PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2024-11-26, 17:20:09
News: If you have a suggestion or need for a new board title, please PM the Admins.
Please remember to keep topics and posts of the FE or casual nature. :)

Pages: 1 [2] 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 ... 22
Author Topic: Lawrence Tseung sent a Prototype to test... any comments?  (Read 342711 times)
Group: Professor
Hero Member
*****

Posts: 3017
 MH posts the photo of Prototype A above in reply 20.  It is fairly obvious that with all the resistors and two LEDs, there is quite a bit of loss, not to mention losses in the toroidal "transformer".  I would appreciate comments on how to reduce unnecessary losses... then we could test the suggestions on the Prototype. 
   
Hero Member
*****

Posts: 520
@all

OK, I'm back on the forum. Staying in guest mode is too freaky. I felt like I was an OU Mutant Ninja crawling around this place and hiding behind this guest mode moniker. Not really my style. So anyways...

@PhysicsProf

Maybe one point to consider in testing this device is to make sure you do not leave the scope probe or volt meter probes connected to the device for any real length of time since just this one additional connection can aid in keeping the battery charged.

It is always very difficult with devices that work on low voltage, low amperage to separate the true effect from any induced effects as mentioned from others. Let's say you put your volt meter on the device and reading 1.47 volts, then put your scope probe on the device and immediately see a jump to read 1.49 volts, these are the type of problems one will run into. Or, the LED is lit, then when you put a scope probe on it lights brighter or dimmer.

wattsup


---------------------------
   
Group: Elite
Hero Member
******

Posts: 3537
It's turtles all the way down
Welcome back Wattsup

Very good suggestion...best to eliminate all stray leakage paths that could inject current.

Sometimes an oxidized connection can rectify AC leakage currents or airborne RF enough to provide a fake charging path.

If using a scope or meter, best to follow POYNTS grounding techniques.

BTW...the power in this circuit is so small the chances of error are surely increased.


---------------------------
"Secrecy, secret societies and secret groups have always been repugnant to a free and open society"......John F Kennedy
   
Group: Guest
MH posts the photo of Prototype A above in reply 20.  It is fairly obvious that with all the resistors and two LEDs, there is quite a bit of loss, not to mention losses in the toroidal "transformer".  I would appreciate comments on how to reduce unnecessary losses... then we could test the suggestions on the Prototype. 

I find your question unusual.  Why should you need to modify a working prototype that was sent to you by the inventor?  There is really not much that you can do to reduce unnecessary losses anyways.  This device is outputting electrical power which is being turned into heat.  The energy being burned off in the output components,  the 10 ohm resistor, 100 ohm resistor, and LED are all that you have to worry about.  You shouldn't have to worry about the heat looses in the wires, the ferrite core, and the transistor.  These types of losses are part and parcel of any alleged over unity device.

There are a few other unusual things.  For starters Lawrence is not counting the main JT output through the LED that's connected to the transistor.  The concepts of "peak-to-peak" COP and "RMS" COP are superfluous and make no sense.  For the "RMS" COP there are two issues that come to mind.  There is a non-linear diode in the circuit so you can't use the RMS data.  The second issue is that the JT circuit might be oscillating at 50 KHz or more and there is very little data provided for "True RMS" multimeters with respect to the sampling rate, bandwidth, etc.  I am assuming that Lawrence is using a "True RMS" multimeter for his measurements.  The COP is the average power out divided by the average power in.  There is no wiggle room for that definition.

The power levels are indeed quite low so the water bath method may not work or it could be modified.  You could put the three components that constitute the output in a Styrofoam cup and only put about one centimeter worth of water in the cup.  You also put a standard thermocouple in the water.  Put a cover on the cup and then gently move the cup with your hand to agitate the water.  Do that for 10 minutes and the water should go up in temperature by a few degrees Celsius.  Then with the very same components in the cup do your pure DC fixed power runs to determine the output power.

