Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Discrete transistor voltage regulator

Status
Not open for further replies.

boylesg

Advanced Member level 4
Joined
Jul 15, 2012
Messages
1,023
Helped
5
Reputation
10
Reaction score
6
Trophy points
1,318
Location
Epping, Victoria, Australia
Activity points
11,697
Could some one please explain to me how the output voltage is set by R3, R4 and R5.
It can have a maximum output of 12V as is but I wish to modify the circuit to get up to 18V.

**broken link removed**


**broken link removed**

Here is my version so far but I can't seem to get it past perhaps about 14V. Output current is excellent though.
 

The clue here is that the schematic "Figure 1" only adjusts up to about half the voltage you put into it. To apply a similar principle to yours, you have to put around 36V in to it from your charge pump.

Incidentally, the current limiting part of the regulator prevents it passing more than about 3A. You can remove Q15 and short out R7 as there is no chance of you ever reaching a current as high as that. I'm not familiar with the simulator but if X1 is a 7.5W lamp used as a load, it is never going to work, the diodes in the charge pump will light up brighter than the lamp !

Brian.
 

The clue here is that the schematic "Figure 1" only adjusts up to about half the voltage you put into it. To apply a similar principle to yours, you have to put around 36V in to it from your charge pump.

Incidentally, the current limiting part of the regulator prevents it passing more than about 3A. You can remove Q15 and short out R7 as there is no chance of you ever reaching a current as high as that. I'm not familiar with the simulator but if X1 is a 7.5W lamp used as a load, it is never going to work, the diodes in the charge pump will light up brighter than the lamp !

Brian.

I have given up on this one. It seemed to work OK at first but now it is giving me constant simulation errors and the output looks something like an audio signal. If it is giving me this much trouble in simulation then there is a good chance that it wont work in a real circuit. I have no clue what made it go from good to crap and I don't think I want to waste time trying to figure it out. This one is consigned to the bin!

This looks more promising however:


Negative circuit already detailed.

When I tack it on to my voltage doubler it seems to have the strange effect of reducing the doubler's output voltage to about 16V, but then I can adjust back up to 18V using the pot in the voltage regulator. Any words of wisdom on this for me?

It might be possible to cut down the voltage regulator to its essential components and incorporate it into the output stage of the doubler some how.
Might reduce its effect on the voltage doubler and increase the output current a bit as well.

R8 and C4 form a low-pass filter at the input with a pole at 12.5 Hz. After that filter, the regulator circuit begins.
Well we can get rid of that bit then - might improve matters.
 
Last edited:

I think a fundamental stumbling block here is you are using a very inefficient voltage doubler and following it by a very inefficient voltage regulator. The 'goodness' of regulation in that schematic is very poor and you might find it has poor temperature stability and it's output voltage changes according to the light level falling on the LED. The R8/C4 statement is only partially true but in any case is irrelevant to the circuit.

As mentioned elsewhere, a much simpler circuit using an inductor will do the job better and would be many times more efficient.

Brian.
 
  • Like
Reactions: FvM

    FvM

    Points: 2
    Helpful Answer Positive Rating
I think a fundamental stumbling block here is you are using a very inefficient voltage doubler and following it by a very inefficient voltage regulator. The 'goodness' of regulation in that schematic is very poor and you might find it has poor temperature stability and it's output voltage changes according to the light level falling on the LED. The R8/C4 statement is only partially true but in any case is irrelevant to the circuit.

As mentioned elsewhere, a much simpler circuit using an inductor will do the job better and would be many times more efficient.

Brian.
The problem is with the boost converters is that I could not figure out how to invert it to boost -12V to -18V and I could not find any examples of an inverted boost converter.

At present, ineficient or not, the doubler and regulator is the best I can do.

