Yep. The Arduino has a 7805 three terminal LINEAR voltage regulator to knock down the input voltage to what the card needs. 12V to 5V is a 7V drop. Power = I*V so the wattage is 7x whatever the current is. For 200mA the device has to dissipate (with only the PCB as a heatsink) ~1.4 W. If the 7805 fried under the load and shorted, the full +12 gets applied to the board and everything cooks.
You definitely had a cross couple power supply issue. If it came in thru Vcc or your I/O pins is the only point.
You need to have level shifters to get to +12 from Vcc (3.3 or 5). ULN2003A is what the stepper motor cards use and they work great for any kind of 12V load while protecting the CPU from damage.
Just looked up the SOT-89 package 7805 and max current is only 100mA.
This is really useful, thanks for the post.
In fact thanks to everyone on here!
I've been testing everything individually with a power supply and multimeter, one of the motor controllers is backfeeding 8.5v to th logic pins when powered with 5v.
It would seem I'd (wrongly) assumed connecting a jack to the arduino would allow it to happily power everything.
This jolted everything when the 7805 blew but fortunately only seems to of blow one motor controller. Does this seem plausable?
hm.
I always thought that they tend to melt (and be open circuit) rather than short circuit. Maybe that is what that killed my Due..
When a semiconductor die overheats and fails, you cannot predict exactly what is going to happen.
Mostly they will go open circuit but they can go short circuit too and the fuses might not go.
Back in the mists of time I do recall finding a fault with a commercial device we were making. Two units being tested caught fire at night which fortunatly was spotted by security and the factory did not burn down.
It transpired that a device was failing due to a design fault and mostly it failed 'safe' went open circuit which cut the power. But now and again when it failed it went short circuit, overloaded the power supply and set fire to the unit.
True that; I've had more than one linear regulator fail short. The current pass transistor(s) are definitely not guaranteed to fail open. Too bad, because if they fail short, Vin appears on the output and mayhem ensues...
This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.