Hysteretics have "built in" hysteresis (notice the name?) in the comparator to keep the thing from endlessly chattering. It's one of the main disadvantages of a hyst design, since it's output voltage ripple is higher due to the fact the output goes up and down a specific amount of voltage between the design set points where the oscillator turns on (Vlow) and shuts off (Vhigh).

I think you are ignoring my point of posting: IMHO the circuit as shown won't work, and it's ridiculous when a simulator shows performance data for a circuit that can't work since it is LYING and telling the user it does work. The schematic doesn't even include a pull-up resistor to turn off the switching transistor. As I said: if you think that dog can run, build it up. Then when you get it working, post the final schematic and compare to the one posted in the OP.

I'm not ignoring your point, but you seem to be ignoring mine. The "built-in" circuit (not comparator) hysteresis (yes, I noticed the name) is provided by the lag of the output LC filter as I previously noted. You can add some hysteresis to the comparator to minimize oscillations in the comparator due to local feedback but it's not needed for the intrinsic operation of this hysteretic converter. The simulator is not "LYING". The circuit does work as shown (albeit with some ideal characteristics).

The comparator in the schematic has a push-pull output drive so it doesn't necessarily need a pull-up resistor. (But the simulation doesn't show that the comparator would likely blow with 12V applied so that is a definite simulator limitation).

You can say the dog can't run but you give no reason other than saying so. You'll have to pardon my lack of agreement.

I don't really have time to build a circuit to prove a point, and even if I did and it worked, I'm sure you'd find some reason to show how it's an exception to your rule that simulators lie.

It's true Bob and I were skeptical of sims, and with good reason: we kept running into a whole batch of idiots who use them as a crutch. Our customers used sims to leapfrog the design phase and then howled when their crap didn't work. No person should ever be allowed to use a sim unless and until they are experienced enough to know where the sim is lying. That sentence can not be emphasized enough. Even some of our "best" designers fell into the lazy trap and got burned using sims for IC design. The so-called best models don't work in many cases and designers get bit.

..............................
In this thread, the problem is pretty clear: the OP doesn't know enough about a buck converter to know where the power losses are and is using a sim to give performance data. I don't see that as helpful.

Then exactly what good did simulating it do? The "data" from the sim can't be taken as accurate, you still have to take real data. The design is not accurate, you still have to make it work. The designer should read an app note, build the circuit, modify until performance is correct, then they are done. In the process they gain the understanding of how it actually works.

.............................

What good did simulating do? Simulating showed that the circuit will fundamentally work. It can save a lot of initial design errors and help develop the design parameters and explore design margins. Then you tweak the final design with the breadboard.

You are sour on simulation because you had experience with a bunch of idiots who don't know how to properly use a good tool so you throw the tool out with the bath water. Better to educate the idiots about how to properly use the tool.

But I understand I'm beating a dead horse here. You don't find simulators useful and I do. So I guess we'll have to let it go at that.

 

----------------------------------------------------------------------
Using the same configuration, I use LM393 comparator, with 1kR pull-up resistor at the output. 12V voltage-source, those two light bulbs are as my R2, just in case it would blow-up.

300uH inductor, Q1 is 2N2222, Q2 is BD140 with 4.7kR pull-up resistor at the base for turn-off enhancement, and the fast diode is MBR1635.

1N4148 as my Vref for non-inverting input, connected to a voltage divider, adjustable from 0V to ~80mV. The Rsense is 150mR.
----------------------------------------------------------------------
I made some rough measurements...

The powerIn is 12.15V*340mA=4.13W, powerOut is 3.3V*250mA=825mW.
Efficiency approximately 20%.

That is so poor. The Q2 gets warm, without heatsink.
----------------------------------------------------------------------

Do you use light bulbs to limit input current? That is generally not so good for switching circuits, the voltage source impedance should be as low as possible.

4.7K pullup for the BD140 might be far too high.

I use 47 Ohm + 180 Ohms for PNP but that's for 50V.
And the 180 Ohm is shorted to ground using smaller NPN.

