Partnered Output Coils - Free Energy

[Jimboot]

There are literally a zillion DC to DC converters

Gah!

[MarkE]

There are literally a zillion DC to DC converters and this link is just an an example not necessarily tuned to this application:

http://www.digikey.ca/en/product-highlight/m/murata-power-solutions/spm1525-series-singleoutput-potted-isolated-dcdc-converters?WT.srch=1&mkwid=sWhX1e14b&pcrid=71384157545&pkw=_cat:converters&pmt=b&pdv=c

I am sure the experts could give good guidance on this.

Something like:

Input range:   8 to 15 volts
Isolated from input to output  (may as well get one that is input-to-output isolated because it is much more versatile.)
Output voltage:  12 volts
Output max wattage:  20 watts
Package:  Just go for a bare board because that the cheapest
Price:  I am going to guess $20 to $25 USD.   You just have to buy a very commonly used and cheap one.  Exactly which one that is I am not sure.

What's inside the guts:  It's something like a "smart Joule Thief" that pulses an inductor at perhaps 30 KHz.   The output of the inductor goes to an output capacitor.  There is an electronic control system that controls the pulsing to keep the output cap charged to the desired voltage.

They can be bucking converters, boosting converters, or Cadillac buck-boost converters.  The efficiency is typically 90% or higher.  Very high efficiency ones cost more.  All that they really do is make sure that the coil is pulsed for a very very short time to avoid all of the i^2R losses.

Note you have all of the buzz for Joule Thieves that has been going on for years.  However, in the electronics industry they have been using "smart Joule Thieves" for years in quantities of several billion per year or more.

MileHigh

There are surely many devices that can do the job, once we know what job it is that we need to do.  I outlined a series of steps before towards that end.  The big issues:

A. Is isolation required?  From the existing schematic, the stator coil circuitry appears only on the output side, and appears already galvanically isolated from the rotor circuitry.  Tinman says no.  But prior experiments where the sope ground of one channel was on the negaive side of the input and the scope ground of the other channel was on the negative side of the output strongly suggests isolation.  In that case we can use a non-siolated DC-DC, or depending on the V-I curve of the output, no converter at all.

B. What do the voltages look like under the boundary conditions:  Turn-on no load, turn on with load, turn off without load, turn-off with load.  If we are going to ask Tinman to plug in some converter, we need to know the voltage stresses that it will have to tolerate.

The stipulated restriction is that we have to figure these things out without the benefit of knowing the precise details of Tinman's circuit, or at least acting as though we don't.

[poynt99]

As I mentioned before, I currently don't have 100% confidence in the measurements thus far, but I've promised Brad that if he faithfully follows the setup outlined below, I will stand by the results 100%.

The key difference from what seems to have been done thus far is that the current is measured with a CSR. I've asked Brad to make sure he captures scope shots of V1 thru V4. (some may find fault with V3; it can be measured either side of the CSR with little difference in the computed Pout)

Having agreement between the scope and DMM would solidify it for me.

[picowatt]

As I mentioned before, I currently don't have 100% confidence in the measurements thus far, but I've promised Brad that if he faithfully follows the setup outlined below, I will stand by the results 100%.

The key difference from what seems to have been done thus far is that the current is measured with a CSR. I've asked Brad to make sure he captures scope shots of V1 thru V4. (some may find fault with V3; it can be measured either side of the CSR with little difference in the computed Pout)

Having agreement between the scope and DMM would solidify it for me.

.99,

I am all for any number of tests that further confirm Tinman's measurements.

However, in reply 3812 on page 255, Tinman did verify his input current using a 1R CSR and his scope.  It looked like there was agreement within 2% or so with what the DMM measured.

Regarding his output, when he performed the "same current thru the load using the bench supply" test in his last video, it seemed there was a 5% or so spread in the voltages measured between the bench supply, scope, and DMM while delivering the same load current  (I believe the voltages variously indicated were 9.4V, 9.6V, and 9.9V)  I cannot recall how close his bench supply ammeter and the DMM ammeter readings compared, but I do not recall anything way out of the ball park.  (I attempted to verify these observations but I see that video has been removed.) 

Given the amount of observed "OU", the measurements seem accurate enough.

It certainly would not hurt to verify the calibration of all measurement devices using a known voltage and resistance.

Has Tinman agreed to perform the test you propose using the scope and CSR?

PW

[poynt99]

PW,

I'd still like to see the tests done again with a non-inductive, 0.25 Ohm CSR. I'd like to see the scope shots as well, particularly across the CSR's. In addition, the CSR voltage measurement is critical and should be made as close as possible to the resistor body.

Brad has not yet responded to my post.