Just a quickie simple test I did, after leaving my machine on for 10 minutes, with the back box opened up. These are the lamp column resistors on a Williams System 7.

www.siegecraft.us/pinball/BK-023.jpg

Well, the average #44 bulb uses 250ma of power, and the average LED uses 20ma. It really adds up fast. I've got about 65 GI bulbs in mine to do, 2/3 of which were right behind the backglass. That adds up to almost 15 amps.

-Hans

There's a lot that can go into that factor, but I don't trust that 8amp number for squat. Do they mean a 2010 Stern machine with a flourescent tube behind the translite? Or your 1990's Williams WPC machine, with banks of 555's? And is that peak draw? or average? I plan on buying a kill-a-watt meter and measuring it for myself, as I've learned a long time ago to never trust numbers that aren't explicitly defined.

In my case, I know I have 65 GI bulbs, drawing .25amps each, which means 15amps of constant draw for the GI circuit alone. This makes perfect sense as the GI circuit also has a 20amp fuse. Yours I would assume has less 555's than I have 44's, because your translite is smaller due to the DMD display and speaker panel taking up about 1/3 of the available space. (You have an Indiana Jones, correct?)

Playfield lighting is a different beast though. Those lights are very rarely ON at the same time, and most are OFF for most of the time. The few times you see them all ON, they are blinking. This drastically reduces the overall power demand that you'll see. Another thing that helps out is that they are fed by a huge capacitor (at least, mine are). This means that in those instances you have a huge power spike, it draws down the capacitor instead of placing the demand on the wall outlet. Just a random guess here is that you're seeing under 5amps of an average power draw on that circuit. My Black Knight has only an 8amp fuse on that circuit, so it can't be too much.

But overall, I'd say as an educated guess, over 75% of the power demand comes from lighting alone. This is why people so strongly suggest using #47 bulbs instead of #44's as the amp draw is .15amps instead of .25..... it's a HUGE difference going from 15amps to 9amps in the GI circuit. Or LED's which take it down to 1.2amps.

-Hans

Here's a quickie voltage/wattage/amperage calculator that I found, sure beats trying to memorize the formulas.
http://www.sengpielaudio.com/calculator-ohm.htm

The switch from 6.3v to 110v would change the total wattage, but the amperage would stay the same as that is dependent on the draw from bulbs not the voltage of the circuit. Being AC or DC voltages don't make a difference in these equations, other than the incandescent bulbs lasting longer on AC power.

Don't worry about it too much if you don't know the theory yet. You can study all you want, but the practical experience is almost always more valuable. I've always felt that repairing busted wiring and circuits helps you to identify good and bad practices in design. And pinball stuff definitely has some bad designs to it.

Particularly the lighting circuits, which are pushed really right to their limits even under the best conditions.

Cooling fans will help, as will going with higher wattage resistors, and also downgrading the wattage used by the bulbs. In this specific case, swapping the transistors connected to those resistors with a compatible MOSFET also helps. I'm, eventually, doing an "all of the above" job to things. First I need to diagnose some other problems though.

There are lots of options available to us now, that weren't in 1980 when this game was made.

-Hans

Yes, the 110v draw is 1.5 amps at the wall socket. Actually, in my case, it averages to about 1.8amps or so..... at 110v. Which works out to about 200 watts. I'm going to be doing a series of power comparison tests for how much LED's affect your power useage, as well as the internal temperature. I did my first baseline test today to measure the KwH useage. It came out to .54 KwH over three hours with all the #44 bulbs running. I'm still trying to find a thermometer I can use for the backbox internal temp, I don't want to use an electronic remote sender inside a shielded box with all that RF noise.

My constant in this set of measurements will be the wattage, with the volts and amps being the variables. I figure that I've been throwing enough random numbers out, without a sense of comparison, and I need to get that solved before you guys are REALLY confused. So, from here on, I'm going to make sure I'm clearly explaining what I'm talking about..... It's bad enough that I made mistakes above already.

And it does get complicated after feeding it through all the different step-down transformers inside the game. The stuff can make your head spin when dealing with all the different voltages inside a pinball machine. IIRC mine has 110vac input and internally it uses 100vdc, 5vdc, 12vdc, 6.3vac, 18vdc, 28vdc.... and I may be missing one or two. The 15 amps I was quoting was for the 6.3v GI lighting circuit, which measures out to about 100 watts. This is 50% of the total power use in the system. Replacing the GI bulbs with LED's I'm estimating the 6.3v circuit draw will go down to 1.2 amps, which is only 7.5 watts. That's a huge power reduction, 46% off the total power used is eliminated.

Not sure yet what the insert lamps are going to see, as they are running that funky 18v strobed system. Same with replacing those TIP42 transistors, I have to find the datasheets to see what their draw is. Easy to find the info, I just haven't gotten around to it. Making the thing WORK was my first priority. Now that it's running reliably, I can get down to business on this sort of stuff.

-Hans

There are a couple options for you that might be worth looking into. First is the #47 incandescent bulbs, which draw .15amps instead of .25amps. That right there is a big improvement when spread out over 100 or so bulbs, and with NONE of the confusion of LED's.

As to the LED's, there are big differences between the ones on the market now, and a few big changes in the past couple of months. Cointaker now offers non-polarized LED's, and they also have a "warm white" color available, decent price too at around 60 cents each if bought individually. Pinballcenter's new Noflix Plus LED's are really neat, they have a full-wave rectifier and a smoothing capacitor built into each LED, unfortunately they are a lot pricier and shipping to the USA is horridly expensive (40 Euros, just for shipping). The Ablaze LED's that pinballlife sells have been upgraded recently too, and now work in controlled circuits, but without a warm-white option I'm not considering them at all.

Not too sure about the WPC power saver settings.... I'm dealing with an older system on mine.

-Hans

LOL This is just TOO rich. I figured I'd run a full set of #44 bulbs in this thing to do some testing, and order up the LED's in a week or two from now. Not even a week after I got all the backbox bulbs working, It's already burned up a brand new GI circuit connector on the power supply board and is causing GI lighting failures. I haven't even gotten a chance to run a temperature test yet.

So, I have a full set of GI LED's on order from cointaker right now. 170 degree warm whites for the backbox, frosted warm whites for the playfield, and 90 degree color matched for the coin slots.

And since it makes NO sense to pay $5 shipping for just a $1 part (that burned connector), I'm also going to be ordering those mosfet's to change out the TIP42 transistors on the driver board.

-Hans

The LED's should totally solve the problem, they should be here today or tomorrow I hope. It's really Williams fault that this happens on System 7 machines, as they were putting 15amps through a connector rated at 12amps. Add in the heat of the box, which reduces how much load the connectors can support, and this result is inevitable without modification.

I got around to running some power usage and temperature tests the past couple days. The machine, with all #44 lights, draws 180watts/1.6 amps at the wall socket.

Temperatures pretty much peaked at the 45 minute mark, but crept up a touch more until the 1hr30min point. It was 82 degrees in the room the whole test, the backglass went to 100 with an IR thermometer, and inside the backbox hit 120 degrees. All of the lamp resistors on the driver board were bumping at 200, and the GI connector was over 300.

Yikes!