The coil might be wrong as well, I suggest to try different one's.

If you get less than 70% efficiency, not a good switcher circuit.

 

To be fair to the circuit you would exclude the losses in R2 [the light globes] from the input source power.

It also looks [from the CRO traces] like oscillation is intermittent - possibly due to the effect of the globes pulling the supply low and the interaction with the supply bypass capacitance.

 

Geez, you guys makes me feeling guilty now.
I never thought it could be this complicated.

................................................
I never meant to cause troubles...

Nothing to feel guilty about. You didn't cause any troubles... Sometimes threads tend to take on a life of their own when some of the posters get into spirited discussion about a tangent issue but that's what makes the forum interesting. I believe some of the discussion was still helpful to you.

 

Adding some deliberate hysteresis would improve the circuit a lot.

 

...
The coil might be wrong as well, I suggest to try different one's.

If you get less than 70% efficiency, not a good switcher circuit.

What did you meant by wrong coil? Could you elaborate more?
And yup, I know the circuit is bad, hence the purpose of this thread...

To be fair to the circuit you would exclude the losses in R2 [the light globes] from the input source power.

It also looks [from the CRO traces] like oscillation is intermittent - possibly due to the effect of the globes pulling the supply low and the interaction with the supply bypass capacitance.

You're right! I forget to take that into account... Thanks!

Nothing to feel guilty about. You didn't cause any troubles... Sometimes threads tend to take on a life of their own when some of the posters get into spirited discussion about a tangent issue but that's what makes the forum interesting. I believe some of the discussion was still helpful to you.

Glad to know that...
Without doubt, all of the responses here are extremely helpful to me.

Adding some deliberate hysteresis would improve the circuit a lot.

I done some Googling on hysteresis, but still can't understand what it is for. Could you explain?

Sorry for all of the troubles...

 

...
I done some Googling on hysteresis, but still can't understand what it is for. Could you explain?
...

Since you googled and know what hysteresis is, in this application it will have the effect of making the output current oscillate between 2 current levels.

So instead of your output oscillating fast and chaotic somewhere around 1 amp, the hysteresis wil make it oscillate more slowly and in a much more stable way, between say 0.9A and 1.1A.

 

I just build the circuit again, with some minor changes...

I replaced Q1 with PNP, the comparator feedback are now reversed, and both Q1 and Q2 has 4.7kR pull-up resistor. The rest are pretty much the same.
The upper trace is suppose to be the efficiency curve...

The built circuit, running.

Close-up.

Scope-shot at comparator's output. Weirdly enough, it's not digital, I wonder why...


The performance seems to improve a lot. The rough efficiency now is at least 60%. I'm running it with 6V battery, with no light bulb in series with it as before. (Thanks to takao21203 and t_n_k)

The transistor also appears to be running cooler now.
The input and output current seems to be pretty close to each other, weird.

 

Since you googled and know what hysteresis is, in this application it will have the effect of making the output current oscillate between 2 current levels.

So instead of your output oscillating fast and chaotic somewhere around 1 amp, the hysteresis wil make it oscillate more slowly and in a much more stable way, between say 0.9A and 1.1A.

Wow, that's super fast reply!

Hmm, could I apply the hysteresis by adding a resistor across the comparator inputs? i.e. a 100kR at +ve and -ve input of the comparator.

Thanks a lot!

 

Well by wrong coil I mean a coil which is causing too much loss, or causing too much heat up inside the transistor.

This can be quite a concern at higher output currents, let say 4 Amps, 6 Amps etc.

For my TL494 project I tried many different cores, changed number of turns, wire diameter, but in the end, I am now using larger cores that I specially ordered for this project. I am lucky they are suitable, and it's even possible to use two in parallel. The wire is thin, tough, so they need exposure to airflow.

See the attached schematic. I used it as starter, modified many things, but the power PNP drive is essentially the same! I mean, see for the voltage divider that is used. And it also works with P-CH MOSFET, without modification!

For 12v the voltage divider needs different levels.

