Re: Motors comparison page updated

> It's actually not a motor - it's a small electric hot water heater element. When
Good thing, the load is not inductive.

> plugged in it produces 1300W, but if I want less than that I need to figure out
> some way to reduce the current, or the voltage... or, alternatively, cycle it on
> and off very rapidly. 1300W on for 1 sec and then off for one sec should
> simulate a power output of 650W, for example. But if the cycle time starts
> drifting up above 1 sec, it will induce thermal oscillations in the system I
> don't want... so switching something like this off/on rapidly seems to be a good
> option.
OK, with a 1 second cycle time, and pure on/off, it is simpler. All you need is
a solid state relay.
>
>> ...to keep it simple and safe - you might try to control the
>> dial of a powerful light gradator with a NXT motor?
>
> That was actually my first thought (in fact using a "dimmer switch" manually
> would be good enough... I really just need a way to set the power output, no
> absolute need for NXT control). The problem is nobody needs to "dim" 1300W light
> sources,
Looks like some people do:
http://www.amazon.com/Leviton-61500-I-Single-Pole-Electro-Mechanical-Incandescent/dp/B003AUDF0Y/ref=sr_1_1?ie=UTF8&s=hi&qid=1294925232&sr=1-1

> and worse, most of these dimmers work by dissipating and dropping the
> power in the dimmer, not the light... meaning in this case you don't save power,
> and the dimmer gets... really warm.
>
Fortunately no, they are not dissipating all the power not going in the load! (a
triac controls load as on/off at mains frequency for a fraction of cycle) But of
course there are somme losses - 2 to 3 V drop at 15A gets a small box quite warm
indeed!

> What I've got now is a wall switch in a small wooden box, with a LEGO framework
> pinned to it with an NXT motor. The motor is used to simply throw the switch
> back and forth at the desired rate. This should work... but it's really noisy,
> and limits the cycle time to around a second (hard to throw the switch faster
> than that under NXT control). I saw your control circuitry, and starting
> thinking of ways to do it completely solid-state (well, or solid + relay state,
> if that makes sense).
Indeed, even an electromechanical relay will play castanets!
Something like this
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=Z921-ND should
be fine and easy to use (screw terminals). You may need to screw it on some
power dissipator to limit heat...

Philo