector to the emitter.

Transisters are really only amps at extremely low voltages.  Above a
certain voltage on the gate and they become solid state switches.

On 7 May 2006 20:05:23 -0700, "tsdave@gmail.com"


Assuming every thing is commonly grounded, it could be as simple
as connecting the output of the small controller to the base of
the MOSFET.

Here is my new plan, two 343 oz-in servos modified for continuous
rotation (these have 2 ball bearings). I will attach a sprocket to the
servo wheel, and the drive wheels, and use about a 2:1 teeth ratio to
double the torque. The wheels will have ball bearings and mounted on a
single fixed axle.

That should be 686 oz-in at 30 rpm. Course have to divide that 686 by 6
for six inch wheels ... thats 114 oz at 6 inches right?, thats almost
10 lb of force ? It should move like 10 inches / second travel max,
right ?

My robot will be 50-60 lbs.

If this sounds sane, does anyone know where i can get sprockets for
servos.

I have found the wheels at Northen Tools "2-Pc. Split Rim, Hub, 60T
Sprocket Wheel Assembly - 10.5in. O.D." with ball bearings. All i
have to do is mount a #35 sprocket to the servo wheel(thinking JB
Weld), and mount the servo to tension the chain.

I know the sensor array I want to use, plan on using two laser pointers
one pointing straight ahead (of the camera) and one on a servo. When i
want to measure a distance, I will start moving the mobile pointer
until the vision system only sees on red dot, then calculate the
distance. A^2+B^2=C^2. Think it will work ? All this and the camera
will be mounted to a pan and tilt mechanism.

I am still debating on the arm system to use. A lynxmotion arm would be
easy and pretty nimble, but very weak. A custom arm would be slow and
probaly only have one degree of freedom (plus bot freedom), probably
just up/down and pinch. But it could do 10 pounds (or more). Maybe Both
! But which first ?

I really want to get to the software side. I havent been much into
hardware in the past.

On 9 May 2006 16:08:12 -0700, "tsdave@gmail.com"


I suspect you will break the servos and be out the $$ as they
probably can't handle the shaft side load of driving the chain
sprocket if directly connected. If you want a cheap strong little
geared motor you can get a 6v B&D cordless screwdriver at walmart
for ~$7.

My desires for using a servo are i can use the ssc-32 servo controller
i already need. Servos can do forward, backwards, proportional movement
internally.

The servos i have in mind for the drive are hs-805BB dual ball bearing.
They pull 1.2A max.
They cost $40 each. I can buy metal gear sets for them if they do strip
out for $20.

If i use motors, i will have to get some kind of controller for them
with a RS232 interface. Then get a power controller. Then get the
motors, then get the gearboxes, and get it all to work. While some of
the motors come with gear boxes and even wheels, the cheapest Ive seen
are $160 and they are relatively high speed. (14ft/sec) I only need an
inch or two/sec.

I looked at the controller mentioned, and its 1/3 the price ive seen
for large ones but still $100 (with speed control extra ?).

To make things worse i dont understand electronics, and dont understand
how a h-bridge works or how to make one porportional, so i cant do it
mayself.

I think ill just try the servos, if they dont seem strong enough, ill
use them for the arm or accessory.

What I need to do it with motors is a motor controller that takes servo
input and porportionally controlls the motor. then i could use some
cordless drills.

On 9 May 2006 18:18:39 -0700, "tsdave@gmail.com"


The below link has inexpensive motor and servo controllers that
probably can be connected to the same serial port. You should be
able to have the motor controller drive the MOSFET transistor in
a variable speed drill.

http://www.pololu.com/products/pololu/#motocon  

Everyone thanks for the advice. I have ordered my parts. I am going to
try the servos first. But since my wheels will use chain/sprocket, i
can change my plan easy. I will probably move to cordless drills with a
controller. I was thinking maybe the scc32 can 'drive' a motor
controller too since it says its output is servo/TTL ? I may have to
learn more about electronics also. Worst case i end up with two extra
MEGA-SERVOS ;-) Sure i can make use of them ;-)

Status report:

frame constructed, wheels mounted.
pincher ...  pinching

I am about to mount my drive servos, i believe that they will work
excellent so long as i dont have to go up steep inclines or run 'over'
stuff. The large (11.5") wheels have ball bearings, and while on carpet
the slightest force to the sprocket on the wheel generates movement. I
have modified one servo to continuous rotation (805BB), it has a tab
you  can remove around the POT, and i cut the anti-rotation tab off.
the POT is adjusted for standstill at 1500, and the servo reassembeled.
The servo provides an amazing amount of torque, easily enough to move
the robot on hard or carpeted flat surfaces. It also turns
proportionally with the difference of the PWM. Being the wheels
themselves are half the weight so far, makes the robot pretty stable,
about like a table. :-)

At first i tried to hot glue the servo wheel to the metal sprocket, but
hot glue wont stick to the metal. So im gonna go back to the original
idea of JBWelding it on, better make sure its right, cause i doubt ill
get them apart. Ill post back later and tell how well he moves with it
all.

Just letting you all know, the servo works fine, it has enough torque
to run over my foot, although running over a foot makes a two wheeled
robot very tipsy ! I hope the servo lasts my experiments. If not i have
found places to buy new gears ~$2 or metal gears ~$20. They can go from
a snails pace to a pretty good speed. At the lower speed the robot will
stop when it run onto something (like a screwdriver) but if you up the
PWM it will run over stuff up to a couple inches although it really
jerks when it comes off !

I am working on a problem a lot like this one.  I decided to go with a
bigger drive system (350 W electric scooter motors @ ~$20 and HB 25 @
$50).

Here is a good site for finding parts:

https://sdp-si.com/eStore/

Any pictures?

tsdave@gmail.com wrote:

Although the servos work as well as i could have hoped so far, thanks
for the ideas for version 2, or maybe a retrofit. But knowing how much
torque and speed the servos can generate, i would be afraid of
seriously breaking something with larger motors.

I guess your using two HB-25s ?


texwes@earthlink.net wrote:

Take a look at our new RX50 h-bridge - it might be just what you are
looking for:

     http://www.bdmicro.com/rx50/

These h-bridges are designed to run in the 30 amp range.  What
voltages are you looking to use?  Note that by using higher voltages
you can handle the same amount of power at lower current.  This allows
you to use smaller wires, connecters, etc and handle the same amount
of power delivery.

-Brian
--
Brian Dean
ATmega128 based MAVRIC controllers
http://www.bdmicro.com/