your no noob to voltage dividers. or.... /a subscriber of yours.

OMG your project is awesome wifi rgb led show me in video :)

@1:27 1.21 Giga Watts

3:05 also

I'll have to move to the UK then lol, where I live there are people going mayhem with plutonium guns and Delorians.

+doug holsted Probably, yes. To be safe make sure you measure the unloaded peak voltage of your rectified DC and make sure that's within the range of the buck converter's max input.

I imagine yes, but you need a feedback IC that could control the 555 timer

Variable boost converter video is a Patreon milestone and will likely come soon. It's impossible to filter out ALL ripple but after a certain point it becomes so small you won't even care. Adding more capacitance will help but adding additional LC filtering is a more cost effective method. You could also add a linear regulator stage.

Don't do that. If you do this you could easily end up shorting the feedback pin to ground or to your high voltage output.

Ah, but of course. I didn't think of that. Thank's for the clarification.

Personally, I would use a fixed voltage divider on the output, and then feed its output to a comparator input; the other input would go to the wiper of a potentiometer; its ends would then be connected to a TL431 or similar voltage reference. The comparator output could then be fed into a pass transistor for a linear power supply (I tested this on a simulator and it works flawlessly) or, I suppose (I could be wrong so don't quote me on that), that it could also go to the feedback pin of a regulator IC such as the LM2678 shown on the video. If, for example, your voltage reference is 2.5V, and you want an output range from 0 to 25 volts, you could use a voltage divider with a 9 to 1 resistor ratio (which would divide the output by 10)

true that

Not really, for small power linear regulators is OK. Another advantage is absence of HF noise caused by switching

MrJohhhnnnyyy plus lm317 is way cheaper than lm2678. So in a lot of projects it'll be more preferable than lm2678

MelectroU Of course, and it's not picky to construct. As you saw in this video, you have to lay out components properly to get good performance. I use switchers too, but sayin' that linear ones is not needed anymore is just... wrong

depends from delta V.

You want what is called a constant current source. You can make a linear one out of an LM317 but you can also make a switch mode one.

Afrotechmods Could that be done with another LM267T-ADJ?

Well ideally you'd want to use a constant current LED driver circuit which would be similar to this buck converter but with current feedback instead of voltage feedback. If you want to hack this design, try using some mosfets or a digital potentiometer to change the effective feedback resistor values to get different voltages.

Afrotechmods hmmmm that makes sense. Thanks I'll look into that.

Afrotechmods hmmmm that makes sense. Thanks I'll look into that.

I was just using my bench power supply

If you think about it, moving the pot to one extreme end or the other will result in a direct connection between the feedback pin and ground or feedback and Vout. You can get away with this in certain situations but I wanted to show a resilient design to my audience.

Ok, thanks. Can you tell, just for curiosoty, why it can be a bad idea to connect this connection directly to Vcc or ground?

szafran Depends on how the feedback circuitry is designed. Could damage it. It's the type of thing you need to read about in the datasheet.

Understood, thanks

I doubt it would hurt to just use a pot for feedback, unless data sheet gives a minimum resistance between output and feedback. This looks like a really interesting chip. wonder if more powerful versions have been since developed.....

Yes! The LM2679 will be very similar.

@@Afrotechmods ok thanks for the answer i will they with the same components or similars thanks and great video

.......the switchode power supply can interfere with itself .... Naw ,, I'm too mature for purile humour. I'll leave it alone .

yes

What should be changed to make 1.5v or even lower?

Or would capacitors do the job of smoothing out the ripples?

Yes, hybrid power supplies like that exist for cleaner outputs.