I am working on a clock that I would like to turn power on and off to our ROKU, so that the roku is only available to my kids during certain hours of the day.
The roku needs 5V at 2.5A if I recall the label correctly.
I have considered using either a relay, or a FET to accomplish this…
My concern with the relay would be having the relay on continuously for the hours that the Roku is available, primarily for heat.
My concern with the FET would be the voltage drop due to the Rds. for a 5V load, I don’t think that https://www.sparkfun.com/products/10349 would be the right choice.
I’m thinking a P-Channel MOSFET is probably the right class of device, but I’m trying to figure out what I need to be looking for to drive the thing with my arduino, 5V logic, with a 5V load, and minimal power / voltage drop across the device…
So, any suggestions for what I should be looking for?
For the component you say is not suitable, Rds on is 0.07 ohms. At 2.5A that gives a power dissipation of (2.5**20.07=) 0.44W which at 0.8W/degrees C means that it should only rise about a half degree above ambient. The Rds stays fairly flat in the current range so the drop from 5V would be (2.50.07=) less than 0.2V. That leaves 4.8V to run on or give it an input of 5.2 and you’ll be right on spec.
Bottom line, I don’t know why you are disqualifying the FQP27P06. It seems like it would work fine in your application.
The Roku draws about 1 amp continuous at full tilt…5 watts.
Probably be better off using an N-channel MOSFET. More selection, better ratings, etc. Take your pick at which one to use. Chances are if it’s a TO-220 case and you put even a small heat sink on it, it’ll work forever. Easier to design to ‘apply a ground’ with an N-channel than to ‘apply power’ with a P-channel.
skimask:
The Roku draws about 1 amp continuous at full tilt…5 watts.
That's 5 watts in the Roku. The transistor only dissipates the power absorbed due to Rds otherwise it's a pass thru. With an Rds of 0.07 ohms and 1 amp draw that's only 0.07 watts in the transistor. At 2.5 amps it's still less than 0.44 watts.
skimask:
That’s 5 watts in the Roku. The transistor only dissipates the power absorbed due to Rds otherwise it’s a pass thru. With an Rds of 0.07 ohms and 1 amp draw that’s only 0.07 watts in the transistor. At 2.5 amps it’s still less than 0.44 watts.
Chip
Well, ya, that's what I meant by the 5 watt draw...really very little if the proper MOSFET is used.
My experience with a Roku 2 XDS is much less than 5 watt draw. Silly experiment way back when…I was able to run the thing continuously for a little under 6 hours from a 4 AA 2500 Mah NiMH pack.
Throw in a couple of fudge factors…
assume that 2500 Mah pack is only actually a 2000 Mah pack
maybe the 6 hours was actually 5 of continuous streaming
meh…an average draw of about 2 watts.
Maybe one day I’ll hook it back up to my uCurrent, log the data, and plot it out for some comparison.
capnregex:
I am working on a clock that I would like to turn power on and off to our ROKU, so that the roku is only available to my kids during certain hours of the day.
How about using parental instructions which if disobeyed has a 100% chance of unpleasant consequences, worse on a successive incident.