If you check http://www.dkpinball.com/ I have power tap products for just this thing.

I do have a Tap product that will fit F-14, i just don't have it up on the web site yet. Check out the Data East power tap and the online installation instructions. The data east is very similar to the system 11. You can see it here: http://www.dkpinball.com/DKWP/?page_id=601

It gives you two fused sources. One for 12v and one for 5v. If your mod shorts out or if you overload it, the fuse will blow on the tap rather than on your game.

And, you're fine with the 12v unregulated for LED strips.

PM me if you're interested and if you have any questions. I have a lot of different solutions for power distribution and control.

You're right to suggest that PWM should be used to vary the brightness of LED's, but altering the input voltage does alter the brightness. They aren't just on or off. The brightness is driven by the current and with a fixed resistor, the current will go up or down based on the input voltage. It doesn't do it linearly, and you can get outside of efficient ranges quickly, so altering the current isn't the preferred way to control LED brightness.

If you put two of the same LED/Resistor combinations next to each other and feed one 4v and the other one 12v, you will definitely see a brightness difference. Real life example: This past week I was fitting cheapo 555 LED's to clear flipper buttons to make an inexpensive lit flipper button. I plugged it in and was really happy with how bright they were. Then I saw the smoke. I had connected the 6.3v pinball LED to 12v. Changing to 5v made the bulb significantly less bright. I swapped out new bulbs because damaged LEDs will appear less bright. It was still significantly less bright.

With 12v strips though, you're not going to see much difference between 11 volts and 13 volts, that's why I said that 12v LED strips on an unregulated 12v supply are fine.

Not necessarily. The short answer is that it's just not that simple. We're used to the experiment where we take an incandescent bulb and we alter the voltage and the light bulb changes brightness.

I know how the math all works but I'm still working on knowing enough about this to teach it. The LED is rated with a forward voltage of 3.3v. Any excess voltage rolls right on through. Resistors don't resist voltage, they resist current. If you had no resistor on the circuit and you fed it 5v, then 1.7v (5 - 3.3) would flow on through and with 0 resistance you would be forcing 1.7 amps through an LED that has a max rating of 30ma per the data sheet. (https://www.sparkfun.com/datasheets/Components/YSL-R542B5C-A11.pdf)

What you need to do is to put a resistor in series with the LED that will limit the current to an acceptable level for the LED. 20ma is really common for a lot of LEDs as the highest brightness with the longest life.

To get the circuit limited to 20ma you need to do this math:

(Source Voltage - Forward voltage) / .02 = Resistor Rating in Ohms.

5v - 3.3v / .02 = 85 ohm resistor.

1.7v * .02amps = .034 watts. So you could get away with a resistor rated at 1/8 watt.

I had a project that required an LED to go on when a 48v circuit was live. The LED had a forward voltage of 2.2v and of course an ideal current of 20ma.

48 - 2.2 / .02 = 2290 ohms (a 2.2k resistor was close enough).

However, 45.8v * .02 amps = .916 watts. So I needed a resistor rated at 1 watt in order to dissipate the heat.

The Blue LED shows a max current of 30ma. So you could squeeze a little more brightness out of it but the lifespan would diminish. Or you could get less brightness by allowing only 10ma through the LED.

It's just not practical to alter the resistance dynamically. You would need a crazy potentiometer that wasn't linear and had a weird lower range.

Using PWM is the preferred way because you're keeping the LED's balanced electronically but even PWM isn't completely linear where brightness is concerned. There are a lot of arduino threads on PWM curves for achieving an affect that is as close to linear as possible.