Power Usage Statistics

[FrizzleFried]

I picked up a Kill-A-Watt device from ebay and it was just delivered today.  For those not in the know,  a Kill-A-Watt unit plugs in to a wall socket...it has a digital read out and you plug things in to it.  It has the ability to report voltage, watt consumption,  amperage consumption,  and a few other things. I went out to the game room and used the device to determine the power consumption of my different arcade cabinets.  The following numbers represent the MAXIMUM value that came up on the read out...our cabinets don't consume electricity at exactly the same rate like a light bulb...theres usually a 2 to 5 or 6 watt fluctuation.  To determine the accuracy of the device I first plugged a 15w florescent lamp in to it.  It read 15w.  I next plugged a 60w lamp in to it.  It read 60w but would flash to 61w every once in a while.  It seems to be pretty accurate.

Here are my results:

Asteroids Deluxe - 145 watts
Asteroids - 149 watts
Star Wars - 238 watts
Tempest - 239 watts
Marvel vs Street Fighter (25" monitor) - 125 watts
Nintendo VS Unisystem - 92 watts
Space Invaders - 127 watts
Multipede (60-in-1 Board) - 95 watts
Double Donkey Kong (Arcadeshop Multigame) - 89 watts
Horizontal MAME (25" monitor) - 210 watts
Vertical MAME (19" monitor) - 205 watts
Cocktail MAME/Juke (15" PC monitor + stereo amp) - 235 watts

Now, as for amperage, can someone explain why sometimes I got a readings like:

Space Invaders 127 watts & 1.27 amps
Marvel vs Street Fighter 125 watts & 1.55 amps

I guess I am not up on electricity enough...because I thought if the wattage increase the amperage increase as long as your plugged in the same 120v outlet?

As for what I read in these results...

Color vectors suck the power compared to raster games. Star Wars and Tempest suck over 85% more electricity than Space Invader or even the 25" JAMMA cabinet I have here and 159% more power than the Nintendo VS cabinet!

B&W Vectors also suck up a bit more power coming in about 20-25 watts higher than the 25" JAMMA cab.

The new multi game boards suck the least amount of power. I was AMAZED to see only 89 watts of power being used by my Double Donkey Kong cabinet. 89 watts powers the marquee, monitor, and runs the board. Same can be said for the 60-in-1 which came in at 95 watts...

Now, everyone should keep in mind that these devices have the potential to DOUBLE in power consumption for an instant at power up. The games I had access to the power switch of were Multipede and the JAMMA cabinet. Multipede ran at 1.02 amps. When I powered down and back up again it bumped to 1.44 amps for second then went back down. The JAMMA cabinet runs at 1.55 amps and when I turned it off, waited a few seconds, then powered up, it jumped to 2.89 amps for a second then went back down.

What that says to me is that it's probably a good idea to power up each cabinet individually if possible as opposed to all at the same time (that is, of course, only if you are close to the limit of the circuit).

[RetroACTIVE]

I guess I am not up on electricity enough...because I thought if the wattage increase the amperage increase as long as your plugged in the same 120v outlet?

General rule is Power = Voltage * Current

With AC load thats not 100% absolute but generally correct...

This is really good info... for those considering building a game room and would like to have an idea how much current is used by these things!

[MonMotha]

Yet another long winded and somewhat technical explaination if you are really interested:

Regarding P=VI, it doesn't fully hold at AC, like RetroACTIVE mentioned.  There's the possibility that the current and voltage may not line up exactly with each other in time, and then you have to account for that loss of power.  Your meter reads RMS voltage and RMS current, which doesn't account for that. For devices using SMPSes ("switchers"), you have to assume a power factor of something like 0.6-0.9.  It can be higher or lower depending on the design of the supply and how heavily it is loaded.  Linear supplies can vary from very good to very poor power factor, depending on design (I've seen <0.3 before).  Then P=VIcos(phi) where cos(phi) is the power factor (V is RMS voltage and I is RMS current).  Complicated explaination I know, but the math is at least fairly simple in the end.

If you just multiply RMS voltage times RMS current, you get "apparent power" (in VA) which is also a useful number as that's how you have to rate transformers.  For sizing circuits (breaker and wire size), you just need to know the expected RMS current.  Note that on devices with SMPSes, current will increase as voltage drops so that it draws about the same amount of power.

Note that the power company bills residential customers based on kWh, so actual energy (power consumed over time) and doesn't consider power factor.  Industrial and some commercial customers are penalized for having poor power factor since it costs the electric company money.  All that extra juice (the difference between apparent and actual power) results in current "sloshing around" in their distribution network heating up wires and going to waste.  There are ways that large customers can correct this if the penalties are high enough to justify it, but for residential customers it isn't really a concern.

This is useful information, though.  As a note, the inrush current on some devices can be several times their rated or steady-state draw, not just double.  My upright cabinet's monitor's rating for inrush current is 30A!  Turning on my Pump It Up machine causes the lights to briefly dim and UPSs on the same circuit to drop the utility line momentarily.  These inrushes are very short lived (and hence aren't dangerous from a wire heating perspective and won't cause breakers to trip in most normally loaded cases), but are something to be aware of.

