|
prong
Experienced Roboteer
Joined: 19 Jun 2004
Posts: 839
|
hmmm i think servo switching is pretty underrated. While i am biased towards servo switching because that is what i use to save money, i do not think it has all the disavantages you mention.
that said proper speed control would be nice, but when i am building a couple of bots i need them to all together to cost less than the IBC, so servo switching is the best way for me.
Servo switching is pretty easy to make, a bit of aluminium, drill a few holes, screw everyhting on. The first one you make will be the hardest while you get all the postitions right, but after that you just copy the first one.
As for prone to mechanical failure, i have not experienced this at all, and i am using 20 year old servos and microswitches, all allready heavily used. I find there is not much to break, the switches might fuse is you try and run too much current through them, but an IBC will blow if you try and run too much current through it, they are both about setting it up correctly for your use. And if a spinner etc hits the actual servo switching, then sure it most likley will fail, but so will anythign else. The most important thing is having a good servo switching design, so nothing can get jammed or damaged from g forces and i have found there will be no issues.
hmmm variable speed control, this is servo switching's real disatvantage, as you can run a motor very slowly with a speed controller. To a small degree you can provide some speed control by rapidly turning your switch on and off, but this is not much of a subsitute. In the end i do not mind though, i have grown up driving non proportional tank control radio control cars, and it is all about the timing, you go full power and let it off at the right time and you can be very accurate, it just takes a lot more practice.
With mixing throttle and steering, after having a think about it i see no reason why you cannot set up your switches to do this, it is more complex than tank control though. That said i prefer tank control, after driving radio control cars with tank control as a kid is just feels like the natural way to drive, and i think it gives you greater control.
Failsafe, while you can buy those hobby ones or cheaply make your own i do not think it has to be that complex. I have a radio control plane that uses a servo to adjust flaps on the V tail. These flaps are not actually hinged as such, the ends just bend, and they are quite springy. At full extend the servo can move the flap all the way but it has to fight the springyness. If you turn off the controller or lose signal the spring in the tail has enough power to push the servo back to almost its starting position. If this idea was applied to servo switching you could easily have a spring etc that when the reciever loses signal then the servo is returned to the centre postition, automatically turning off the servo switching. This also has the bonus that it would return the servo even with reviever power failure, and because the servo has to work harder to move it will help damp slight interference. Unless i missed something here this gives a very simple and cheap failsafe that passes the failsafe test.
With braking i guess you mean electric braking? as in shorting of the motor wires when they are not being driven to give breaking and more accurate control? My servo switching setup atm does electronic breaking, how it is wired means that is the motor is not being powered it has its contacts joined. This is just done using standard 3 pin microswitches, two per servo. I have found that the electric breaking gives much greater control of the robot.
I do not understand what you mean by no weapon control? Why can't the servo switching turn a weapon on and off the same as it does to the drive motors? I use 4 channel servo switching on my robot, two of those channels are two way weapon control, allowing me to raise and lower the drill arm and run the drill forward or reverse. The only robot i can think of off the top of my head that would benefit from proportional weapon control would be pinscher, to accurately control the grip speed. I am sure there are others but it seems that most weapons are on or off type things, or driving something slow where you only want full speed anyway.
I think that the only real disatvantage of servo switiching is lack of speed control. Also in theory they are more vunerable being a mechanical system but i have yet to experience any problem. While speed control would be nice, i think the money spent on one speed controller would be better spend on making a few extra bots. Also in no way am i against the IBC, it is a very succesful speed control system, just for my use i find servo swithcing best, so i thought i would give my reasons why
|
Mon Oct 25, 2004 4:44 pm |
|
|
|
|
|
|
|
|
|
Spockie-Tech
Site Admin
Joined: 31 May 2004
Posts: 3160
Location: Melbourne, Australia
|
As Jake said, I dont mind how you do it.. As long as you can demonstrate that your robot will not do anything without an active radio link I'm happy..
Spring returns on the servo's, Micro controlled Relays, Commercial Hobby Failsafe units, PCM Radio's (with built in Failsafe), Electronize Controllers, Drill Triggers, whatever..
The test will be for you to activate your drive and weapon and with them active, turn off the radio Tx.. the drive and weapon should immediately shut down and stay shut down, if you can pass that test, then you are fail-safe..
Just to clarify a point, despite being the designer, I personally dont care whether people buy an IBC or not.. I get a small per-unit royalty from Jason (who does all the hard work of production and distribution etc), but its a tiny amount and even if sales of them tripled I still wouldnt be excited by what they earn me. As it says in the manual, we designed them because we wanted them and then a few other people asked for one so we started selling them.. its not part of my retirement plan believe me..
So dont worry about any ulterior motives forcing people into using IBC's.. In fact, I'd almost prefer they didnt since then I wouldnt have lots of people relying on a widget I designed to win a comp for them putting pressure on me.. Although its nice to think that I've helped others get started in Bots in some small way..
The Failsafe rule is there because thats the international standard for competing robots, as well as being a damn good idea. There have been some occasions that I have been very grateful that rule exists I can tell you.. _________________ Great minds discuss ideas. Average minds discuss events. Small minds discuss people
|
Mon Oct 25, 2004 6:59 pm |
|
|
prong
Experienced Roboteer
Joined: 19 Jun 2004
Posts: 839
|
I think the problem is people do not understand that the failsafe does not have to be expensive, or be any sort of electronic controller. The failsafe rule is a good rule, any robot fights with any sort of insurance should make having a failsafe a must, and in general it is a good idea for everyones safety.
Personally i have nothing against the IBC, it seems to be a very succesful unit, but for my sort of use it is not viable because of the cost, I undersand Brett that you do not make the IBC to try and make profit, i am a pennyless uni student so i tend to make my every dollar count, and for me servo switching with a cheap simple mechanical failsafe is what i will use.
The reason i keep mentioning a mechanical failsafe, as in spring loading the servo horn, is that until i posted about it the other day it did not seem to have been considered or talked about, and i think it is a very simple and cheap way to failsafe servo switching, better in fact than hobby failsafe units. It also means that building a legal to compete servo switched robot can be done very cheaply, increasing in my opinion the appeal of servo switching for anyone wanting to build in a tight budget.
|
Mon Oct 25, 2004 7:15 pm |
|
|
|
|
|
|