Author Topic: General Purpose Power Supply Design  (Read 203221 times)

0 Members and 1 Guest are viewing this topic.

Offline amspireTopic starter

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
General Purpose Power Supply Design
« on: March 20, 2012, 06:33:17 am »
This is a designed inspired by Dave's REV C PSU design, but it is opamp/transistor based rather then IC regulator based.

I did the design really as an exercise to see what could be done with some brilliant idea that Dave has used in his RevC power supply design. There are both benefits and negatives in using a LT3080 based supply, so this is just a different approach. I am pretty happy with the results.

First, just to put the design into context. I have designed a number of opamp-based supplies before, and usually you decide on the input voltages, the output voltage range, currents, etc and you then design a supply to suit those specs.

You then have to spend a lot of effort into stabilizing the control circuit - that is the hardest part of the design. Using the LT3080 bypasses any stability issues, but if you do an opamp designed, you are guaranteed to need compensation to get it unconditionally stable with all loads.

Now typically this design will only suit the set of specs the supply is designed for, and the moment you change the voltage ranges or input voltage, you have to change the values of components and then start from scratch with the stability compenstaion.

The ideas from Dave's supply that change all this are:

1. a low dropout design
2. single input supply
3. supply works to full specs with a wide range of input voltages
4. the regulation of the supply sets the output voltage to be the same as the opamp sense input (ie no voltage divider or voltage references involved in the supply regulation).

The result of #3 and #4 is the result is super flexible as you can feed the supply with whatever source supply voltage you like, and it will output whatever voltage you give it on the voltage control input. So a fully stabilized supply can be designed, and you can then make it output any voltages you like between about 0 and 30V without any change to the supply design at all. I do not have to redesign the control loop just because the voltage divider ratio has been changed.

I have never seen anyone produce a generic power supply board, so I decided to give it a try. I am hoping that it may be useful alternative to the many bad LM317 based supply design attempts.

The idea that really got me interested though was Dave's  Rev C addition of a pre-regulator so that the linear regulator never had to dissipate much heat. This is only possible if the supply can work fully with a very wide range if input source voltages and if it has a low dropout. Heatsinks and buckets of wasted energy have always gone hand in hand with linear supplies. Dave has done away with that.

I think people may not get how brilliant and revolutionary Dave's concept is. We are totally used to hand held meters, but we are still stuck to the idea that a power supply has to be anchored to a wall socket on the bench, and the idea that you do development on the bench, rather then the place the design is actually going to be used - whether that is on the roof, half way down a canyon, in the car, on a boat or wherever.

If a supply efficiently uses the energy in a lithium battery, the supply can easily last for days powering a typical modern low power circuit. Given that, why would you want a power cord at all? So a huge congratulations to Dave.

So on to my design.  Here were my design specs:

3V to 30V input source voltage range
1.5V dropout voltage
True regulation down to zero volts
1A output current with 0 to 1A current limit
High level of protection from loads.
Transient performance similar to the LT3080
No-load to full load output voltage regulation  better then 0.001% (Wiring and track resistance is the biggest problem)
Loads like a battery can be connected to the supply while the supply is off without causing any problems to the supply or the battery.
Design based on "garden' components - potential to make it very cheap.
Possibility of a wide range 1A supply using only board mounted devices and no heatsink

I will get onto the design in the next post. This design exercise is not to make something similar to Dave's supply. I have left out any attempt at this stage in a uCurrent type device, and I am only looking a very low cost high performance regulator that could be controlled by just plain potentiometers.

Richard.
 
The following users thanked this post: BravoV, LeoGray

Offline amspireTopic starter

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: General Purpose Power Supply Design
« Reply #1 on: March 20, 2012, 06:34:18 am »
Here is my initial design. If anyone is interested in using it, please feel free. I designed the compensation to be stable under all loads including huge pure capacitive loads, and I hope I got the numbers right. 



