Bedini's Linear Current Amplifier

[Nityesh Schnaderbeck]

Mr Bedini talks about servo amp, summing amp, and comparator. but his Op Amp is only 8 pins so the most he can have is 2 op amps unless one is hidden on the board that you cant see, or he is preforming the summing or servo current sensing an other way.

"Ecancanvas" You know the previous circuits with the op-amp for current sensing linear amp regulator.pdf

Guess what "it is a current source too" yes, no op-amp current sensing needed, the output transistor is the current source. So 2 op-amps.

So the circuit would look something like this Linear Amp Regulator-npn.pdf with 2 op-amps.

Have a look at these links http://sparksandflames.com/p15.html

http://sparksandflames.com/p16.html

"Joster" Use star grounding. This means run all grounds to a single physical point, for your PCB design. This is also mentioned in the first URL. Star grounding eliminates "earth loops" and stabilizes the circuit.

A star connection, at the negative pin of the pre-filter cap, and a thick track (thickened with solder) from "battery negative" to the "negative pin of the pre-filter cap".(the 15,000uF one).

A star connection, at the positive pin of the pre-filter cap, so you don't have "earth loops".

A star connection at op-amp ground, this eliminates unwanted noise entering the high impedance inputs of the op-amps.

A star connection at the emitter of the beta multiplier would not hurt, and a nice "thick as short as possible track" (thickened with solder) between the "emitter of the beta multiplier" and the output device.

If you wish to make extra measures, have star connections on all of the op-amp inputs.
All the resistors connecting to inputs/"summing points" of the op-amp/s connect to a single point at the input pin/s of the op-amps.

This is like designing a circuit board for an audio amp.

I will indicate the star connections on my future schematics.

[Nityesh Schnaderbeck]

This is a another current source, this is very similar to the current sensing in my previous schematic, except in my schematic I had a high side current sensing and this circuit in this link uses a low side current sensing.

http://www.analog.com/en/circuits-fr...cn0099/vc.html

This circuit could gives some hints of the linear amp design, too.

[Nityesh Schnaderbeck]

Linear amp Regulator, Regulators as audio amps

I have just been reading the "140 amp solar tracker" thread and John Bedini was talking about linear regulators as a class A audio amp.

Check it out, http://www.electroschematics.com/668...dio-amplifier/ super cool. And this circuit works and sounds really good, I had it working on a 12V computer power supply, I changed the 5K adjustable resistor to 1K, and the 10ohm resistor to 39ohm. And used my laptop to supply the music.

So now to think of the regulators,op-amp and output device as an audio amp. The plus and minus regulators, as phase splitters, to feed the inverting and non inverting inputs of the op-amp. The op-amp drives the output transistor. The feedback from the battery/load is feed into the input of the regulator phase splitter.

In order to achieve negative feedback between the input and output, The amp needs to be an inverting audio amp.

So I should be able to make regulators,op-amp and output device into a class A audio amp. Something like this example circuit

Audio Regulator Amp.pdf

Current limiting is achieved by R8, (the emitter resistor), Because R8 is inside the negative feedback loop, the voltage drop caused from current, is turning off the output device, in order to limit the current.

Why do I want to make an audio amp, because it's fun. And this is a big step toward success.

And here is a link for Class "A" audio amp design. http://hyperphysics.phy-astr.gsu.edu...nic/npncc.html

This also means looking at the load line of the output device/transistor.

[Nityesh Schnaderbeck]

I have a voltage regulator phase splitter working

Very good news I have just figured out how to bias the voltage regulators, to make a phase splitter.

Here is the circuit, Voltage regulator phase splitter.pdf Using LM137 and LM337 voltage regulators.

I have it driving an 8ohm speaker. Just need to replace C1 and C2 for resistors, for use in the "linear amp regulator".
If the voltage regulators were a matched pair, then the distortion will be less. Still sounds good.

D1,D2,D3 & D4 can all be replaced by a 2.5V zener (in reverse bias) or the LM336 2.5V precision voltage reference IC.


Be thankful the small successes, and this will lead to the big success.

I am working on the "linear amp regulator" stage be stage.

[Ecancanvas]

sorry,
been out for a while, totaled my car, i'm ok and recovering. Looks like I have lots of catch up to do.
have not had to buy a car in over 10 years, ouch!
Ed

[Nityesh Schnaderbeck]

sorry,
been out for a while, totaled my car, i'm ok and recovering. Looks like I have lots of catch up to do.
have not had to buy a car in over 10 years, ouch!
Ed

It's good to know you are ok, happy thoughts for your recovery..It don't matter about the car, your life is more important.

I am in the process of making, the regulators, op-amp and output transistor and beta multiplier into an audio amp. I am going through each stage to the "linear amp regulator" I have gone the wrong direction a couple of times, and had to reset/reassess my thinking, back to "I know nothing" and start again.

I have made a simplified diagram to help me and others understand, more about the "Linear Amp Regulator"

Simplified Diagram.pdf

Please note, this is only the interpretation of a mere mortal.

This is missing the comparator, so this diagram may be updated, as I learn about the comparator stage.

[Ecancanvas]

Why did you use R5 and R6 in your Voltage regulator phase splitter
circuit. such low ohms for feedback. so you are sending feed back to pin 1 and pin 3. Im building up your circuit for audio, looks like fun. what are your ideas for attaching the op amp. just spitting out thoughts. I used a 2.5v zener in a circuit simulator and it gives quite different voltages from the 4 diodes.

[Nityesh Schnaderbeck]

Voltage regulator phase splitter.pdf
I think the 2,5V zener is the better than my 4 diodes connected in series, R5 and R6 are not feedback resistors (R3 and R4 are the feedback resistors), they are to bias the output with the correct voltage, the original circuit from the internet, that lead me to this circuit had 10ohm for this resistor, I changed it to 47ohm, so I don't have to have the regulators on a heat sink. The 2 regulators are operating in class A. One positive "class A" and one negative "class A", they are opposite of each other. And I put the speaker between the outputs, to prove that the outputs are opposite to each other. So I know it is a phase splitter. R5 and R6 are low values only to set the correct bias to drive an 8ohm speaker.

When this stage is inside the "Linear Amp Regulator" R5 and R6 will be something 1K, 2K, 5K, etc.

In short this phase splitter circuit, converts an unbalanced signal to a balanced signal. The op-amp converts the balanced signal back to an unbalanced signal.

I think that in the "Linear Amp Regulator" the plus and minus regulators are phase splitting, so that the feed back from the battery is directly coupled to the op-amp, because the op-amp is operating with balanced inputs on a plus and minus supply of +15V and -15V or so. When the battery is charged and floating then the positive of the battery is the same as the op-amp ground. So it is always trying to pull the battery voltage to op-amp ground. (op-amp ground is held between 30V, so op-amp ground will be 15V same as the battery)

I could use op-amps and bias them in class A, http://tangentsoft.net/audio/opamp-bias.html then I would need to make current sources which will increase the component count, the regulators already have everything inside, just need to get past that internal zener. Just need to cancel out the internal zener with an external zener. The LM317 and LM337 have 1.25V Zeners inside the 2.5V external zener cancels out both these internal zeners. As you could imagine, the better the external zener is matched with both the zeners inside the regulators, the less distortion. The regulators are very much like a class "A" op-amp.

Have a look at this page about two thirds down the page you will see the LM317 internal schematic http://www.m0ukd.com/homebrew/alkali...rging-circuit/ When we know whats inside, it will help us design the phase splitter.


have a look at these links http://www.electroschematics.com/668...dio-amplifier/

http://www.electroschematics.com/670...ge-regulators/

I have been also working on a modified version of this circuit.

"Ecancanvas" Can I ask you what is the simulator you are using?. I have never used a simulator before. I have "protel 2004" with spice, but too many errors to resolve. I think I need to change software. At the moment my imagination is my circuit simulator.

[Nityesh Schnaderbeck]

This is an update to the simplified diagram with a comparator, so to add voltage control.

Simplified Diagram with Comparator.pdf

To me it would make more sense to do like this. Anyway this is the block diagram, of my circuit goal, at this point in time.

[Nityesh Schnaderbeck]

Complimentary Regulator Phase Splitter Ver.2

Here is another version of the "Complimentary Regulator Phase Splitter"

Regulator Phase Splitter Amp Ver2.pdf

This circuit is easier to modify to amplify DC, when this stage goes in the "Linear Amp Regulator" the capacitors C1 and C2 will be removed. The capacitors are there so I can just use a 12V single rail supply, to make testing easier.

[Ecancanvas]

Nityesh, Its free but they restrict the times you can use it. They only want you using it in evening and not at work. the program is called Yenka its down side is i have not been able to find voltage regulators, but for working out how the circuit is working and the amps and voltage its pretty good

[Nityesh Schnaderbeck]

Regulator Phase Splitter Amp Ver3

I have discovered that this circuit works without the diodes.



It seems to work better without the input capacitor, it also seems to work without the output capacitor. I have the 470uF cap in place, just an extra measure to make sure DC does not flow through the speaker.

The ratio of resistors values in this circuit seem to give the best sound.

[Nityesh Schnaderbeck]

Regulator Phase Splitter Amp Ver4

This design can be simplified even more, and this also works. Looking at the LM317 schematic, I see that the output pin is connected to the emitter of the internal power transistor. So this amp is using an emitter follower as the output stage. Because of this, the two 47ohm resistors are load resistors and the feedback resistors. Here is the simplified schematic, and it even works.



"Ecancanvas" you were right once again, without knowing it, in this circuit the two 47ohm resistors are feedback resistors, Congratulations, good on you and thankyou

Some notes: If you drive it too much it will distort, too much bass will make it distort, if the load resistors are incorrect for the speaker load it will distort, and if R1 and R2 resistor values are too high or too low it will distort. But when you get the bias right it sounds good. And this amp is not very loud. The output capacity can be increased by making the two 47ohm resistors a lower value, then the regulators will need a heat sink, but this is not the point of this experiment since, this amp will not be delivering any power anyway once it is in the "Linear Amp Regulator" circuit.

[Nityesh Schnaderbeck]

Complimentary Regulator Phase Splitter Ver.5 with offset adjust

Here is the same circuit with an offset adjustment, I believe John Bedini had a similar or same offset adjustment on his "Linear Amp Regulator"



Note: R2 and R3 are 3.3K (on the picture it looks like 33K)

Reasons why I think that these regulators are used as like an audio phase splitter.

In the posts I read John Bedini, says that the regulators have to be matched. If the regulators were used as static voltage references, then they would be not need to be matched, just simply adjust the opposite voltages the same.
But as a phase splitter, the more identical the plus and minus regulators are in characteristics, the more accurately it will phase split the signal.

John Bedini is also talking about regulators as class A amplifiers, and that the industry standard had taken away the control pin. But if you bias it correctly, these regulators can be used as low power Class "A" amplifiers.

[Nityesh Schnaderbeck]

Regulator Phase Splitter Amp Ver6, Sounds awasome now

I had a look at the internal schematic of the LM337 and I saw my big mistake with all the previous versions of the "Regulator Phase Splitter Amp"

Both Lm317 and LM337 have NPN power transistors inside.

Inside the LM317 the emitter of the internal NPN power transistor is connected to pin2 (+Vout)
Inside the LM337 the emitter of the internal NPN power transistor is connected to pin2 (-Vin)

So the following circuit has the internal NPN power transistors from both regulators connected as "emitter followers"

This new circuit has the true high fidelity sounds, nice bass and nice treble, and is louder than the previous versions of this circuit.



note: Both R2 and R3 are 3.9K (looks like 39K on the picture)

"Ecancanvas" you will love the sound of this amp now, very nice.