555 Boost

[mikeinfodoc]

I want to create a boost circuit, to take a 9v regulated wall power supplies, and increase the output voltage to 12V.

I intend to drive 100mA worth of tiny efficient LED's (via an LED strip).

I have drafted up this attached simple sim to start with.


There are so many variables I should consider, such as:

• duty cycle consideration
• resister values on the 555 timer (which is shown on the left of the schematic
• inductance value
• storage cap value (the E-cap on the right of the schematic)

I am aware of some of the math, but because there are so many variables, I was not sure where to start.


I welcome your thoughts on recommended values and parts.  (or any other feedback you may have for a beginner)

[Coceth]

If I had to do this, I'd probably get a boost converter IC. The datasheet should have a guide on choosing component values to set the desired output voltage. Maybe this part would work? https://www.digikey.com/product-detail/en/diodes-incorporated/AP3012KTR-G1/AP3012KTR-G1DICT-ND/4505297

Good luck!

[bd139]

Use an MC34063 IC (basically a 555, comparator and switch in a single package). You want to drive LEDs with current rather than voltage and you can set it up as a current source. Plenty of calculators on the web to tell you what parts to use. The advantage with this solution is if you add another strip in series it will still drive it (until you go over voltage and the 34063 explodes)

[IanMacdonald]

R3 being so high will waste a lot of power. Instead put a series resistor in the 9v supply for test purposes so as to prevent overcurrent, and adjust the 555 mark/space to give the desired LED current. Once you have the correct current, remove the series resistor and you have an efficient circuit.

This assumes the LED turn-on voltage will always be higher than the 9v supply. If not, then using a ferrite cored transformer may be better. Or a commercial boost/buck unit. Maybe $5 at most from Ebay.

Oh, and C4 should be after the diode. As it is, it shorts the FET.

[mikeinfodoc]

Thank you,  Good catch on the C4 after the diode.

I adjusted a few parts and values, and got it closer to my specs.

I am attaching my latest schematic/sim. 

Notice however that, my Vout is near 40V, which is much higher than my desired 12v out.

I welcome thoughts about how to adjust part values to make the inductor loop generate lower Vout.

[Dachpappe]

Well I also wouldn't recommend a 555 for this purpose but a few years ago i built a similar thing as I only had 555s laying around. It was a small powerbank, where 3,6V were converted to regulated 5 volts. I can not find a schematic at the moment, but I still have this thing flying around somewhere. I will try to find it and redraw the schematic for you.

[StillTrying]

If you can afford to lose another 0.65V, you might be able to control the average LED current this way, it'll be a bit rough and vary a bit with temperature, and you have to be careful a spike doesn't blow the base-emitter.
There's a 555 boost on here somewhere, probably one of Hero999's!

[Dachpappe]

Here's the powerbank from an era when you could not buy one at every corner 

Was a bit messy inside, but it could deliver up to 1,4A and the voltage regulation was acceptable. I'm not sure if you can go to 12V with the values in my schematic, but I think you don't need to change much more than the pot in the feedback section. Additionally you can try to connect the reset pullup resistor to the input voltage directly.

[b_force]

Why using the 555?
Just for fun? Because there are hundreds of better boost converters out there.

[Zero999]

I echo the comments about not using the 555 timer.

If you must use this circuit, you can get rid of D1 R3 and C4 and let the eye do the low pass filtering, like a Joule thief. It will be more efficient without the power losses in those components.

[mikeinfodoc]

Use an MC34063 IC (basically a 555, comparator and switch in a single package). You want to drive LEDs with current rather than voltage and you can set it up as a current source. Plenty of calculators on the web to tell you what parts to use. The advantage with this solution is if you add another strip in series it will still drive it (until you go over voltage and the 34063 explodes)

So, for my desired LED strip driver what would you recommend for Freq min?  (is there any advantage of slower or faster freq?)

I assume the noise/interference is one of the factors to my question above, and also the size of the inductor I have laying around.  But please enlighten me, as I suspect the Freq speed decision is probably more involved

[mikeinfodoc]

I echo the comments about not using the 555 timer.

If you must use this circuit, you can get rid of D1 R3 and C4 and let the eye do the low pass filtering, like a Joule thief. It will be more efficient without the power losses in those components.

if I did remove those, what values would you recommend for the other remaining components?  (R1, R2, C1 and L1)

...to get to at or near 12V on the output

[mikeinfodoc]

Why using the 555?
Just for fun? Because there are hundreds of better boost converters out there.

Mainly because I have about a dozen of them, and they are dirt cheap. 

Also, because I was hoping to do many various boost circuits using recycled/discarded wallwart power supplies of varying amounts (whatever I find at goodwill).

