PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2017-05-22, 20:18:12
News: Forum TIP:
The SHOUT BOX deletes messages after 3 hours. It is NOT meant to have lengthy conversations in. Use the Chat feature instead.

Pages: [1] 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 ... 22
Author Topic: Lawrence Tseung sent a Prototype to test... any comments?  (Read 141992 times)
Hero Member
*****

Posts: 2102
   I was contacted by a colleague of Lawrence Tseung's, and after some correspondence, I have accepted to look at his toroid-based, solid-state device.  Unlike the ###### motor, there are no moving parts.

  LT claims COP>1.  He sent me a simple prototype, called his "Prototype A" at OU where he seems to like to post (I do not like that forum very much...)  You can see photos of the device and claims there.

   I propose to look at the input power, from a single AA battery, and the output power, and compare the two.  I will probably use oscilloscopes... realizing there is some danger in correctly assessing the actual (integrated) input and output power.  I will be working with a colleague here... we will check each other.

   Anyway, the prototype arrived via mail to my hands today.  Lots of fun... I should say.  I am new at this effort, and welcome comments both on LT's claims AND on pitfalls etc. so that I can do a thorough and accurate test.

Thanks.

   

Group: Elite
Hero Member
******

Posts: 1852

   Anyway, the prototype arrived via mail to my hands today.  Lots of fun... I should say.  I am new at this effort, and welcome comments both on LT's claims AND on pitfalls etc. so that I can do a thorough and accurate test.

Thanks.



You said it.... "thorough and accurate". The other thing to watch for is ..... test as the inventor intended THEN test it the way you think it should be done. Having both data can be a very good thing.


---------------------------
"As far as the laws of mathematics refer to reality, they are not certain; as far as they are certain, they do not refer to reality." - Einstein

"What we observe is not nature itself, but nature exposed to our method of questioning." - Werner Heisenberg
   
Hero Member
*****

Posts: 2102
You said it.... "thorough and accurate". The other thing to watch for is ..... test as the inventor intended THEN test it the way you think it should be done. Having both data can be a very good thing.

Excellent point, WW. 
The challenge will be what I'd call "energy bookkeeping", total energy in versus total energy out.  Integrated power (over time).
   
Hero Member
*****

Posts: 2102
 Interesting  -- Bob Boyce has just stated at the OU site that he rec'd his Prototype (C) from Lawrence today also...  after I made the announcement there, and here.
http://www.overunity.com/index.php?topic=8825.1680

  So its a bit of a horse race, it appears, to see who can test his prototype first??    I mean, I'm not convinced this thing will work (OU) in the first place, but I'm willing to test it. 

  Interesting that Lawrence sent to Bob Boyce the "better" unit, the one he claims has much higher COP as well as tunability with the variable resistor.  (My prototype has all fixed resistors...)  Now if I can figure out where to put that variable R, I might try it.

  Dang, this is interesting.

  Does anyone know who Bob Boyce is?  is he a serious engineer-type, or scientist?
   
Sr. Member
****

Posts: 297
Interesting  -- Bob Boyce has just stated at the OU site that he rec'd his Prototype (C) from Lawrence today also...  after I made the announcement there, and here.
http://www.overunity.com/index.php?topic=8825.1680

  Does anyone know who Bob Boyce is?  is he a serious engineer-type, or scientist?
Bob Boyce is a very serious player indeed. He is the originator of the Boyce 101 plate electroliser for cars. Actually,
it does not electrolise; it fractures the water molecules by hitting them with 3 frequencies, 42.8khz and the
two lower octaves. The gas does not come off at the plates but between them:
http://www.free-energy-info.co.uk/Chapter10.pdf
page 10 (et seq).
   
Group: Elite
Hero Member
******

Posts: 3044
It's turtles all the way down
If you could supply more information on the device I / O requirements I would be happy to assist in designing a first cut simple test setup that would not require a DSO, or even an oscilloscope, and could be done with very simple measuring devices.

The DSO would be valuable later when you need to split hairs.


---------------------------
Just because it has a patent application or is patented does not always mean it really works.
   
