Results from running the motor by itself; 2 power meters (KillAWatt = KAW and Intertek = Int)
KAW Int
Power: 199 213 W
PF: 0.24/0.25 -- (could not find this on the Int meter)
VA: 788 -- " "
Volts: 120.9 120.9
Amps: 6.53 7.05
First question: Pls instruct me -- how can I "trust" the Power reading when the PF is 0.25?
Got this from the web don't know if it helps you.
Very few USA residential customers' bills include any charge for power factor, regardless of how low (less than 1.00) their power factor is. this is usually a charge to commercial or industrial classes of consumers. While some household appliances such as an aircon, 'fridge, washing machine, flourescent lights, (and, yeah, a ups) have "poor" power factors, the average for the whole service is generally about 0.8 or 0.75. However, to supply a power factor less than 1.00 requires a greater current by approx. 1/PF. This additional current causes increased wattage (which you are charged for) losses in the wiring, and the increase is by the square of the increases current level. So, you do pay extra for a low power factor. Utilities will add capacitor banks in their systems to improve the system power factor. Customers-not the utility-will add cap banks at their service entrance or at the "offending" appliance (usually a motor) to "correct" their power factor to 1.00 or near 1.00 to avoid power factor charges.
I want to congratulate Gabriel for a device that hums beautifully after a rather tedious re-build! USing skype, he guided me in re-building the "friction heater" today. O0 We saw various mistakes when the mysterious "they" put it back together... kinda funny really.
Putting that aside, Gabriel knows his stuff and guided my hands expertly. Its running! Here's the first data:
1. Starts at about 1150 W and drops to about 930 W after about 12 minutes -- as the oil warmed and viscosity decreased.
2. PF = 0.82 approx. during these first 12-15 minutes.
3. Heat rise roughly 1 deg F per minute in the oil (I stuck a thermocouple probe down through the vent hole, about 3" deep into the oil).
Quote from: PhysicsProf on 2013.03.08, 18:41:54
Results from running the motor by itself; 2 power meters (KillAWatt = KAW and Intertek = Int)
KAW Int
Power: 199 213 W
PF: 0.24/0.25 -- (could not find this on the Int meter)
VA: 788 -- " "
Volts: 120.9 120.9
Amps: 6.53 7.05
First question: Pls instruct me -- how can I "trust" the Power reading when the PF is 0.25?
Your Intertek and KAW meters
read real power in "Watts", not apparent power. They do allow you to calculate apparent power by simply multiplying V x A e.g. 120.9 x 6.53 = 789.48 Watts (apparent power). Now multiply this by the power factor (0.25) and the result is very close (197.44) to the readout of real power, which is 199 as read on the KAW meter. Same applies for the Intertek.
You don't need to do these calculations, as it is built in to the "Watts" readout of the KAW and Intertek meters and can be trusted for mostly sinusoidal waveforms. Priorly, I tested these units for accuracy with phase angle fired "light dimmer" controlling the load and they performed very well with such a chopped waveform.
BTW, On the Intertek, Power Factor is displayed in the lower readout when in amps mode capital "A", third press of the function button. The Intertek does not display "VA" (the KAW does), if needed you multiply Apparent Amps times Volts for VA.
To sum up, just use either device and read out "Watts" which is already considering the PF in it's computation.
You need not concern yourself with VA readings unless you are curious.
Regarding spinning disc wattmeters as used in most older homes ( I believe you have one in your lab), from my tests these also seem to read real power.
I have not tested the newer digital type home power meters.
A "True Grit" Moment there !!
Good to have it up and running!!
Thx
Chet
Fantastic to read that you got it running ! O0
With the discs being out of true and nuts and screws all over the place that was surely no light task.
Also thanks ION, that was good to know about the PF corrections in a KAW meter.
For the digital Smart Meters, the best power correction method is to hit it hard with a hammer and go 100% solar ;D
Quote from: Slider2732 on 2013.03.15, 02:11:16
Fantastic to read that you got it running ! O0
With the discs being out of true and nuts and screws all over the place that was surely no light task.
Also thanks ION, that was good to know about the PF corrections in a KAW meter.
For the digital Smart Meters, the best power correction method is to hit it hard with a hammer and go 100% solar ;D
Yes, thank you, Ion.
Excellent point about the smart meters, slider!
So, today (Thursday that is) I ran side-by-side tests of Gabriel's device and a control (a standard resistance heater w/fan). I had one big cardboard box, but I wanted TWO IDENTICAL boxes -- so I went instead to two identical child's tents - these worked just finel IMO.
But I have more tests to run so not yet done, so hold the phone. Sorry, it just takes a while.
so what was the test out come????
OK... I would use the same tent or box, not 2 different ones. that is what I did.
