Exploring the Inductive Resistor Heater

[gmeast]

Welcome to this thread. This thread explores one variation of an Inductive Resistor Heater.

[gmeast]

Some details of the hardware

Hi all,

The tests and test protocol I've outlined are based on an academic investigation of a particular mode of energizing an Inductive-Resistor Heater.

I have performed an analysis of the Heater's performance. The analysis DOES NOT rely on the use of expensive test equipment and can be replicated by anyone with the patience and diligence to follow through. A comprehensive testing protocol is outlined at the end of a Video Slide Show I have prepared. The Video Slide Show can be viewed on my YouTube channel at:

Inductive Resistive Heater: OU Possibilities - YouTube

You must put the following in context AFTER viewing the Video Slide Show. The analysis is of the 2nd circuit configuration below:

The PWM and MOSFET Gate Driver are powered from their own support battery common'd at GND to the circuit being tested. This is to isolate the controlling circuitry from the MOSFET Switch and the Heater Element. Therefore, the PWM and Gate Driver are being treated as sort of a 'friction-less commutator' for the purposes of this investigation. However, it is vital to consider the effect the Driver has on the entire setup. For this reason I performed several tests to determine if the Driver was contributing any appreciable energy in heating the Inductive-Resistor. The tests were performed for two circuit configurations with similar outcomes. Below are the two circuit configurations followed by scope captures showing MINIMAL (if any) contribution from the Gate Driver.

The difference in their voltage drop between Loaded and Not Loaded is 0.9mV across a 0.1 Ohm x 1% non inductive CSR ...SHDriver. The support battery runs nominally at 13VDC. That translates to 0.117 Watts and for the duration of my tests (8-hours) translates into 0.93W-Hrs of energy.

So ... the contribution from the MOSFET gate driver is negligible, but has been considered against the observed performance of the circuit and noted here. Also, the PWM's input to the Gate Driver is logic level and wouldn't be considered in the scheme of things any more than you'd consider the power a function generator is drawing from its wall socket.

The Test Circuit Configurations:


The Gate Driver Power Draw (Element and Diode disconnected from battery (B+):

The Gate Driver Power Draw (Element and Diode connected to battery (B+):


Thanks for viewing,

Greg

[gmeast]

My Batteries and Chargers

Just in case someone will want to know what batteries I'm using, I shot some photos of them. I also took a shot of the chargers I use. The larger battery is a SEC1075, 12V, 7Ah AGM (Absorbent Glass Mat) and the smaller one is the SLA0810, 2V, 6Ah, AGM. I use two 12V and one 2V in series. The battery charger is a really cool 6V / 12V Charger, P/N SEM-1562A. I have two of these chargers. One I use for charging the 12V batteries in parallel. The other charger is used to charge the 2V battery, so I have two more 2V to make up the total of 6V, so those charge in series. This is so I can charge both 12Vs & 2V at the same time ... here they are:

Batteries:

Charger:


Enjoy,

Greg

[gmeast]

Test Protocol Outline

I feel it is very important to stress some details about using batteries in conducting experiments. The points raised here refer specifically to the Lead-Acid Batteries I am using in these analyses, but I'm sure some of this will apply to other Lead-Acid Battery species and maybe other battery types in general.

Before I began my critical analysis of the Inductive Resistor Heater, I felt it was important to make an attempt to characterize the batteries I was using. First of all, my choice of batteries hinged mostly on the Amp-Hour rating of the batteries. I chose a rating based on the amount of current (I estimated) my circuit was consuming and the time interval I wanted to dedicate to a single test. I know I wanted to stay between 1/4 and 1/3 of the Amp-Hour rating for the batteries, and I wanted a test interval of 8-hours. So, I chose 7 Amp-Hour batteries because that would be .350Amps over 20Hours and my circuit (was estimated) to consume somewhere between 150mA and 250mA depending on Duty Cycle and other factors. This would give me a reasonable battery draw-down slope for my analyses (edit added) "and not degrade the battery".

So, to characterize the batteries, I put a 'purely resistive' load on them and documented the voltages over time. To my surprise, the batteries exhibited what I regarded as some peculiar behavior ... that is to say ... under load, the batteries' voltage dropped as expected and then stopped dropping and 'hovered' at a lower voltage, but then began to rise. The voltage rose to a higher voltage and 'hovered' there for a while and then began its final plunge. This peculiar behavior is simply a result of the electrochemical activity of the batteries coming to some equilibrium. The period of this 'cyclic' behavior is a function of the load. For a heavy current load, this behavior may not be as noticeable. For a large battery under a light load, it would seem as though the battery is charging or, at least, NOT discharging ... an incorrect conclusion might result.