Hero Member
*****

Posts: 2102
If you could supply more information on the device I / O requirements I would be happy to assist in designing a first cut simple test setup that would not require a DSO, or even an oscilloscope, and could be done with very simple measuring devices.

The DSO would be valuable later when you need to split hairs.

That would be deeply appreciated, ION.  I agree that the DSO could be used later...

Lawrence Tseung is quite open with his descriptions of Prototype A (hereafter, A) which he gave to me, here --

http://www.overunity.com/index.php?topic=8825.1530

See post #1538 on page 103 of the thread above -- shows photos of A.  Posted 20 Dec 2010.  Other nearby posts give his instructions/suggestions on how to test  the prototype.

   
Group: Elite
Hero Member
******

Posts: 3044
It's turtles all the way down
That would be deeply appreciated, ION.  I agree that the DSO could be used later...

Lawrence Tseung is quite open with his descriptions of Prototype A (hereafter, A) which he gave to me, here --

http://www.overunity.com/index.php?topic=8825.1530

See post #1538 on page 103 of the thread above -- shows photos of A.  Posted 20 Dec 2010.  Other nearby posts give his instructions/suggestions on how to test  the prototype.



After reviewing the information you have supplied I hereby respectfully withdraw my offer to help. We have been down the blocking oscillator path many times before. Now it is called a Joule Thief and a rather poorly designed one at that. Contact me by PM for more details.

Maybe someone else will pick up the gauntlet, MH perhaps.


---------------------------
Just because it has a patent application or is patented does not always mean it really works.
   
Group: Guest
I had a look myself.  It looks like a Joule Thief with an additional secondary winding.  The standard JT output is driving one LED and the secondary winding is driving a second LED.  Besides the fact that there is nothing to be found there, if you use LEDs as the loads then you have the agonizingly difficult task of measuring the power across the LEDs unless you go the full DSO route.

Lawrence Tseung's reputation is another thing that merits doing some due diligence on.  It would not surprise me if he insists that you must use LEDs as the loads.

I will pass on any involvement here, thank you.

MileHigh

   

Group: Elite
Hero Member
******

Posts: 1852
PhysicsProf,

Since this device wound up as a Joule Thief with LED loads, I will add to my earlier suggestion...

You may be aware but I'll mention these anyway...

LEDs may act as photo cells.
Many LED types will produce light from A.C. electric fields (as in the noisy switching of a blocking oscillator with spurious harmonics running into VHF).
The same types will produce a small DC current when they are immersed in a high frequency electric field (more than the expected rectifier with antenna leads).
In some applications LEDs create their own EMI.
Each light emitting mode/method has different performance characteristics and values.

Due to the above and things like the junction voltage drop / Zener-like action (clamping - mentioned by MH in another thread)...

Make sure you aren't talked into using a light meter or the visible indication of light intensity as a means of measuring P-Out.
Some of these circuits may run for months seemingly perpetual.


This time two years ago I had one that ran for 3 months off a AAA NiCad. As long as you didn't touch it, it would run down in a week or so. If you wanted to keep it running all you needed to do was make a change in a resistance, disconnect the battery and reconnect or make body contact to the right conductor or move some of the windings. The thing would 'perk-up' with almost any intervention.

These circuits take many folks including hook, line & sinker.
 


---------------------------
"As far as the laws of mathematics refer to reality, they are not certain; as far as they are certain, they do not refer to reality." - Einstein

"What we observe is not nature itself, but nature exposed to our method of questioning." - Werner Heisenberg
   
Hero Member
*****

Posts: 2102
  Wow, not a very popular approach, then.  

  My colleague and I had a look at it using an oscilloscope hooked to his laptop (so we can preserve the trace) and a few other instruments.

The output voltage was measured across the leads from the secondary winding on the toroid.  
The paired-primary windings showed about 0.44 micro-Henrys each, while the secondary winding was 215 micro-Henrys.  So not surprising there is a step-up in the output voltage from the battery (which sat at 1.45 V).

The output voltage showed an interesting small-h oscilloscope trace, with a rapid high pulse (the left leg of the h) that itself was quite ragged.  How in the world does one measure the output power in a case like that?

Can I ask you guys to set aside for a moment your aversion to such a device and give me some of your experience on POWER MEASUREMENT?  