Another option is to simply gob the thermocouple, 10 ohm resistor, the 100 ohm resistor, and LED together into a ball of white thermal paste and suspend it in the air and do the open air tests like Ion said.  It should take only about five minutes for that to reach thermal equilibrium.  You could also use a non-contact infrared thermometer if you wanted.

Best of luck with whatever you choose to do!

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

Posts: 3537
It's turtles all the way down
I wil agree with MH....we are not here to redesign the inventor's device so he has yet another complaint of why the testing failed.

POYNT and I went through testing such JT devices a few years back. Based on this, I have little faith in Lawrence Tseung's prototype.



---------------------------
"Secrecy, secret societies and secret groups have always been repugnant to a free and open society"......John F Kennedy
   
Group: Professor
Hero Member
*****

Posts: 3017
  We have built a similar device, from scratch, that we're playing with....   Should I ask for your permission, fellas, to make some changes on this one?
I'm finding it educational, actually.

   
Group: Professor
Hero Member
*****

Posts: 3017
  (Note:  I was having trouble posting and hit clicked on "unlock", but I wonder if somehow I inadvertently locked the thread... sorry if so. Not intentional.)
 
This posted by LT at OU yesterday may be of interest:
Quote
Prototype A and C have arrived at the hands of the intended verifiers.  The first successful FLEET prototype was demonstrated on July 13 (reply 434 on page 29).  Much historical information is available on this thread - thanks to Stefan.  The diary form of posting does have its value.

Bob Boyce and I communicated via phone today.  When I get back from the ski resort on Jan 2, 2011, we shall work more closely.  Confirmation of COP greater than 1 with videos, scope shots, computer analysis results etc will be out soon.  He also has the Bedini three magnet wheel. 

We may combine the two in a future conference.  The participants will have two working COP > 1 devices to take home.  When hundreds or thousands of people have and can build such prototypes, the Energy Crisis will be over.

The bonus may be the HHO device with much higher output power from Bob.
   
Hero Member
*****

Posts: 520
@PhysicsProf

In answer to your question above as to should you make modifications, I would answer yes, but only after you have used your best efforts to analyze the device as is. There is a question of giving the device the justice it deserves based on the efforts the maker has made to put the device in your hands.

I think the hardest thing for the more traditional engineers (I am not one of them) is to both look at the device in the traditional sense, but thereafter, put those notions aside and try to see the device through the makers eyes in order to understand what the maker is trying to convey regarding any functional underpinnings.

wattsup



---------------------------
   
Group: Professor
Hero Member
*****

Posts: 3017
 
@PhysicsProf

In answer to your question above as to should you make modifications, I would answer yes, but only after you have used your best efforts to analyze the device as is. There is a question of giving the device the justice it deserves based on the efforts the maker has made to put the device in your hands.

I think the hardest thing for the more traditional engineers (I am not one of them) is to both look at the device in the traditional sense, but thereafter, put those notions aside and try to see the device through the makers eyes in order to understand what the maker is trying to convey regarding any functional underpinnings.

wattsup



Thanks, wattsup, and I agree the device should be tested as-is -- with one caveat:  The resistors are set into the device by Lawrence (1 ohm after the battery and 10 ohms on the output) with the intent to have the tester measure input and output voltages and currents using oscilloscope(s) and calculating power in and out that way (multiplying...IV).  As I explained above...  and I don't trust the power calculated in this way because the voltage traces are so very non-uniform (like an h, as I said).  So I'm obliged already to modify the circuit somewhat in order to get a reliable measure of the input and output power -- something I can believe.

I did it "his way" (Lawrence's) and found COP >1...  but now I'm moving on to better measurements, something I can hang my hat on.  

In parallel, we have built a similar toroidal device and we're playing with variations...  I'm not a conventional engineer either...  ;)
   
Group: Guest
Wattsup:

The notion of looking at the device in a "traditional sense" and then from the "maker's eyes" is false.  You hope it will work and you hope that some sort of alternative way of looking at it or analyzing it will show that it works.

