STEORN DEMO LIVE & STREAM in Dublin, December 15th, 10 AM

[teslaalset]

@Omnibus,
I've looked at some excel data from your Radio Shack transformer and some questions came up:

- It seems that the input current is perfectly square shaped, while the input voltage is not.
I would expect the other way around. Normally these pulse generators are voltage driven not current driven. Can you check this?
- You mentioned core losses. Core losses only occur when you full cycle the B-H curve. In this case you only pulsed with positive current, so no real core losses.
(Only the upper part of the right half side of the B-H curve is used when this transformer is pulsed)

The only core losses that may occur are due to the viscosity of the B_r value.

[teslaalset]

You are welcome....
Peer comments on this method of measuring pulsed energy would be valuable. I guess no one has peer reviewed steorn so far. Unfortunate that.



Teslaalset, I made a separate thread for my replication when you were away. Hows your replication going ? You have the holy Metglas, throw us some bone please.

Omega, yes, I noticed you started a separate thread. Very good, and a good overview of the links you follow.
I monitor the same list (I have no additional ones)

Yes, I've received several MetGlas cores, which I have investigated intensively.
I've no motor replication yet, since I had some ideas on a solid state version, that up tilll today appear non working w.r.t. COP>1.
I also obtained a 100 Mhz DSO that I coupled via Visual Basic to Excel, so this enables me to perform the integrating power measurements.
Great stuff, but very time consuming.
So, my progress is very slow, but solid.

[Omnibus]

@Omega_0 and @teslaalset,

The peer-reviewed journal will have to wait. Here's the problem:

As you can see, I'm using an air core as the load. Unfortunately, I thought this would be a pure active resistance load of negligible inductance. Thus, as my Excel data shows, I was dividing the voltage across that air coil by its Ohmic resistance. That's incorrect. The measured voltage across the coil consists of two components only one of which is due to the active resistance of the coil. Therefore, in taking the entire voltage value across the coil and dividing it by its Ohmic resistance I have overestimated the value of the current causing the Ohmic heating.

In saying the above, I should add that Steorn do that measurement quite correctly because they use a dedicated current probe and don't rely on a poor man's solution of measuring current indirectly as I do.

To solve this problem now I have replaced the air coil by an active resistance which allows me to measure the true current. So, everything now is as expected ... underunity, that is.

@teslaalset, regarding the pulse generator -- I had the same concern, expecting it to be voltage driven and not current driven. However, that's a moot point when you consider that what matters are the actual data which you get from the properly calibrated oscilloscope. The distortion of the signal (otherwise perfectly square) is due to various factors stemming from the attached transformer but what you see on the scope screen and download from there are true values when the measurement is done correctly.

As for the Visual Basic you mention which you intend to use with Excel to automate the calculations, I'd recommend to use the beautiful macro capability Excel has, which is actually VBA and you don't need and external Visual Basic medium. Unless you intend to crunch all the 125,000 (let alone the million) points and won't settle for the REDUCED mode whereby you can only tackle something on the order of 5210 points. Excel itself cannot handle more than 65000 and something points. So, if you have an idea how to place all of the 125,000 points in arrays and crunch them in this way I'm all ears. Otherwise, Excel is just fine with its own VBA built into its macro mode.

Anyway, with your DPO coming we seem to be building a strong team which hopefully will bring the quest for OU to a new level. Also, let me mention that I already got my metglass cores but haven't done a thing with them yet because of that business with the RadioShack transformer. If nothing else, that transformer project confirms the viability of the methodology and its readiness to test  claims in the area of OU rigs having to do with electricity, if not more.

[teslaalset]

@Omega_0 and @teslaalset,

As you can see, I'm using an air core as the load. Unfortunately, I thought this would be a pure active resistance load of negligible inductance. Thus, as my Excel data shows, I was dividing the voltage across that air coil by its Ohmic resistance. That's incorrect. The measured voltage across the coil consists of two components only one of which is due to the active resistance of the coil. Therefore, in taking the entire voltage value across the coil and dividing it by its Ohmic resistance I have overestimated the value of the current causing the Ohmic heating.

Agree, always take the complete impedance as load value.

@teslaalset, regarding the pulse generator -- I had the same concern, expecting it to be voltage driven and not current driven. However, that's a moot point when you consider that what matters are the actual data which you get from the properly calibrated oscilloscope. The distortion of the signal (otherwise perfectly square) is due to various factors stemming from the attached transformer but what you see on the scope screen and download from there are true values when the measurement is done correctly.

Clear point. Scoop readings are the valid results,no matter how you drive the circuit.
It's no issue calculating the input / output power, but I wanted to double check with you because it looked like you mixed up input current and input voltage, which is not the case.

As for the Visual Basic you mention which you intend to use with Excel to automate the calculations, I'd recommend to use the beautiful macro capability Excel has, which is actually VBA and you don't need and external Visual Basic medium. Unless you intend to crunch all the 125,000 (let alone the million) points and won't settle for the REDUCED mode whereby you can only tackle something on the order of 5210 points. Excel itself cannot handle more than 65000 and something points. So, if you have an idea how to place all of the 125,000 points in arrays and crunch them in this way I'm all ears. Otherwise, Excel is just fine with its own VBA built into its macro mode.

Anyway, with your DPO coming we seem to be building a strong team which hopefully will bring the quest for OU to a new level. Also, let me mention that I already got my metglass cores but haven't done a thing with them yet because of that business with the RadioShack transformer. If nothing else, that transformer project confirms the viability of the methodology and its readiness to test  claims in the area of OU rigs having to do with electricity, if not more.

I use VBA and Excel to control the DSO and process the data as well.
It is already up and running.

Control is done with the use of VISA drivers from National Instruments (free of use).
My DSO is VISA compatible, which is a great advantage.
VISA is acting as a software interface between VBA and the driver DLL's.
It took me quite some time to get this stuff going (I am not a software guy), but it is good fun learning this kind of programming as well.

As for the limitations of Excel, I use Excel 2007, which has following limits:
Worksheet size     1,048,576 rows by 16,384 columns.

Here is nice overview of alternative spreadsheet programs:
http://en.wikipedia.org/wiki/OpenOffice.org_Calc
(a bit further down the page)

[Omega_0]

@Omega_0 and @teslaalset,

The peer-reviewed journal will have to wait. Here's the problem:

As you can see, I'm using an air core as the load. Unfortunately, I thought this would be a pure active resistance load of negligible inductance. Thus, as my Excel data shows, I was dividing the voltage across that air coil by its Ohmic resistance. That's incorrect. The measured voltage across the coil consists of two components only one of which is due to the active resistance of the coil. Therefore, in taking the entire voltage value across the coil and dividing it by its Ohmic resistance I have overestimated the value of the current causing the Ohmic heating.

Does that mean that the previous results are also incorrect ?

I still think that you will get different results at different pulse frequencies. The reactance is a function of frequency, and it is very high at high frequencies. When the frequency is low, the resistive part of the output coil + probe coil will dominate and vice versa.

E-orbo operates at frequecies below 100 Hz (my current setup goes max 40 Hz), so I think steorn's measurement is still valid.