View Single Post
Old 05-02-2010, 09:23 AM
FuzzyTomCat's Avatar
FuzzyTomCat FuzzyTomCat is offline
Silver Member
Join Date: Jun 2008
Posts: 620
Send a message via Skype™ to FuzzyTomCat
Originally Posted by Harvey View Post
Thanks Glen,

As always I am impressed by your work

I was trying to do some basic calculations on how long your two batteries can sustain a 5.5 watt load. I come up with about 104 hours, does that sound right? They are each 12Ah batteries so there is 24Ah of charge in them. A basic DC breakdown is 5.5W / 24V = 0.229A. 24Ah / 0.229A = 104 hours.

So all we need to do now is run for more than 104 hours on those batteries and we have pretty good proof that we have extra energy coming from somewhere else And that's not even counting the lost energy in MOSFET or CSR to heat. Good Stuff!

ETA: Oh, I almost forgot - if we conclude that those Gel-Cell (edit: wait, those or Liquid Acid?) batteries are discharged when they reach 10V each, then that would be a drop of 4V over the 104 hours. That would give us a 0.0385V (38.5mV) drop per hour. So for the 5 hours we would have expected a minimum of 193mV drop not counting the energy spent on the MOSFET and CSR. Our results show only 110mV drop in that time frame, 83mV short of the linear projection. So you can see why we think we are getting energy from somewhere. Either that, or our battery discharge is not linear And BTW, it only gets better for us if we conclude the battery voltage should be lower than 10V when discharged (of course we all know that the battery voltage needs to be measured under specific load conditions)

Hey Harvey,

I'm sorry it took so long to do a detailed overview of the "LIVE" broadcast I did in the "Open Source Research and Development" channel on the January 9, 2010 5 Hour non stop video recording.

This video as you are aware is one of the best ever recorded representation of the preferred mode of operation but only in a non stop 5 Hour video. I'm sure that many members and guests don't realize the difficulty in capturing this effect for the purpose of recording the data properly and if given the time looking at the recorded video everyone can see the problems that we face in getting accurate data.

The constant 24 volt battery bank voltage fluctuations going up and down the Mosfet "drain" spike oscillating from 500 to 900 volts, battery voltage down the Mosfet spikes, battery voltage up the Mosfet voltage to normal operating range, back and forth over and over.

I have tried to get as close to this mode of operation in Test #13 which was used in the IEEE submittal Open Source Evaluation of Power Transients Generated to Improve Performance Coefficient of Resistive Heating Systems the team including yourself did, and in Test #22 but never being able to record the data scientifically correct because of the circuits complex oscillating waveforms. I don't think everyone, members and guests understands that the Test #13 was done with a Tektronix TDS 3054C which has a maximum resolution of 10K of data spread over a 10 x 10 grid or divisions so each one has 1k of data samples separately for each of the 4 channels. The data collected in Test #22 was with a Tektronix DPO 3054 which has a maximum resolution of 5M of data, but I used the 100K which is spread over the same 10 x 10 grid or divisions so each one has 10k of data samples separately for each of the 4 channels ..... ten ( 10 ) times the data of the TDS 3054C used in Test #13.

The problem being we need to find a method of capturing the data continuously in real time, there's nothing wrong with Tektronix TDS 3054C or the DPO 3054 these are the finest instruments I've ever used and are extremely accurate, but if you push the acquire button at the wrong time you can appear to get conflicting or skewed data, not the case .... were you before the spike, during the spike or after the spike when the data was collected. I had a allotted dedicated set time to record the data, It was the time frame I used with the 6 minutes or as fast as the data could be physically collected with the finest equipment I had at my disposal.

I am in total agreement with you that something "good" is happening in the Mosfet Heating Circuit and can be plainly seen in the recorded videos, we just need to somehow get a streaming real time data recording. Maybe by somehow obtaining a Real-Time Spectrum Analyzers from Tektronix or some other method to verify the data findings as you suggested, the equipment I previously used as good as it is, just isn't enough to totally capture what is occurring during the preferred mode of operation.

Best Regards,
Reply With Quote