I was thinking I could simply create a several 555 astable circuits (with adjustable pot), each of which could be used as the timing/switching source for any of my boost circuits (which will likely have varying purposes, and V & I output needs)

Is there aany fundamental issues with this vision?  (aka: any inherent weaknesses to the astable timing aspect of the 555?)

[T3sl4co1l]

Why does the OP schematic show four farads across the transistor?  That's rather counterproductive...

Some ideas:
https://imgur.com/gallery/SoYtht4

If you're heart set on using a 555, you can use two with some support components to make a peak current mode controller, which will be well behaved under all conditions: startup, inrush and load step (including heavy loading).  It's nowhere near as convenient as using a proper device.


If you're concerned that you won't learn anything using an integrated controller, then build one out of discrete parts: op-amps, comparators and logic.  TL494 and UC3842 both have detailed enough block diagrams that they can be built from these components, right off the datasheet.

Tim

[b_force]

Why using the 555?
Just for fun? Because there are hundreds of better boost converters out there.

Mainly because I have about a dozen of them, and they are dirt cheap. 

Also, because I was hoping to do many various boost circuits using recycled/discarded wallwart power supplies of varying amounts (whatever I find at goodwill).

I was thinking I could simply create a several 555 astable circuits (with adjustable pot), each of which could be used as the timing/switching source for any of my boost circuits (which will likely have varying purposes, and V & I output needs)

Is there aany fundamental issues with this vision?  (aka: any inherent weaknesses to the astable timing aspect of the 555?)

Efficiency and total surface area, parts count.
Also, with a 555 you still need an external MOSFET.
For the specs you're using you can rather get a dedicated boost converter.
In the end that's more reliable, more robust, more efficient and cheaper

Even dedicated 12V LED PSU's are so cheap nowadays.
Depending how many you're gonna use, but it will pay itself back since the efficiency is much better.

[T3sl4co1l]

Just for S&Gs, I did it in one 555:

https://www.seventransistorlabs.com/Images/555%20Boost.pdf

Good for about 5W at the values shown.

Gimmicks:
- Peak current mode control (see UC3842, etc.), constant off-time oscillator
- Uses 555 for current comparator, latch and gate driver
- Error amplifier regulates output voltage precisely
- TL431 provides accurate voltage reference

Pitfalls:
- 555 has one adjustable comparator threshold, and the other is half that.  Namely, when V(THR) >= V(CTRL), the output is turned off.  This is good for a peak current mode controller (if we simply put V(current) on THR, we have the entire output stage solved!), but terrible for a current sense (V(CTRL) is typical 10V!)
- I used a common base amplifier (noninverting voltage gain) to solve this.  With a gain of about 10, the current shunt resistor sees a peak of 1V amplified to a ~rail sized voltage.
- The CTRL pin has internal resistors loading it down.  Beefy resistors, 3.3k Thevenin equivalent.  This sucks for driving, and the LM358's crossover distortion and current output limitations are evident.  Just have to grin and bear it.
- The TRIG threshold is half of V(CTRL).  This is the only way to begin an output cycle (there are no other "trigger" inputs).  This is the logical place to start for closing the loop and making it oscillate, but the variable threshold frustrates stable timing.  We could add a second 555 as clock source, and not have to worry about poor timing.

Thinking about it, I noticed, if the timing capacitor could be charged up to V(CTRL) every cycle, and discharged towards ground with a timing resistor, then that would solve the timing problem, to first order*.

The time constant is then given by the time taken for the capacitor to discharge by half, i.e. from V(CTRL) to V(CTRL)/2, where the factor of 1/2 comes from the 555's internal divider resistors.  This is ln(2)*R*C.  R10 and C4 are the timing components, giving a 3.2 microsecond delay.

*(Higher orders, with respect to time, are problematic: CTRL is being used as a control input, meaning its value will generally change over time.  That means, so will the timing.  Timing wouldn't change if the capacitor's instantaneous voltage also varied proportionally with V(CTRL), but that's a trick we can only pull in SPICE.  Relatively over-damped compensation is probably desirable, to limit how fast V(CTRL) varies, preventing weird interactions between oscillation frequency and line and load ripple.)

The real situation is not ideal, however: Q2 cannot charge the capacitor fully to V(CTRL).  A diode drop is lost, which hastens the delay.  In particular, when V(CTRL) < 1.2V, the delay goes to zero.  The condition arises at low supply voltage and high load (usually during startup).  The result is maximum output duty cycle, which tends to brown out the supply and melt the transistor!  D2 was added to help prevent this.

This gives a switching frequency around 140-200kHz, depending on supply and load.