I know it takes more time but that way everything is the same. ... well for the most part. you all know this will come up and we will have to redo the testing.
1 box or tent
These two tents were identical -- and the crucial feature is that the ambient, the room temperature for instance and humidity and pressure -- these were the same, because the two tents were just ten feet apart in the same room. It is the latter that is most important to be duplicated, once one uses identical tents.
If I go over today and switch tents, then the room conditions, the ambient, differs from yesterday -- the air temperature is different, for example.
Surely you are not saying that the two tents from the same manufacturer at the same price and the same identical design purchased from the same store can differ so much that a comparison using identical tents is invalid -- are you?
OK -- let me refresh our collective memory by quoting the summary by ION from another thread (referenced at the top of the quote):
Quote from: ION on 2013.02.07, 18:12:55
So we have a delta 64 degrees F for the resistive heater vs. delta 70 degrees F for the friction heater or roughly 8.6% gain over resistive.
Friction heater started higher, ended lower in wattage so it is difficult to find the average without a time / temp curve, but we can use 745 Watts for now.
Due to the changes in power used, and difficulty in exact measurement, it might be premature to declare a clear winner but this is a rough COP =1.086
I'm optimistic that with a little tweaking we can perfect our measurement technique. Thus far the results look encouraging as a reasonable test.
The above uses a cardboard box for the testing, and the result is "roughly 8.6% gain over resistive." In thought I recalled more than that, but that was the result.
Roughly 8.6% is not much...Next I post early results from tests here. I just have to provide the data as they come in, and the method and the interpretation can be discussed. I may have made some mistake, or perhaps the re-construction of the device was not actually getting us back to Gabriel's design-parameters. That is quite possible.
For these tests, I ran the DUT (device undergoing testing) and the control in the same room and at the same time; one control test was run right after the first tests were completed. The tests were performed in a large storage room where the ambient air temp was quite constant, at 19.8C (varied in the tenths of a degree).
I used the equilibrium heat-flow to ambient method used before by Chet and Gabriel, advanced IIRC by ION. The method seems sound. The DUT and the Control were placed in identical children's tents for these experiments. I measured input power using two kill-a-watt power meters, identical models, and I previously checked that they agreed by running the DUT and measuring Pin using both meters simultaneously. They agreed within a couple of watts out of about 1100 watts, and generally gave the exact same reading.
The temperature probes were identical and showed the exact same reading for the room temp prior to the tests. The probes were placed in the same location of the tents (to the best of my ability). For the final test, I took the additional precaution of moving the temp probe used for the DUT over to the Control.
Yesterday (14 mar 13) and the day before, I ran the DUT and acquired data.
After running the DUT and letting it warm up the oil to 78C, I placed a tent over the DUT and an identical tent over the resistive heater (control). Note that the PF for the DUT was about 0.82 during these tests; that is accounted for (by the meter, not by hand) in the "watts" power meter reading. With the tenting systems, I found that equilibrium was reached quite quickly -- in about 1.3 hours for the DUT (which had a lot of oil to heat up) and in about 25 minutes for the control. Clearly, more ENERGY was put into the DUT to reach equilibrium; but here we are testing for the Pinput at equilibrium temperature (both the tent-air and the oil reached an equilibrium high-temp), so that heat FLOW was balanced, heat going into the tent stemming from the electrical Pin and = heat flowing out so that the air temp was very close to constant.
DUT results:
Time Watts in Oil-temp C Tent-temp C
19h52 830 75.7 32.3
20h00 791 83.0 33.7
21h01 725 114.5 44.9
21h11 714 117.2 46.7
Here, I turn on the resistive heater for the control. I noticed later that this drops the voltage in the shed, so that the power also drops to both heaters -- but the power input is measured so this is accounted for.
21h16 654 118.7 47.1
21h 21 642 119.7 47.5
21h30 637 120.4 49.9
21h41 " " 49.0
21h43 636 " 48.8
Power off. I noticed that the tent-air temp immediately rose to 56C (due to the hot oil) and then began dropping slowly (as the DUT cooled). The DUT motor has a fan which pulls air down from above, so this cools the tent air and this fact needs to be considered. For comparison with the control, I suggest using the max ambient temp of 56C for the DUT at input power 637W.
Resistive heater control warms up very quickly (no oil to heat) and reaches equilibrium temp quickly:
Watts In Tent-temp C
21h16 704 63.
21h 21 705 64.5
21h30 709 64.8
21h41 710 64
21h43 709 64
Final test, I plugged the resistive heater (still in the tent) into the KAW meter and then plugged that into a Variac. I used the temp probe from the DUT tent, moved it over for this final test. After about 20 minutes,
Watts In Tent-temp C
638 57
630 55
So the control provides reasonable agreement with the DUT which showed the max ambient temp of 56C for the DUT at (immediately after) input power 637W.
