Controlling a 2n3055 power transitor from a microprocessor
LoopyByteloose
Posts: 12,537
I know I could use power MOSfets, but I have a few 2N3055 transistors available.· I suppose that these bi-polars are becoming dinosaurs, but I'd prefer to use up what I've got.
I saw a schematic where these 15amp NPN transistors were controlled by a microprocessor through a 1k resistor.· I was considering using a 2n2222 to create a Darlington pair, but now I am reconsidering.
I cannot seem to Google any design tutorial for DIY Darlington pairs.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
PLEASE CONSIDER the following:
Do you want a quickly operational black box solution or the knowledge included therein?······
I saw a schematic where these 15amp NPN transistors were controlled by a microprocessor through a 1k resistor.· I was considering using a 2n2222 to create a Darlington pair, but now I am reconsidering.
I cannot seem to Google any design tutorial for DIY Darlington pairs.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
PLEASE CONSIDER the following:
Do you want a quickly operational black box solution or the knowledge included therein?······
···················· Tropically,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan
Comments
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
www.fd.com.my
www.mercedes.com.my
I guess the 2N4401 and 2N4402 are suggested because of high current output is needed by the 2N3055 to get the most output. The 2n2222 is about 100ma tops.
I've no idea of how to size the pull-down resistors. I'll Google around.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
PLEASE CONSIDER the following:
Do you want a quickly operational black box solution or the knowledge included therein?······
The pull-down on the input to the power transistor is about the same, and ideally it should only be enough to keep any leakage current from turning on the power transistor without diverting too much current from the power transistor when the small-signal is on. This might be 1K-10K, depending on how much current you want to sink through the power transistor.· If the power transistor tends to stay on a bit then you need to go for a lower value pull-down (and thereby get that base closer to ground.)
I pull-down the small-signal transistor·to keep it from ever floating, it's not "essential" (to some that's an ill-affordable extravagance, but that's how·it should be done.)· The pull-down on the small-signal output, which is also the power transistor input helps turn off the power transistor quicker.· If the power transistor tends to stay on a bit then you need to go for a lower value pull-down (and thereby get that base closer to ground.)
The Darlington multiplies the beta of the two transistors.· Let's say the small-signal beta is 100 and the power is 20, then the effective beta is 2000.· The current possible from the power is a product of the current available from the small-signal.· In an ideal situation, if you have 1mA of small-signal IB then you could get 1mA X 2000 [noparse][[/noparse]2A] out.
PS -- I am nutshelling all this.
I have a Taiwan buddy that is wasting a fortune as a tube audiophile and thinks he is learning all and everything of electronics. I am trying to get him to look at transistors in contrast [noparse][[/noparse]current amplification versus the good old voltage amplification].·
He still isn't paying any attention to digital - too tiny. This might begin to wake him up.
I did understand the the pull-downs are to stablize the response to control signals. I'd like to eventually use PWM, so it would be nice to do the math and really have a solid, efficent scheme. I think Wikipedia mentioned the Darlingtons tend to float due to a bit more time delay in signal movement. I doubt if the PWM will get high enough to matter, but the square waves will certainly be crisper.
At first, I wasn't even considering pull-downs. But they are a wise addition. I got some 2n4401s today, so it looks like quick work to have a couple of hefty motor drivers. I don't think I'll ever get to 15amp motors. But having the module built opens up possiblities - maybe a motorized bicycle with digital control features.
I am shooting for 10ma will handle 10amps. Am I too ambitious? I'll try to do the math. If it falls short, the 2n3055s will at least run cooler.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
PLEASE CONSIDER the following:
Do you want a quickly operational black box solution or the knowledge included therein?······
Post Edited (Kramer) : 3/22/2008 4:57:35 PM GMT