Claimed OU circuit of Rosemary Ainslie

[ION]

In your Quantum article of October 2002 it is stated:

"Measurement of temperature rise was enabled through the use of a platinum-based temperature probe fixed to the hollow wall of the resistor as shown in fig 2 and fig 3. This instrument was chosen because it is not affected by the applied high frequency. This in turn was linked to a digital device that displayed instantaneous readings of temperature change (in degrees Celsius)."

Is the name or manufacturer of the platinum probe available?. Was it a bobbin wound or thin film probe?  Who manufactured the measurement instrument and what was the model and serial number.

I have considerable experience in the manufacture of platinum RTD probes and associated amplification, signal processing, and digital display. It is not easy to filter all frequencies, provide leadwire compensation and prevent self heating effects due to capacitive reactance of the elements, even if they are non-inductively wound.

What tests were performed to insure the instrumentation was not affected by the 143kHz to 200kHz spurious oscillation mode?

Did the platinum probe use 3 wire or 4 wire lead compensation? The compensation circuits can limit CMRR and be a pickup for noise leading to erroneous measurements.

I believe this issue was addressed in an earlier post, but I would like to dig a little deeper to put this to rest in my own mind.

Thank you kindly for any of this information and if it was already covered please point me to the page.

[Rosemary Ainslie]

Hi ION.  Not really able to help you out here.  I forwarded all the measuring tools for replicators on the Energetic Forum.  This included those probes.  Nor did we record the make and type in that paper.  But the principle is simply that the thermocoupling positioned inside the hollow of the resistor body was the chosen reference - and was also used for establishing the heat profile of the resistor.  And our assurance was that this material would be least effected by the frequencies we were measuring.  But frankly - I am open to correction.  I just followed the advices given us by the academics who discussed this and suggested that they were the preferred, if not required components.

I'm no expert here ION.  And sorry I can't give you more information regarding this.

I would add that regardless of the cause of the 'heating' there was a decision made that the 'effect' as it relates to the caloric profiling - would be considered as empirical evidence of the energy dissipated regardless as to whether it was caused by capacitance, inductance or whatever.  What we were assured is that the digital display would not be erroneous as a result of RF or magnetic interference resulting from high frequency transitions.  That, as I understood it, was what we needed to guard against.

[poynt99]

An excerpt of an on-going discussion with Dave from Tektronix. This first response received 2009-NOV-10. More to follow where he touches on the differential probes among other things:

First of all, you should be using a current probe, not a P6139A and 0.25 Ohm resistor.

What is the amperage across the 0.25 Ohm resistor?

Is one end of the resistor at ground (the same exact ground as the scope ground)?

Is the sense of current negative and the voltage positive? 

Is the sense of the current positive and the voltage negative?

Those would give negative power readings.

You probably have a ground loop involving two or more of the following:

The 2 P6139A ground clips, TDS3054C, 0.25 Ohm resistor, battery leads, load leads, load ground.



Best Regards,

Dave McDonald
Tektronix Technical Support Center
ph 503 627 1279 or 800 833 9200 press 3 
e-mail david.m.mcdonald@tektronix.com
http://www.tektronix.com

I agree.

Used properly, the current probe will resolve half of the problems being encountered with these power measurements. The grounding issues he raises would be resolved with the use of some true differential voltage probes. That's the other 50% resolved.

.99

[fuzzytomcat]

An excerpt of an on-going discussion with Dave from Tektronix. This first response received 2009-NOV-10. More to follow where he touches on the differential probes among other things:


I agree.

Used properly, the current probe will resolve half of the problems being encountered with these power measurements. The grounding issues he raises would be resolved with the use of some true differential voltage probes. That's the other 50% resolved.

.99

Hi David,

I understand a XxXxXx XxXxXx from XxXx XxXxXx XxXxXx in XxXxXxX XxXxXx XxXxXx has been asking questions on a "Device Under Test" of mine using a Tektronix TDS 3054C on XxXx to me for a second time and now through November that was set up by Lisa Bieker the Product Marketing Manager at Tektronix.

There has been a posting at a alternative energy forum by XxXxXx XxXxXx with answers to some unknown questions he submitted there .....

http://www.overunity.com/index.php?topic=7620.msg208540#msg208540

**********************************************************************

First of all, you should be using a current probe, not a P6139A and 0.25 Ohm resistor.

What is the amperage across the 0.25 Ohm resistor?

Is one end of the resistor at ground (the same exact ground as the scope ground)?

Is the sense of current negative and the voltage positive? 

Is the sense of the current positive and the voltage negative?

Those would give negative power readings.

You probably have a ground loop involving two or more of the following:

The 2 P6139A ground clips, TDS3054C, 0.25 Ohm resistor, battery leads, load leads, load ground.



Best Regards,

Dave McDonald
Tektronix Technical Support Center
ph 503 627 1279 or 800 833 9200 press 3 
e-mail david.m.mcdonald@tektronix.com
http://www.tektronix.com

**********************************************************************

The circuit in question has been posted at forums in numerous places ..... here is a link to a complete test (TEST #5) including test images, data, YouTube video links, set up photos and circuit schematic that has also been forwarded to Lisa Bieker for comment and review several times ..........

http://www.overunity.com/index.php?topic=7620.msg206788#msg206788

I am curious about the discrepancy between the approved use of the existing TDS 3054C probes doing what is required and as recommended by Lisa Bieker and now you recommend a different set of probes indicated and required for doing measurements on this "very" small circuit by you on my testing not XxXxXx of this device.