For the energy analysis, there is only one way of looking at this device.  You have to try to measure the energy supplied to the device as accurately as possible and measure the energy output by the device as accurately as possible.

There is no 'tricking' the system.  It has to be a level playing field no matter what the device.

For example, I look at the Bedini stuff and many people repeat what they have been told by John Bedini and Rick Friedrich and Aaron Murakami.  They say, "It's really a COP > 1 device when you factor in the mechanical output from the rotor."  The reality is that there is no mechanical output from the rotor.  The spinning rotor in a Bedini motor only represents lost power and represents part of the inefficiencies inherent in the design.  So the "alternative" viewpoint about the spinning rotor is false in this case.  There is only one way to do an energy analysis for a Bedini motor.

A Joule Thief is just a transformer with a feedback loop that turns it into an oscillator.  Lawrence Tseung has added an additional secondary winding to this transformer-oscillator and believes that he has created an over unity device.  The measurements will speak for themselves.

MileHigh
   
Group: Guest
Thanks, wattsup, and I agree the device should be tested as-is -- with one caveat:  The resistors are set into the device by Lawrence (1 ohm after the battery and 10 ohms on the output) with the intent to have the tester measure input and output voltages and currents using oscilloscope(s) and calculating power in and out that way (multiplying...IV).  As I explained above...  and I don't trust the power calculated in this way because the voltage traces are so very non-uniform (like an h, as I said).  So I'm obliged already to modify the circuit somewhat in order to get a reliable measure of the input and output power -- something I can believe.

I would advise you to proceed with caution here.  For starters, the power supplied to the device should be measured after the one-ohm resistor.  If you use a battery eliminator/power test circuit like Ion showed, then you don't need the one-ohm resistor at all.

If you change the 10-ohm resistor for a different value, then you change the overall impedance of the circuit as seen from the perspective of the power supply.  Changing the value of the 10-ohm resistor will have an impact on the power input and the power power output of the circuit.  It will also most likely change the frequency the JT operates at.  So if you modify the circuit you will completely change the dynamics of the circuit.  Assuming you measure under unity Lawrence will claim it's under unity because you modified his design.  This is "standard practice" in this realm and Lawrence is fully aware of this.

If you were to not modify the circuit then the viable options for measuring the power output are to use a DSO or to do it thermally.  In my opinion, my variation on Ion's method is the easiest and fastest.  Gobbing the three output components together with thermal paste and then suspending that in the air and letting it reach thermal equilibrium would be very easy to do.  After measuring the temperature from the running circuit, it would take another four or five runs at different DC power levels to be able to deduce the output power of the device.  The whole thing could be done in a single afternoon.

MileHigh
« Last Edit: 2011-01-01, 23:22:40 by MileHigh »
   
Group: Professor
Hero Member
*****

Posts: 3017
MH: 
Quote
"If you were to not modify the circuit then the viable options for measuring the power output are to use a DSO or to do it thermally.'

Your thermal method would have to be calibrated -- I think this is the hardest part, frankly, with your method.  How would you calibrate it? 

Back to the DSO method for a moment, please.  We have a lap-top based oscilloscope and can capture the traces easily.  So how would we measure power out using this method?  (I'm seriously trying to learn as we go along here.)
   

Group: Tinkerer
Hero Member
*****

Posts: 3055
...

A Joule Thief is just a transformer with a feedback loop that turns it into an oscillator.  Lawrence Tseung has added an additional secondary winding to this transformer-oscillator and believes that he has created an over unity device.  The measurements will speak for themselves.

MileHigh

The basic circuit can be improved upon in innovative ways to
enhance switching efficiency.  It is true that the Blocking Oscillator
configuration (bipolar transistor version) is not highly efficient
as it is.

I applaud the efforts of those who are pursuing an accurate means
of evaluating pulsed energy input/output on the small scale which
this type of circuit affords.  The PhysicsProf is doing what he does
best and will surely arrive at an equitable solution in due time.

