|
Spockie-Tech
Site Admin
Joined: 31 May 2004
Posts: 3160
Location: Melbourne, Australia
|
The Medium Motor Controller is mostly the same as an O.S.M.C. controller (
http://www.robot-power.com/products/osmc_info.html
with heatsinks and a switchmode regulator added.
It should handle around 200 amps fairly reliably, and several prototypes have been built and tested,, but we havent gone into production with it, since the arena's for heavyweights and larger dont exist yet, so we (or anyone else) havent needed them..
Its not going to be cheap though, since it has 4x the amount of fets, heatsinks etc of the IBC, and its a single motor controller, so you need one per drive channel (side) of your bot. At a very rough guess, I'd expect them to be around $700-$750 for a set of two, with the uMOB needed to drive them
(the uMOB is built into the IBC, its the micro brain bit that converts the radio signals to PWM drive for the H-Bridge Fets) _________________ Great minds discuss ideas. Average minds discuss events. Small minds discuss people
|
Mon Aug 16, 2004 11:06 pm |
|
|
|
|
|
|
|
Spockie-Tech
Site Admin
Joined: 31 May 2004
Posts: 3160
Location: Melbourne, Australia
|
Hi-Power Motor controllers are a bit like Race-Car engines. the more power you want them to deal with, the more everything has to be spot-on or they will break..
Driving a full H-Bridge of Fets (so they can reverse) at high currents and (PWM) Pulse Width Modulating them on and off hundreds to thousands of times a second without getting "Shoot Through" (where the upper and lower Fets on the same side of the bridge are on at the same time) is quite an art.
It seems like it should be simple enough at first glance, the Fets should only do what the micro tells them to shouldnt they ? So if there's no shoot through in the micro-code, then the fets should be fine ? Uh-Uh.
They behave capacitively and take time to start and stop their huge surges of power, so require either a lot of theoretical work to figure out the current flows in and out of the gate capacitance to charge and discharge their switching, or a fair bit of "practical experience" (read blown up Fets) tuning the drive circuit until you hit the sweet spot.
Do a google search for the "miller effect" as an example of one of the unsuspected areas of complexity that can bite you.
If you want to save yourself a lot of trouble and benefit from the experience of others rather than remaking the same mistakes, read through the OSMC group (on Yahoo groups) archives and you will get a running start on making your own controller.
Other than the Fet-Drive bit (which is the important one), A controller is fairly simple in basic concept.. Time the width of the servo-drive pulse (1ms - 2ms long) with your micro, scale it and turn it into a bi-directional variable Pulse Width Output to drive the output stage.
Adding features like mixing and failsafe takes a bit of extra work, but they can be added after you get it moving.
Remember though, if your primary goal is to battle reliably, bite the bullet and buy a controller, I guarantee it will save you a lot of time, money and frustration in the not-too short term while you debug your own design (and get your bot beaten up in the process). If you are interested in learning "just cause you want to know", then read up on the OSMC group and ask myself or them any questions you would like an opinion on..
Which PIC should you use ? If you are just building one controller, get the beefiest, fastest one you can. preferably with multiple hardware PWM outputs.
The more hardware grunt you have (for a few $ extra), the less time you will have to spend getting the code to do things which will minimise your development time. If you are building for a high-volume production environment, then you always use the micro that has *just* enough to do the job (With a safety factor). just like most engineering.
I cant give you specific model #'s though, since my PIC experience is limited to 16f84's and the Pic-Axe range, and they bring out new ones all the time. Jake/Valen might be able to. _________________ Great minds discuss ideas. Average minds discuss events. Small minds discuss people
|
Sun Sep 05, 2004 12:39 am |
|
|
|
|
|
|
|
|
|
|
|