If the DUT is producing "anomalous heat", then we may see an effect in the rise of temperature of the oil in the vessel as the device is running.
Basically for this test I'm using the oil-filled vessel itself as a crude calorimeter. Note that the heat loss in the motor is not considered, but its efficiency is assumed to be at least 60%. Also, heat lost to and through the steel is not considered -- just the heating of the oil itself. Given these caveats, if the oil-heat-Q exceeds or even approaches the Einput-electrical, then this approach may show something interesting. So I went ahead with it. Much easier than dipping the whole DUT in water!
There are nearly 2 gallons soybean oil = 15.36 lbs = 7.0 Kg. I started with two gallons and measured the weight of that NOT used = 774g. So the mass in the container is 7,000g - 774g = 6226g. The specific heat for soybean oil is 1.97, so the equation for Qheating-oil is:
Oil Q = 1.97 * 6.23kg * delta-Temp (result in KJ)
I think I will forego a tedious table of data -- available in my logbook.
The results:
Test 1 -- 11:34-11:47, delta-T = 69.1C - 48C = 21.1C in 13 minutes.
Oil Q = 1.97 * 6.23kg * 21.1C = 258.8 KJ
Input energy approx. Ein = 1127W *13min * 60 = 879KJ
cop ~ 30%
Test 2 -- 11:50 - 11:59, delta-T = 11.5C in 9 minutes
Oil Q = 1.97 * 6.23kg * 11.5C = 141 KJ
Input energy approx. Ein = 927W *9min * 60 = 500.6KJ
cop ~ 28%
Nothing to write home about. Note from the other thread that I had cop (just for the oil heating) ~ 50% with a blender beating soybean oil; there the blender container was an insulator, plastic.
Personal note, sorry about the results as they stand, Gabriel.
Your best data IMO come from the fact that you have used the DUT for home heating, with lowering of your heating bills.
It may be that the device was altered in some way -- during dis-assembly or re-assembly.
Steve
Good work so far, Steve
Using a strict calorimetric measurement method you need to also consider the mass of the steel and it's specific heat, as it is an additional heatsink to the oil.
Also losses to ambient need to be considered since the container is uninsulated it will skew measurements.
If you can insulate the vessel and include the mass and specific heat of steel, the COP will rise over the present readings.
In the end this may become difficult to capture all losses including the motor, and an insulated vessel may rise in temperature too quickly and may become dangerously overheated.
IMHO use the box method, which allows the device to radiate normally and allows for all heat sources to be tallied. Then using the same box, run the test using the electric heater. While not an exact method it is an worthwhile COP comparative method if run to saturation.
Agreed that the box method is the better test. The steel does not rise in temperature as fast as does the oil; and that method has the caveats I mentioned.
Will relent and run electric heater separately in the exact-same box, replacing the current box with its twin. But the room-air conditions are (as I indicated before) different from before.
Quote from: PhysicsProf on 2013.03.15, 19:31:18
Agreed that the box method is the better test. The steel does not rise in temperature as fast as does the oil; and that method has the caveats I mentioned.
Will relent and run electric heater separately in the exact-same box, replacing the current box with its twin. But the room-air conditions are (as I indicated before) different from before.
Ambient air changing should not present too much of a problem as you are recording ambient and considering rise over ambient for each run.
The box will represent a fixed thermal resistance to ambient. The simple model is a voltage source (heat generating device) feeding the resistive divider (thermal paths) to a second voltage sink (ambient).
Did you say the temp rise is only 1deg F per min?
As you seen in my videos it should be in the likes of 6 to 7 F per min.
I can only speculate why my numbers are so much different then yours. I clearly need to get my hands around the heater.
One more thing the oil temp in the friction heater should be over 200F before you start testing, Like we talked about on the phone that it needs to run.
But it should be rising in temp much faster.
Quote from: Oilpiggy on 2013.03.15, 20:23:47
Did you say the temp rise is only 1deg F per min?
As you seen in my videos it should be in the likes of 6 to 7 F per min.
I can only speculate why my numbers are so much different then yours. I clearly need to get my hands around the heater.
Please note for example the data posted:
QuoteTest 1 -- 11:34-11:47, delta-T = 69.1C - 48C = 21.1C in 13 minutes.
That is 21.1C/13min = 1.6 degC per minute = 2.9 deg F per minute, not 1 degF per minute, but still less than "the likes of 6 to 7 F per min."The ~3 deg-F rise per minute is when the oil is rising from 48 to 69.1 C; that is, from 118F to 156F, the rise rate slows down as the oil gets hotter.[/b]
Quote from: Oilpiggy on 2013.03.15, 20:27:49
One more thing the oil temp in the friction heater should be over 200F before you start testing, Like we talked about on the phone that it needs to run.