There is much to be said for pulsed energy systems.  Even Nikola
Tesla was intrigued by their properties and came to some very
interesting conclusions in his time.



---------------------------
For there is nothing hidden that will not be disclosed, and nothing concealed that will not be known or brought out into the open.
   
Group: Elite
Hero Member
******

Posts: 3537
It's turtles all the way down
Perhaps I should have explained in more detail.

The thermal bridge "null" balance method is like the old time beam balance weighing scale, where the unknowns are put on one side and are "balanced" or "nulled" with a known quantity on the other side of the scale.

A system which converts electrical energy into heat by dissipation in resistors, transistors and other components is weighed against a known heat generating system.

The differential thermocouple system performs the function of the beam in the beam balance.

We are using a power supply and resistor on the known part of the bridge which we adjust  to develop heat equal to that which our DUT is outputting (the unknown side of the bridge).

When the system is balanced we can simply read the power on the balancing power supplies meters.

Normal precalibration of the power supply used to provide balance, and the source power supply to the DUT is assumed.

We can get more detailed, and match the thermocouples, make sure we measure the value of the heat generating resistor as a double check etc. but this is normally unnecessary except for extreme precision.

This method can be broken down into the two step method MH has described. I just tend to prefer null balance techniques...


---------------------------
"Secrecy, secret societies and secret groups have always been repugnant to a free and open society"......John F Kennedy
   
Group: Guest
MH: 
Your thermal method would have to be calibrated -- I think this is the hardest part, frankly, with your method.  How would you calibrate it? 

Back to the DSO method for a moment, please.  We have a lap-top based oscilloscope and can capture the traces easily.  So how would we measure power out using this method?  (I'm seriously trying to learn as we go along here.)

There is no requirement to calibrate the system.  You can measure the output power by empirical observation.  For example, suppose that you run the device and the three output components are suspended in mid air by the wires and scrunched together and covered with white thermal paste like previously described.  You wait five minutes and you measure a temperature of 27.3 degrees.   To be sure, you wait another five minutes and you measure again and the temperature is still 27.3 degrees.  You are now certain that the the mass has reached thermal equilibrium.

You then disconnect the the JT circuit from the three suspended components and connect up the battery eliminator/power test circuit so that you can power it with pure DC at a measurable voltage and current flow.  Suppose that you take a guesstimate that the Lawrence Tseung circuit was supplying 0.1 watts to the load.  So the first thing that you do is dial-up 0.1 watts of power with the power test circuit by carefully adjusting the current output (preferably) from the variable power supply and monitoring the DC voltage and current readings until you are supplying exactly 0.1 watts to the load.  Then you wait five minutes and take a temperature reading.  Supposing you measure 27.3 degrees.  We will also assume that the ambient temperature in the room has not changed and during both tests there has been no appreciable airflow around the hanging thermal mass, just the normal ambient airflow.  Assuming that everything is true then your guess was dead-on and you now know that the power output of the Lawrence Tseung device is 0.1 watts.  Naturally there is an error tolerance for this measurement but I am not going to discuss that.

With respect to your laptop-based oscilloscope I would not really put much faith in that as an accurate measuring device.  If you are running software on the laptop that makes use of the built-in sound card's line input, it's doubtful that the setup is good enough to make the required measurements.  I am just giving you my gut feel here.  I am assuming that the sampling is limited to 8-bit quantization at 44 KHz which would not cut it.

If you were to use a "real" DSO of an older generation then you could export your voltage and current waveforms into Excel and do the multiply-accumulate function to calculate the average power over one or multiple cycles, making sure you satisfy the "enhanced" Nyquist sampling rate requirement.  With a very modern DSO you could record the waveforms and push a few buttons and the DSO itself would calculate the average power for you.

MileHigh
   

Group: Administrator
Hero Member
*****

Posts: 3217
It's not as complicated as it may seem...
Perhaps I should have explained in more detail.