The peak-to-peak voltage (Vpp) on the output side was 11.7 volts.  The Vpp for the 10 ohm resistor on the output circuit was 0.343 V, suggesting a current reading of 0.034 A.   It pains me to multiply these together, given the weird shape of the output pulse (which was however uniform over the couple of hours we played with the thing) -- yields about 0.4 W out.  And a Pout/Pin > 1.   But I don't believe it.  Yes, one could integrate the oscilloscope trace somehow, the product of the Vout and the current.  Not sure how to do that reliably.

My colleague suggests that we charge a rechargeable battery with the output, which is basically pulsed-DC, compare with the drain on the input battery...  That is a way to integrate the power out and the power in... perhaps not very elegant.  But I like the simplicity of it.  He says others are doing this in this community of researchers.  (He's been at this for a while.)

Suggestions, please!  I think RELIABLE, BELIEVABLE measurements are going to be a key in this field, and frankly, the simpler the better.
The Prof is learning... it ain't so easy.
« Last Edit: 2010-12-31, 06:04:10 by PhysicsProf »
   
Hero Member
*****

Posts: 949
@PhysicsProf
Quote
The output voltage showed an interesting small-h oscilloscope trace, with a rapid high pulse (the left leg of the h) that itself was quite ragged.  How in the world does one measure the output power in a case like that?
A small "h" you say, it may be due to the fact that the transistor base(trigger) and primary winding are both inducing the output winding which is not normally the case for a standard JT, could you post a picture?.

Quote
Can I ask you guys to set aside for a moment your aversion to such a device and give me some of your experience on POWER MEASUREMENT?  

I use the black box method where we assume any device is simply a black box with storage capacitors on both the input and output, I think this is acceptable in this case as long as the output winding is electrically isolated from the rest of the circuit which includes the input capacitor. Personally I do not like batteries or oscilloscopes because the results can be riddled with errors and to a large extent are open to interpretation. With input/output capacitors the result is a simple voltage measurement of each capacitor at start and finish and a few simple calculations however it relies on the fact that the output capacitance cannot couple to any other significant capacitance in the system.
Regards
AC


---------------------------
"Great minds discuss ideas; average minds discuss events; small minds discuss people." - Eleanor Roosevelt.

There is infinitely more that we do not know, than we know.
   
Group: Elite
Hero Member
******

Posts: 3044
It's turtles all the way down
 Wow, not a very popular approach, then.  


Can I ask you guys to set aside for a moment your aversion to such a device and give me some of your experience on POWER MEASUREMENT?  [/b]

Suggestions, please!  I think RELIABLE, BELIEVABLE measurements are going to be a key in this field, and frankly, the simpler the better.
The Prof is learning... it ain't so easy.


For something like this, and as a first cut for power out, I like to use the thermal RMS into a known resistor. This can be measured with a small thermocouple taped to a resistor, since temperature and power can be easily correlated.

The resistor can be characterized by putting known power levels in and recording temperature rise above ambient, which is also recorded.

Then the resistor is put on the output of the device and temperature is noted and compared to the known temperature rise vs. power curve you earlier created.

A more sophisticated differential thermocouple technique can be used to null ambient, and a readout in simple millivolts will equate to power, once your load resistor is "calibrated". You must use a non inductive resistor.

This is a foolproof thermal RMS technique that does not have crest factor limitations commonly found in true RMS instruments. I have decades of experience and success using these methods.

For power input, I generally use a current source power supply feeding a capacitor. I adjust and note input current required to produce the desired operating voltage, then multiply the current X voltage for power.

Alternately, for power input, you can use a voltage source power supply with shunt resistor of one ohm feeding the capacitor. Volts across the shunt resistor equates to Amps of power drain, then this is multiplied by voltage across the capacitor for Watts input.

A null balance technique can also be used where a differential thermocouple setup is used to sense the temperature rise of the load resistor under test as compared to an equivalent load resistor hooked up to a power supply. Adjust the power supply feeding the control resistor until a "null is obtained (the thermocouples are wired in inverse series and their millivolts will "null" when the temperatures of the load resistor under test and the "control" resistor are equal)

These techniques eliminate the tricky process of trying to use capacitors or batteries on the output and trying to measure charges stored over time.