I also believe that many misleading conclusions have been drawn by novice experimenter that think their circuits are charging their batteries when in actuality they are observing the behavior I have just described. I also think this has possibly been used to lure unsuspecting investors into believing a falsehood.

This behavior is the reason for items .01 through .12 in the outline at the end of the Video Slide Show. The entire outline from the Video Slide Show is shown below:





Thanks for stopping by,

Greg

[gmeast]

next steps

Hi anyone following this,

I recently began the next phase in showing COP>1 operation of my heater configuration. I'll try and clearly explain it below:

I am attempting to move away from relying on the discharge profile(s) of my batteries as the only criteria for determining the COP of my heater. What I'm now going to do is to determine the energy required to recharge the batteries after running the heater for a test interval. Luckily, the manufacturer of my batteries recommends a 'Constant Voltage' charging procedure which makes this less difficult than using other alternatives. I've included the data sheet for those batteries here.

After several attempts to get it right, I now can consistently determine the Energy required to recharge the batteries after being discharged ... in Watt-Hours. (W-Hrs). Using the same method for determining the Energy Output of the Heater Circuit, as outline in the video slide show, I need only compare the Heater's Output (in W-Hrs) against the Energy required to recharge the batteries (in W-Hrs) immediately following. Hopefully, the Output will exceed the Input.

This will only be possible if the recharging efficiency is high. Many commercial chargers just don't meet the efficiency requirements. Luckily all I need to do is use my Precision Power Supply since the recommended procedure is 'Constant Voltage Charging' ... so say the Manufacturers.

The tests outlined in the slide show suggested a COP = 1.25. So to prove my Heater runs at a COP>1, I need a charging efficiency ABOVE 80% in order to determine anything definitively.

So, the below attachment shows a plot of the data showing a rather 'qwerky' looking Battery Discharge curve from a 200-Ohm resistor load (maroon color), and on the same graph and time scale, the Recharge curve showing a classic Current curve (no particular vertical scale other than being proportional to the actual charging current data), (Navy Blue color). The results show a 'Discharge to Charge' efficiency of around 88% using my Precision Power Supply as the Constant Voltage charging source. So this is "doable".

(NOTE: Energy calculations were determined using the Straight-Line approximations of the curves shown ... ie. sum of the energies calculated for the data point pairs. This method does NOT require any graphical representation ... like in the video ... just number-crunching ... the more data points, the higher the resolution and more accurate the results)

The graphics:


The Battery Data Sheet:


Thanks for watching,

Greg

[gmeast]

updated protocol

For those interested,

I'll start with some history. The graph below is from the research presented in my video Slide Show wherein my Inductive Resistor Heater circuit is suggested to be displaying an operational efficiency of 125%, ie. a C.O.P. of 1.25. That's wonderful and all, but those determinations involved a little 'indirectness' in determining the Circuit's energy consumption, or energy Input because it involved the use of plotted data to determine what the energy Input "Must Have Been" (after the fact) instead of what it "Is or Was", specifically. Though I took great care in recording and processing the data, another, more 'direct' protocol was pursued. And we go to the next Graph:

From my YouTube Video Slide Show:


The following graph is already here on-site, but it also belongs grouped here as part of the technical progression. The graph below shows the results of an effort to 'directly' determine the amount of energy required to replenish the energy used from a battery after expending some of its stored assets, such as was done by powering a 200-Ohm resistive load for a period of time. In the end, it was hoped that an efficiency determination could be made regarding Energy Discharge vs. Energy Charging. This test provided that efficiency figure. The test results returned a Discharge vs. Charge efficiency of 87% to 88%. Previous and subsequent tests were all consistent and agreed to within a narrow margin.
(NOTE:Graphing the data is no longer needed as a tool to interpret, or reduce the data returned. It is nice to see it represented in this format though). And we go to the next Graph:

Battery Discharge Energy vs. Charge Energy protocol:


This final graph shows the results of the most recent Circuit test and uses the protocol referred to above to determine the Heater Circuit's efficiency. In this test, my circuit powered my Inductive Resistor Heater using a battery source, and noting the equilibrium temperature maintained by the Heater during the period of the test. At the conclusion of the test, the heater was powered by a precision D.C. power supply and adjusted to power the Heater at the same equilibrium temperature as during the Circuit test. The product of the voltage and current displayed on the power supply represents the Circuit's Output. The batteries were charged using a Constant Voltage method as recommended by the battery manufacturer.

