Actual watts consumed, TC, laptop

tony · Post by **tony** » Mon Nov 17, 2008 12:23 am

Hi
One of the benefits of a thin client is the purported low power consumption. I've yet to receive the TC I won on ebay to test. Meanwhile I plugged my laptop which is powering my MJ and an extended monitor (and nothing else) into a energy meter. Once stabilized, the meter is showing a steady state consumption of 22 watts.

At $0.0655/kWh that's 3.5 cents per day or $1.04/month for electricity. Add $20USD/$25.20CDN per year for MJ and my monthly cost is $3.14/month (or $2.49 USD/month) for unlimited LD.

I'll report back when I get the TC.

booyah5 · Post by **booyah5** » Sat Nov 22, 2008 7:20 pm

I have one of these, and I just plugged my modified 5700 into it, so I'll know in a few days how much it's costing in electricity.

tony · Post by **tony** » Sat Nov 22, 2008 10:22 pm

Hey, you've got a T5700. What did you do to set it up for MJ? Do you have 256/256 memory or 512/512? I just got a T5700 and am waiting for the 512/512 memory to arrive before I start setting it up.

thanks.

booyah5 · Post by **booyah5** » Sun Nov 23, 2008 12:14 am

I got 512 of ram and bought the CF to IDE adapter and used a 4GB microdrive. If I were to do it over, I'd just get the flash DOM due to my inability to get XP embedded to work on the microdrive due to some licensing issue. I had to go with full-blown XP.

HolmanGT · Post by **HolmanGT** » Sun Nov 23, 2008 4:49 am

booyah5 wrote:I have one of these, and I just plugged my modified 5700 into it, so I'll know in a few days how much it's costing in electricity.

Why do you have to wait a few days to know how much power it consumes.

I have a P3 also and my T5730 running MJ consumes 16 watts of true power.

One thing I did find interesting is the switching power supply (the block supplied with the unit) has a terrible power factor mine measured around 0.57 so the power company is taking the total load of about 32 watts apparent power (V*A). Glad I don't have to pay for that portion of the power consumption.

I don't know why the power factor for switching power supplies are so poor. I guess I am going to have to go research the "why of it". I know that the energy is stored in a reactive component but I just never considered it would be as high as it apparently is (no pun intended) I have used switchers for years and just never gave power factor a thought and just enjoyed the fact that they run cool as an ice cube compared to an equivalent analog power supply, which if you claim to be an Electrical Controls Engineer is a little embarrassing.

I do have some concerns about how accurate the P3 power meter is when it is reading a non sinusoidal signal. The only thing that will handle that accurately IMHOf is a Bolometer and I am sure a $16 dollar P3 does not have anything as sophisticated as that on the inside.

Tony, what type of power meter do you have? And 22 watts for your laptop with a monitor is pretty damn good. I would have expected the monitor to pull at least 20 watts all by itself.

Regards, - it is way to late (or is that early) for a conversation that pulls on knowledge from AC theory 101.

booyah5 · Post by **booyah5** » Sun Nov 23, 2008 2:53 pm

HolmanGT wrote:I do have some concerns about how accurate the P3 power meter is when it is reading a non sinusoidal signal. The only thing that will handle that accurately IMHOf is a Bolometer and I am sure a $16 dollar P3 does not have anything as sophisticated as that on the inside.

George, assuming you're right about the P3, what would the outside range of its error rate be? Like 10-25% or more?

Here are my stats for my modified 5700 at rest without making a call:

0.12 amps
7.0 watts
0.14 kwh

When making a call, the watts jump up to 9 watts.

I'm using the PS you first turned us all onto.

tony · Post by **tony** » Sun Nov 23, 2008 3:28 pm

tony · Post by **tony** » Sun Nov 23, 2008 3:29 pm

I'm using a retail UMP EM100 power meter. Nothing fancy.

Link http://www.upm-marketing.com/products

I should clarify, the 22 watts is for the laptop with the monitor screen off. The laptop is driving an external monitor with the monitor's consumption not counted in the energy consumed (it's not getting its power from the power meter).

I plugged in the T5700 and in steady state once it's settled down, it's drawing 9 watts.