I have breadboarded it and it does work. The only change I had to to was to add capacitor C3 - without it, Q3 and Q1 had managed to form a very nice 108MHz sinewave oscillator - I was actually quite impressed by that. That actual power transistor I used was a MJE3055 - the TO220 version of the rather sluggish 2N3055 and I didn't think it was even possible to get it to generate a 108MHz sinewave.

The supply as drawn is for an output voltage range of 0 to 22volts with a 0 to 2.2 volts input. The current control regulates to 0 to 1A with a 0 to 1V input. The supply source should be at least 1.6 volts above the output voltage and 2.1 volts would give a small margin. My breadboard version actually regulated down to a 1V dropout voltage, but this is exceed the specs promised by the LM324.

MOSFET M1 on the output acts as a overvoltage clamp that will start to turn on between 25V and 30V, and it is also the reverse voltage clamp diode. Diode D10 prevents the regulator input pin of the LM324 going to -0.6 volts which is about the point you can get the IC to latch up. I didn't bother protecting the inputs of U3 from negative voltages. Tests seemed to indicate that with the 100K series resistance, you couldn't casue latch-up at -12V. and M1 will keep the negative voltages to -1 volts or less.

D5 stops a voltage on the output from damaging the regulator transistor. The last part of the protection is the LM324 itself - its inputs can be safely taken up to 32 Volts, and the MOSFET circuit will see that the inputs never get to 32 volts.

This supply does not need a bypass diode back to the input supply, so you could safely turn the supply off with a 12 or 24 volt battery on the supply output terminals.

I can talk more about component choices later, but as I said before, I see this as a design exercise, and there are things I would like to add. In particular, I would like to get rid of the 1K discharge resistor and replace it with an active circuit instead. Still thinking about the best way to implement that. Some kind of load discharge is needed, otherwise the regulator just cannot even regulate properly.

Richard.
 

alm

  • Guest
Re: General Purpose Power Supply Design
« Reply #2 on: March 20, 2012, 06:57:00 am »
Cool project! I think using a more complex circuit (both in design and in components) with cheaper and more readily available components might be the right one for a design people all over the world might build, as opposed to a kit.

Any idea how well it will reject the (relatively) high frequency switching noise from the pre-regulator? Would you expect similar performance to the traditional lab supplies with huge heatsinks or do you trade off noise for efficiency? Most three terminal regulators have a fairly limited ripple rejection once the frequency gets much higher than 120 Hz.

I'm aware of one well-designed universal variable power supply, a design by German electronics maganize ELV. There's a link to a PDF of the article on the product page, unfortunately the text is in German. This design is quite flexible in input voltage and current range, and I haven't heard any complaints about stability. It is a more traditional design intended to use with transformers, however.

I would welcome a well documented design of a power supply to be available in English, most that I've found suck in some way (eg. the poorly implemented LM317 designs that you mention).

I might make the over voltage protection adjustable, so you can set it to 6V if that's the absolute max for your components, but I realize that this may be outside the scope, and that feature creep may screw up the intended purpose (simple, cheap).

Interesting that you were able to get the MJE3055 to oscillate at 108 MHz, just make C3 switchable and you have a portable power supply and radio transmitter in one ;).

Intuitively I would be worried about the effect of D5 on the stability at low currents, since you're introducing a component with a dynamic resistance inside the feedback loop, but I don't have time to go over the circuit in detail, and I trust you've done your homework correctly.

I agree about replacing R3 with an active down programming circuit. How about driving M1 in linear mode, to represent a moderate resistance to keep dissipation in check, if D5 is not forward-biased?
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 11771
  • Country: us
Re: General Purpose Power Supply Design
« Reply #3 on: March 20, 2012, 07:16:24 am »
That circuit goes way over my head, but it sure looks pretty  :)

On the pre-regulator aspect, what are the possibilities of making a switching pre-regulator that works at a low frequency like 100 Hz to avoid switching noise getting into the output? I'm thinking along the lines of a big filter cap as found in a normal linear supply, but switching current into it to top it up to just the right voltage. What size inductor in the switching circuit would adequately handle the low frequency, high current pulses?

