I have a linear power supply build that is a variable voltage supply. The circuit I need is for the ripple left over after the filter capacitors. The supply uses power transistors as the output which are controlled through the base voltage with a variable resistor. I want to hook up another transistor to invert the ripple. If I sample the ripple as a control on the base and get the inverted voltage output on it's collector resistor and use an isolation capacitor to the base of the power transistor it will work. The down side is that the ripple will show up varying through the range of voltages and it will be about 43 degrees advance out of phase. So my question is how can I set up my transistor to invert and remove the ripple of 120 cycles evenly across the variable voltage range. Is there an inovative way of doing it or a cheat or something. I know it can be done. I just have not figured out the way to do it. Anyone with any advice?

Edited November 17Nov 17 by Patch Cabbage

2 hours ago, Patch Cabbage said:

I have a linear power supply build that is a variable voltage supply. The circuit I need is for the ripple left over after the filter capacitors. The supply uses power transistors as the output which are controlled through the base voltage with a variable resistor. I want to hook up another transistor to invert the ripple. If I sample the ripple as a control on the base and get the inverted voltage output on it's collector resistor and use an isolation capacitor to the base of the power transistor it will work. The down side is that the ripple will show up varying through the range of voltages and it will be about 43 degrees advance out of phase. So my question is how can I set up my transistor to invert and remove the ripple of 120 cycles evenly across the variable voltage range. Is there an inovative way of doing it or a cheat or something. I know it can be done. I just have not figured out the way to do it. Anyone with any advice?

Circuit diagram ?

If you google capacitance multiplier circuit you will see that your idea is sound although you need to make a few changes to your circuit. The circuits shown are usually not variable but you could add a potentiometer to the output terminals.

Edited November 17Nov 17 by OldTony

4 hours ago, Patch Cabbage said:

Well a few numbers might be useful on you psu, don't you think ?

As it is not clear whether you are talking about an emitter follower type pass transistor or a collector follower here are a couple of pages to cover either or both eventuality.

It would have been much more helpful if you had stated you expected current draw and voltage adjustment range.

Since you are using power transistors I assume substantial current draw and therefore you will require all the base drive current to pass through that one (variable) resistor.

There are several methods of overcoming this. One is to use a darlington configuration for higher power transistor gain.

The other is to add fixed end resistors in seroies with the variable one to ensure that one end of the track is not burnt out.

An emitter follower type (circuit 1 in my attachment) is not short circuit proof in respect of the load so again, especially if high powers are envisaged, some protection is best.

The collector follower type (circuit 2 in my attachment) is inherently short circuit proof.

You should add a high value bleed resistor across your filter/reservoir capacitor.

This will discharge a subtantial energy source safely when the supply is switched off.

It is unknown what your resources are or if you are using up old stock (if so well done) so further discussion of this would be fine if you have any more questions.

As a matter of interest we have a very clearly marked engineering section so why did you post this in other sciences ?

Edited November 17Nov 17 by studiot

Hey thanks. What book is that? I'm starring at that second diagram. Can it even work. Looks to me like the polarity of the main transistor is backward and can't work. Where is the positive voltage coming from for the positive output. As for power on my circuit it is 30 amps. It is already regulated as it's variable and able to be tuned to the load without a zener constant. I just need to remove the 120 cycle positive ripple. I still need the ripple inverter. And for protection of the circuit it has a fuse before the transformer and a circuit breaker directly after it. The power transistors are fully capable and if not I could always stack some more. As for the base it requires very little current and is not even an issue. RIPPLE REMOVE TRANSISTOR CIRCUIT NEEDED. If you are wondering why I posted here instead of engineering I did it as a cheat, a hack, because in engineering I would get answers like circuits that are already on the market. Needs 9 chips to do it as an example. I needed to ring up some scientists or something to get an effective and simple circuit using the a single transistor. Not a my book says kind of answer to a different question that wasn't even asked. I will maybe try to help you with your homework in the future if you were to ask. What is the name of your trainer book and who is the author. He may have been broke at the time and is an idiot.

Studiot, I don't mean to sound like I'm coming down hard, I'm not. This is the simplest of basic circuits. It's like asking if an npn silicon amp or switch circuit is an emitter follower or not and what might the base voltage on be and needing a schematic to understand it. It is one of the simplest of circuits. In theory or use you will need to understand circuits without schematics like some unit without the book or papers. You find the main components and draw at least common circuits from them making a usable schematic. Everyone always asks for a schematic on a linear power supply. Why it just a transformer and a bridge. It can even have a filter section on it and even a choke or resistor. It may also have fuses or circuit breakers and a main switch. But, and big but, it is still a simple( transformer output gets rectified) circuit. Everyone then asks for a diagram. Why do they need a diagram for two moving parts. Is it that hard. It's a linear supply. It has a filter. It has a control. It does not use a zener diode. It still needs a ripple invert circuit to hook up to the control. AND. I won't find it here. Thanks.

