"Friction Heater" by Gabriel

[PhysicsProf]

Well
I Passed out ......

Seems like the rest of the crew Too

Even Steve seems to have run out of lead in his pencil?

So Steve.....Whats the oil Q

2 gallons soybean oil = 15.36 lbs = 7.0 Kg.  The specific heat for soybean oil is 1.97, so the

Oil Q = 1.97 * 7.0Kg * 6.1 degC = 84.1 KJ

Add to Qwater-heating:

Qliquids = 2206 + 84 = 2290KJ

The efficiency, electrical energy to Qliquids, is
2290/2900 = 79%, which is reasonable and a good start also.

[Chet K]

Steve
My Oil was 30 W {Recycled}
The Oil I tryed to buy for the run was going to be almost 40 dollars .
One thing is certain Gabriels device Gets The heat out to the environment much better than the smooth drum I am running. [yes I know]

I am picking up supplies today for a few projects ,If I get a chance I will stop By a friends who has scrap  Elec motors [very large] with cooling fins built in.
Perhaps a large case with a Gabriel Disc stack inside will be the next Test !

Thx
Chet

[Gabriel]

I ran the my Friction heater while I was sleeping with the same setup for the test runs. ( same oil, same 45 liters of water, same 55 gallon drum )
Run time was 8 hours 17 minuets
Water start temp:           17.777 C
Water end temp:            44.444 C
Total KWHs:                   7.500
start ambient air temp:    11.666
end ambient air temp:     13.333
Watts start:                   1390- 1400
Watts end                     850- 825

{I did not check the oil temps in side the core at start and stop.}

Thank you
Gabriel

[Gabriel]

Steve
I am picking up supplies today for a few projects ,If I get a chance I will stop By a friends who has scrap  Elec motors [very large] with cooling fins built in.
Perhaps a large case with a Gabriel Disc stack inside will be the next Test !

Thx
Chet

Chet,

If you are going to use a motor case you will have to make sure that the ID will match the OD of the static discs. In other words you can not have oil passing by the out sides of the discs.
The motors I have around the house with cooling fins on them are cast Iron and if you ask me not a really good way to dissipate heat, more made that way for the cost of production.
Is there a local metal  Fab shop around you? If so, look in there scrap bin you will be surprised what they recycle, and they sell it cheap. I get a lot of my metal supplies that way.
To have a 8" OD roll up is cheep as well.

I would still like to see the test done with just the emergent heater in the 45 liters of water.

Thank you
Gabriel

[Chet K]

Gabriel
Thx for the cylinder tolerance info!


In my other testing on Heaters ,it is routine to run resiistive controls ,I only work in water ![so far   ]

However, as you wish..........

As Steve says all data is good.

My gut tells me it will be very telling ! [one liter water raised 120 F with 0.1 KWH input is my line in the sand for the happy dance !}

I will start in ten minutes [I just emptied that Vessel

I'm going to do a static no stir test first and then setup a stir test [automated].

Keep you posted.

Chet

** 45 liters resistive heat   water only     special request test
Ambient 12 C
water    11 C

Input 1020  [odd last nights was 978??]
Input down to 945 W  at 0.14 KWH [volts 115 were 119 at start up]

This experimenter having "compulsive stirring issues" [so far I have resisted..........?]i

[Chet K]

Data for 45 liter 0.8 KWH input resistive heater [950 Watt ]  Water only  immersion test. [no stir]
Ambient  start  12C         stop      12 C [big shop  ]
Water     start  11 C        stop      22 C

Run time 50 Minutes

[PhysicsProf]

I ran the my Friction heater while I was sleeping with the same setup for the test runs. ( same oil, same 45 liters of water, same 55 gallon drum )
Run time was 8 hours 17 minuets
Water start temp:           17.777 C
Water end temp:            44.444 C
Total KWHs:                   7.500
start ambient air temp:    11.666
end ambient air temp:     13.333
Watts start:                   1390- 1400
Watts end                     850- 825

{I did not check the oil temps in side the core at start and stop.}

Thank you
Gabriel

Glad to help out (was gone all afternoon helping others; sorry for delay).