But it should be rising in temp much faster.
The testing for the equilibrium temperature was done (see my post above) when the oil temp in the heater was approx
120.4 deg C = 249 deg F, well above the 200F you mention above.
OK -- I ran the control run (resistance heater) at essentially the same power input (635 +/-1) watts as I had run the DUT last evening:
http://www.youtube.com/watch?v=NqZPXWQuU_4&feature=youtu.be
QuoteBrief description of a control run, to compare with my tests of an unusual spinning-disk heater. I'm using a constant-heat-flow-to-ambient method, which allows the device to radiate normally and allows for all heat sources to be tallied. The tent rather quickly reaches an equilibrium temperature at which the heat flow in (from the entire device) matches the heat flow out to the room. First I ran the device shown at the end of the vid. Then using the same box, I'm running a control using a simple electric heater. Input electrical power measured with the same power meter; also same tent and same temperature probe -- and same input power as measured on the power meter (635 +/-1 Watts).
The temp reached by the control is about 54.4C, or 54.4-16.8 C = 37.6C above the ambient room temp 16.8C.
Yesterday the temp reached by the DUT was about 56 C, or 56-19.8 C = 36.2C above the ambient room temp of 19.8 -- in the same tent, same temp probe, same power meter for both runs.
Quote from: Oilpiggy on 2013.03.15, 20:23:47
Did you say the temp rise is only 1deg F per min?
As you seen in my videos it should be in the likes of 6 to 7 F per min.
I can only speculate why my numbers are so much different then yours. I clearly need to get my hands around the heater.
Do you have a record, or vid, that gives the actual numerical temperature rise per minute when you were running your device?The scientist in me just likes to see a record. Maybe this device is just not "worth" further testing? -- that
6-7 F per min rise is a first and straightforward test!
Yea its on my YouTube channel, or it should be.
As far as anymore testing goes, that is up to you.
BTW, the tent was a good idea, much better then the box.
If you get time give me a call.
Thank you
Gabriel Peterson
Consider that a "new" friction heater will load the motor and more of the horsepower will be converted to heat through friction. As the device "breaks in" the surfaces of the metal plates become more polished and less turbulence results in less friction.
Automotive engines used to have a break in period whereby it was not recommended to run at high rpm's for extended periods. This would allow metallic wear surfaces to become finely polished, thus reducing excess friction. Failure to observe these precautions could lead to catastrophic failure of certain wear components such as sleeve bearings or piston rings or walls.
Without accurate data of power input and heat output for extended periods of operation, we can only guess at what the break in period would be for such a friction heater. This would also probably depend on the lubricating properties of the oil used, especially at higher temperatures.
This is a fairly exact science for internal combustion engines and rotating machinery in general. I am not skilled at these calculations but we should be able to research the topic for more insight.
I submit that the reduced temperature rise may be due in part to this factor. Only repetitive testing using the same method would allow compiling enough data to plot such a curve.
I'm sure Chet could add a few lines to this, as he has had aircraft maintenance experience
ION
Yes there are many factors to consider,I did speak a bit with Steve About the Control runs I did to simulate
Gabriel's device last Month [pryor to shipping]
I had done many tests over several days with resistive heaters in Oil [as per his spec and volume ]
at 500 watt 750 watt and 1000 watt inputs ,static tests and stirring.
I put 30 ilbs of steel inside the stainless test vessel to simulate his steel mass as well.
all tests did fall short of Gabriels Marks [not by much].
I am out of town for a few,when I get back I'll pull the data ,Steve was most interested in the Time to Heat vrs input Curve [oil]
Thx
Chet
Thanks, Chet.
I did one more quick control test today, room air still 16.8 C. I switched the resistive heater to 1500 W; wanted to see what it would equilibrate to. Actually pulled 1358W per the KAWatt meter, and the temp reached in the tent was 94.4 degC = 202F.
With 634 W, the same device reached (equilibrated at) 54.4 degC = 130F. So, yes, the equilibrium temp rises markedly with input power, as expected.
DUT was tested once more before sending for repairs. Gabriel had noted that the oil had to be about 80 F before the electric motor could handle it. (Viscosity down as temp up). I found it would not start (keep the electric motor going before auto-shutoff) until 101F! Suggests internal problems. Further, At that temp, the max heating was about 3.4 deg-f, dropping to about 1.5 deg-F/min around 200F (oil temp). Much less than Gabriel's 6-7degF/min heating at roughly the same input watts.
So it needs repairing; hope Gabriel is not getting discouraged! Will get another chance near future, let us hope.