I have seen an example in some application notes where an active SCR switching rectifier/pre-regulator was used, but I couldn't clearly follow how the circuit worked. It seemed to switch just enough of each AC half cycle to top up the reservoir capacitor to the desired voltage. What puzzled me is that this is clearly a very non-linear system and I was interested in how it was made to work in a stable and effective manner.
 

Offline amspireTopic starter

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: General Purpose Power Supply Design
« Reply #4 on: March 20, 2012, 07:43:21 am »
Any idea how well it will reject the (relatively) high frequency switching noise from the pre-regulator? Would you expect similar performance to the traditional lab supplies with huge heatsinks or do you trade off noise for efficiency? Most three terminal regulators have a fairly limited ripple rejection once the frequency gets much higher than 120 Hz.
I will see if I can test that with my breadboard. Setting up a power MOSFET as a source follower and feeding a signal generator in the gate should be able to modulate the supply input. I will let you know how it goes. Trouble is there are a number of issues with the breadboard version - I really need a PCB version to see accurately how it behaves dynamically.

It is an issue with the LT3080. It has 50dB rejection at 100KHz but it falls off like a stone after that. Most of the modern switching regulators work at 200KHz or higher. The one Dave used in has recent LTSpice simulation of a tracking regulator runs at over 1MHz.
Quote
I'm aware of one well-designed universal variable power supply, a design by German electronics maganize ELV. There's a link to a PDF of the article on the product page, unfortunately the text is in German. This design is quite flexible in input voltage and current range, and I haven't heard any complaints about stability. It is a more traditional design intended to use with transformers, however.
A different approach totally, but it is interesting to see what they have done. They have followed the standard HP method of the floating +/- supply for the opamps with the zero volts connected to the supply positive output. That is the way to get the best performance, but it needs a mains transformer, and it is really hard to find a suitable multi-winding transformer these days.
[quote

I might make the over voltage protection adjustable, so you can set it to 6V if that's the absolute max for your components, but I realize that this may be outside the scope, and that feature creep may screw up the intended purpose (simple, cheap).
[/quote] It certainly could be done, but the protection I implemented was for the power supply protection. For circuit protection, it is probably hard to beat an adjustable crowbar protection using a nice big indestructible SCR.
Quote

Interesting that you were able to get the MJE3055 to oscillate at 108 MHz, just make C3 switchable and you have a portable power supply and radio transmitter in one ;).
I could add one of those voice simulation chips and I could transmit the supply status to any FM radio in the neighborhood.
Quote
Intuitively I would be worried about the effect of D5 on the stability at low currents, since you're introducing a component with a dynamic resistance inside the feedback loop, but I don't have time to go over the circuit in detail, and I trust you've done your homework correctly.
I believe it is OK, but I will admit I left it off the breadboard version.  I can dig up a suitable diode and test it.
Quote
I agree about replacing R3 with an active down programming circuit. How about driving M1 in linear mode, to represent a moderate resistance to keep dissipation in check, if D5 is not forward-biased?
Not a bad idea. I will think on it.

Richard.
 

alm

  • Guest
Re: General Purpose Power Supply Design
« Reply #5 on: March 20, 2012, 08:04:58 am »
It is an issue with the LT3080. It has 50dB rejection at 100KHz but it falls off like a stone after that. Most of the modern switching regulators work at 200KHz or higher. The one Dave used in has recent LTSpice simulation of a tracking regulator runs at over 1MHz.
There are some three-terminal regulators specifically designed to be used with a switching pre-regulator, but this is an issue with the switcher + linear combination in general. Those 200 kHz/1 MHz frequencies are just the fundamentals, the edges produce much faster transients, since the MOSFET gates are usually driven quite hard to improve efficiency. It may be acceptable to aim for a higher noise level than the traditional linear supplies (eg. Agilent E36xx) if the advantages are efficiency, no heatsink and battery-powered operation. Many circuits don't need the <= 1 mV level of noise.
 

