Graham Gunderson's Energy conference presentation Most impressive and mysterious

[Spokane1]

Looks nice Spokane1! Love the LeCroy.

Are you using readings from that green current transformer in your calculations? Have you corrected for any phase shift in the current transformer readings?

Dear TK,

"Oh Great Master of Measurement bestow upon this humble and unworthy novice the secrets of professional analysis. Show me the path to accurate and meaningful evaluation of the True and Pure knowledge of Performance. I have only my hands and a few crude tools to sacrifice to this higher cause."

I haven't used any of the observations for calculations nor have I corrected for any phase shift since I don't know how to do that. I was busy just attempting to get the circuit to run. The photos are from last Sunday evening and are of the primary circuit only. Now I have the secondary circuit operational and can actually harvest a little on the back end. After making adjustments and changing gate resistors I was able to collect a whooping 1.75 V across the output capacitor using a Harbor Freight DVM.

The next step is to add some additional wide band current transformers on the secondary in two places to see what might be going on there and how the harvest circuit works.

I did notice that once I started to harvest secondary energy that my tank voltage became lopsided with the initial negative swing being larger than the following positive swing. (No photos since I left my cell phone at work). Also the current trace is not very close to what Graham had, then again I'm glad that this circuit functions at all.

I just purchased a new Import LCR meter and the instruction book is all in Japanese or something, so I can't follow the calibration procedure. So, I'm not sure my static inductance measurements of the transformer are correct or not. Also I need to do some DC resistance measurements of the transformer since there is significant losses there, much more than what Graham had with his huge Litz wire windings.

Anyway this is a place to start.

Spokane1

[TinselKoala]

Looks good as always! I'm really glad you are using the logic board rather than an external arrangement of FGs.

As far as the phase shift of the Pearson current monitor, here's some info from Pearson:
http://www.pearsonelectronics.com/phase-shift

And from another site:

Phase angle errors are more complex. The current transformer introduces a phase shift (or time delay) in the AC current signal, relative to the actual current. This is commonly measured in degrees and varies from 0.2 degrees (or better) for highly accurate CTs to as high as 6 degrees. At and near unity power factor, CT phase angle errors have little effect on the measured power and energy. However, at lower power factors, such as 0.7 or below (especially below 0.5), even small phase angle errors can cause large errors in the measured power and energy.

(emphasis mine)

Correction will involve using the scope's "deskew" or channel delay function to compensate for the phase delay of the CT. Measuring the phase delay in order to cancel it by the delay setting would involve taking a Vdrop measurement across a known noninductive resistance with a Kelvin-type probe arrangement and comparing that with a simultaneous measurement from the CT in the same current loop, done at the operating frequency range of interest.


Also... the mosfets. The IRFPG50 is slow and has a high resistance of 2 ohms when fully on. I've done a lot of work with that particular mosfet... heh heh... and I don't think I'd want to use it in a high-frequency H-bridge or an energy harvesting circuit.

I'd be tempted to try IRFP260n instead, since you are operating at relatively low input voltage. Just to see if they can take the strain....         

[Spokane1]

Also... the mosfets. The IRFPG50 is slow and has a high resistance of 2 ohms when fully on. I've done a lot of work with that particular mosfet... heh heh... and I don't think I'd want to use it in a high-frequency H-bridge or an energy harvesting circuit.

Dear TK,

I figured as much for those MOSFETS, But the price is right when you have them on hand. These will probably be the first things to go as I get this circuit going as good as it is going to get using an iron core. At 3 kHz their slowness is most likely not going to impact a basic understanding of what is going on. Right now I believe that we have at least three proposed approaches for a harvesting protocol. This circuit might determine which is best.

As far as measurements go, can't we just measure the usable energy in and the usable energy out and dispense with Kelvin probes (which I probably can't afford). I have placed a current sensing resistor at the battery. Is there some method that you would recommend where I could see how much power is coming from the battery and compare that to how much power is being consumed by the load?  To me what happens in-between is not all the relevant to our goal. (except for engineering judgments) Both ends are fitted with large smoothing capacitors.  An accuracy within 5% would be good enough for me. If this system can't produce a COP of at least 5.0 then it is probably not going to have much commercial value anyway.

Dealing with AF and DC mixed for measurement purposes with $5 DVM's is not my forte, but it is probably what most researchers have to work with.

Spokane1

[Spokane1]

Dear TK,

Here is a measurement question for this morning. I'm attempting to observe the secondary current in my iron core transformer.

I'm using a Pearson Model 6164 Wide Band Current Transformer that has a sensitivity of 0.05 Volts / Amp. It is a 50 Ohm device so I set the scope input parameters to this. I projected that there was not going to be much current to observe so I wrapped five (5) turns of the secondary lead around it and reconnected it. The transformer leads are about 16" long to start with.

What I observed was not expected since none of my simulations displayed the response. The red trace is the observed current amplified by about 5X. The green trace at the bottom is the harvest pulse. In this case it is about 3 us long since this is about as fast as my slow MOSFETS can go. As you can see a large ringing effects is noticed. The frequency of this decaying burst is about 166 kHz. I really doubt that the core could respond that fast - if at all. This means that the ring must be coming from the current probe or the transformer winding.

The yellow trace is the primary current.

Is this real or a measurement artifact? If it is an artifact then how do I get rid of it?

I have set my harvest pulse to take place when the secondary was at peak current, however the $5 HF DVM goes to almost zero as I approach the peak. Perhaps because more AC is being produced rather than DC.

Also it appears that I can afford those IRFPG260n MOSFEt'S you recommended since they go for about $1.50 each in lots of 20 on eBay with free shipping. I have a batch of them coming in next Monday.

Thanks in advance for the advise.

Spokane1

[TinselKoala]

Well.... hmmm.

Now I'm confused. In previous posts you've said that the harvest pulse is turning the mosfets _off_ for a very brief time (see the diagram attached below) but now your mosfet Gate trace (CH4) looks like it's going HIGH and turning the mosfets _on_ for a very brief time and leaving them off for the remainder of the cycle.

So is the substantial current you are reading here, actually going through the mosfet body diode? What am I missing here?

What is that 50R, 10W load resistor? Is it a wirewound resistor?

Yes, I'll bet that ringing you are seeing is a "real artifact" in the sense that it is really happening but is something you don't want happening and may be caused by stray inductances due to wiring length or maybe the inductance of that load resistor.