Note yesterday's numbers for comparison:

0.8 KWh input energy = 2900 KJ.

Output heat picked up by the water:

Q = 4.19 *45kg *11.7DegC = 2,206 KJ output to water only.

The efficiency just to the water is 2206/2900 = 76%, which is reasonable and a good start!

Today you have 7.5 KWh input, over 9 times the input 0.8KWh yesterday; same 45 liters of water; but temp rise is less than 3 times that of yesterday (not 9 times) -- so efficiency is way less.  Here are the numbers:

7.50 KWh input energy = 27,000 KJ.

Output heat picked up by the water:  (temp rise 26.7 degC)

Q = 4.19 *45kg *26.7DegC = 5,034 KJ output to water only.

The efficiency just to the water is  = 5,034/27,000 = 19% much less than yesterday's result...  Probably a lot of loss of heat to the air/environment during the long run.

[PhysicsProf]

Data for 45 liter 0.8 KWH input resistive heater [950 Watt ]  Water only  immersion test. [no stir]
Ambient  start  12C         stop      12 C [big shop  ]
Water     start  11 C        stop      22 C

Run time 50 Minutes

OK, Chet, let's see:

0.8 KWh input energy = 2900 KJ.

Output heat picked up by the water:

Q = 4.19 *45kg *11DegC = 2,206 KJ output to water only.  Yesterday the temp rise was 11.7degC... today 11 degC, so the eff will be less today:

The efficiency, energy  to heat the water is 2206 /2900 = 72%.

(Why not stir?)

Both of you -- please consider adding INSULATION around the water vessel, such as bubble-wrap.

[Chet K]

Steve
Last nights Oil and water run [which you already analyzed]

Repost here:

Open top run ,constant hand stirring through out test [both Oil and water]

45 liters
Start
Start TEMP       water 7 C    Oil 23.6 C        Ambient 14 C
                   0.1 KWH 8C          35C
                   0.2         10.1        43C
                   0.3         13           53
                   0.4         13.7        58
                   0,5         13.9        58.6
                   0.6         14.1        62.1
                   0.7         15.3        63
                   0.8         16.6        63.8

Ambient Temp  14C
total run time   50 minutes
total input     0.8 KWH
watt draw  978 W

Today Gabriel asked me to do a Water only run with the same conditions
I raised the water  11 C today
Yesterday I raised the 45 liters water 9.6 C plus raised 2 gals  of oil 40 C at the same time?
and my Eff was higher than today?

Just seems Odd to me , I expected much more of a temp rise?
I suppose as you say the loss to ambient must be huge?{same ambient temp as yesterday though??]
No stirring ?

Thx
Chet

[PhysicsProf]

Chet,
I think you are losing heat mainly to the ambient environment, yes.  There may also be smaller measurement-accuracy issues (much smaller I expect).

The way to test that HYPOTHESIS, insulate your vessel and run again.

To get a better handle on these issues, I ran this evening two simple tests, using a PLASTIC vessel for my new (cheap) Osterizer as the "calorimeter vessel", bought for the purpose of such experimentation.


Test 1:  300 W immersion heater in 1600 ml water, input energy = 10 Watt-hours (0.010 KWh) = 36,000J.

Temp rise from 14.6C to 19.8C measured with my TK probe (thermocouple type) = 5.2C temp increase.

Then QtoWater = 4.19 * 1600g * 5.2C = 34,860J

And Eff = 34860/36000 = 97%.  Very reasonable, and indicates low losses to the environment over a few minute run.

(That's another thing you can do Chet -- reduce the run time, to reduce losses to the environment)


Test 2:  Using the blender itself, the spinning blades, as a "friction heater."  Why not, as a test?
As before,  1600 ml water, input energy = 10 Watt-hours (0.010 KWh) = 36,000J.


Temp rise from  19.8C to 22.5C measured with my TK probe (thermocouple type) = 2.7 C temp increase.
Running at high+blend speed (quite high for this blender), just in water.  Around 300W so run length about the same as with the immersion heater "control".  (This is also a type of control run, using the blender to convert electrical to mechanical energy, then mechanical energy to heating the water.)

