Controller Current Draw

[qwertymodo]

I'm trying to build a controller detector circuit based on a current shunt resistor on each controller's Vcc line (high-side current sense circuit), but to figure out the math, I need to know the max current draw to expect on each port's supply line. I already measured the current draw of a single OEM gamepad, and it's about 300 microamps during normal gameplay, but that's a single gamepad. I'm sure other peripherals have higher draw. At the very least, a fully-loaded multitap should be roughly 4x that. Maybe something like a light gun or one of the IR wireless receivers might have a higher draw, I don't know. I'll be testing everything I have, but my SNES accessory collection is far from comprehensive... If nothing else, I'll probably assume a max of 10-100mA or something on the high end like that, but I'd like to get some real numbers if at all possible. The circuit works fine even if I estimate too high, it just wastes power and drops the controller's Vcc a bit, but better numbers up front means less chance for problems down the road.

[lidnariq]

Maybe try using something with a non-linear V-I curve, such as a diode, or a self-biased FET?

[nocash]

The SNES mouse is rated 50mA.
The Multitap is rated max 17mA per slot. That would be 68mA for all 4 slots (plus the power consumtion of the Multitap itself, so it might be around 70...100mA in total).
That are the official ratings, I don't know how far the actual hardware is following those values.
Nintendo appears to have been strict about not to exceed that ratings, but unlicensed controllers might be completely ignoring those limits.

[tepples]

How much current does a PS/2 mouse for PCs draw? I'm trying to figure out whether a PS2 to SNES adapter with a microcontroller inside would be feasible.

[MottZilla]

You could always use an external power adapter like Copier Devices used.

[lidnariq]

How much current does a PS/2 mouse for PCs draw? I'm trying to figure out whether a PS2 to SNES adapter with a microcontroller inside would be feasible.

As a data point, my optical PS/2/USB one is rated for maximum 100mA. Non-optical mice will be much lower, but making it clear to the user that they need to use a crappy cheap ball mouse might be difficult. A random source claims that the original PS/2 port has a fast-blow single-time fuse set to 275mA, although later ones at least used a polyfuse instead.

[qwertymodo]

Now that I think about it some more, I don't really need to know the maximum current so long as the shunt resistor is small enough as to not limit the current below that amount. I don't need to design the current sense circuit around the max current because all I care about is the minimum threshold, and anything above that doesn't matter. Higher current = higher voltage drop across the sense resistor = higher output from the current sense monitor... however, the current sense monitor, being a real, non-ideal amplifier, should clip the output to the supply rail, so really all that matters is the MINIMUM current draw, because the maximum is likely to be 2, if not 3 orders of magnitude larger than the minimum, so there's no way to fit the maximum within a linear amp range, but I don't have to because of the clipping. The differential input voltage will be in the millivolt range, and the common-mode voltage is 5V, so that's all within reasonable specs. I have it hooked up and it's working, only problem now is that I'm getting a nasty voltage spike on the output pin at power-on and immediately after connecting the controller, roughly twice the supply voltage, which I didn't think was possible, since the amp should be clipping at the supply voltage. 10V spikes into a microcontroller ADC pin probably isn't going to go over very well... I'm really not a fan of analog problems... :/

[lidnariq]

Add a 5V zener diode as a voltage clamp?

[qwertymodo]

Yeah, that's the plan, but I don't have any, and I'm too broke to buy any :/

[lidnariq]

You can mostly use stacks of ordinary diodes or LEDs to clamp outputs too, they just have less good V-I curves (larger small-signal resistance) once they're in the conducting zone. And they're a little harder to tune.