The thermal methods give real power readings in real time while the unit is in operation without adverse loading or peak charging effects of batteries or capacitors on the output nor the necessity of gauging how much charge was actually stored in your battery.

While these techniques may seem simple, they are actually quite sophisticated, and thermal RMS meters are very expensive, and not fooled by irregular waveforms. All power is integrated as heat rise. This is a budget approach to the same.

I'll see if I can sketch something up for you.
« Last Edit: 2010-12-31, 12:52:17 by ION »


---------------------------
Just because it has a patent application or is patented does not always mean it really works.
   

Group: Elite
Hero Member
******

Posts: 1852
Can I ask you guys to set aside for a moment your aversion to such a device and give me some of your experience on POWER MEASUREMENT?

I agree with ION.


---------------------------
"As far as the laws of mathematics refer to reality, they are not certain; as far as they are certain, they do not refer to reality." - Einstein

"What we observe is not nature itself, but nature exposed to our method of questioning." - Werner Heisenberg
   

Group: Administrator
Hero Member
*****

Posts: 2606
It's not as complicated as it may seem...
I would add that in order to be fair, the DUT should in addition to the load resistor, include a thermal-couple on the transistor.

You are right Prof, for output power don't even bother multiplying V x I with anything other than a pure sine wave or clean pulse train.

.99
   
Group: Elite
Hero Member
******

Posts: 3044
It's turtles all the way down
I would add that in order to be fair, the DUT should in addition to the load resistor, include a thermal-couple on the transistor.
.99

Actually, this is not necessary as all circuit losses will be easily computed as the difference between power input and power output.

Temperature of the circuit components(transistor, core, bias resistors) is part of the internal "black box" losses.

Input power and Output power are considered external to the black box.

Guaranteed, with the blocking oscillator power output will be less than power input.

« Last Edit: 2010-12-31, 16:00:27 by ION »


---------------------------
Just because it has a patent application or is patented does not always mean it really works.
   

Group: Administrator
Hero Member
*****

Posts: 2606
It's not as complicated as it may seem...
I should have clarified.

I think if one is using a control compared to the DUT, then you would want to account for dissipation in the switch as well, because the load will not be receiving all the available power. This of course assumes that the input power is set to an equal amount for both tests.

Is this not correct?

.99
   
Group: Elite
Hero Member
******

Posts: 3044
It's turtles all the way down
I should have clarified.

I think if one is using a control compared to the DUT, then you would want to account for dissipation in the switch as well, because the load will not be receiving all the available power. This of course assumes that the input power is set to an equal amount for both tests.

Is this not correct?

.99

In this differential "null" circuit, the control resistor is passively driven directly from a separate power supply, which is used to balance the thermal bridge. The power required to balance the thermal bridge must equal the power in the load resistor attached to the DUT.

Power In (total) minus Power Out = Power consumed by DUT

No need to account for transistor power as all components in the black box are accounted for as part of DUT power consumption.

Do not confuse this method with the specialized circuit I developed for the Ainslie resistor, whereby duty cycle must be taken into account if a control is actively driven.

 I will have to draw another circuit for this to be clear.
« Last Edit: 2010-12-31, 16:56:11 by ION »


---------------------------
Just because it has a patent application or is patented does not always mean it really works.
   
Hero Member
*****

Posts: 2102
Thanks so much, ION, .99, all -- I'm reading and understanding.  Busy day today -- more later.
   

Group: Administrator
Hero Member
*****

Posts: 2606
It's not as complicated as it may seem...
ION,

May I offer a hypothetical scenario:

Once all has been adjusted as per the instructions and the null meter is balanced, we have a DUT input power of 10W and a control input power of 10W.

What would be the resulting conclusion?

.99
   
Group: Guest
Great comments gentlemen:

I was thinking of going the budget route for a thermal approach myself, very low tech.

For starters I think it would be prudent if PhysicsProf checked with Lawrence to find out what the claimed COP is for "Prototype A."  This device consists of a series of components.  So the other piece of information that PhysicsProf would need to get from Lawrence is precisely which components constitute the "output."  Note that there are two LEDs, some resistors, etc.  If PhysicsProf could get those two questions answered by Lawrence and then post his reply in this thread, then we have a basis for doing a thermal evaluation of the energy in and energy out of this device.