Do you have some hints for me as to how to invert an inductor based boost converter? Is it even possible given that I have not found any examples of it?

light level falling on the LED
I didn't think ambient light had any effect on LEDs. It will ultimately be in a closed box at any rate. I could always put a couple of series rectifier diodes in place of the LED if ambient light is a problem for it. I heavy duty rectifier diode with a Vf of 1V and a lighter diode with a Vf of 0.7V might do the trick.
 

I assume you would prefer to start with +12V rather than -12V although the schematic remains basically the same. Google for MC34063A it will turn up several manufacturers data sheets, the TI and OnSemi ones are best (and curiously almost word-for-word identical). Download the data sheet then go to this web site : **broken link removed** and enter the following in the boxes:
Vin = 12V
Vout = -18V
Iout = 200mA
Vripple = 10mV
Fmin = 20KHz

Then click on "calculate", it will draw the schematic and tell you all the values to use. It will give about five times the current output of your present design with lower ripple and will be fully stabilized and overload protected. All in 9 Components and none of them are expensive. Where are you located? I may be able to help you find a supplier.

Brian.
 

I assume you would prefer to start with +12V rather than -12V although the schematic remains basically the same. Google for MC34063A it will turn up several manufacturers data sheets, the TI and OnSemi ones are best (and curiously almost word-for-word identical). Download the data sheet then go to this web site : **broken link removed** and enter the following in the boxes:
Vin = 12V
Vout = -18V
Iout = 200mA
Vripple = 10mV
Fmin = 20KHz

Then click on "calculate", it will draw the schematic and tell you all the values to use. It will give about five times the current output of your present design with lower ripple and will be fully stabilized and overload protected. All in 9 Components and none of them are expensive. Where are you located? I may be able to help you find a supplier.

Brian.

Well actually, purely for interests sake but not necessity, I would still like to know how to do a -12V to -18V conversion.

And since I bought a transistor kit from Jaycar I would also like to be able to do it with discrete transistor rather than an IC. Both to make use of the transistors and to gain some level of understanding of how these things work and how to design them.

I am not designing precision circuits here, so it does not bother me that much if the discrete solution is not as efficient as you can get.

Call me stubborn but I have a bee on my bonnet that just wont die!

I suppose I could take the internal structure of this chip and used it as a basis to design a discrete transistor version, all be it less efficient.

Plenty of examples opamp comparators and I have a dozen or so opamps that I have salvaged from tvs.

I know how to build an oscillator.

I could have a stab at building a discrete transistor Darlington pair - the datasheet specifies what value resistor to use at least.

What is a ref reg exactly?

What is the rectangle with SQR in it?

I presume that other thing is a logic AND gate - plenty of examples of those implemented with diodes and and trannies.

Once I get my make shift chip sorted I could then use the specific application circuits in the datasheet.

Tell me, can a buck converter do below 0 voltage reductions or only above 0 ones?

- - - Updated - - -

Hmmmmm!

High-efficiency step-down switching regulators for positive voltages are very common, however negative step-down switching regulators (negative voltage in, negative voltage out, common ground) are not as well known, even though they are often needed. Although they are not difficult to set up, literature on how to build them is rather scarce. This article analyzes the architecture and detailed operation of the negative buck topology. It will also discuss actual circuit implementations for the topology, from a system perspective down to the building of the needed circuit blocks, and include examples on how to build a voltage translator circuit, a key block in implementing a negative buck regulator using readily available boost ICs.

I have been searching for the wrong 'string' for starters. Should have been 'buck a negative voltage' rather than 'boost a negative voltage'.

- - - Updated - - -

Just had a thought regarding the inefficiency of my circuit.

As I understand it thus far, inductors and capacitors are used in much the same way. You push a current through them one way, they store energy, you ground one end and the stored energy comes out as current the other way.

So why couldn't I just replace the charge pump capacitors in my doubler with inductors plus any other required alterations?
 
Last edited:

As I understand it thus far, inductors and capacitors are used in much the same way. You push a current through them one way, they store energy, you ground one end and the stored energy comes out as current the other way.