31 minutes ago, Patch Cabbage said:

Hey thanks. What book is that? I'm starring at that second diagram. Can it even work. Looks to me like the polarity of the main transistor is backward and can't work. Where is the positive voltage coming from for the positive output. As for power on my circuit it is 30 amps. It is already regulated as it's variable and able to be tuned to the load without a zener constant. I just need to remove the 120 cycle positive ripple. I still need the ripple inverter. And for protection of the circuit it has a fuse before the transformer and a circuit breaker directly after it. The power transistors are fully capable and if not I could always stack some more. As for the base it requires very little current and is not even an issue. RIPPLE REMOVE TRANSISTOR CIRCUIT NEEDED. If you are wondering why I posted here instead of engineering I did it as a cheat, a hack, because in engineering I would get answers like circuits that are already on the market. Needs 9 chips to do it as an example. I needed to ring up some scientists or something to get an effective and simple circuit using the a single transistor. Not a my book says kind of answer to a different question that wasn't even asked. I will maybe try to help you with your homework in the future if you were to ask. What is the name of your trainer book and who is the author. He may have been broke at the time and is an idiot.

Studiot, I don't mean to sound like I'm coming down hard, I'm not. This is the simplest of basic circuits. It's like asking if an npn silicon amp or switch circuit is an emitter follower or not and what might the base voltage on be and needing a schematic to understand it. It is one of the simplest of circuits. In theory or use you will need to understand circuits without schematics like some unit without the book or papers. You find the main components and draw at least common circuits from them making a usable schematic. Everyone always asks for a schematic on a linear power supply. Why it just a transformer and a bridge. It can even have a filter section on it and even a choke or resistor. It may also have fuses or circuit breakers and a main switch. But, and big but, it is still a simple( transformer output gets rectified) circuit. Everyone then asks for a diagram. Why do they need a diagram for two moving parts. Is it that hard. It's a linear supply. It has a filter. It has a control. It does not use a zener diode. It still needs a ripple invert circuit to hook up to the control. AND. I won't find it here. Thanks.

Thank you for wasting my time.

I just looked in but I think I will go back to watch the ending of the late night film, whilst you have a nice discussion with the moderator about your attitude.

Though it may seem like attitude it isn't. I'm just different from the people you already know. I was brought up minimalistic with nothing. I was told to learn the things around me in the world. I wasn't given anything. Electronics was one of the harder things to learn about but I learned it. I appreciate the time and effort you went through to look things up. I was told if I don't know something to look it up. There are no libraries with the answers sometimes, even in the reference sections. Libraries near colleges are the best though. Schools were never good to me. Probably genetics though they don't want to admit that. I'm second generation russian. You know, the bad guys. School are a form of govenment spying and indoctrination and making records on all the new people. These are your school records and they stay with you for life along with military records and police records. So from a government standpoint you may be a stupid militant crook or something. With me add commie. Not a commie but argue with schools. Hey looking things up wake up the brain cells it's good for you. You know use it or loose it. It sure beats watching the tube on fox and drinking a beer, which not only damages brain cell but also makes you fat. And finally the circuit I was looking for probably doesn't exist yet. That's why most regulated power supplies use the zener or have a million chips in them. The thing needs a better scientist, something nobel or something. Have a nice one. Bye.

6 hours ago, OldTony said:

If you google capacitance multiplier circuit you will see that your idea is sound although you need to make a few changes to your circuit. The circuits shown are usually not variable but you could add a potentiometer to the output terminals.

@Patch Cabbage , stop feeling sorry for yourself, and follow this advice. We're not doing brain surgery here.

This should work.

On 11/18/2025 at 2:40 AM, Patch Cabbage said:

That's why most regulated power supplies use the zener or have a million chips in them.

They have “a million chips” (seriously, not quite) because modern power supplies are usually “switching power supplies”..

https://en.wikipedia.org/wiki/Switched-mode_power_supply

They are more efficient (lose less energy) than traditional power supplies (from your sketches) with a transformer.

They generate less heat.

They are smaller.

They are cheaper to manufacture (contain less expensive copper).

