No announcement yet.

Do PC's really use that much power?

  • Filter
  • Time
  • Show
Clear All
new posts

  • Do PC's really use that much power?

    I've been planning to make a car PC for a long time but have been putting it off because lots of people seem to be having power problems. The opus is 150Watts and this is just not enough when using a cheap full size mother board (I say this only after using one of those web sites which gives you an estimate of power you would use.)

    I have this gadget at home that calculates how much power an electrical item uses. I plug in my PC which has an Intel Celeron 2.4Ghz. When I boot the system it hits 80Watts. When the system settles it reads 34watts. When I push a CD in it reads about 43watts.

    The actual device takes Power Factor into account. The power factor when plugged into the PC is 0.67. If I shut the PC down the power factor goes to 1, and only reads a few watts.

    If I click on the function button it reads 235volts and 0.22amps.

    Should I trust the readings on this device? - It seems that an OPUS would have no trouble running this full sized tower, contrary to what the power estimator web site says.

    (In the pic 34 watts 0.67 Power Factor. The 11:35.06 is the time)
    Attached Files

  • #2
    Just read something by Numbers on another thread

    I'd say the opus 150 would suit you just fine, but so many folks seem to be overloading the 5v rail even though their overall consumption is under 150 watts.
    How do you overload 5v rail? What does it mean?


    • #3
      A PSU outputs multiple voltages (12V, 5V, 3.3V)... Each has a current(or power) rating. I.E. You can have like 2Amps or 10Watts on the 5V line(rail) and 3A on the 12V rail... If you were to overload the 5V rail you would exceed its current/power rating. This would not happen from just running the PC, but rather having a bunch of 5V devices connected to the PC(GPS, Rear View Cam, etc)

      Note: All numbers stated here are strictly for example.. I dont know how much current/power typical power supplied are rated for per rail
      B Smoov
      Project Status: 90% Complete

      Next Step: TIVO for Radio

      Ampie Case :: MII10000 Mobo :: M1-ATX PSU :: 512MB RAM :: 2.5" HD, 60GB, 5400rpm, 16MB Buffer :: DWW-700H :: Centrafuse


      • #4
        Originally posted by RMIM

        (In the pic 34 watts 0.67 Power Factor. The 11:35.06 is the time)

        Play a DVD full screen and tell me what it draws.
        MPEGBOX - Plexiglass Computer


        • #5
          Originally posted by zootjeff
          Play a DVD full screen and tell me what it draws.
          OK I did that:

          As soon as you right click and hit play using Power DVD it jumps from 38watts to 70watts. It then settles at 50watts playing at full screen.

          Remember Im just measuring what the tower consumes. (not monitor.)

          To me my full size computer does not draw that much power as I have been let to believe on this forum by using the power calculators.

          The most my system draws is when it boots. - It shoots to 80watts but settles to 35-38watts when you're not doing anything.

          My question is - are the reading I'm getting from my power gadget thing accurate?

          If I had known that a computer would only be drawing about 80watts max, I would have bought an opus 150watt long ago.

          If my gadget is accurate - why are people having problems with the 150watt opus when I doubt their system are drawing anything near 150watt? (In total - I don't really understand the separate rail issues yet.) - Where do we get the figures for the max power a rail can handle? What rails do USB device use? 5v? Are people loading there system with too many USB devices and the Opus can't handle this?



          • #6
            I wish I could just post this once and it would do, but I seem to have to post it every week or two...

            Computers DO NOT draw what everyone is led to believe by power calculators. Your gadget is reading correctly. Based on overall power consumption, the opus 150 really should run most of our setups. I have gatherred that in some cases where it is insufficient, it is due to excessive 5v rail draw, but I have no solid evidence for that as I do not own one. I get tired of repeating myself. (no offense to you) Search the board for posts by me containing "consumption" or "opus" or "draw" or something.


            • #7
              What is this "power factor" that your device is using? That seems to affect the wattage rating on your device. You should understand that better before deciding that the readings on the device are accurate.