[RetroACTIVE]

Yet another long winded and somewhat technical explaination if you are really interested:

Regarding P=VI, it doesn't fully hold at AC, like RetroACTIVE mentioned.  There's the possibility that the current and voltage may not line up exactly with each other in time, and then you have to account for that loss of power.  Your meter reads RMS voltage and RMS current, which doesn't account for that. For devices using SMPSes ("switchers"), you have to assume a power factor of something like 0.6-0.9.  It can be higher or lower depending on the design of the supply and how heavily it is loaded.  Linear supplies can vary from very good to very poor power factor, depending on design (I've seen <0.3 before).  Then P=VIcos(phi) where cos(phi) is the power factor (V is RMS voltage and I is RMS current).  Complicated explaination I know, but the math is at least fairly simple in the end.

If you just multiply RMS voltage times RMS current, you get "apparent power" (in VA) which is also a useful number as that's how you have to rate transformers.  For sizing circuits (breaker and wire size), you just need to know the expected RMS current.  Note that on devices with SMPSes, current will increase as voltage drops so that it draws about the same amount of power.

Note that the power company bills residential customers based on kWh, so actual energy (power consumed over time) and doesn't consider power factor.  Industrial and some commercial customers are penalized for having poor power factor since it costs the electric company money.  All that extra juice (the difference between apparent and actual power) results in current "sloshing around" in their distribution network heating up wires and going to waste.  There are ways that large customers can correct this if the penalties are high enough to justify it, but for residential customers it isn't really a concern.

This is useful information, though.  As a note, the inrush current on some devices can be several times their rated or steady-state draw, not just double.  My upright cabinet's monitor's rating for inrush current is 30A!  Turning on my Pump It Up machine causes the lights to briefly dim and UPSs on the same circuit to drop the utility line momentarily.  These inrushes are very short lived (and hence aren't dangerous from a wire heating perspective and won't cause breakers to trip in most normally loaded cases), but are something to be aware of.

...what he said 

[FrizzleFried]

This little device has a Volt button,  which is easy to understand,  a Watt button...then if I push the watt button again it goes to "VA"...then there is a Hz button that if pushed twice it goes to PF,  then KWH and Hour.

Which measurement is more accurate,  the WATT or VA setting?

[MonMotha]

They measure different things.  WATTs are what he power company bills you for using.  WATTs transfer actual, useful energy to the device.  VAs are what the power company generates.  VAs include power that actually gets sent back to the power company, but sloshes around heating up wires.

In terms of how much it costs to operate your cabinet, you want WATTs.  In terms of how big you want your circuit to be to run your cabinets, you need to look at CURRENT.  If you want to use VAs for sizing circuits instead, you can.  Just divide by nominal line voltage (120 in the USA) to get current.  In other words, a 15A circuit can support about 1800VA.

BTW, PF on those meters refers to the power factor that I mentioned earlier.

[richms]

These things are woefully inaccurate on low power factor switching power supplies, like found in PCs etc because they only pull current at the crest of the ac wave, so averaging that over the whole ac cycle makes for an under reporting of VA and watts - they can only give good readings on resistive/inductive loads like lights and motors etc.

To get a proper reading needs an expensive piece of kit that calculates many 1000 values per cycle rather then taking the average voltage and average current and multiplying them.

[ChadTower]

Frizz, thanks for doing that.  Wish you had a couple of pins to do the same. 

[solderguy1]

Over the last couple of years, I've changed my attitudes on power consumption.  Some people talk about standby consumption like it's killing the planet (ex an off TV using 5W while it wait for the remote).  But anytime you're running a heat source, that 5W (which turns mostly into heat), means you're furnace or whatever works that much less.

There's a couple caveats, electric heat usually costs more than gas, and phantom consumption is working against an AC in summer, but still i've stopped feeling as bad.

My aquarium has a 18W UV sterilizer on the filter line to kill bacteria and that's really free electricity because the tank heater works that much less to keep the tank at 77F.  And any heat the tank loses......warms the house! 

[Level42]

Frizz, you have an upright SW right ?

Does this have a fan inside ? And the Tempest ?

The Cockpit uses a big fat VERY powerful fan (the first one I've seen that actually really needs a finger guard) which probably eats quite some power.

Else, the vector power is drawn by the deflection transistors of course....

[FrizzleFried]

Yah,  both Tempest and SW fast noisy fans that sounded like a jet engine when starting...I've since replaced them with much quieter replacement fans...maybe a few watts at the most.

[Level42]

Ow......hope they move enough air to keep thing cool !

The original fans are putting out a small storm !

Also, all the older games use lineair power supplies so the readings should be correct for those at least.

[ChadTower]

Odd.  My Tempest doesn't sound like there is any fan at all.  I'll have to look at that.

[Level42]

Mmmm, my Star Wars manual is for the upright. I don't remember seeing a fan drawn in there, neither do I remember seeing one on a friends' upright SW I'm working on. Not that I looked specificaly for it though....gotta re-check....

[FrizzleFried]

The replacements are 110v fans from radio shack.  You can still hear them,  but they are easily 1/4 as noisy.  They do push a good amount of air too.  Keep in mind that my cabinets only run when they are being played and during parties...not 24/7 like they were designed to run.   The Tempest fan looks to have been added aftermarket...