Please advise what is going on as this relates to my extensive testing and evaluation which all the information has been publically posted and not XxXxXx XxXxXx who has done some testing on this device but has not published any actual original data and only spice software simulations.

Thank You,
Glen XxXxXxXxXx
(aka) fuzzytomcat

[poynt99]

Glen,

If you recall, some time ago I informed the group that I invited David to the EF topic before I was pressured into leaving, and expectantly, he did not join in. I doubt very much he would be interested in joining the discussion here either, so I have not even bothered him with this request. So addressing him here will not likely elicit a response from David.

I am curious what Lisa Bieker's background and technical know-how is in terms of being able to provide the detailed technical support and recommendations as David does on a daily basis as a technical support engineer? I am not insinuating Lisa is not qualified, but it should not be assumed either. A Product Marketing Manager is not necessarily someone that has the technical expertise of an experienced electrical engineer specializing in measurement.

You are obviously curious (and perhaps doubtful I even sent one?) what I asked David, so here are my emails to him so far:

Hi again David.

I wonder if you might be able to help me out with a measurement problem we're having.

There are two groups making power measurements on a MOSFET switched-inductor circuit. I am one and am using a TDS784A oscilloscope, and the other group is using a TDS3054C oscilloscope.

The challenge we are facing right now is obtaining a usable measurement of the instantaneous and average power being delivered by the 24V battery supply to our switching circuit. We are placing one P6139 probe on the battery line, some distance away from the actual battery + terminal for reading the battery voltage, and we are using a second P6139 probe on a low-side 0.25 Ohm shunt in an attempt to get a good measurement of instantaneous current through the circuit.

With these two sampled wave forms imported into Excel, they are multiplied together, then averaged over the 10,000 samples or so. The frequency of operation is about 450kHz, and the time base used for sampling vary from 2us to 40us runs.

The problem arises when the results are examined for various runs over several hours. See the table below:

Battery Power AVG (W)

Hour 2µs             20µs             40µs

1 1.514827571 -2.479456000 1.500064000

2 -3.205999153 -4.045888000 -5.239488000

3 -5.533148312 -5.207520000 -1.636576000

4 -2.350759808 -1.969696000 -3.160768000

5 -2.070294717 -3.493984000 -4.054848000

6 -1.257904431 -4.098176000 -2.052608000

7 -2.550643907 -4.765152000 -3.565344000

AVG -2.207703251 -3.722838857 -2.601366857

Notice that 19 out of the 21 results are negative. Also the results seem to vary all over the map.

Can you make a suggestion as to what may be causing the wide variation and the negative values? Is there something wrong perhaps with the measurement method?

I reference this Tektronix article for consideration of the above:
http://www2.tek.com/cmswpt/tidetails.lotr?ct=TI&cs=afs&ci=14789&lc=EN

Hi Dave.

Thank you very much for the reply.

I've attached a schematic/block diagram of the test setup in the pdf file for Figure 1. Ignore the 5k pot for the Gate resistor, it is there for experimental purposes. I've also attached a scope shot of my wave forms for VP1 (supply), VP2 ((Drain), and VP3 (Shunt). and also for the other group's tests. You can see two scope shots are very similar in that the shunt voltage goes negative during the inductive kickback of the 10 Ohm ceramic 100W wire-wound resistor.

In my tests, after importing the sampled data into Excel and doing the number crunching, I consistently get a negative power for the MOSFET, and too large a power for the load resistor. The other group as you could see from the table of results in my last email, was getting mostly negative powers for the supplied power.

I have also attached the data dump related to the scope shot dated 24 Oct 2009, but note that the data was trimmed for one complete cycle not shown in the scope shot. You see in the Excel spreadsheet that after all the number crunching, I get 64W for the power in the load resistor, 20W for the supplied power (verified ok with my filtered shunt shown near the supply), and -44W power for the MOSFET

There is definitely something off with the measurements.

In my own tests, I did calibrate the probes, but I did not make any skew adjustments because when I had all 3 probes connected to the same rising edge, they looked bang on in alignment.

I did notice however, that the probe ground leads were very influential in this setup, in terms of how the wave forms change by moving the leads around, or even grabbing the load resistor. Star point grounding and 2.5" ground leads were used for two probes, and a 4" for the other (VP3) (which made a huge improvement), but still the inductance form the leads seemed to be having a large effect.

The apparent voltage across the 0.25 Ohm shunt as you can see peaks negative at about -1V and goes positive to about 0.5V.

Each probe had a ground clip and they were starred to P4 on the diagram. I feel the problems we are having is a combination of wire inductance and probe ground leads which is causing a false increase in the supply voltage at the peak, and inductance in the shunt probe ground lead skewing the wave form, which when multiplied by the huge voltage spikes computes to an artificially large value.

The supply voltage should be measured right at the supply + and - terminals, but that means attaching that probe's ground lead directly to the supply. Then we have the other two probes with their grounds attached to the low side of the shunt. Now there is a ground loop.

Seems to me several if not all of these problems could be solved by a) As you say, to use a current probe, and b) to use HV differential probes such as the P5205.


Your input is greatly appreciated, thanks Dave.

So, as you can see, I gave him information on your tests and my own as well. The discussion is on-going, and he is out of the office today.

.99