I did a computer controlled mechanical drummer with four small analog VU-meters but they just don't hit very hard...
Next version will be based on a bunch of hard drives from a slaughtered server, plan is to have around 10.
So far I've only tried hooking up one to my arduino using a 2N3904 transistor as mentioned here: http://arduino.cc/forum/index.php/topic,56159.msg403641.html#msg403641 seems to work fine can't help but wonder if it wouldn't be best practice to include a diode to help with any potential backward current spikes like it's commonly done on similar devices like solenoids and motors?
Can't detect any spikes with my cheap multimeter but it might not be up to the task either, same thing with normal forward current - I measure 30mA with my circuit for one Voice Coil so it seems to suggest that I should be able to go with external power on a Duemilanove or even USB with 10x30mA but then again I'm not sure if my multimeter can catch the peaks.
Get a scope - www.dpscope.com, $70 + $10 shipping, probes included.
If there are spikes, you will see them when you switch off the current and the voice coil tries to keep it going.
If they approach 40V, you will blow the transistor eventually.
I've borrowed a Fluke 124 scope which let me measure the voltage spikes easily. They were going up to the 70V range so I put in a 1N4004 diode I had around as a flywheel which brought the top spikes down to around 9V.
Can't get current metering working with the Fluke. When I try to put it in series it just breaks the circuit, not sure if special probes are needed or if I just can't set it up correctly.
Not sure how to think about dimensioning the power supply without being able to measure, how much margin I should have. The transistor gate will only be on for a few ms(10 max?) a time.
The meter may have a blown fuse in the ammeter circuit.
Do you have a really small value resistor you can put in series and measure the voltage across that?
The way I've had it operating the transistor doesn't even get warm which sort of make me want to think it's relatively safe operation. I just tried PWM:ing it though and that way it surely wouldn't last long.... got hot quickly.
Anyway I could easily choose a higher rated transistor but I think the root of the problem is my understanding of analog circuit theory is poor enough for me to not really know how to make sure I design a "safe" circuit.