Bedini's Linear Current Amplifier

[Joster]

Hi All,

This is really cool....not really sure how it works but I suppose it coverts the voltage from the charged cap to constant current?

http://youtu.be/be7_M_Xf5Iw

[Tom C]

yes that is what it does. We are working on pricing and availability with John right now.

Tom C

[aln]

JB sure knows how to keep us coming back for more, i'll start saving... again...

[Ecancanvas]

Been working on a linear charger/generator. Not sure what john did to control the Transistor. I just worked out the filters then put on a TI battery charger circuit on the end. I have ordered some parts to start working on this project. Just looking for some tech input, better way to do it. got to start some where.
thanks Ed

[Nityesh Schnaderbeck]

Very nice to see another schematic, of a charger circuit, Here is a schematic I have been working on, I hope this can help, it is still in development stages, modification may be necessary, like resisters values and stuff.

The feedback resistor R13, C3 (at the base of the beta multiplier), will need to be performance tuned. Also the damping factor is also worth investigating for performance tuning.


linear amp regulator.pdf

[Nityesh Schnaderbeck]

Been working on a linear charger/generator. Not sure what john did to control the Transistor. I just worked out the filters then put on a TI battery charger circuit in the end. I have ordered some parts to start working on this project. Just looking for some tech input, better way to do it. got to start some where.
thanks Ed

I had a closer look at your design, I can see in the video "part 37" John Bedini, did not have the main current flowing though the regulators. I know this because the regulators were not attached to a heat sink, it was the beta multiplier that got hot, not the regulators. And I believe he used a plus and minus regulator, eg Lm7812 and Lm7912. With your design you will have to have the LM338 on a heat sink, as the main current is passing through it. If you have a look at the John Bedini's "Linear current regulator" you will see he has 3 devices on a heat sink, 2 x 2n3055 and a transistor, it looks like the amplifying transistor is a pnp, because the positive output to the battery is connected to the center pin of the transistor, which is the collector. The 2 big resistors are big because they have to carry lots of current. One of the big resistors, the one that balances the emitter is 0.1ohm, I believe this is the current sensing resistor, the voltage drop across the 0.1ohm resistor, is very small. And needs to be amplified with a servo amp, and mixed with the battery voltage. With a summing amp, so the summing amp has a voltage that represents the impedance of the battery. Which is used to control/(change the impedance of) the regulators and the beta multiplier.


I think that this "Linear Regular Amplifier" is an impedance regulator, and it regulates the impedance of the battery, not just the voltage but the voltage current ratio. Maybe it could be helpful to think Impedance regulation instead of voltage regulation. And it is the Impedance of the battery that determines the voltage. So if that was true then, each different battery would require a different adjustment, because different batteries will have different impedance's at 15V.

If you get a flat battery and try to charge it, it draws not much current, and can go strait to 15V, if you were only measuring voltage you would think it was charged, only to find that the battery doesn't run anything. But if you were measuring a signal that was a mix between the current into the battery and voltage across the battery, you will know the battery if the battery is charged or not.

OP amps and comparators work better with a dual power supply, eg +- 15V. A voltage divider, with caps across the resistors can half the voltage, and have plus and minus 15V, with respect to the center of the voltage divider. The center of the voltage divider, can be your ground for your op amp/comparter circuit. The ratio of the resistors in the voltage divider, may have to made different to compensate for the other resistors the are connected to ground. (op amp ground).

Here is another interesting thing, So you have the cap charging to double voltage of the battery, and the output of the beta multiplier is double battery voltage, then the positive terminal of the battery will be zero with respect to op-amp ground, so I am thinking that the op-amps/comparators are always trying to zero (servo) the battery voltage with respect to op-amp ground. I also think that the feedback that is going into the beta multiplier, is telling the beta multiplier always be double, the battery voltage., but an average of the double voltage, as the capacitor (I think this value of this cap is important for performance tuning) at the base of the beta multiplier is also filtering any high frequency fluctuations from the battery.

I also think that the op-amp ground is held between the emitter of the beta multiplier, and the negative of the pre-filter cap (the 15,000uF one). Because the op-amp ground has a smooth average, from the beta multiplier, the high frequency impedance fluctuations appear as small plus and minus voltage differences between op-amp ground and the battery voltage, these fluctuations are used to zero the battery voltage with respect to op-amp ground, controlling the output device/amp to do it.
It is almost as if the battery is controlling it's own charging, and telling the linear amp regulator how to do it.

So the ions in the battery are controlling the "linear Amp Regulator".

[guyzzemf]

been awhile N_techo';s GOOD TO SEE YOU HERE

[Nityesh Schnaderbeck]

yes, I have been spending all night and all day figuring out this linear amp regulator.

I have tried a number of circuit configurations, all so far load the ssg.

I have a hall sensor, that detects the magnet going past, and the pulses are fed into a frequency/tachometer which I purchased from ebay at $12.00. So I can very accurately see the speed variations. This will tell me if the linear amp design is working, to find out how not to load the wheel.

I did not want to take the pulses from the ssg transistors or the coil, in case of oscillations which will throw my magnets per second reading out the window.

[Nityesh Schnaderbeck]

It is so good to talk electronics with someone else. Designing these circuits is more fun than testing batteries, forever and ever.

[Nityesh Schnaderbeck]

If you hook the linear amp regulator to a normal ac transformer, I bet it would not work. Charge a cap with normal power it charges slowly, and when you short it out, the spark is orange red yellow, but if you charge a cap from the ssg it charges fast and when you short it out it has a green in the spark, and that "green spark electricity" is charging your battery. Did you know that electrons have mass, and normal physics say that anything with mass has inertia. So maybe electrons have inertia too. I think that a coil puts the electrons out of phase with the holes. Because electrons have mass and holes are mass less charge, a charge with no mass can travel faster than the speed of light.
And when this mass less charge charges a cap the electricity is different, a mix between positive electricity and negative electricity, with a special phasing arrangement between electrons and holes. Could it be that electricity can have many colors, and each color has it's unique properties.

[Ecancanvas]

Thanks Nityesh for spending the time on this. I am working my way through your design. i tried what i posted and it just pulls my system down till my wheel stops I see where your going. I see your dumping the extra load not used from the 30 zener through transistor. Why not just the Zener?

[Joster]

Wow! I have been banging my head against a wall trying to figure out what is going on in this circuit. You guys obviously have a lot more electronic design experience than me. These schematics are an amazing start. You guys rock! I learned alot from Lman...he also pointed me to the 160Amp solar tracker thread. John points out a few details about whats going on there too

[Nityesh Schnaderbeck]

Originally Posted by Ecancanvas

Thanks Nityesh for spending the time on this. I am working my way through your design. i tried what i posted and it just pulls my system down till my wheel stops I see where your going. I see your dumping the extra load not used from the 30 zener through transistor. Why not just the Zener?

In my first circuit I did that, only to find my zener gets damaged, So I designed the "clamping voltage reference circuit" to handle more current. Also you want to make extra measures that this zener does not go open circuit. If this happens the voltage from the beta multiplier will raise to 80V, the neon on the ssg will flash. Speaking from experience. I put a clipping LED in the circuit so I can see when it is clamping, this happens when there is no load, when the battery is charged, and if the battery is very high impedance. When the pre-filter cap (the 15,000uF one), is not getting dumped the voltage will try and raise above 30V. anything above 30V gets dumped into the clamping circuit. To protect the linear amp regulator from overvoltage.

Just like the neons protect the ssg transistors, this clamping circuit protects, the linear amp regulator.

Linear circuits don't like 80V, comparators/op-amps a maximum of 30V and about a maximum voltage drop of about 40V between the input and output, for the 3 terminal regulators.

[Nityesh Schnaderbeck]

Wow! I have been banging my head against a wall trying to figure out what is going on in this circuit. You guys obviously have a lot more electronic design experience than me. These schematics are an amazing start. You guys rock! I learned alot from Lman...he also pointed me to the 160Amp solar tracker thread. John points out a few details about whats going on there too

Yes it it very difficult when you have holes in your knowledge, I have lots of those, lol.

Servo amps, is a key, http://en.wikipedia.org/wiki/Servo_drive

In the case of the "Linear amp regulator" the charging the battery is like driving the motor, and the speed sensor, is like the impedance sensing circuits in the linear amp regulator. Just like being able to control a motor to a certain speed. The linear amplifier regulator can charge the battery to a certain impedance, set by a command voltage.

YesCommand voltage, voltage references, comparators/op-amps, summing amps, drive circuits, sensors or sensing circuits, feedback amps, and performance tuning. lol and all that good stuff.

[Nityesh Schnaderbeck]

linear amp regulator.pdf

I will explain how this circuit works, starting from C1. C1 (15000uF) is the pre-filter. R2,R1,ZD1,Q3,Q4,LED1 and C2 is the clamping and voltage reference circuit, and provides a zener controlled, voltage reference to the Beta Multiplier. R3,C3,Q1 and Q2 is the beta multiplier, which is a second filtering stage, and protects the linear circuit from radiant spikes. C4 just takes out the ripple a little bit more. Reg1,Reg2,RV1,IC1A,R4,LED2,LED1,R5,RV2,R7 and ZD2 is a plus and minus adjustable voltage regulator, that is controlled by the feedback signal, via pin 2 of IC1A. RV1 Controls the plus and minus offset with respect to op-amp ground. R7,ZD2 and RV2 provide an adjustable voltage reference from 0V up to 16V.
RV2 sets the impedance the battery charges to (Voltage adjustment for battery).
IC1A compares the difference between the adjusted voltage reference set by RV2 and the
feedback signal. And sends the difference to the control the regulators (Reg1 and Reg2).

Between R10,C5 and R11,C6 is op-amp ground.

IC1B is a summing amp and a difference amp. Pin5 of IC1B is the summing point for resistors R6,R9,R16 and R13. The summing point (Pin5 of IC1B) is compared the summing point of pin6. Pin6 is the summing point of R8 and R12. R14 limits the current going into Q5. R15 is the current sensing resistor, it turns current flowing through it into a small voltage drop that represents the current flowing into the battery. The voltage drop across R15 is amplified by IC2A, and fed into the summing point pin5 of IC1B. R13 is the feedback resistor, this looks at the voltage of the battery. The voltage signal from R13 is mixed with the current signal from R16. The mix of the voltage signal with the current signal gives a voltage that represents the impedance. This impedance feedback signal is fed into the Beta multiplier (through R6), to IC1A which controls Reg1 and Reg2. R1 and R6 make sure the output of the beta multiplier is held to double battery voltage, but an average of double voltage because of C3 at the base of the beta multiplier (Q1 and Q2 is the beta multiplier)

Q5 and Q6 drives the charging to the battery controlled from IC1B.

The output of the beta multiplier (the emitter of Q1) is double battery voltage, but an average of double voltage, because of C3 at the base of the beta multiplier (base of Q2), which is filtering out any high frequency jitter, from the impedance feedback signal.

The output of the beta multiplier is smooth double battery voltage Lets say 30V because the battery is 15V. Between (op-amp ground) R10,C5 and R11,C6 the voltage is divided into half 15V. But this is a smooth 15V with no jitter. IC1B is always trying to zero the battery voltage with respect to op-amp ground and driving/switching q5 and q6 to do it.

Imagine a graph of a charging battery, that is being pulse charged, if you zoom in you will see some jitter, high frequency jitter. Ok so you have also monitored the battery through a filter (Resister capacitor filter) The second graph through this filter will have no high frequency jitter. Take the difference between the filtered charging graph and the unfiltered charging graph and you get left with the battery jitter, which represents what the ions in the battery are doing.