Prior to running the Circuit, the batteries were loaded for about an hour and then recharged at a constant voltage of 29.5V until the current was down below 0.100A. The batteries were removed and let to sit until they settled at 28.84V. The power supply was hooked back up and adjusted to 28.84V and left for several hours. The current draw fluctuated from 0.000A to 0.003A, indicating a stable resting voltage for the batteries. 28.84V was designated as the starting point for the test, and also as the end resting voltage signifying the completion of the battery recharging cycle. The latter procedure was incredibly important to establish the proper baseline references for the tests.

The Circuit powered the heater and the batteries were drawn down and the voltage data was recorded. Then the circuit was unhooked and after 2-3 minutes the 'unloaded' battery voltage was recorded. Why this step ? Why care about the "unloaded" battery voltage at the end of the test? We want to know how "solid" our voltage data figures are. I have found that if the DIFFERENCE in the starting and ending "loaded" voltages is nearly the same as the DIFFERENCE in the starting and ending "unloaded" voltages then "solid" data has been returned. It makes sense if you live in my head. It should make sense to you too.

Looking at the results: The tests concluded that the Circuit performed with a 111% efficiency. OOOHHHH NOOO ... that's not as good as shown in the video Slide Show ... that showed 125% efficiency. Actually it's almost exactly as good as the video Slide Show. Remember, our charging efficiency is only 87% to 88%, leaving 12% to 13% that was wasted away. So if you take the 111% efficiency of the Circuit and add that 'lost' 13%, you get 124%. So the data agrees almost everywhere you look ... especially from test to test ... and that's a good thing.

There's not much more to say at this point other than to acknowledge some future hurdles on the way to any practical application. One goal is to get the output of the circuit to increase. Another goal is to get the thing to self-oscillate. Yet another is to decide whether to use switching or transformer technology for a mains-line-driven charging circuit, and probably more. The latter most goal has some potential buzz-kill aspects to it regarding inefficiencies ... it adds some. For this reason, it seems most vital to try and get the Circuit's output power to increase ... significantly.

Circuit Energy Output vs. Battery Recharge Energy Input:


Thanks for watching,

Greg

[aaron5120]

I am following your experiments, Greg. And this protocol seems to give more robust results than the previous one. Congratulations! Now may I ask : what will be your next steps? As a suggestion, I would prefer to eliminate the battery due to its charging inefficiency, and replace it with an ultracapacitor bank. Maybe my approach is wrong, due to the charge and discharge characteristics of a cap, which leave half of the input energy within inutilized. P=0.5CV'2
aaron5120

[gmeast]

ultracapacitor

I am following your experiments, Greg. And this protocol seems to give more robust results than the previous one. Congratulations! Now may I ask : what will be your next steps? As a suggestion, I would prefer to eliminate the battery due to its charging inefficiency, and replace it with an ultracapacitor bank. Maybe my approach is wrong, due to the charge and discharge characteristics of a cap, which leave half of the input energy within inutilized. P=0.5CV'2
aaron5120

Hi aaron,

I agree ... my instincts are to use an ultracapacitor bank in place of the batteries. Now, since I don't know precisely the source of the extra energy nor the actual mechanism(s) involved, I can't say if a cap bank will return any OU results at all. People like Bedini seem to imply that the phenomenon stems from the ionic chemical nature of lead-acid batteries. That hasn't been proven to my satisfaction however. In order to get the sort of discharge capacity and characteristics needed, a cap bank would need to be fairly large. Ultra caps are also ionic in nature, so they may work. I'm ready to transition to ultra caps, but I cannot afford to buy them.

One very important goal I must not lose sight of is coming up with some way to make the system self-oscillating, thus getting rid of the PWM and possibly the gate driver overhead. The scheme would have to produce the same waveform as I presently have.

Thanks for your input.