They use MOSFETs (on your sketches there are NPN/PNP).

Edited November 23Nov 23 by Sensei

58 minutes ago, Sensei said:

switching power supplies

I recall reading in an electronics book a while back the author strongly urging its readers not to attempt to build a switching power supply but to purchase a commercial version instead. It is my understanding that there is no isolation between the powered circuit and the mains power.

One of the things I have learnt over the years is that real electronic circuits are more complicated than ideal circuits because the non-ideal behaviour of electronic components can cause problems that need to be countered by additional circuitry. A simple example is a switch. An ideal switch transitions power from OFF to ON or from ON to OFF. A real switch bounces, which can be a problem for digital circuitry.

17 hours ago, KJW said:

I recall reading in an electronics book a while back the author strongly urging its readers not to attempt to build a switching power supply but to purchase a commercial version instead. It is my understanding that there is no isolation between the powered circuit and the mains power.

One of the things I have learnt over the years is that real electronic circuits are more complicated than ideal circuits because the non-ideal behaviour of electronic components can cause problems that need to be countered by additional circuitry. A simple example is a switch. An ideal switch transitions power from OFF to ON or from ON to OFF. A real switch bounces, which can be a problem for digital circuitry.

Some switched mode supplies have no transformer, the better ones do, the sort of swmpsu you and I are used to, are relatively low power and may be implemented simply as a capacitor store type.

Many of the cheaper ones do not in fact meet the UK wiring regulations.

But the OP type is a 12 volt 30 amp+ supply. these are available commercially ( I have one) that are used by radio amateurs, marine folks, anyone who wants a mains supply to replace a 6/12/24 volt Lechlanche battery. Higher amperage ones are used as vehicle starters and fast chargers. in garage workshops.

One way to achieve this is to put controlled rectifiers in the bridge, but at these current levels any design is always fighting the short time when positive current can flow into the reservoir, compared to the period of the ripple voltage, necessitating large value capacitances with robust rating to take the inrush current that may need to be several times the average current (so 30 amp might become 90 - 150 amps pulses ) .

The ripple on commercial units is often acceptable because they are only powering lighting or machinery, both of which have inertia of their own.

I have a very strong opinion of switching supplies. I hate them. They work good for computer stuff, which makes them even worse in my opinion. They are dangerous because of high voltages. High voltages on a transistor board are like some of the old tube transistor hybrid boards. Also if you look at the capacitors on the switching supplies you see at first a 25 for voltage level. It's not 25 it's 250. A beginner could make a serious mistake there when working on one for the first time. I once replaced a transformer on a portable tv with a crt with a resistor on a heat sink and a diode. I made sure to insulate everything since high voltage would be present at the junction. It was way lighter and the vertical sync worked just fine. Switching supplies to me are the same kind of circuit only with more parts. I like older circuits where the voltage levels never exceed 25 volts. Switching power supplies are a dangerous design. Also don't get me started on lithium batteries. Anything but lithium because it reacts with water. Hey if you design something, try to make it safe.

Edited November 25Nov 25 by Patch Cabbage

10 hours ago, Patch Cabbage said:

I have a very strong opinion of switching supplies. I hate them. They work good for computer stuff, which makes them even worse in my opinion. They are dangerous because of high voltages. High voltages on a transistor board are like some of the old tube transistor hybrid boards. Also if you look at the capacitors on the switching supplies you see at first a 25 for voltage level. It's not 25 it's 250. A beginner could make a serious mistake there when working on one for the first time. I once replaced a transformer on a portable tv with a crt with a resistor on a heat sink and a diode. I made sure to insulate everything since high voltage would be present at the junction. It was way lighter and the vertical sync worked just fine. Switching supplies to me are the same kind of circuit only with more parts. I like older circuits where the voltage levels never exceed 25 volts. Switching power supplies are a dangerous design. Also don't get me started on lithium batteries. Anything but lithium because it reacts with water. Hey if you design something, try to make it safe.

Let's try again.

I asked for a few numbers.

I understand you have a 120v mains in transformer capable of supplying the 30 amps and rectifiers to suit.

So what is the output of the transformer ?

Does it have any other secondaries that might supply a negative rail at low current ?

You say you want a variable voltage DC supply.

Between what voltages ?

You do realise that your design is unable to go right down to zero volts ?

I understand you have a reservoir capacitor, or bank of them.

What are their characteristics ?

Have you thought about protecting both the load and the psu from overload - 30 amps is a lot of current ?

