How much are our switches costing us?

[BraveSirRobbin]

I think the answer to the question I'm about to ask is 'yes', but it would be interesting to see if the current/power those switches used was a 'linear' addition to the overall load power when the lighting was on. In other words, what would be interesting is to take a light load bank without a switch in place, see what the power draw was, then put the switch in place and see if the 'switch' load linearly adds to the overall load.

The reason for this interest is in calculating overall cost. If the load wasn't adding linearly, then you would have to account for a reduced cost value of the switches' draw during the time your lighting was on.

Sure it's probably insignificant, just mentioning it out of curiosity.

 

[apostolakisl]

At 40 cents per KWH you should absolutely look at adding solar. I did all of the calculations two years ago when I built my house. The city and feds were going to pony up about half the cash of install. I then looked at the cost of money (interest rates) and cost of electricity (about 11 cents kwh) and determined that it was a break even. Since I couldn't really figure out where to put the panels that wouldn't be ugly, I skipped it. Since that time the cost of money has dropped about 1 percent and the cost of electricity has stayed the same.

At 4 times my electric rates, you should be able to come out ahead even without subsidies. I did also take into account the fact that we get good sun here in Texas. If your hours of strong sun aren't as good, it will be different. My cost of install was going to be something like $15 k (30 total). I don't recall the production of that setup. It may be that in 2 years the panels are more effecient and or less costly.

I have 5 Insteon switches coming in the next few days. I have been wanting to wire up a bunch of those in parallel and do what you did. The best test would be to disconnect the house and wire them straight to the actual meter. Whether this is more accurate than a kill a watt. . . ? It is definitely the most accurate when you are talking about the bill. However, I think my wife would kill me if I told here I was going to shut the electricity off for 12 hours.

 

[drozwood90]

So...don't tell her?

 

[tmbrown97]

Now you guys are going to get me going... if you're REALLY curious, I can do some real-world comparisons, and maybe take a switch or two and hook them up to a light-socket and compare the difference. I think I have more 1140's than I'll install this weekend, so whatever's left I may be able to do some experiments with.

Lou - I'm definitely interested in doing solar - but we're working a few things with our finances right now, and it's just not the time at the moment. When we do decide, I'm in Northern California - we get an a LOT of sun and very hot days too - so it'd be nice to offset those horrible August A/C bills.

 

[jaysonc]

So 0.8W per switch seems ok to me as well, but here is a question.

How much power do the switches pull if your turn off the LEDs? (Set them to Always Off on the options tab)

 

[apostolakisl]

Radio shack sells some standard LED bulbs that they spec at 30ma 2v. That's like nothing in watts (.06). I doubt shutting off the leds will make much difference. Maybe a .1 watts? But if you can measure it directly, go for it. I suppose that could amount to 10% of the total use, however, if total use is .9 watts.

 

[tmbrown97]

OK - so I'm revisiting.

I'll skip straight to the spoilers, knowing I type a lot...

• There is no real measurable difference with the LED's on or off - even when running 10 switches to increase the measurable load.
• There is no measurable difference between the 1140's and the 240's.

I rigged up the 4 US240's that I received today, along with a single 1140 - in the exact same rig as yesterday with all 1140's. There was no difference in power consumption.

So, I decided to go bigger, and just do all 10 together. Here are the readings I got for all 10 switches combined:

• 100VA
• 6 Watts
• .82 Amps
• Power Factor .06

It hasn't run long enoguh to get stats on cost for a year - but it was looking exactly the same across the board, like we'd expect.

I did take the time to program every switch and test them with Green LED, Blue LED, and No LED. No difference between Green and Blue; and with No LED, the difference was barely measurable, and only in the VA measure - it dropped down to 98.

So - 10 switches is close to where a lot of smaller installs will be; 60 like I'll have is definitely on the high-end - and my electric rates are astronomical; so if it's costing me $107.82/year, you're almost guaranteed to be way lower - a buck or two - maybe 5 max per month; which is hopefully offset by the savings they offer by keeping the lights dimmed and keeping them from being left on.

Dan was saying he's seen tests of other switches (and technologies) being much higher. All I can say is, the more time I spend working with the SAI switches, the more I'm glad I standardized on them. If there's anything else I should test, let me know... if all goes well, a lot of these will be installed tomorrow.

 

[apostolakisl]

Great work with the 10 switches. By my calc, you are looking at 5.256 kwh's of energy per swtich per year. For you at $.42 per kwh (you should move! ) that's $2.21 per year per switch. For most other people it is going to be between 1/4 and 1/2 that cost.

Now if someone would do that for Insteon. Like I said, I have 5 switches on the way. I just don't have a kill-a-watt.

 

[nov0798]

Remember Watts=volts * amps only for DC power not AC power.


Im not sure this statement is correct. I thought Ohm law is the same for both AC, and DC! If its not then Ive been doing my AC calculations wrong for many many years.

Example:
If I have a heater rated at 1500 watts, and the line voltage is 120AC, then the amps would be 12.5A.