[gmeast]

cap bank

Just an update:


I'm still going to replace the batteries with a capacitor bank at some point, but the bank will have to be huge just to get a small enough voltage drop on the caps to allow for a a decent test duration ... hopefully 8-hours. If you examine the discharge curve for any capacitor you see it's the inverse of it's charge curve. On discharge, the cap voltage drops very rapidly over time for a given load. So the cap bank must be very big (wide - that is "parallel") if the voltage is to drop only .5VDC to no more than 1VDC over the test duration. As the voltage supply drops much below 24VDC to 25VDC, the performance drops off considerably. Lower voltage also affects the gate driver circuitry.



I'll first build a single 'series' bank for the supply voltage requirement (for 30VDC) and load it as I did during the battery tests ... around 3.2Watts. From this loading I'll be able to determine the rate of discharge of the bank for the particular capacitor specie. It then should be only a matter of arithmetic to scale the 'real' capacitor bank based on the same specie of capacitor. It could take well over 150 capacitor$ to get a sustained 3.2Watt draw for 8-Hours with only a 1VDC(max) voltage drop. It's possible to figure that out now just from the above requirements, but actual data is always best to have.


That's all for now ... only takes money.


Again, my video slide show "Preliminary Study of The Inductive Resistor Heater" is on my YouTube Channel:
gmeast - YouTube


Thank you

[gmeast]

still on track

Hi all,

Well I'm still planning to continue my Inductive Resistor research. The only thing delaying it is my lack of funds to purchase the Ultracapacitors I plan on using in place of batteries. As you will recall, I am taking this route in order to avoid the "Battery Effect" ... as it has been called. In reading the available technical information on Supercaps and Ultracaps, their discharge/charge efficiencies are not that superior to lead-acid AGM batteries, so that efficiency factor must still be considered in evaluating the C.O.P. of the system. The difference is, the discharge/charge curves are consistent and don't contain voltage plateaus and reversals. It seems that Aluminum Oxide electrolytics are the more efficient though it would take literally thousands of that type to equal a supply of 30VDC and 7Ah of storage with only a 1VDC drop over 8-hours.

As an IMPORTANT aside: It seems that many of the OU devices that have shown promise demonstrate OU performance because they are Isolated systems ... that is to say, they are NOT grounded or plugged into the power grid. I have realized this for a very long time, and it seems to be a requirement that the system indeed be isolated. This implies that a system that: 1) operates OU must 2) operate on batteries or temporary storage of some kind and 3) operate cyclically in a Discharge-Charge cycle. I indeed proved this by the conclusions of my last tests wherein I compared the Battery Discharge Energy against the Heater circuit's Energy Production and then corrected for the Battery Discharge/Charge efficiency.

Over Unity (OU), or Over Ordinary (OO) power production is REAL. What's wrong with the rest of the world guys (& gals)???????

Again my YouTube Video Channel is:
gmeast - YouTube

Regards,

Greg

[gmeast]

Merry Crhistmas

Merry Christmas to all here. I hope the New Year Brings cooperation in the promotion of Free Energy for all.

[gmeast]

forging ahead

I will be continuing my tests using Aluminum Electrolytic capacitors ... they seem to have the lowest ESR around ... next are Tantalum capacitors. In order to achieve an acceptable voltage drop across the duration of the tests, the 'size' of the required Capacitor Bank is enormous as far as the required Farads and as well as enormous in cost ... even for a 1-Hour test duration. I have set a requirement for the Cap bank to have only a 0.5VDC drop from start to finish of the test. This has set the cost of the Capacitor Bank at nearly $2500.00 for a 1-hour duration test ... OUCH !. Then I can recharge the Cap bank with little or no "Battery Effect" and see how 'REAL' the Over Ordinary or Over Unity performance really is. With this testing, I will be able to quantify the charge/discharge efficiency of the Capacitor Bank and then compare the energy to charge the Caps against the Energy in Equivalent Heat produced during the Bank's discharge by the Circuit and Heater Element.

Any real-world application of this technology (if proven valid) would be a Cyclic System ... IOW: Charge the Capacitor Bank ... and then Discharge the Capacitor Bank through the Circuit and Heater Element. Please refer to previous posts here and to my YouTube presentation for details of the Circuit and Heater Element at:

Preliminary Study of The Inductive Resistor Heater - YouTube

Regards,

Greg

[wayne.ct]

Current Sense Resistor

@Greg I wish to commend you on your efforts, the time spent and the dedication you have demonstrated. I have followed some of your earlier posts but did not read this thread until just now. I admire your attention to detail and the quality of the documentation you have assembled. I hope the lack of contributors to this thread has not been a discouragement to you. I wish you well and hope you have continued success in this your pursuit.