Offline markus_b

  • Regular Contributor
  • *
  • Posts: 115
  • Country: ch
Re: General Purpose Power Supply Design
« Reply #6 on: March 20, 2012, 08:39:49 am »
1. a low dropout design
2. single input supply
3. supply works to full specs with a wide range of input voltages
4. the regulation of the supply sets the output voltage to be the same as the opamp sense input (ie no voltage divider or voltage references involved in the supply regulation).

The result of #3 and #4 is the result is super flexible as you can feed the supply with whatever source supply voltage you like, and it will output whatever voltage you give it on the voltage control input. So a fully stabilized supply can be designed, and you can then make it output any voltages you like between about 0 and 30V without any change to the supply design at all. I do not have to redesign the control loop just because the voltage divider ratio has been changed.
I very much like those aspects of Daves design. It is really very simple and can be used with a potentiometer, PWM or DAC as control input. Brilliant !

So on to my design.  Here were my design specs:

3V to 30V input source voltage range
1.5V dropout voltage
True regulation down to zero volts
1A output current with 0 to 1A current limit
High level of protection from loads.
Transient performance similar to the LT3080
No-load to full load output voltage regulation  better then 0.001% (Wiring and track resistance is the biggest problem)
Loads like a battery can be connected to the supply while the supply is off without causing any problems to the supply or the battery.
Design based on "garden' components - potential to make it very cheap.
Possibility of a wide range 1A supply using only board mounted devices and no heatsink

I have been toying with a variant of Daves power supply with very similar specs, still based on a LT3080 (I do not have the analog experience to replace it). I think we can live with 10-30V input. But I've based my design on a xMegaA3U because it has a 12bit ADC & DAC, so we need no other digital components. It also has USB with a boot loader, so people can upgrade the Firmware in the field easily (no programmer needed).
Markus

A good scientist is a person with original ideas. A good engineer is a person who makes a design that works with as few original ideas as possible.
 

Offline amspireTopic starter

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: General Purpose Power Supply Design
« Reply #7 on: March 20, 2012, 09:46:09 am »
That circuit goes way over my head, but it sure looks pretty  :)

On the pre-regulator aspect, what are the possibilities of making a switching pre-regulator that works at a low frequency like 100 Hz to avoid switching noise getting into the output? I'm thinking along the lines of a big filter cap as found in a normal linear supply, but switching current into it to top it up to just the right voltage. What size inductor in the switching circuit would adequately handle the low frequency, high current pulses?

I have seen an example in some application notes where an active SCR switching rectifier/pre-regulator was used, but I couldn't clearly follow how the circuit worked. It seemed to switch just enough of each AC half cycle to top up the reservoir capacitor to the desired voltage. What puzzled me is that this is clearly a very non-linear system and I was interested in how it was made to work in a stable and effective manner.
Definitely can make a switching pre-regulator at 100Hz like the SCR one in the Linear Technology app note, but the high frequency switchers are just so small and efficient, and the ripple can be pretty good - you tend to get a fair amplitude of ripple at 100Hz unless you go to huge capacitors. 

For me, the key to low power is to reduce the ripple amplitude from the switching regulator to minimize the switcher to linear regulator differential, and for that the high frequency switcher is going to be the way to go. 100Hz switching designs were ideas from the 1970's electronics mags. Time to move on.

Richard.
 

Offline markus_b

  • Regular Contributor
  • *
  • Posts: 115
  • Country: ch
Re: General Purpose Power Supply Design
« Reply #8 on: March 20, 2012, 12:39:38 pm »
Yes, I agree, a high frequency switcher to pre-regulate is the way to go. We probably should leave 2V for the linear regulator, this limits its power dissipation to 2W/Amp, something we can live with.
Markus

A good scientist is a person with original ideas. A good engineer is a person who makes a design that works with as few original ideas as possible.
 