1500 / 120 = 12.5

If I have the same heater, and need to find the wattage, but only know the line voltage(120), and the Amps(12.5) then.

120V * 12.5A = 1500W

 

[Photon]

Remember Watts=volts * amps only for DC power not AC power.


Im not sure this statement is correct. I thought Ohm law is the same for both AC, and DC! If its not then Ive been doing my AC calculations wrong for many many years.

Example:
If I have a heater rated at 1500 watts, and the line voltage is 120AC, then the amps would be 12.5A.

1500 / 120 = 12.5

If I have the same heater, and need to find the wattage, but only know the line voltage(120), and the Amps(12.5) then.

120V * 12.5A = 1500W

It's the power factor thing that confuses the issue. The heater in your example is a purely resistive load, so PF doesn't come into play. If you have a load with capacitive and/or inductive reactance, then the magic kicks in.

 

[apostolakisl]

Remember Watts=volts * amps only for DC power not AC power.


Im not sure this statement is correct. I thought Ohm law is the same for both AC, and DC! If its not then Ive been doing my AC calculations wrong for many many years.

Example:
If I have a heater rated at 1500 watts, and the line voltage is 120AC, then the amps would be 12.5A.

1500 / 120 = 12.5

If I have the same heater, and need to find the wattage, but only know the line voltage(120), and the Amps(12.5) then.

120V * 12.5A = 1500W

You are correct. P=VI applies to AC and DC. The trick with AC is that the appearant V (what you measure with a multimeter) may not be the utilized V. Power factor corrects for utilization that does not match the Vrms (root mean square). With AC current the voltage is constantly changing from negative 165 v to positive 165 or so. The listed voltage is an average of that calculated by the rms method. When a pure resistive load is put across that voltage then rms is the actual utilized voltage. But inductive loads don't draw evenly across the sine wave resulting in a utilized voltage of something less than the rms. The amount less is defined as the power factor where PF x 100% = the percentage of rms utilized. So a PF of .5 means that utilized voltage is 50% of rms.

So two devices could be plugged into the same 120 outlet in your house and drawing the same power (watts) but be drawing different volts and current to get to that power. The volts and amps can be changed just so long as the product of those two equals the same watts.

The trouble with low PF devices (as far as the electric co is concerned) is that it messes with the current and volt balancing on the system. The electric co needs to maintain a proper sine wave with a proper Vrms because that is what its customers expect and require for their equipment to work properly.

 

[IndyMike]

Hi Lou,

Unfortunately I've lost a bit of data on this subject due to a server crash (mirrored drives don't help much with a controller failure - we won't go there).

I put the following together based on a series of posts I had on other forums. The one column highlighted "yellow" is highly suspect (outlier). In contrast to Work2Play's method of averaging multiple units, I used long term tests to average the consumption and improve the accuracy. I did have an accuracy analysis that accompanied this, but that's on the server drive(s).

I'm running a "long term" test on a ApplianceLinc now. If there is a device you're interested in, let me know. I'm gearing up far a basement install and may have something on the shelf.

Updated:

1) My Appliancelinc finished up at 1.11 W after 99 hours of accumulated measurement. Based on the display resolution alone this measurement is roughly +/- 5%.

2) The real surprise is the V5.15 SL dimmer - it doesn't register on the Kill A Watt at all. I substituted many other devices (some that registered at .3W) to make myself believe the Kill A Watt was still functional. Based on the display resolution, I'll need roughly 240 hours before anything will register on the kWh meter - not sure I'm that patient. The switch itself is completely operational (led's on - I even linked to it with my ISY). The switch body feels dead cold - sanity check.

As much as I would love to believe that SH made a "significant" reduction in power consumption, I'm having troubles with the degree of improvement. Could someone else verify these measurements?

IM

 

[apostolakisl]

Hi Lou,

Unfortunately I've lost a bit of data on this subject due to a server crash (mirrored drives don't help much with a controller failure - we won't go there).

I put the following together based on a series of posts I had on other forums. The one column highlighted "yellow" is highly suspect (outlier). In contrast to Work2Play's method of averaging multiple units, I used long term tests to average the consumption and improve the accuracy. I did have an accuracy analysis that accompanied this, but that's on the server drive(s).

I'm running a "long term" test on a ApplianceLinc now. If there is a device you're interested in, let me know. I'm gearing up far a basement install and may have something on the shelf.

Mike,

You are quite the organized man. Very impressive stuff.

The 2476d is probably the most common switch used on the Insteon world. I would be curious to know what they consume. It would also be interesting to know if the new ones use the same juice as the old ones.

Lou

 

[IndyMike]

Lou,

I have a 2476D V5.15 (date code 1024) sitting in front of me. I'll put it in the "on deck" position.

 

[apostolakisl]

Good deal. They are shipping v 5.something presently, could be 5.15. Everything in my house is now less than 2 years old as I went through and replaced everything older than v 3.2 (I think that is the number) for the paddle issue. It was a lot of work but at least my 6 year old system is now only about 1 year old on average. So far, everything is functioning perfectly. I am praying that Insteon really does have the QC issues behind them. It seems to be the case so far.