Then QtoWater = 4.19 * 1600g * 2.7C =  18,100J

And Eff = 18100/36000 = 50%.   (See how this works?  its easy really.)

I think a 50% efficiency in heating the water with the spinning blades is reasonable -- here most the "lost heat" is lost in the motor getting hot, I think.

Would attaching Gabriel's disks to this motor and running (in water, or in oil) give higher yields??


Can we see how a SIMPLE device such as this, with Gabriel's disks in place of the blades and other needed modifications, could produce a conclusive demonstration?  The simpler the better IMHO.

Note that Gabriel's device yesterday showed a calculated Eff of about 79%, which is higher than the simple "spinning blades" device (a blender) which I tested, eff ~ 50%.


I may try heating vegetable oil instead of water tomorrow, and do the same two tests as above...
Lots of fun...  we're learning as we go and that's great!

[Chet K]

Summary
Gabriel test [post 44] input 2900 KJ    Total Transfer    2260   Eff   79.%  

Control Test[post43]  input 2900KJ     Total transfer     2364  Eff    79.05%

Steve please Check
Thx
Chet

[PhysicsProf]

OK -- thanks for the data, Chet!  Let's look just at the water-calorimetry part for starters, see how we do.
0.8 KWh input energy = 2900 KJ.

Output heat picked up by the water:

Q = 4.19 * 45,000g (H2O) * (16.6 - 7)C  -4.19*45kg *9.6DegC = 1810KJ output to water only.

The efficiency just to the water is 1810/2900 = 62%, which is reasonable and a good start!

Perhaps fins on the oil vessel would be good...  But a comparison with the Gabriel device can be made also!

Chet, your oil (post 43) went from 23.6 to 63.8 C = temp rise of 40.2 C.

2 gallons oil = about 7 Kg.  The specific heat for the oil is approx 2, so the

Oil Q = 2* 7.0Kg * 40.2 degC = 563 KJ

To water, there was 1810KJ in your test (my post 45), so total =2373 kJ
and
Eff = 2363/2900 = 82%.

Which is close to your 79% (I may have approximated something in my calc?) -- the agreement is good.

[PhysicsProf]

Summary
Gabriel test [post 44] input 2900 KJ    Total Transfer    2260   Eff   79.%  

Control Test[post43]  input 2900KJ     Total transfer     2364  Eff    79.05%

Steve please Check
Thx
Chet

In summary, my calcs show about the same as your summary above, with

Gabriel's test (post 44) Eff 79%  
and
Chet's control (post 43) Eff 82%.

Note that my control with immersion heater in water in a plastic blender-vessel showed Eff = 97%.
My test with the blender blades heating the water in the same vessel, showed Eff = 50%.

[Chet K]

Steve

The heat loss at the motor is not in this figure [could be as high as 50% of input ],The real point here is the immersion heaters have been shown to run
at 97% .
So we know what they do,regardless of leaky vessels and such ,if the parameters are similar between the two tests
[in this case 3000 miles apart],this test will give a good indication of Efficiency in Gabriel s device down in the oil [W/out Motor calc].

If we were testing the efficiency of our vessels ability to retain heat ,        well... that's another test.

All things considered
Good stuff
Thx
Chet

[Chet K]

Well
It has been suggested that I collect more info ...
The surface area of all parts at heat transfer points between water and oil ...between air and water.
there will also be a point where there are loss's at the oil /air interface [top of oil vessel].

to underestimate these heat transfer points and their ability to effect the calibration between these two runs would be unwise
in the long run.
That being said ,I will do a more precise mock up    even the water  vessel material ["control" was plastic, Gabriel "steel" drum]
will be the same.
The MFG info on the motor will be needed for Calcs on motor Loss to heat?

If it is decided that there is a big disparity here I will do another Run for a better calibrated Control]

The run will include "SLOW"spinning stir disc of similar mass to Gabriels device in the oil [drive mechanism will not be part of Caloric measurements].