I can probably nix the LM358 as well, by using the TL431 directly.  Having an open collector output is rather inconvenient, so I'd have to add a complementary follower to give it a proper sink-and-source output.

To convert this to an isolated flyback supply, wire the error amplifier as a unity gain inverter, and wire up the optoisolator and secondary-side TL431 (error amp) as usual.  (The primary side TL431 is no longer necessary and can be replaced with a voltage divider.)

Tim

[b_force]

Neat circuit!

But this is what I mean, it can easily be done with just one single little chip and an handful of passives.

[mikeinfodoc]

T3sl4co1l, NOW I know what is meant by using a boost chip to lower the BOM.

Your feedback is very much appreciated.

I was going to head down the path of adding an LM358 (as I have a few of those in stock also) for a feedback loop, but your circuit is far more complex than I could have dreamed up.

I will study it and your notes however

[mikeinfodoc]

Neat circuit!

But this is what I mean, it can easily be done with just one single little chip and an handful of passives.

What IC would you recommend for my circuit?

I saw mentions of these earlier

• TPS61040
  UC3842
  TL494
  mc34063a
  

do you have a preference for my use case?

ps: I do pic/atmel microcontroller programming, as I have a programming background, but I feel like I am cheating, by not doing things the hard way  :-)

[b_force]

Neat circuit!

But this is what I mean, it can easily be done with just one single little chip and an handful of passives.

What IC would you recommend for my circuit?

I saw mentions of these earlier

• TPS61040
  UC3842
  TL494
  mc34063a
  

do you have a preference for my use case?

ps: I do pic/atmel microcontroller programming, as I have a programming background, but I feel like I am cheating, by not doing things the hard way  :-)

Most of them are rather old.
I think the TPS61040 actually doesn't swallow 9V?
Just go to the website of TI, Linear technology or so, maybe just digikey and have a look what suits best.

[T3sl4co1l]

Programming is no shortcut -- you then need to understand DSP and hard-real-time control systems.  (Scary phrases for actually pretty simple things, look for introductions on the subjects.)

And there's the hardware: I'm not sure if PICs can do it, but ATmegas really can't.  Xmegas I think can, or are borderline; STM32 definitely can.  The keys you're looking for are very short time delays between an external event (such as current rising past a threshold) and a response to it (like a comparator propagation delay resetting a flip-flop).  Or for an average current mode control, you need high enough sample rate.

BTW, by "can't", I mean that, of course you can; you could with the very earliest microprocessors, even; but because the response time is long, the operating frequency must be disappointingly low, in the audible range, and the inductor needs to be massive.  This is determined by the propagation delay and therefore how much current rises above threshold, or the sample rate, or clock frequency, or the amount of time taken for the CPU to compute the next loop -- whichever is worst.

And again, you can synthesize something like a '3842 using a '393 comparator, CD4011 (wired as RS flip-flop), complementary emitter follower (for gate drive), and LM358 and TL431 (for error amp and reference).  It's a good exercise!

For actual chip recommendations, I would suggest UC3842, or more particular, 3843.  The family has options for duty cycle (useful for forward converters) and supply voltage (the -42 and 44 require 16V to start up, while the 43 and 45 work at 6V).  These are kind of annoying for low voltage operation, because the current sense comparator has a 0-1V threshold.  That means you're losing 10% of your 10V supply, in the worst case.  Still, it's a good introduction, and a 70 to 80% efficient converter is still better than a linear one, or none at all (since you don't have the supply voltage to run the LEDs in the first place!).

Once you've built that, you can consider tweaks: you can improve efficiency by replacing the shunt resistor with a current transformer.  That looks like this:
https://www.seventransistorlabs.com/Images/Mag_Amp_PSU.png
Note how everything is pretty similar to the standard application (and recognizable as a boost converter), and note that there's a current transformer reading drain current, and passing that (divided by turns ratio) to the current sense pin.  The 2.2 ohm burden resistor dissipates a fraction of the power (that fraction being the turns ratio) a shunt resistor would.

Tim

[mikeinfodoc]

Right, I was just looking at the TPS61040 which seems to have max Vin of 6volts. (so that one would not work)

http://www.ti.com/product/TPS61040

Now, looking at an MC34063A calculator.
http://www.nomad.ee/micros/mc34063a/

Is the attached schematic realistic? 

if so, is it really THIS easy?

[mikeinfodoc]

Any thoughts about the LM2588?

See attached

[T3sl4co1l]

Whazzat, one o'dem Simple Switchers?  Those are perfectly fine too, not much better than MC34063 really, but definitely better, and way simpler, and fairly priced.

Tim