For reference, here is a picture of the device:



My budget route for the thermal approach would be as follows:

Apply black paint to the LEDs.  Apply insulating lacquer to the exposed contacts of the components that constitute the "output."  Get two medium-sized Styrofoam coffee cups and put one in the other to act as a thermally isolated container.  Get a Popsicle stick to agitate the water and put insulating Styrofoam on the end that you will hold between your fingers.  Fill the container with water to a fixed level and insert the "output" components.  Power the setup with a power test circuit like Ion showed or equivalent.  Monitor the temperature over time as you very slowly agitate the water with the Popsicle stick.

Once this is done then all that you have to do is do some more runs where you put some resistors in the water that are dissipating fixed amounts of DC power.  By comparing the temperature profiles of the water over time you will easily be able to determine what the output power of the "Prototype A" circuit is.

All that you need is a good variable power supply, preferably one that can operate in both voltage source and current source mode, and a good thermometer with at least 0.1 C resolution.  The lower the thermal mass of the thermometer the better.

If you are using a medium-sized coffee cup to hold the water and the COP claim is greater than 1.5 you should be able to prove or disprove the claim very easily like this.  The thermal mass of the water will be so much larger than the thermometer or anything else such that the results will be irrefutable.  My suggestion for a datum for the test runs would be the amount of time it takes for the water to rise by 10 degrees Celsius.  Hopefully this would take around 10 minutes and you should be fine.  You just have to use your common sense here to find a happy medium for running the temperature trials.  They may need to be tweaked (eg: reduce the amount of water in the container) to get excellent results.  Note that you need to agitate the water very slowly and do it exactly the same way for all trial runs.

Personally, I would have to get Lawrence to commit to the COP performance and "output" components in writing as I originally stated above before I would start any testing.

MileHigh
   
Group: Guest
Just one more comment about my proposed setup.  In this case the heat supplied by the components immersed in the water is being supplied at a fixed steady rate.  That means that over a period of 10 minutes you should expect to see a nearly straight-line rise in the water temperature.  It's worth it to record the temperature every 30 seconds (for example) so that you can see a nice near-linear rise in the temperature.  That would be very satisfying.

Even if the temperature vs. time line starts to take a slight curve where the slope starts to decrease, the data is still 100% valid.

If you assume that the water is thermally isolated from the outside world with the two Styrofoam coffee cups and the water evaporation over the time of the test run is insignificant, then the temperature vs. time line should be almost perfectly straight.  The equivalent electrical circuit for this thermal system is a fixed current source (the output components dissipating heat) charging a capacitor (the thermal mass of the water).  When a fixed current source charges a capacitor the voltage (temperature) in the capacitor will increase linearly over time.

Naturally you should start the test with the water at room temperature.

MileHigh
   
Group: Elite
Hero Member
******

Posts: 3044
It's turtles all the way down
ION,

May I offer a hypothetical scenario:

Once all has been adjusted as per the instructions and the null meter is balanced, we have a DUT input power of 10W and a control input power of 10W.

What would be the resulting conclusion?

.99

In that hypothetical situation, you have an overunity device, because some power must be dissipated in the DUT. There are several methods to arrive at total power dissipated by the DUT, not including load resistor.

Can you guess how this can be done using the same test setup with a slight re-arrangement?

I designed this test strictly for black box devices as a first cut trial. Most DUT's will be quickly vetted as underunity and not require more time be spent. I'm sure this will occur with the rather poorly designed blocking oscillator of LT.

The black box itself is exactly that, a small enclosure that will contain the circuit and allow input / output connections.

I chose the open air, resistor with taped thermocouple because it is quick responding and has fairly linear losses to air compared to the insulated water bath technique, which is more of a power integration method.

Both the "loss to ambient" method described by myself and the "insulated water bath" method described by MH are valuable and each has different merits.


---------------------------
Just because it has a patent application or is patented does not always mean it really works.
   

Group: Administrator
Hero Member
*****

Posts: 2606
It's not as complicated as it may seem...
In that hypothetical situation, you have an overunity device, because some power must be dissipated in the DUT. There are several methods to arrive at total power dissipated by the DUT, not including load resistor.