Comments welcome [as always]
thx
Chet

[PhysicsProf]

Chet asked me to analyze these older data, which I will do just looking at the heating of the 2 gallons of oil:

Some very Raw Data

The device run was started at 90F                           Control [  500 watt resistive
all runs 0.1KWH input on EZ Killowatt meter        =     Same as device
oil volume 2 gals Type 304 SS vessel                 =    Same as Device +30 lbs steel
Start    90F    Stop  110     elapse time 5 min               start  47       stop 66
start    110     stop   135      5 min 40 sec                             66             91 [little time Prob]
Start   135     stop   160      6 min 30 sec                             91            108
start    157     stop   187      7 min 15 sec                            108           125
start    180     stop   206     6 min 50 sec                             125           135
start    201     stop   231     5 min 40 sec                             135           150
start    225     stop   260     7 min 10 sec                             150           167
start    250     stop   283     8 min 15 sec

total temp rise 197 F                                                total temp rise  120 F
total input 0.8 KWH                                                  Total input 0.8 KWH                                            
total time  51 min                                                     Total time 97 min
Ambient Temp 54F                                                 Ambient Temp 36 F  
           

197F rise = 109.4 C rise

0.8 KWh = 0.8* 3600KJ = 2880 KJ input energy

Output with device for the above 51 minute run, looking only at the oil:

Q = 1.97 * 7kg * 109.4C = 1,509 KJ

Eff = 1509/2880 = 52%

The control had a temp rise of 120F = 66.7 C

Q = 1.97 * 7kg * 66.7C = 920 KJ

Eff = 920/2880 = 32%.

INTERESTING.  These data suggest the following --
1.  Pls go back to this basic set-up
2.  Insulate the vessel so that it does NOT lose heat to the environment (as little as possible) -- eg., wrap it in fiberglass insulation if you have some.
3.  Re-run the CONTROL run.
If you can at least DOUBLE  the observed efficiency, conversion of electrical to heating the oil, to 64% with the immersion heater, or so, then you are ready to re-run the device in the same container.
4.  Re-run the device, and pls let me/us know the numbers, that is the temperature rise for both the control and the device.

(This test is not quite definitive, due to conditions of the disks present in one case and not the other, but it will be progress -- I hope encouraging progress.)

[Gabriel]

yes.... will do butttttt

The Powers that be { the wife }  wish for the heater to be back into the house. Happy wife = Happy husband 
Like the bible says " so let it be written so let it be done "  It's back in the house 
I am sure after a week or so I could withdraw the heater to the shop once more for some more testing.

I seen that you sent me an Email at my other address. Use Frictionheater@hotmail.com

Thank you
Gabriel

[Chet K]

Gabriel,
The is strong consensus that a fixed loss to ambient run needs to be done,I am setting up a 1000 watt and 1500 watt control [either or?]
in a large box out side on an extra cold night [actually will be enclosed in my outside shop free from wind ]
I will see if we can get the start temp "hot" and proceed from there.[I will ask?]
this way you can just take the unit outside with out having to shut down the heater and cool your whole house down.

It should not take long ,you may need to pick up another thermometer for the box too

I will see how that goes and get the info to you,we may find with the temp in the teens a simple fridge size box will do?

Thanks for all you are doing here !!
Chet

[Gabriel]

This will be a week or so on my end before I will do this. How big of a box are you going to be testing? I have a 1500 watts space heater I could run both tests on my end if you like?
I just picked up an old dryer box that is  31"x28"x47" tall.

[PhysicsProf]

The "big box" method is meaningful, when done in comparison to a control.

Meanwhile, I've replaced the water in my vessel with soybean oil and learned a few things that may be helpful as we go forward.

I used my 300W immersion heater immersed in the oil, stirring it with the heater while adding 10 Watt-hours = 36,000 J.
Start temp of 1600ml (1472g) soybean oil = 19.0C.

What I noticed when finished was that -despite some stirring - there was a LARGE temperature gradient between top and bottom of the oil (in the plastic vessel):
Top:  38.4C
Bottom: 21.1C

Taking these at face value (incorrect!) but for instruction purposes, we would find that the heat out Q and efficiency are:
Top:  38.4C -19C = 19.4C rise  =>  56260J out =Q  => Eff = 156% !
Bottom: 21.1C -19C = 2.1 C rise =>  6090 J output =>  Eff = 17%

A huge difference, and neither is correct!  so we see a CAVEAT when measuring temps especially in oil -- a large gradient of temp is found in the oil.