What are you actually driving (supplying) - the requirements of the load make a real different to suitable circuitry ?

(Good) engineering is requires drawing a balance between competing requirements and resources.

Edited November 25Nov 25 by studiot

On 11/25/2025 at 2:59 AM, Patch Cabbage said:

I have a very strong opinion of switching supplies. I hate them. They work good for computer stuff, which makes them even worse in my opinion. They are dangerous because of high voltages. High voltages on a transistor board are like some of the old tube transistor hybrid boards. Also if you look at the capacitors on the switching supplies you see at first a 25 for voltage level. It's not 25 it's 250. A beginner could make a serious mistake there when working on one for the first time. I once replaced a transformer on a portable tv with a crt with a resistor on a heat sink and a diode. I made sure to insulate everything since high voltage would be present at the junction. It was way lighter and the vertical sync worked just fine. Switching supplies to me are the same kind of circuit only with more parts. I like older circuits where the voltage levels never exceed 25 volts. Switching power supplies are a dangerous design. Also don't get me started on lithium batteries. Anything but lithium because it reacts with water. Hey if you design something, try to make it safe.

What a load of nonsense..

A mains-powered supply - whether switching or transformer-based - always has the same dangerous high voltage on the input side. The difference is in efficiency and design, not in the presence of high voltage. The real difference between a linear and a switching power supply is not the presence of high voltage, but the operating frequency and the way energy is transferred. Both are mains-powered and both have dangerous primary-side voltage. SMPS just use high-frequency conversion to make the transformer smaller and more efficient.

18 hours ago, studiot said:

I once replaced a transformer on a portable tv with a crt with a resistor on a heat sink and a diode.

Well, that's just total nonsense. A transformer changes the AC voltage to a higher or lower one, while a resistor and a diode only cause a current limitation, rectify the sine wave to pulses, and still leave the voltage unchanged. The pulse frequency before and after remained the same.

On 11/25/2025 at 2:59 AM, Patch Cabbage said:

Also if you look at the capacitors on the switching supplies you see at first a 25 for voltage level. It's not 25 it's 250.

You can use a 250 V-rated capacitor in a 25 V circuit, as long as the capacitance value is correct. The voltage rating just indicates the maximum voltage the capacitor can safely handle. Sometimes manufacturers use components with higher voltage ratings simply because they had them in stock and/or they were cheaper. It doesn’t mean the design is ‘dangerous’. This is not a rule. Just because someone uses 250V in a 25V circuit does not mean that there is high voltage there - you would have to measure it in real time on a working device.

BTW, to rectify 230/240 VAC, capacitors rated at 400 V (!) are used, not 250 V. So you're wrong anyway..

Edited November 26Nov 26 by Sensei

9 hours ago, Sensei said:

On 11/25/2025 at 12:47 PM, studiot said:

I once replaced a transformer on a portable tv with a crt with a resistor on a heat sink and a diode.

Well, that's just total nonsense. A transformer changes the AC voltage to a higher or lower one, while a resistor and a diode only cause a current limitation, rectify the sine wave to pulses, and still leave the voltage unchanged. The pulse frequency before and after remained the same.

No I didn't say that.

You must have been getting out of bed the wrong side recently, your responses have been so grumpy which may have been what has led to several inaccuracies, quite unlike your more usual input.

This is a pity because Patch, although also needlessly grumpy, at least want to discuss hard technical matters ; he is not one of the flood of fantasists we get too many of.

So please don't frighten him off.

20 minutes ago, studiot said:

No I didn't say that.

I know that. It must be forum software bug, because I quoted Patch Cabbage, not you..

22 minutes ago, studiot said:

You must have been getting out of bed the wrong side recently, your responses have been so grumpy which may have been what has led to several inaccuracies, quite unlike your more usual input.

I don't know which part of my post was grumpy. Was it when I wrote “This is nonsense” to summarize his post.. ?

I copied and pasted his post into ChatGPT (without suggesting my answer), just in case I had made a mistake in the translation, and it agreed with me that the post didn't make sense.

For example, ChatGPT: "CRT televisions rely on careful HV safety design. That modification could have been extremely dangerous, even if it “worked”. " (about replacing transformer by resistor+diode)..

"The person is expressing real safety concerns, but much of their reasoning is incorrect or based on misunderstandings."

"Their message reflects anxiety, personal anecdotal experience, and some misconceptions—not an accurate technical assessment."

36 minutes ago, studiot said:

So please don't frighten him off.

OK. I will try..

Edited November 26Nov 26 by Sensei