Can you guess how this can be done using the same test setup with a slight re-arrangement?

I designed this test strictly for black box devices as a first cut trial. Most DUT's will be quickly vetted as underunity and not require more time be spent. I'm sure this will occur with this rather poorly designed blocking oscillator.

The black box itself is exactly that, a small enclosure that will contain the circuit and allow input / output connections.

I chose the open air, resistor with taped thermocouple because it is quick responding and has fairly linear losses to air compared to the insulated water bath technique, which is more of a power integration method.

Sure,

Once the test is nulled, take the thermocouple that was on the DUT resistor, and place it inside the DUT box. Do not adjust the DUT input power, but adjust the control power to again null the meter.

The hypothetical test situation I gave would indeed indicate OU. Now, a slight adjustment to the scenario; what if the DUT input power indicates 15W, and the control power is still 10W. What would be the conclusion from these results?

.99
   
Hero Member
*****

Posts: 2102
In that hypothetical situation, you have an overunity device, because some power must be dissipated in the DUT. There are several methods to arrive at total power dissipated by the DUT, not including load resistor.

Can you guess how this can be done using the same test setup with a slight re-arrangement?

I designed this test strictly for black box devices as a first cut trial. Most DUT's will be quickly vetted as underunity and not require more time be spent. I'm sure this will occur with this rather poorly designed blocking oscillator.

The black box itself is exactly that, a small enclosure that will contain the circuit and allow input / output connections.

I chose the open air, resistor with taped thermocouple because it is quick responding and has fairly linear losses to air compared to the insulated water bath technique, which is more of a power integration method.

I prefer the simplicity of the open-air method of ION, especially for a quick test.  But the water-bath method of MH is rather good -- I may have access to a precision calorimeter at the university...  that would be fun.  Note below, though -- the output power is small... for this prototype.

Thanks for posting the photo, MH.  Your questions:
Quote
For starters I think it would be prudent if PhysicsProf checked with Lawrence to find out what the claimed COP is for "Prototype A."  This device consists of a series of components.  So the other piece of information that PhysicsProf would need to get from Lawrence is precisely which components constitute the "output."  Note that there are two LEDs, some resistors, etc.  If PhysicsProf could get those two questions answered by Lawrence and then post his reply in this thread, then we have a basis for doing a thermal evaluation of the energy in and energy out of this device.

I think Lawrence posted the results on the OU forum, but anyway I have this from him:
The output is between the bottom two copper wires in the photo (he numbers them 5 and 6), connecting to the output coil.  I checked -- there is no direct coupling between this coil and the other wires in the prototype.  The components on the output, starting at the bottom in the photo, are:  10 ohm resistor, 100 ohm resistor, and LED.

The claimed COP -- is done two ways by him, peak-to-peak (PP) and RMS.  He measures the voltage output of the battery and the voltage drop (using an oscilloscope, as the DC is pulsed) across the 1.o-ohm resistor (connected to the black lead from the battery), both PP and RMS.
  Then he measures the voltage across the output leads (5 and 6, described above) and to get the current, the voltage drop across the 10 ohm resistor (bottom of photo).
OK -- Input power PP is 0.117 W, RMS is 0.0023 W.
        Output power PP is 0.37 W, RMS is 0.023 W.

Thus he calculates COP to be 3.2 = 0.37/0.117 from PP measurements,  and from the RMS measurements, 10 = 0.023/0.0023.
 
As I said above, I don't like the multiplying of V*I when the voltage trace is so non-uniform, nothing like sinusoidal.  And Lawrence intrinsically admits such when he says "the more accurate value from the integration of the power curves should be used."  So let's be fair on that point.  

Later, in a 20 Dec 2010 posting at OU, he says, "in the Workshop at Hong Kong University on Oct 9, 2010, ..one student, Felix, produced a prototype with COP > 280."  
Probably calculated in a crude way, but that is not stated.  
Hope that helps MH-- you see the output power is very small with this prototype A, would not warm water very fast at all.   ;)





   
Pages: [1] 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 ... 22
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2017-05-22, 20:18:12