So why couldn't I just replace the charge pump capacitors in my doubler with inductors plus any other required alterations?
Both are storing energy but they are not working or used in the same way. Their operation principles are opposite:

Vinductor = dI/dt
Icapacitor = dV/dt

According to different operation principle, capacitive and inductive converters are implemented in quite different circuits.

I would still like to know how to do a -12V to -18V conversion.
I have been searching for the wrong 'string' for starters. Should have been 'buck a negative voltage' rather than 'boost a negative voltage'.
I don't know which literature you have been reviewing, but -12 to -18 refers clearly to a negative voltage boost (or step-up) converter.

I think, there are two aspects of the question:

- in principle, you can design the said a -12 to -18 converter by flipping the polarity of all components in a +12 to +18 converter. In so far there's no basic problem.

- in practice you won't find dedicated negative boost switcher ICs. You can possibly implement the converter based on switcher ICs for positive input voltage, or switch mode controllers and external transistors. That's just daily engineer's work.
 

To answer specific questions:

The reference is a stable voltage generator. All regulators work by comparing the output, or a proprtion of it, with a reference voltage. The difference "error voltage" is used to adjust the regulator to bring the voltage back to where it should be.

"SQR" is a flip-flop, a kind of logic circuit not unlike your multivibrator but the transistors need a kick to make them change state. "Q" is one of the flip-flop outputs, "S" sets "Q" high, "R" resets it low. There is no connection to the other output in this particular flip flop but like your multivibrator, there is internally another point with the opposite to "Q" on it.

Yes, the other thing is a logic AND gate.

As FvM states, you can't just swap inductors and capacitors, for one thing, inductors appear as very low value resistors to DC so they would short out the voltage! The trick to making efifcient buck or boost converters, whether inverting or not, is to make most use of stored energy and to control how much energy is actually stored. A linear regulator can have very low efficiency, in some cases only a a few % but a good switching regulator can reach 95% or better.

Even with your transistor pack it should be possible to build a simple switching inverter with fewer components than your original design. Bear in mind though that the MC34063A has 79 transistors inside it so you need to think of more advanced circuits to match it's performance.

Brian.
 
  • Like
Reactions: FvM

    FvM

    Points: 2
    Helpful Answer Positive Rating
Both are storing energy but they are not working or used in the same way. Their operation principles are opposite:

Vinductor = dI/dt
Icapacitor = dV/dt
Similar to a transformer where you convert a lot of current with little voltage to a lot of voltage with little current.
I think I sort of get it.

And my inefficient circuit, as it currently is arranged, is only driving up to about 15mA through the capacitors.

According to different operation principle, capacitive and inductive converters are implemented in quite different circuits.



I don't know which literature you have been reviewing, but -12 to -18 refers clearly to a negative voltage boost (or step-up) converter.

I think, there are two aspects of the question:

- in principle, you can design the said a -12 to -18 converter by flipping the polarity of all components in a +12 to +18 converter. In so far there's no basic problem.

- in practice you won't find dedicated negative boost switcher ICs. You can possibly implement the converter based on switcher ICs for positive input voltage, or switch mode controllers and external transistors. That's just daily engineer's work.

I did find a discrete transistor positive boost converter - the originl circuit I was fiddling with.

But no matter what I did with inverting everything and switching npns or pnps, I could not get it to work in multisim. So I gave up and tried the charge pump systems.

Which never the less brings me back to this circuit which I gave up on:


The key point is the inductor is connected directly across the power supply with no resistors so that it gets max current.

Well I can certainly replace the charge charge pump capacitors in my circuits with inductors or 'an inductor' connected similarly across the supply.

And I already know how to invert the multivibrator part of my current circuit such that it still 'vibrates'.

Then I would only have to figure out how to invert the right hand side of the above circuit minus the feed back zenner diode etc. That might be easier for me to figure out .
 
Last edited:

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top