This is my observation and question for your consideration. When you measure the voltage drop over your current sense resistor (CSR) SH3 in particular, what kind of meter are you using? And, when you compare the inductive resistor setup to the battery characterization setup, are you considering the different signal characteristics over the CSR? My concern is that with a non-sine wave pulse on the CSR, you might not be getting the accuracy you presume.

The cost of doing this type research is also one of my concerns. To be clear, I wish I had more resources to throw at my own experiments! For what it is worth, in some of my experiments, I put one or two large capacitors in parallel with my batteries. The idea behind that is to provide a low resistance source of electrons to the intermittent loads that I put on the power supply. I am doing something similar to what you are doing, in a way, because I am driving a MOSFET with a 555 pulse generator. When the MOSFET turns on, the internal resistance of the battery becomes a large portion of the total resistance in the entire circuit. The capacitance acts to bypass some of that internal battery resistance.

I hope that all makes sense and perhaps gives you an idea or two to think about.

[gmeast]

@Greg I wish to commend you on your efforts, the time spent and the dedication you have demonstrated. I have followed some of your earlier posts but did not read this thread until just now. I admire your attention to detail and the quality of the documentation you have assembled. I hope the lack of contributors to this thread has not been a discouragement to you. I wish you well and hope you have continued success in this your pursuit.

This is my observation and question for your consideration. When you measure the voltage drop over your current sense resistor (CSR) SH3 in particular, what kind of meter are you using? And, when you compare the inductive resistor setup to the battery characterization setup, are you considering the different signal characteristics over the CSR? My concern is that with a non-sine wave pulse on the CSR, you might not be getting the accuracy you presume.

The cost of doing this type research is also one of my concerns. To be clear, I wish I had more resources to throw at my own experiments! For what it is worth, in some of my experiments, I put one or two large capacitors in parallel with my batteries. The idea behind that is to provide a low resistance source of electrons to the intermittent loads that I put on the power supply. I am doing something similar to what you are doing, in a way, because I am driving a MOSFET with a 555 pulse generator. When the MOSFET turns on, the internal resistance of the battery becomes a large portion of the total resistance in the entire circuit. The capacitance acts to bypass some of that internal battery resistance.

I hope that all makes sense and perhaps gives you an idea or two to think about.

Hi wayne.ct,

Thanks for your input and sorry for the delayed response. I very much appreciate that you carefully studied my research specific to these tests. You have no idea how nice it is to see someone who thinks before they speak. The CSRs I use are non-inductive. I ran tests early on before the published tests to confirm that the current being read was accurate. What I did was use the family of non-inductive resistors and recorded the temperature and power both pulsed and steady-state, and found the numbers to agreed within 2% - 3%. I acknowledge that "pulse" energy measurements are very hard to quantify at times, but my pre-tests with the non-inductive resistors was the best I could do. I did not publish the data from those tests because they were performed to satisfy my OWN doubts ... maybe I should publish those tests.

The 'capacitors in parallel with the batteries' is a sensible approach but the reality is: 'the batteries are still there'. That's why I will eventually transition to a capacitor-bank-only for the power source ... talk about "expense"!!!!!!

Thanks again.

Greg

[gmeast]

just an update

Hi all,

I posted an updated video that simply corrected the slide presentation. There were 2 corrections made. 1st was a correction to the first datasheet wherein I showed the power output of the circuit/heater-element when it had not yet been determined (in the very bottom line). That's only because I generated final data sheets from the raw data after the completion of the tests. Then I decided to make the video slide show and simply converted the documents into slides and didn't edit things properly. Now things match the chronology/sequence. 2nd, in the charging/start-up sequence outline at the end of the presentation, I showed that I exited the sequence at step 0.05 which was in the middle of the sequence but should have been at the end ... step 0.12 ... like I actually did.

Don't ask ... don't know why I put it down that way initially. I obviously ran the tests after the batteries stopped spazzing out ... which was the reason for the entailed "start-up procedure" in the first place.

You can go to my channel and see the corrected slide show ... no music ... your welcome ... plus the other stuff I've shared.

gmeast - YouTube

Regards,

Greg