Offline amspireTopic starter

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: General Purpose Power Supply Design
« Reply #9 on: March 20, 2012, 03:21:53 pm »
Yes, I agree, a high frequency switcher to pre-regulate is the way to go. We probably should leave 2V for the linear regulator, this limits its power dissipation to 2W/Amp, something we can live with.
I will take a look at the switching regulator soon. Right now, I am seeing if I can get a second version of my design working. It is almost stable, but there is a little ringing on transient load changes. Got to eliminate that - ringing means something is not right yet.

Richard.
 

Offline markus_b

  • Regular Contributor
  • *
  • Posts: 115
  • Country: ch
Re: General Purpose Power Supply Design
« Reply #10 on: March 20, 2012, 03:43:29 pm »
I imagine that the entire thing is powered with a power supply like this one:
Universal laptop power supply (24V)

I like the 1V range of the inputs, my pet micro (xMega32A4U) for this has a 1V internal reference.
Markus

A good scientist is a person with original ideas. A good engineer is a person who makes a design that works with as few original ideas as possible.
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 11771
  • Country: us
Re: General Purpose Power Supply Design
« Reply #11 on: March 20, 2012, 04:21:25 pm »
Definitely can make a switching pre-regulator at 100Hz like the SCR one in the Linear Technology app note, but the high frequency switchers are just so small and efficient, and the ripple can be pretty good - you tend to get a fair amplitude of ripple at 100Hz unless you go to huge capacitors.

For me, the key to low power is to reduce the ripple amplitude from the switching regulator to minimize the switcher to linear regulator differential, and for that the high frequency switcher is going to be the way to go. 100Hz switching designs were ideas from the 1970's electronics mags. Time to move on.

OK, I understand time marches forward. But my thinking was that larger amplitude, lower frequency ripple might be easier for the linear regulator to filter out than small amplitude, high frequency ripple. You mentioned for example that the LT3080 ripple rejection falls off dramatically above 100 kHz. If a very clean supply is desired, is there a balance to be found somewhere above 100 Hz and below 100 kHz?
 

Offline amspireTopic starter

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: General Purpose Power Supply Design
« Reply #12 on: March 20, 2012, 09:16:51 pm »
At 100hz, you need a 10000 uf cap to reduce the ripple to 1v peak to peak.

If the regulator is at maximum and you short the load, the regulator had to dissipate all the stored energy in that big capacitor. Remember the whole idea of using a preregister is to reduce the power dissipation in the regulator.

You want to stay away from the rest of the audible frequencies unless you like listening to your regulator. 20khz to 100 khz is workable. For 1v ripple at 100 khz, all you need is a 10uf cap. Much better!

The alternative is a high frequency switcher followed by a LC filter network to eliminate the high frequency. Definitely possible but it might take a few PCB revisions to get it working right.

Richard
 

Offline markus_b

  • Regular Contributor
  • *
  • Posts: 115
  • Country: ch
Re: General Purpose Power Supply Design
« Reply #13 on: March 20, 2012, 09:38:40 pm »
In Daves latest video he looks at a switching regulator in series with a LT3080. The switching regulator in question runs at 750kHz and this is in the official LT3080 data sheet. The TL3080 seems to have no problem with the ripple of the switching regulator there.

I seem to read data sheets too rapidly, at leat I overlooked that part before, looks very interesting !
Markus

A good scientist is a person with original ideas. A good engineer is a person who makes a design that works with as few original ideas as possible.
 

Offline fmaimon

  • Supporter
  • ****
  • Posts: 165
  • Country: br
Re: General Purpose Power Supply Design
« Reply #14 on: March 21, 2012, 12:18:03 am »
If I´m understanding correctly, you are feeding the pass transistor (Q1) with a current of 10 mA generated by a constant current source (Q3, D1, D2, R1, R2 and C3) and you regulate the output voltage by bleeding this current with Q2/R2, right?