I did a test of my home office: laptop, 2 external monitors, desk phone, PDA cradle, cable modem, router, Hawkings booster, laptop docking cradle, couple of USB hubs, and probably some misc AC adapters I've overlooked, and the power consumption is 38 watts, steady state.

So lets, see, dropping to 9 watts from 22 (replace slave laptop with TC), I save 13 watts. At $0.0655 per kWh and $117 CDN invested in the TC (excluding my time), my breakeven point when energy saved equals what I spent to save the energy will be 137,405 hrs or 5,725 days or 15.9 yrs (360 days/yr turned on). I will update in 16 years when I hit breakeven. Oh well, must discount for the educational component, that's got to be worth something.

-Tony

HolmanGT · Post by **HolmanGT** » Mon Nov 24, 2008 1:09 am

booyah5 wrote:
HolmanGT wrote:I do have some concerns about how accurate the P3 power meter is when it is reading a non sinusoidal signal. The only thing that will handle that accurately IMHOf is a Bolometer and I am sure a $16 dollar P3 does not have anything as sophisticated as that on the inside.
George, assuming you're right about the P3, what would the outside range of its error rate be? Like 10-25% or more?

Here are my stats for my modified 5700 at rest without making a call:

0.12 amps
7.0 watts
0.14 kwh

When making a call, the watts jump up to 9 watts.

I'm using the PS you first turned us all onto.

First on the power supply you are using, I have seen that brand in some sealed HP shipments. So they must think pretty highly of that brand also. And if you look at the certification stamps on the Hipro units they are certified by every agency I have ever heard of and then some.

No as far as the accuracy on the PS they claim .2%. Yes I know, I have a little trouble with that claim myself.

But the .2% is stated here on their WEB site, for what it is worth.

http://www.p3international.com/products ... 00-CE.html

I do have some expensive Fluke equipment but I would have to use an extension cable that I would not mind splitting out the wires to get the clamp-on amp meter on.

The best way, but I can't make myself trash a power cable, would be to measure on the DC side. That way we wouldn't have to worry about things like Power Factor and Switching Power Supply noise/frequency.

booyah, what did you use to measure your thin client?

I ran your numbers and I come up with =>

0.12 amps - 120 volts * .12 amps = 14 watts (assuming a PF of 1).
7.0 watts - should be 14 watts.
0.14 kwh - should be .014 kwh

Am I missing something or just having another one of my senior moment?

HolmanGT · Post by **HolmanGT** » Mon Nov 24, 2008 1:26 am

tony wrote:I'm using a retail UMP EM100 power meter. Nothing fancy.

Link http://www.upm-marketing.com/products

I should clarify, the 22 watts is for the laptop with the monitor screen off. The laptop is driving an external monitor with the monitor's consumption not counted in the energy consumed (it's not getting its power from the power meter).

I plugged in the T5700 and in steady state once it's settled down, it's drawing 9 watts.

I did a test of my home office: laptop, 2 external monitors, desk phone, PDA cradle, cable modem, router, Hawkings booster, laptop docking cradle, couple of USB hubs, and probably some misc AC adapters I've overlooked, and the power consumption is 38 watts, steady state.

So lets, see, dropping to 9 watts from 22 (replace slave laptop with TC), I save 13 watts. At $0.0655 per kWh and $117 CDN invested in the TC (excluding my time), my breakeven point when energy saved equals what I spent to save the energy will be 137,405 hrs or 5,725 days or 15.9 yrs (360 days/yr turned on). I will update in 16 years when I hit breakeven. Oh well, must discount for the educational component, that's got to be worth something.

-Tony

tony,

looks like the same type of unit that I am using. But I will check back with you in about 16 years to see if your calculations are accurate.

I irony of all this is in the range of power consumption we are dealing with here most people use that much power and more just to make a pot of coffee, albeit coffee is in the category of required energy consumption. Besides if we stopped buying South American drugs our government would probably have to subsidize their economy and in the long run that would end up costing a hell of a lot more from each of us than one thin client and and the power to run it. No - please don't ask what this has to do with the original topic, I don't have a clue. It might be something induced by some of those Brazilian drugs (I am talking about coffee of course).