The thermal bridge "null" balance method is like the old time beam balance weighing scale, where the unknowns are put on one side and are "balanced" or "nulled" with a known quantity on the other side of the scale.

A system which converts electrical energy into heat by dissipation in resistors, transistors and other components is weighed against a known heat generating system.

The differential thermocouple system performs the function of the beam in the beam balance.

We are using a power supply and resistor on the known part of the bridge which we adjust  to develop heat equal to that which our DUT is outputting (the unknown side of the bridge).

When the system is balanced we can simply read the power on the balancing power supplies meters.

Normal precalibration of the power supply used to provide balance, and the source power supply to the DUT is assumed.

We can get more detailed, and match the thermocouples, make sure we measure the value of the heat generating resistor as a double check etc. but this is normally unnecessary except for extreme precision.

This method can be broken down into the two step method MH has described. I just tend to prefer null balance techniques...

I agree with this except for one issue:

In accordance with your diagrams, you are only placing part of the DUT on the balance beam, that part being the resistor.

Also, let me repeat part of a previous post that may have got overlooked:

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


---------------------------
"Some scientists claim that hydrogen, because it is so plentiful, is the basic building block of the universe. I dispute that. I say there is more stupidity than hydrogen, and that is the basic building block of the universe." Frank Zappa
   
Group: Guest
I agree with Poynt here.  Ion's proposed system allows you to tweak a test setup such that the absolute temperature of the resistor under test and the control resistor are the same.  However, for the two resistors, the thermal resistance to the outside world is not necessarily the same unless you are very careful in how you set things up.  The ultimate temperature that the two devices stabilize at is a function of the power dissipation and the thermal resistance to the outside world.

Of course, if you are testing two objects that are manufactured in exactly the same way, then you can expect the thermal resistance to the outside world for each object to be approximately the same.  In this particular case, we are talking about three tiny components.   It might be difficult to replicate the three tiny components such that they had the same thermal resistance to the outside world as the device under test.

Not to say that you couldn't get creative here.  If you put the three components under test into a small plastic case filled with thermal paste, and compared that with the identical plastic case filled with thermal paste containing a single resistor, then you have two objects that have approximately the same thermal resistance to the outside world.  You should get very good results like this.  (At least when I do the thought experiment in my head!)

MileHigh
   

Group: Administrator
Hero Member
*****

Posts: 3217
It's not as complicated as it may seem...
Hi MH.

I think you may be close, but that's not quite what I am getting at.

.99


---------------------------
"Some scientists claim that hydrogen, because it is so plentiful, is the basic building block of the universe. I dispute that. I say there is more stupidity than hydrogen, and that is the basic building block of the universe." Frank Zappa
   
Group: Professor
Hero Member
*****

Posts: 3017
  Today we spent some time on the toroidal prototype.  We actually came quite close to your null balance, ION, but not quite.  We only had one thermocouple + meter system available (this is a small town) and so we did the balance a piece at a time.  We found that we could match the heating of the output resistor from the DUT with 0.04 Watts of power over the same resistor from our battery+variable resistor (which substituted for our power supply -- again, a very small town).  The thermocouple was taped to the resistor...  It took some time for us to home in on the matching input power... but we feel the calibration is believable.

So great idea, ION -- it worked to this extent.\

1 -- Here is a potential problem I see with your null-balance drawing, ION...  When we look at the INPUT voltage coming into the DUT, even though the "power supply" right now is a battery, the trace shows that the circuit is gating the input in a complex way, so that the INPUT voltage has the shape of a small "h", with a high peak followed by a shoulder.  Now here is the question to you -- can I TRUST the power supply (PS) to give an accurate reading of the power (from V*I) given this actuality? 
I can use/borrow some good power supplies (one for each side of the balance) from the university... if I can trust what they are telling me.
This is an important question.  The main problem now is -- determining the input power.