A couple questions.

1. Shouldn't the dropout voltage be about 2.5V, as you need 1.4V for the current source feeding the base, another 0.7V for the Vbe of Q1 and 0.4V for D5?

2. With a hfe of 20-70, there is no way of getting 1A with an MJE3055 without increasing the current from Q3 (decreasing R1 to 12 ohms), right?
 

Offline amspireTopic starter

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: General Purpose Power Supply Design
« Reply #15 on: March 21, 2012, 02:44:32 am »
In Daves latest video he looks at a switching regulator in series with a LT3080. The switching regulator in question runs at 750kHz and this is in the official LT3080 data sheet. The TL3080 seems to have no problem with the ripple of the switching regulator there.

I seem to read data sheets too rapidly, at leat I overlooked that part before, looks very interesting !
At 750kHz, the LT3080 has only about 13dB of line ripple rejection, and most of that would be the 2.2uF output capacitor that they specify. That means that if you have 500mV ripple going into the LT3080, you will have about 100mV coming out.

So the reality is if you want to use a 750kHz pre-regulator, you have to remove the switching noise before it gets to the LT3080. A good filter could probably get rid of more like 60dB of switching noise.

Richard.
 

Offline amspireTopic starter

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: General Purpose Power Supply Design
« Reply #16 on: March 21, 2012, 03:01:14 am »
If I´m understanding correctly, you are feeding the pass transistor (Q1) with a current of 10 mA generated by a constant current source (Q3, D1, D2, R1, R2 and C3) and you regulate the output voltage by bleeding this current with Q2/R2, right?

A couple questions.

1. Shouldn't the dropout voltage be about 2.5V, as you need 1.4V for the current source feeding the base, another 0.7V for the Vbe of Q1 and 0.4V for D5?

The dropout of the current source is something like 0.7V as the transistor can turn on to saturation. For the shottky, I just used one in the LTSpice library, but you can get schottky's with 0.3V drop.

Quote

2. With a hfe of 20-70, there is no way of getting 1A with an MJE3055 without increasing the current from Q3 (decreasing R1 to 12 ohms), right?
I tested the gain of the one I had and it was 120, and I did use a lower R1. But there are transistors available with gains of 200 to 350 at 1A, and that is the ones I would go for in a final design. High gain transistors eliminate most of the heat in the driving transistors.

Anyway, I may be close to getting a PMOS based design working with a dropout of 1V. I have a stable design, but I have to check the margins of the stability. A lot less parts then the transistor based design. This would be great if I can pull it off as there are fabulous PCB mounted MOSFETs available. Much easier to find then the super gain PCB mount medium power transistors.

It would be much easier to scale the design up to higher currents with MOSFET power devices, but at the moment, I am a fair way from a circuit that would allow me to swap to different MOSFETs and remain stable.

Richard.
 

Offline IanB

  • Super Contributor
  • ***
  • Posts: 11771
  • Country: us
Re: General Purpose Power Supply Design
« Reply #17 on: March 21, 2012, 05:42:01 am »
How do devices like the LT3080 "miraculously" remain stable with different loads? I had the presumably naive impression it was all down to placing a suitable amount of capacitance at the output? (Qualitatively, the capacitor dominates the properties of the output system and isolates the load from the controller. The feedback loop only "sees" the capacitor, so as long as the loop is stable with the chosen capacitor it will be stable with all reasonable loads.)

In my world I have never had to do the kind of frequency domain stability analysis that EE's often have to do. It is interesting to see the problem from that perspective.
 

Offline BravoV

  • Super Contributor
  • ***
  • Posts: 7547
  • Country: 00
  • +++ ATH1
Re: General Purpose Power Supply Design
« Reply #18 on: March 21, 2012, 06:14:23 am »
Design based on "garden' components - potential to make it very cheap.