Maybe I will try to rebuild your circuit, using only one small PNP, and only for smaller currents. Alternative to MC34063/LM2576, as I use 12V electronic transformers, and often need 3V for circuits. What I have here is LM358.

 

Hmm, could I apply the hysteresis by adding a resistor across the comparator inputs? i.e. a 100kR at +ve and -ve input of the comparator.

To add hysteresis you need a small amount of positive feedback. Add a resistor in series with the + input and a larger resistor between the + input and the output.

If the output waveform of your comparator is not square, then it means it is trying to drive a lower impedance load than it can handle. What comparator are you using?

 

After tinkering with it for quite long, I guess the circuit really not worth any more time.
I just can't get the efficiency more than 65%.

I just came across THE_RB's 'Black Regulator' and find it quite neat. I guess I will start from there.

And, if I ever give up, I would give MC34063 a try.

Thanks again guys for all of the help...

 

You don't need to give up on your comparator circuit it should work quite well provided you get some hysteresis in the switching so it switches as a slower freq clean square wave.

 

MC34063, LM2576 and the like are pretty easy to work with if they are needed for a circuit. I get these working almost always immediately.

Only if you need high power, then you need to design a circuit.

Or to research/for experiment.

For my part I don't understand why people still use wall adapters with different voltages, and linear regulators. Above ICs can replace this, and all can work from 12V cheap boxed electronic transformers.

If there is any kind of noise for instance circuits which will vary frequency depending on output voltage, these circuits likely won't be used for commercial designs nowadays.

Power Integrations and Texas Instruments have application notes as well their own forums, also National Semiconductors / International Rectifiers have App notes on their web pages. Microchip has some documents about SMPS as well.

 

Hi,

I have just about done the tour of this and similar threads here so have some idea what I need to do.

My aim to build an MPPT regulator for 240W of PV panel, so looking at a max current circa 20A. Open cct max <40V.

My basic idea was something like OP started with but using NMOS to drive the inductor.

However, the last post got me wondering whether I could use something like the 2576 to provide the control logic but use a more powerful discrete device to increase the current.

http://www.datasheetcatalog.org/datasheet/BayLinear/mXsqqqr.pdf

The output voltage will be varied using something like an Atmel uP to ensure correct charging of lead acid battery load and to track the max power from the panel.

This could be done by interfering with the feedback loop. Designing a stable fixed regulator is probably the trickier part.


Otherwise the block diagram in the application notes would probably make a good starting point.

Does using the 2576 like this seem sensible?

Thanks.

 

Hi,

I have just about done the tour of this and similar threads here so have some idea what I need to do.

My aim to build an MPPT regulator for 240W of PV panel, so looking at a max current circa 20A. Open cct max <40V.

My basic idea was something like OP started with but using NMOS to drive the inductor.

However, the last post got me wondering whether I could use something like the 2576 to provide the control logic but use a more powerful discrete device to increase the current.

http://www.datasheetcatalog.org/datasheet/BayLinear/mXsqqqr.pdf

The output voltage will be varied using something like an Atmel uP to ensure correct charging of lead acid battery load and to track the max power from the panel.

This could be done by interfering with the feedback loop. Designing a stable fixed regulator is probably the trickier part.


Otherwise the block diagram in the application notes would probably make a good starting point.

Does using the 2576 like this seem sensible?

Thanks.

Hello. I have boosted a LM2576 using a p-ch MOSFET + gate transformer.
To start up a 12V 50W bulb I had to patch it a lot because there is internal current sensing built in. I got it working but only marginally.

The LM2576 is not so good for this because the internal current limiting. There seems to be more than one limiting scheme built in.

Better use a TL494, or MC34063 (which both have external current sense).

 

It must the internal power dissipation limiter that is triggering. Why is that so when your driver is external ?

I'm not sure exactly what version of your cct this refers to but if you have something like 47R driving the transformer that is the likely cause.

I have not seen this sort of coupling used for high side switching (probably expensive and bulky) is it tested as being stable or is it just an idea you tried?