2.  It is interesting that this value for the output power, using the thermocouple, is BETWEEN the simple V*I calculations (PP and RMS) I discussed above, which gives about 0.39W (using PP values) and 0.027W (using RMS values for input and output).  So a firm conclusion is that we need something better than just simple oscilloscope values and V*I...  as we all apparently suspected.

3.  MH and .99 are probably wondering -- how did you get ONE output resistor when the prototype has two resistors (in series) AND an LED on the output side?  What a pain to work with three separate elements like that...  We could try to put all three in a thermal paste as you suggest MH, but where to put the thermocouple to get a good measurement??  And how to arrange a null resistor for comparison?

It was easy to replace the two resistors in series with one resistor...  But just with a resistor across the output wires (from the output winding), LED out, the trace of the output signal was much different than with the output-LED in place.  We figured the LED had a diode function, but it also was pulling out quite a bit of power, compared to that from the two resistors (now replaced by one).  So we tried -- one resistor of equivalent resistance (close enough) to the 2 resistors AND the LED  -- plus a diode which draws essentially no power loss...  Now the output trace was essentially the same.

I know, I know, Lawrence can complain that we "changed his circuit so the results are invalid" -- maybe he will, but we're getting results that we can understand at least...  And then we may go back to the original circuit later, perhaps.  Meanwhile, we have checked that the output signal-trace is very much the same replacing the LED with a simple diode and inserting the appropriate resistance in the output circuit.

4.  Yes, we saved all important OSC traces from our various tests... but pls don't ask that I post these here yet...  We're still learning.  And I wish to get the MEASUREMENT methods down before talking about detailed results..

   
Group: Guest
PhysicsProf:

Quote
3.  MH and .99 are probably wondering -- how did you get ONE output resistor when the prototype has two resistors (in series) AND an LED on the output side?  What a pain to work with three separate elements like that...  We could try to put all three in a thermal paste as you suggest MH, but where to put the thermocouple to get a good measurement??  And how to arrange a null resistor for comparison?

I am going to comment without considering the issue of a null resistor.  The central issue that I will address is about getting valid temperature data from the thermal mass.  The thermal mass may include the three components in the original setup, or the choice of components would be at your discretion.  Because there is an non-linear LED or potentially a diode in the thermal mass, the higher the excitation voltage then proportionally higher power will be dissipated in the resistor(s) as compared to the LED (or diode).  Therefore there is an inherent asymmetry in the relative heat production between the resistive components and the LED or diode relative to the excitation voltage.  That means the relative heat production between the resistive components and the LED or diode is waveform dependent.  If the thermocouple is placed very close to the resistive components, for example, it can induce errors in the temperature measurement when you compare the waveform-based and the DC-based heat production of the load.  To be as precise as possible you want to reduce the waveform dependence affecting the temperature reading of the thermocouple.

So, the easy way around this is to increase the thermal mass and keep the thermocouple relatively far away from the individual components.  The first idea that comes to mind is as follows:  Get a metal thimble and put the load components in the bottom of the thimble.  Make sure that they are at the bottom of the thimble. Then fill the thimble up with thermal paste.  Then embed the thermocouple in the thermal paste near the top of the thimble in the center of the thimble.  Put a piece of tape on the top of the thimble to secure everything in place and then suspend the whole thing by the wires.  This should guarantee you enough thermal low pass filtering to ensure that your temperature readings are completely waveform independent.  It may take a few extra minutes for this larger thermal mass to reach temperature equilibrium.

Just some ideas, you can use them if you choose to or perhaps not.  At this point I am officially switching over to lurker mode on your project and wish you good luck.  I look forward to seeing the results of your testing.

MileHigh
   
Group: Professor
Hero Member
*****

Posts: 3017
  Thanks for the comment, MH... 
  My thought is that when the COP for a device is sufficiently large, it will not require heroics to demonstrate that the COP>1.  We may have to for this one.

It is interesting to me that the output power is as large as it is for this little toroidal device... still not convinced that the COP >1 at all, though...
The problem has to do with power measurements as noted.