I will get onto the design in the next post. This design exercise is not to make something similar to Dave's supply. I have left out any attempt at this stage in a uCurrent type device, and I am only looking a very low cost high performance regulator that could be controlled by just plain potentiometers.

Richard, thanks for starting this initiative especially on not choosing exotic part, really appreciate it.

Love the idea on using "garden" components, please keep it this way should this design evolved into more complex circuit. Eagerly waiting to see the pre-regulator section, hopefully you will use those "garden" switcher as well.  :)

Btw, how does this design scale at higher current rating ?

Offline amspireTopic starter

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: General Purpose Power Supply Design
« Reply #19 on: March 21, 2012, 06:45:18 am »
The intuitive thing is that extra capacitance, but capacitance is the most difficult load of all for stability.

I will try and keep it simple. Power supplies use feedback to regulate the output which means you have a high gain amplifier with negative feedback. The negative feedback will change the output voltage until the difference between it and the reference voltage is zero.

Now add frequency to the equation. Due to internal capacitance, amplifiers loose gain as the frequency increases. What goes along with that is phase shift. If at any frequency a control loop has a gain more then 1 and the total phase shifts reaches 180 degrees, the control loop will oscillate.

Opamp designers have known about this forever. What they do is to make the input and output stages of opamps really fast, and then slow the middle gain stage down with a single RC time constant. This has a phase shift of 90 degrees from about 10Hz up to 1MHz (or whatever the opamp goes to). With one single RC dominating, the phase shift through the opamp's whole frequency range is around 90 degrees so it is stable.

Just to make this clear, if you had a 1KHz signal going into a unity gain opamp, the output would look pretty much in phase with the input. But if you looked at the microvolt AC signal between the inverting and non-inverting inputs, that signal would be leading by 90 degrees. Simply to have a zero phase lag 1KHz coming from the output, the actual input signal between the input pins has to be leading by 90 degrees.

Now we come to power supplies. you still have the opamp with its 90 degrees phase shift, but you have extra stuff too -  power transistors, driver transistors, protection circuitry, etc, and no matter how hard you try, these all add extra phase shift. Ok so say you keep all these other things to an extra 45 degrees - the supply is stable?

Unfortunately for the designer, stupid people actually want to put loads on these power supplies. What happens when you put a massive capacitor on the output of a supply? You have another 90 degrees phase shift and it oscillates. The bigger the capacitor, the more likely the supply will oscillate. 

So the power supply designer and the IC designer has to do the impossible - if you disconnect any filter capacitor from the output, the control loop has to have less then 90 degrees phase shift all the way from Dc to the 100kHz+ bandwidth of the loop. You can follow the opamp trick of making everything blindingly fast except for one RC constant and then you do some tricks to slow the roll of to less then the 6dB per octave of the RC time constant, and the result is you get a little less then 90 degrees. This is the technique probably used in the LT3080.

For the discreet componect supply designer, you usually want to use an opamp that has the 90 degrees RC time constant built in, so you have to do tricks with external RC networks to cause a bit of phase lead to cancel out the excessive phase lag. It is not easy which is why Dave went for the LT3080 solution. If he did design a discrete supply, he would probably be on video blog number 23 by now - still fiddling with the compensation.

The conclusion is that for most power supplies, if it is stable with a 100,000uf capacitor under all output voltage and current conditions, it is probably going to be stable for most other loads. Lots of power supplies out there are probably not stable with a huge capacitor, as the voltage is hardly changing on the capacitor - it is the control circuit in the supply going nuts. Most people wouldn't notice. This may seem OK, but it is not, as it will affect regulation, the current limit control, and a few other things.

Richard.

« Last Edit: March 21, 2012, 07:11:39 am by amspire »
 

Offline amspireTopic starter

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: General Purpose Power Supply Design
« Reply #20 on: March 21, 2012, 06:58:44 am »
Design based on "garden' components - potential to make it very cheap.