              Also, when you used the power calculator did you put in the Dell motherboard you show in the photo, or a different one. You should make like comparisons before drawing conclusions.

              You are also measuring a Celeron. I understand that the Pentiums tend to draw the most power.

              In addition, many people run their screens off of the PSU. It provides clean and regulated power to the screen. Although they don't draw very much power, they do add a bit.
              Originally posted by ghettocruzer
              I was gung ho on building a PC [until] just recently. However, between my new phone having internet and GPS and all...and this kit...Im starting to have trouble justfiying it haha.
              Want to:
              -Find out about the new iBug iPad install?
              -Find out about carPC's in just 5 minutes? View the Car PC 101 video


              • #8
                Originally posted by Bugbyte
                What is this "power factor" that your device is using? That seems to affect the wattage rating on your device. You should understand that better before deciding that the readings on the device are accurate.
                The gadget device reads a Power factor of 0.68 when you're not doing anything with the PC. If the PC is off the Power Factor reading changes. So the gadget actually reads true power consumption and does not simple multiply voltage with current to work out the wattage. I think the device can measure the actual power consumed, it also can measure the voltage and current. With these pieces of info it calculates the power factor I think. The device does not use a fixed power factor. (Only guessing here, not even 100% if I understand power factor.)

                I have tested my device with 2 full size computers. The dell settles at about 35watts and jumps to 80 while booting.

                I have just tested my AMD Athlon XP 2600+ (1.9sGhz) 3 hard-drives. AGP graphics card. PCI tv tuner. And it hits about 90watts when booting and settles to 75watts when idle (Power factor 0.60, the dell was 0.68)
                I will go and check what the power calc web site says.

                I get 266watts peak when using

                Now going with the 266watt I would never buy an opus even if it is a peak reading. But according to my gadget its 90watts when booting. Might go a bit higher if I burn a dvd. But that no where near the Opus 150watt limit.


                • #9
                  Originally posted by RMIM
                  Power Consumption
                  The real problem is how your PC splits up the power. With the OPUS, They have 10 amps on the 5v, 10 amps on the 3.3v and 5 amps on the 12.

                  Some systems only use the 3.3 and the 5 for the whole system and then only a little bit on the 12. Also, newer P4 and AMD mobos use a ton on the 12 and much less on the 5. The 150 is the total of all the rails. If I had a power supply that put out 30 amps on the -12 and 1 amp on the +5 and +12. It would be a 370 Watt power supply, but wouldn't power a thing!

                  You really have to disect the numbers. The other thing your not getting from your Power Meter is the Peak power. over the course of a second, a rail may draw 50 watts for 0.1 seconds, and 5 watts for the remaining 0.9 seconds.

                  Your Meter would measure something like 8 watts or whatever for the whole time, while the computer would have crashed if it was starved for that 0.1 seconds.

                  These are exagerations, but show my points..

                  MPEGBOX - Plexiglass Computer


                  • #10
                    Originally posted by zootjeff

                    You really have to disect the numbers. The other thing your not getting from your Power Meter is the Peak power. over the course of a second, a rail may draw 50 watts for 0.1 seconds, and 5 watts for the remaining 0.9 seconds.

                    So what you're saying is that the power calculators are useful after all?

                    I used the power calculator and got a figure of 266watts.
                    I used my gadget thing and got a max reading of 90watts.

                    So you don't think that a 150watt opus would survive the above computer? (Although my unit claims it's only using a max of 90watts.)

                    Damn it! - Im back to square 1


                    • #11
                      You would need to measure the consumed power on the 12v & 5v side of your PSU to get a proper accurate idea for car use.
                      And it would be better to use a scope that shows instant power useage and records max readings rather than one that just tries to measure an average.
                      Your meter is telling you what your PSU is drawing, not the actual mobo & drives etc.

                      The 240v thats entering your PC is split into 3.3, 5, & 12v, as has been said above, the OPUS splits its 150w into 3 outputs (not sure of the ratio's). Your PC might be drawing 79w at boot from the 12v rail, and 1w on the other 2, and if OPUS only supplies 60w of 12v then its no good.

                      Other thing to note is if you have a small engine and small battery, a high powered desktop will put a large load on your cars electrics & battery and will affect fuel consumtion. Better save your money for a Pentium M or an Via SP13000, VIA will work out much cheaper/cooler/more efficient. And Via will run just fine on an OPUS 90w with a 5400rpm 3.5" HD, USB: GPS/WIFI/DVDRW/Rev Cam/Bluetooth/Touchscreen, 512mb ram, and an ECU scanner.

                      I know cause thats what I've got

                      Good luck, and don't be put off if your current kit won't do the job, sell up and buy different kit, thats what I did, on my 4th motherboard now
                      My Nissan 200SX @ Silverstone
                      NOW SADLY WRITTEN OFF!

                      My Car's spec list + pics

                      Stealth CarPC fab'ing...


                      • #12
                        When in doubt revert back to ohms law, im in somewhat the same boat as you. But i just have to keep reminding myself that with the P=IE that current and voltage are inversly proportional. So if voltage goes down say from 120volts at 1 amp -- to say 12 volts then were looking at 10 amps. But there is also the whole ac dc thing that really messes this up. Also that is what makes some amplifiers so funny they have a 10A fuse but say 600w when clearly if they are 12v 10A then if they are 100% efficient they could only then be 120W So if you aint got a 50A fuse you most likely dont have a 600W amp sorry went off track. Back on track then, so i was thinking if i got a second battery and a few voltage regulators and molex plugs made a 1.5 3 5 and 12 volt plugs fed it with 1 guage feed i could potentially make a 300+Watt PSU but deffinately have to upgrade Alternator.

                        1994 Mustang GT Punch 240.4 2 12" Kicker Comp SOLD :(

                        2006 Suzuki Aerio Sx Kicker KX 650.4 RE 10

                        2003 Eddie Bauer Expedition DVD - Navigation Package Traded in for 2008 XL-7 Limited

                        Car PC


                        • #13
                          Power Factor

                          It is difficult to explain "power factor" without getting into some heavy math and electronics, but I'll explain how it relates in the real world.

                          Essentially, power factor is a measurement of how close (in phase) voltage and current draw are in an AC circuit. A purely resistive load (such as a light bulb) will give a power factor of 1 since voltage and current will be in phase. in an inductive circuit (anything with a coil or transforner) current draw will be out of phase with voltage. The larger the inductance the greater the difference in phase. The larger the difference in phase (essentially, the larger the inductance) the lower the power factor.

                          Power Factor is important because low power factors from highly inductive loads result in power generation and transmission losses. Fortunately, residential electric meters measure only usage and do not take power factor into consideration. Commercial electric meters, however *do* take power factor into account, so commercial customers are penalized for loads with low power factor. This is because (at least historically) a large percentage of commercial power draw was for large motors with low power factors.

                          The only time power factor plays a role in carpcs is when you are driving your PC from an inverter, but even PC power supplies with a relatively low power factor won't have a substantial impact.



                          • #14
                            Originally posted by RMIM
                            The most my system draws is when it boots. - It shoots to 80watts but settles to 35-38watts when you're not doing anything.
                            The high current drawn on boot up is due to HDDs/DVDs/FDDs/fans and some other electronic parts inside the PC initialising. Those motor may not start to spin all at the same time so it be hard to measure the real peak current even at boot up. For example if the HDD start to spin its motor, a high current is then registered for a secs, once the HDD motor reach its correct speed current drawn goes down. At this point the DVD may spin and then settle down, then we have the FDD and the fans. Remember you still have other items that can draw more current when being used and this does not happens during boot up.

                            So basically an accurate of peak reading will be difficult to obtain unless you can be sure that each device will draw their peak current all at the same time.

                            But you are right, you can power a PC with a much lower PSU rating. For reliability sake the worst possible case should always be taken into consideration.

                            As for the power calculator, best is to find the exact power consumption of each item you have installed in your PC. Not all brand/model will consume the same amount as assumed on the power calculator.


                            • #15
                              Perhaps you could have two power supplies running to take the load off?