I do like ION's idea of using good power supplies to keep track of the input power... although I have some questions about this as delineated above.
See you all Monday...
   
Group: Elite
Hero Member
******

Posts: 3537
It's turtles all the way down
Regarding some of the questions I have been asked about the thermal bridge method.

The method and schematic supplied is a general purpose design for most devices where it is desired to have an estimate of input / output power into a resistor load. DUT losses can be easily calculated separately.

e.g 10 Watts input power and 8 Watts output power = 2 Watts lost in the DUT circuitry.

Be creative, the method I supplied is a rudimentary starting point for power measurement on a budget. It may not be an exact fit for the Tseung device, but can be adapted to it.

The case presented by POYNT of 10 Watts input and 10 Watts output is unique and unrealistic. In this case the device can be assumed to have some overunity and the output power plus DUT losses / input power will equal the COP. We have not come very close to needing to test for this unique case yet. It is, however easily accomplished as I stated in an earlier post.

Yes, in the matched load resistor bridge, the losses to ambient must be made equal on both sides of the bridge......I thought this was common sense and would not have to be explained.

Quote
Here is a potential problem I see with your null-balance drawing, ION...  When we look at the INPUT voltage coming into the DUT, even though the "power supply" right now is a battery, the trace shows that the circuit is gating the input in a complex way, so that the INPUT voltage has the shape of a small "h", with a high peak followed by a shoulder.


You say when you are "looking at the input voltage" and you see a small "h"...this should not be happening if you have a good battery that represents a good "voltage source". I would guess your battery has a higher than normal  impedance (internal resistance) and so is acting as a partial current source. You can parallel the battery with a large electrolytic capacitor, such as used to stiffen car stereo systems. This should eliminate transient sag in your test setup. Chances are that you will not need a cap that large to do the job.

More info would be helpful (battery type, battery condition etc.)

Quote
Now here is the question to you -- can I TRUST the power supply (PS) to give an accurate reading of the power (from V*I) given this actuality? 

Depends on the design of the power supply, but generally it should be accurate enough.

Please supply a schematic and photo of your test setup, then we can comment further.


---------------------------
"Secrecy, secret societies and secret groups have always been repugnant to a free and open society"......John F Kennedy
   
Hero Member
*****

Posts: 520
In many instances replacing the battery with a power supply is not a good idea since you will be losing any potential throughput that may be required for the device to operate properly.

If the device is producing any level of flyback, that 1.5vdc battery will not be able to handle it all and simply pass it through to the other side where this effect could be unknowingly aiding the overall device to work. Sorry if I am saying this is a crude manner.

@MH

The above is what I mean by thinking out of the box on these devices. You see it in a conventional manner and treat it as such which is not bad in essence, but "maybe" and I say it again "maybe" the actual effect is occurring via some method that you would not have considered. That's why when we see devices, we try to look at it from the makers point of view and often enough the maker does not even know why it works in the manner it does.

@all

On these low voltage systems, is there a way where you can use a high enough mF capacitor with a germanium diode that together have been "pre-quantified'. I just made that word up but what I mean is that you already know in advance that to hold a voltage at x level in y seconds would required z joules, with a good range of known values. Then you switch the device inline to the cap tank and measure how long it takes for the voltage to stabilize.

Well, I don't really know fully how to do it and also it may again play against the overall function of the device.

The other option may be to get the coil ohmage values and wire guages and make a new device but with a larger toroid and comparable windings but run it at a higher voltage/amperage, high enough to be able to determine any OU without the low voltage hassles.

wattsup


---------------------------
   

Group: Administrator
Hero Member
*****

Posts: 3217
It's not as complicated as it may seem...
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

As no one has yet taken a stab at this, I'll explain where I was attempting to go with it.

ION, you mentioned this:

The method and schematic supplied is a general purpose design for most devices where it is desired to have an estimate of input / output power into a resistor load. DUT losses can be easily calculated separately.

e.g 10 Watts input power and 8 Watts output power = 2 Watts lost in the DUT circuitry.

Be creative, the method I supplied is a rudimentary starting point for power measurement on a budget. It may not be an exact fit for the Tseung device, but can be adapted to it.

The case presented by POYNT of 10 Watts input and 10 Watts output is unique and unrealistic. In this case the device can be assumed to have some overunity and the output power plus DUT losses / input power will equal the COP. We have not come very close to needing to test for this unique case yet. It is, however easily accomplished as I stated in an earlier post.

I understand your points, however imho there are some assumptions being made there that could be throwing the baby out with the bath water.

First and foremost, when we are vetting any "black box", we should not make any assumptions regarding whether or not the DUT is OU, nor what degree of OU it may exhibit, no matter what prior tests on similar devices may have shown. We are simply making a purely objective measurement and assessment on the device based on the outcome. From the purely scientific standpoint, we can not make any assumptions or pre-judgments on the device we are about to test with the black box method.

My example of 10W input and 10W output should be considered a possibility. Judging it as "unique" and "unrealistic" is irrelevant to the test, if we are using the scientific method and we are being objective and unbiased.

Quote
e.g 10 Watts input power and 8 Watts output power = 2 Watts lost in the DUT circuitry.

This is also an assumption, which by its nature, eliminates another possibility. In the similar example I gave at the top of this post, I asked what the conclusion of these results would be. Did most conclude that the DUT was under-unity in this case? If so, that would be an erroneous conclusion. We did not measure the DUT as a whole, therefore we can not assume that we obtained the "whole" picture regarding what the total output power is for the DUT.

Black box testing assumes all possibilities, therefore we can not exclude the possibility of an "OU device" before we commence testing any DUT. In my example where 15W was measured as a DUT input power and 10W at the DUT resistor, we are tempted to conclude that 5W is lost in the DUT circuitry, but we have not yet captured the whole picture. It is entirely possible that there is significantly more than 5W of power being dissipated in the DUT circuitry, such as 20W, 50W, or even 100W for example. This of course would represent poor power transfer to the resistor (especially if the goal was to heat that resistor as much as possible), however that is irrelevant.

The point being this: In order to obtain a true representation of the DUT's output power and in keeping with the scientific method and objective black box testing, the entire DUT as a whole must be measured, not just the output element itself. In the case of the device in this topic, and with reference to ION's test setup, this would involve either encasing the DUT circuitry and placing a second series DUT thermal couple on the enclosure, or placing all the DUT components inside a single enclosure, including the output resistor, and measuring this enclosure with a single thermal-couple.

And that, in essence, was the reason for my original post suggesting the requirement of a second thermal-couple on the DUT switch.

Respectfully,
.99


---------------------------
"Some scientists claim that hydrogen, because it is so plentiful, is the basic building block of the universe. I dispute that. I say there is more stupidity than hydrogen, and that is the basic building block of the universe." Frank Zappa
   
Hero Member
*****

Posts: 520
@poynt99

OK I understand about trying to measure the heat dissipation but how can that be included into a determination of OU or not. Seems to me that the thermocouple positioning will always be left to some speculation. If the hot point surface is 2 cm square and is 400 degrees versus another that is 25 cm square and all the surface area is at 300 degrees, will you say the first one dissipates more heat energy? This is the part I do not understand.

I think the first criteria is to listen to the maker of the device. If the maker says  the device is OU without factoring in any heat dissipation, then this is what we should be checking out. If the device proves to be OU, then factoring in the heat is pointless. Of course that would increase the OU factor but it is not the makers point. If the device is showing unity, then the heat factor would push it over to the OU side. But if the device is under OU, the heat factor could push it to unity. If the device is way below OU on its own, then heat or not does not matter anymore.


---------------------------
   
Pages: 1 [2] 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 ... 22
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2024-11-26, 17:20:09