I will get onto the design in the next post. This design exercise is not to make something similar to Dave's supply. I have left out any attempt at this stage in a uCurrent type device, and I am only looking a very low cost high performance regulator that could be controlled by just plain potentiometers.

Richard, thanks for starting this initiative especially on not choosing exotic part, really appreciate it.

Love the idea on using "garden" components, please keep it this way should this design evolved into more complex circuit. Eagerly waiting to see the pre-regulator section, hopefully you will use those "garden" switcher as well.  :)

Btw, how does this design scale at higher current rating ?

I am close to a simpler MOSFET design that does scale really well. The problem I have now is picking some MOSFETs to focus on. There once was a time when there were only about 20 common MOSFETs and life was extremely simple. Now the numbers are ridiculous.

The transistor design I posted probably works best at 1A. You could push it to 2A by using the really high gain Zetex transistors. That is about it without without a redesign.

Richard.
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 37626
  • Country: au
    • EEVblog
Re: General Purpose Power Supply Design
« Reply #21 on: March 21, 2012, 07:31:56 am »
For the discreet componect supply designer, you usually want to use an opamp that has the 90 degrees RC time constant built in, so you have to do tricks with external RC networks to cause a bit of phase lead to cancel out the excessive phase lag. It is not easy which is why Dave went for the LT3080 solution. If he did design a discrete supply, he would probably be on video blog number 23 by now - still fiddling with the compensation.

Bingo, I'm not as silly as I look  :P

BTW, I'm not sure why everyone thinks my supply is locked into using the "exotic" LT3080. You can substitute for an LM317, but then it only goes down to 1.2V which is not a big deal in most cases. Just have to offset the drive voltage by 1.2V. And that aspect can be fixed if you really want.

Dave.
 

Offline T4P

  • Super Contributor
  • ***
  • Posts: 3697
  • Country: sg
    • T4P
Re: General Purpose Power Supply Design
« Reply #22 on: March 21, 2012, 08:08:45 am »
Dave ,
what's your thoughts on the numerous articles out there claiming the LM317 to have a similar performing opamp as a LM741 "They claim it's bad , but then i looked , it's a audiophoolery site" ?
Another case of them wanting overkill for the opamp ? I don't know .  ;)
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 37626
  • Country: au
    • EEVblog
Re: General Purpose Power Supply Design
« Reply #23 on: March 21, 2012, 09:54:38 am »
I am close to a simpler MOSFET design that does scale really well. The problem I have now is picking some MOSFETs to focus on. There once was a time when there were only about 20 common MOSFETs and life was extremely simple. Now the numbers are ridiculous.

Yup.
Is the world better or worse off as a result? Sometimes I'm not sure...

Dave.
 

Offline EEVblog

  • Administrator
  • *****
  • Posts: 37626
  • Country: au
    • EEVblog
Re: General Purpose Power Supply Design
« Reply #24 on: March 21, 2012, 10:31:32 am »
I think people may not get how brilliant and revolutionary Dave's concept is. We are totally used to hand held meters, but we are still stuck to the idea that a power supply has to be anchored to a wall socket on the bench, and the idea that you do development on the bench, rather then the place the design is actually going to be used - whether that is on the roof, half way down a canyon, in the car, on a boat or wherever.

If a supply efficiently uses the energy in a lithium battery, the supply can easily last for days powering a typical modern low power circuit. Given that, why would you want a power cord at all?

Just on the battery part.
I started out with the idea for 3 x 18650's because that's what I fitted in the small case, and I wanted maximum capacity.
Then I dropped to 2 x 18650's because it was cheaper/easier on the charging side.
Now I think I'll be dropping it to 1 x 18650 (or optional larger cell) and going with a proper robust charging solution that gets maximum life out of the cell and better handles the ability to charge and power the supply at the same time. It also adds the USB charging feature back.
The idea is that it's a mains supply most of the time, but when you need it portable, just disconnect and it's ready to go for a few hours or days depending upon the load.

Maybe that belonged in the other thread, but oh well...

Dave.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf