Quote from: TinselKoala on May 14, 2012, 03:40:45 AM
@GL, you are welcome, I'm glad I could oblige. Can you explain the significance of our findings?

Also, I hope you will continue on and explore the negative voltage bias mode with oscillations and without Q1 turning on,  too. I already measured this mode using the oscilloscope last week and I found a small AC ripple on top of the usual 180 mA main DC current or so that I like to run at. In this mode it seems to me that the bias ammeter and the main ammeter will read the same, I think, and the current is DC but with that little ripple on top from the oscillations. This mode seems to put the bias supply voltage in series with the main battery voltage, I think.

TK,

Yes, that is my next thing to check, the oscillation mode, or AC mode as I call it.
But it has to wait until the weekend. I may find time before that. Don't know yet.

GL.

Quote from: TinselKoala on May 14, 2012, 03:29:06 AM
Uh..oh.
I just noticed something.

We have feedback oscillations, call them parasitic or resonant or whatever, they are the same thing as the squeal on a PA system when the mic is too close to the speakers.

And we have negative feedback, a different thing that is causing the main current to decrease when the Q2 is in, during the positive bias mode.

And we have negative bias, a different other thing that causes the oscillations to happen in the first place during the negative bias mode.

And we thought that the confusion between power and energy was bad.............



This circuit cracks me up, I'm telling you. I can hardly keep a straight face.

TK,

I think the reason for oscillation is that the circuit will be an Colpitts oscillator when you have
the 50 Ohm function generator at the Source of the MOSFET. When the function generator
is at the negative pulse then we pull the Source negative, thus getting a positive bias for
the Gate, relative, since the Gate is grounded. I think I explained this earlier in this thread
by some drawings. The negative feedback needed to keep the oscillation going is through
the Drain Source capacitance. The LC tank circuit needed is the RLOAD and in the wires.

GL.

TK,

Here is my analyze of the TK setup when NOT in oscillator mode.

Current through circuit from the +24VDC to minus = 24 / (10,3 + 2,0 + 0,2) = 1,92 Ampere.
This is without the BIAS current calculation.

BIAS current:

The current measured was 0,09 Ampere.
Now voltage drops to find the voltage over RdsOn:
Rmeter = 1,8 Ohm. Rdiode = 1,8 / 0,09 = 20 Ohm. Rfg = 50 Ohm.
Volt Rmeter = 0,162 Volt.
Volt Rdiode = 1,8 Volt.
Volt Rfg = 4,5 Volt.
Volt over RdsOn = 13 -(0,162+1,8+4,5) = 6,462 Volt.
RdsOn Ohm = 71,8 Ohm.
Total bias load = 71,8 + 1,8 + 20 + 50 = 143,6 Ohm.
Check: 13 Volt / 143,6 Ohm = 0,09 Ampere.

Main current with bias current taken into consideration:

The current measured was 1,78 Amperes and the battery voltage, running, was 23,8 Volt.
How to explain the current reduction when the bias current also was running through RdsOn?

I have no idea right now. And we still have a paradox going. The main current sees a
RdsOn of 2 Ohm but the bias current sees a RdsOn of 143,6 Ohm. So .99 if you are
reading this, can you explain this?

GL.

Groundloop:

The reduction of the main loop current is because when you have bias current running through the Q2 MOSFET, that causes a voltage drop across the Q2 MOSFET.  That voltage drop looks like counter-EMF to the 24-volt main supply, and works against the main supply.  Therefore the current in the main loop is reduced.

To the main supply, the Rds looks like (6.462/1.78) = 3.63 ohms.   To the bias supply, the Rds looks like (6.462/0.09) = 71.8 ohms.

In reality the Rds of Q2 is less than 3.63 ohms.  In both cases there is a counter-EMF effect taking place because of the voltage drop across the Q2 drain to source resistance.

Take the example of the bias supply current loop.  Let's say that the real Q2 Rds is 3.5 ohms.   What the +13-volt bias supply 'sees' for the Q2 MOSFET drain-to-source voltage drop is ((3.5 ohms x 0.09 amps) + (3.5 ohms x 1.78 amps)).  The second term is the counter-EMF generated by the main current of 1.78 amps flowing through the same MOSFET.

So this added counter-EMF created in the bias current loop caused by the main current loop will act to reduce the current flowing in the bias current loop.  That makes the Q2 Rds appear to be higher.  It's not actually higher, you just have to imagine that there is an extra voltage source 'inside' the Q2 MOSFET that is working against the +13-volt bias supply.

So the apparent resistance of the Q2 Rds is approximately equal to ((3.5 ohms x 0.09 amps) + (3.5 ohms x 1.78 amps))/0.09.

This stuff would have to be checked with some measurements and calculations but I think it's right.

What happens is each of the currents act to reduce the other current because of this counter-EMF effect.  That's why I said determining i1 and i2 is not trivial.  i1 affects i2.  That change in i2 affects i1, and it goes in circles.

MileHigh

Needless to say this is another "surprise" for Rosemary and the NERD team.  All of this stuff is just basic circuit analysis.  You measure your currents and voltages in the circuit and then understand and explain them so that you can get a basic understanding of how the circuit actually operates.

This measurement and subsequent analysis was clearly outside the capabilities of Rosemary and the NERD team.  It's likely that the function generator output current was never measured by them when they had their setup running and they were never aware that this was happening.

If I was on the bench I would have checked the function generator output current when the output was high.  I would have been expecting to measure microamperes at the most.  Seeing milliamperes was a surprise so you go and do your investigation.  After the investigation is done you realize that it wasn't a surprise after all, and the circuit was doing exactly what it was supposed to be doing.

This 'unexpected' current flow is simply the result of what the circuit really is - an accidental miswiring of some MOSFETs.  All of the hot air for the past year has it's origin in this accident.

MileHigh