Analog power supply output question

[Baluncore]

It works good. Thanks.

I would like to see a diagram sometime, showing how you are connecting the "ripple inverter transistor".

 

[Patch Cabbage]

It will probably just sample the line ripple and amplify it and invert it through a dropping resistor on its collector. Just a standard circuit. I wish I could find a better way of doing it since the ripple invert has to follow a variable output. Thanks.

 

[Baluncore]

It will probably just sample the line ripple and amplify it and invert it through a dropping resistor on its collector. Just a standard circuit.

There are very many ways of doing it. I am interested seeing exactly how it is being done. Can you give me a link to the circuit on the web?

 

[berkeman]

I'm setting up an analog power supply. I have the transformer bridge and filter capacitors so far. The transformer puts out 30 volts. I am currently using two parallel power transistors and a variable resistor to set the output. It also has a meter to monitor voltage. The question is this. How do I set up a single transistor to remove whats left of the ripple after the filter capacitors. It has to vary along with the control transistors to be constant in its ripple removing. The bases of the two power transistors to the potentiometer is currently the control.

No problamo, as I have learned nothing from this thread.

Well, it's the old GIGO principle I guess. But maybe we can still help you a bit, since this part of your circuit makes no sense to me:

The two transistors are not connected as any kind of voltage regulator. Their only function that I can see is to convert current into heat, and lower the efficiency of your unregulated power supply. Did you mean to describe some other connection of these transistors?

 

[sophiecentaur]

this part of your circuit makes no sense to me:

He seems to have connected the transistors as diodes. Just a more expensive solution.

 

[Baluncore]

He seems to have connected the transistors as diodes.

I drew the diagram that way, as it fitted the written description. That forms a current regulator, as the resistor current is multiplied by beta, then passed to the output, and through the load.

 

[sophiecentaur]

That forms a current regulator,

A regulator has a reference. What reference is in that circuit. (I realise that things can be hidden or inferred)

 

[sophiecentaur]

No problamo, as I have learned nothing from this thread.

PF's advice is free. Learning is supposed to be an active process. The excuse is not really GIGO because a lot of the input is far from garbage.

Do your invented circuits normally work OK?

 

[Baluncore]

What reference is in that circuit.

The resistor is selected by the operator to adjust the load current, and so the output voltage. However, the "ripple inverter transistor" wherever whatever is not shown.

 

[sophiecentaur]

OK. I was looking for a voltage regulator - which is normally what one wants.

 

[Baluncore]

Maybe the title of this thread should have been;
"Linear power supply, ripple reduction".

 

[DaveE]

That forms a current regulator, as the resistor current is multiplied by beta, then passed to the output, and through the load.

But then what regulates the resistor current? If the voltage applied changes, so does the current. Just semantics, I guess. It works more like a resistor, as you described. More like a current limiter.

 

[Averagesupernova]

Max 30 amps? 15a each? 0-40 on the volts.

Hmmm. So a bit of quick math without a schematic. Suppose you want to supply 5 volts out at 30 amps. That's within the spec right? That means that since you have a max output of 40 volts, the input to the pass transistors will be at least 40 volts. However, you want 5 so there will be at least 35 volts across the pass transistors while they are conducting 30 amps if you want your supply 5 volts to a 30 amp load. That means the pass transistors need to dissipate over 1000 watts. How do you plan on getting rid of that amount of heat?

 

[Patch Cabbage]

This should work.

 

[Averagesupernova]

If you cleaned that schematic up to make it more readable......
-
.....it still won't work.

 

[DaveE]

Normally I reserve this post for people that don't provide a schematic, but in this case I'll make an exception.

 

[DaveE]

This should work.

View attachment 367686

I think this circuit will work the same if you replace all of the transistors with a piece of wire (OK, maybe a forward biased diode). Why would your transistors ever turn off or decrease their collector current? How would that happen?

 

[Patch Cabbage]

Hey, you know, that's a really good question. My first thought response would be that the variable resistor is really big, but that leads to the ripple invert is biased to be slightly on all the time. Not for sure if it will ever go down to zero. It should but it might not. Also the ripple invert is fixed and the voltage output is variable so that's not so perfect either. Then again if I had all of the answers I wouldn't have to ask someone how to do it on a physics forum. If it works, fine. If not, it's back to the old drawing board. I'm not proficient in daw and I draw still with a pen and paper and not a mouse so sorry about all the sloppy schematics.

 

[Averagesupernova]

None of that makes any sense to me.

 

[DaveE]

I'm not proficient in daw and I draw still with a pen and paper and not a mouse so sorry about all the sloppy schematics.

This is one of my first real designs for money as an EE. It was drawn by hand (early 1980's) and is a #$@! sight clearer than your attempt. The reason you don't draw readable schematics is probably that you don't care enough about communication with other engineers to put in the effort required.

This is a quick schematic sketch I drew by hand about 40 years later.

Graduate Post in thread 'Why doesn't transformer core flux change when load power varies?'

Apr 8, 2021

For a slightly different take on transformers, here is the lumped element model that EEs use for circuit analysis. The same material, different jargon. EEs that work with this will speak of "coupling" and "flux" in conversation (because we did actually take some physics classes); but, in my experience, they really work with "leakage inductance", "magnetizing inductance", voltages, and currents. The ideal transformer with imperfect coupling is modeled as a "T-section" shown below:

Then, applying KVL, we get this solution:
$$ \begin{align}
v_1 &=~ L_1 \dot {i_1} + L_m ( \dot...

• DaveE

• 

My first thought response would be that the variable resistor is really big, but that leads to the ripple invert is biased to be slightly on all the time. Not for sure if it will ever go down to zero. It should but it might not. Also the ripple invert is fixed and the voltage output is variable so that's not so perfect either.

I don't really care much about your "first thought response" at this point. You aren't putting in the effort necessary to make progress here. These words don't mean anything to me. You need to study more. This is the only useful advice I can give at this point.

 

[berkeman]

I'm not proficient in daw and I draw still with a pen and paper and not a mouse so sorry about all the sloppy schematics.

Get a copy of LTspice, it is free from ADI.

@Patch Cabbage -- it's time for you to start working on your toolkit and skills for working with circuits if you want to keep going down this path. As @Baluncore says, LTSpice is a free download and it includes a schematic capture feature. Please download LTSpice for free and start to learn how to draw schematics using it. That is an important step on your journey to learning more about electronics and EE.

https://www.analog.com › resources › ltspice-simulator\

 

[sophiecentaur]

This is a quick schematic sketch I drew by hand about 40 years later.

. . . . and you were careful not to do your diagram whilst riding on a bus or with a mouse. There is little excuse for not using a free drawing app (not a painting app), which makes lines straight and level.

A quick tiffle with contrast and brightness would render that photo as good as new.

 

[Averagesupernova]

@Patch Cabbage do you realize that a correctly configured pass transistor(s) act to reduce ripple? Apparently not and apparently the experiments you have done have not proven this out. It is impossible to get anywhere farther than you have without understanding schematics. Arm waving and talking about pass transistors, rectifiers, and transformers and being able to point them out on a physical power supply is not the same as understanding how they work and why they are configured and connected the way they are.