Note:  I performed several runs like this getting "strange" high and low Eff values, until I realized that the result was depending on WHERE I position the Thermocouple temp probe.  Clearly I needed to do thorough stirring before temp measuring, and still check temp top and bottom.

So then I quickly stirred using a plastic knife with motion up to down and back up, trying to mix the oil thoroughly.
Now the temps were about the same, 30.5C top and 30.3C bottom.

Then we find:
Stirred temp:  30.4C
Stirred results:  30.4-19C = 11.4C  so Q = 1.97 * 1472g * 11.4C = 33060 J out
Eff = 33060/36000 = 92%, which is reasonable for the control and shows how to do it (IMO).  

[Chet K]

Fixed loss to Ambient test happening here tomorrow for the Gabriel device [his data already in]
"Control" being assembled in AM for afternoon test.
Thx
Chet

[PhysicsProf]

Great, Gabriel and Chet!

It would be best to measure the power consumed by the control (and by the device separately), and not just assume that a 1000W heater is actually pulling 1000W.  That should be easy enough.

Also, say the device draws 800W and the box reaches a steady temp of 60F (for example).  It still would be important to circulate the air inside with a fan, so that the temp is fairly uniform and the temp probe provides a decent sampling of the temp.

Then, ideally, one would run a "control" with a regular resistive heater such that it reached (about) the SAME temp as the device reached at equilibrium (60F in this example).  This can be done by varying the power of the heater.  Hopefully a heater will have settings, e.g., 750W, 1000W, 1250W, 1500 W -- one still needs a power meter to measure the input power accurately.  To get finer input-power control, one can use a Variac and plug the power meter directly into the variac (I've tested accuracy down to about 70VAC on the Variac output -works fine).

To continue the example, if the resistive heater requires 1200W to reach the same equilibrium temp in the big box as the device at 800W, then you really have something!

Hope these ideas help.
Looking forward to results.

[Chet K]

Running a tad behind on Control ATM [my 750 Watt control run]

Gabriel had a 4C disparity,   44C vs 48 C on his control  experiment

Device input 745 Watts temp 44C [max against ambient]

he used a 1500 watt heater [all he has] for control same box same ambient

1500 watts input 48C [max against ambient].

I have been running behind over here [getting bits and pieces shipped out for the other projects]

Later
Chet

[Gabriel]

This is what I am going to to do.... I cant find a 800 watt heater but they do sell 1000 watt and I have my 1500 watt heater, with my Friction heater running around 750 watts this is not a match... So, I will make my Friction heater run @ 1000 to 1500 watts to make the match for this "Ambient loss" test. Are you guys ok with this?
I have been out of town for the last week or so and I am leaving again out of town this Friday for the weekend. I have 3 days to make this happen so let me know if this will work.


Thank you
Gabriel

[PhysicsProf]

This is what I am going to to do.... I cant find a 800 watt heater but they do sell 1000 watt and I have my 1500 watt heater, with my Friction heater running around 750 watts this is not a match... So, I will make my Friction heater run @ 1000 to 1500 watts to make the match for this "Ambient loss" test. Are you guys ok with this?
I have been out of town for the last week or so and I am leaving again out of town this Friday for the weekend. I have 3 days to make this happen so let me know if this will work.


Thank you
Gabriel

Yes -- matching input powers  and measuring temp reached in the box (at temp-equilibrium), makes a good test -- as long as you measure the actual input power using a kill-a-watt meter (or similar) and have some sort of fan to circulate/MIX the air to avoid temp-gradients in the air.

The other test I would recommend if you have time, is to first run your friction heater as is (at around 750 Watts; measure the input power) and note the equilibrium temp of the air reached in the box.  Then run run the 1000 W heater (again measure the input power) and see what temp it reaches (i.e., at equilibrium).  Is this temp higher or lower than reached with your friction heater?  This test alone will tell us much!!

Good luck, Gabriel!  "fingers crossed..."