I would swap the optocoupler with more high frequency ones, so that the waveform is more like a square wave. This reduces losses inside the MOSFET, thus gain up the efficiency quite a bit!
Julian .. Small problem solution - add a 'fine tune' pot in series with the adjuster pot. Also wondering if the 50Hz environment is affecting the coil. I like this series.
I enjoyed this Julian. Going through each components and finding little efficiency gains could be time consuming but ultimately satisfying. I wonder how high you can get the average efficiency? I should organise a sweepstake!
Would like to see an efficiency versus output current chart at fixed input and output voltages. That's what they seem to give in data sheets anyway. Would give you an idea of how it stacks up versus commercial ones.
30% for 11v - will the same numbers hold if it was from 24 to 13? or are we also expecting 30% drop from 24v to 1v? would be interesting to see if it would make sense to have two bucks if you do a large drop... say if you were going from 24v to 1v would it make sense to do 24v to 10v and then down to 1v
@Julian Ilett, If you are ever bored, it would be neat to see you do a similar test on the QD188 to see how efficient it is, especially at the 5v usb outlet since it uses 4s2p setup. Also, I freaking love your vids mate. Always look forward to the next one that comes out.
could this be easily made to cc CV buck converter.? and what would the max output current with 22-28 v in and 12.8v out? and if I made this would it be efficient enough for daily use. in a solar powered lighting system. lights 12 v. solar and batts 24 v. thanks
How did you decide on a "reasonable" switching frequency? Do you actually like that sound CRT TVs used to make, or is it rather meant as a joke to annoy someone else?
My question is if i have a source 10v and i put two stepp down to 5v in series and on the other end again in parallel so i converting 5v to 5v on each steppdown so a have a eficiency of 100%??
Can you please run the same tests at much lower switching frequency, such as 100 Hz? Semiconductor switching losses will be affecting measured efficiency. And what percentage of the losses are also the result of inductor core losses? Your meters will also be introducing measurement errors, because of the non-sinusoidal waveforms. Finally, can you check the currents by using the same meter on input, then output? That would eliminate the differences in readings between two meters, neither of whose errors are exactly linear wrt current levels. Thanks.
I can't see the point of this. Without a capacitor after the inductor, all the measurements from your watt meter and multimeter are dependant on how they treat discontinuous signals viz are we talking true rms averaging or what. I have never seen a SMPS that doesnt fit a small cap to reconstitute a true dc output value.
I want to find out which components affect efficiency. After the capacitor tests, I'll change the Shottky diode for a regular silicon diode to see what effect that has. Then I'll replace the diode with a MOSFET and attempt to make a synchronous rectification buck converter.
Julian is not making a production SMPS here more an experiment to build one from first principles. Besides I bet a lot of Chinese switchers have no qualms omitting that output cap.
it does not increase the current, must be the inductor and freq used. The L used is commonly 150khz on most converters that increase the current when bucking
Julian don't so be happ because your measurement is not correct. This is not efficiency vs. output voltage, but you also measured it with different output power. So please repeat this video with 4 W load on all output voltage.
Or to be fair with a constant load current (say, 1A) otherwise you will be drawing 4A @ 1V wich would be very very inefficient. DC-DC converters are rated at max input or load current, wich can be exceded before reaching max power
Can anyone provide me with the name/link of this device at 3:21 of this video, the title is given below "Aukey PB-T11 30,000mAh Quick Charge Lightning Power Bank" in which he said I've got a quick charge 2.0 compatible test unit.
As the voltage output goes down, the current wants to go up, so you should match the impedance to conserve the power input/output. Think of class D amplifiers.
The Muppet 2 project maybe this MMPT solar panel bucket bust converter with higher then the 12v volt possibilities looks to be something to watch, also because my solar panels are going to 68V open and, there are not many thing out there that goes so high and when I want to charge a 48 Li-ion (13*4.2 = 54.6V float) battery there are almost nothing useful, so in My case it will be very useful, I don't know about you other people ?? What do you think ? cds.linear.com/docs/en/datasheet/8490fa.pdf Best Jørgen Thomsen Denmark DK
You don't really need to waste time attempting exact integer voltages, just note the values and the power ratios and plot what you see. It'll still be a straight(ish) line. Also, whenever I see 0.69, I think "Hmm. That's Ln(2) again. I wonder ..."
You stepped power down with voltage, how bout you try to keep power level constant while stepping voltage down (one step at a time), changing two values at the same time does not paint the same picture...
My poor ears!!!!!! as soon as i saw a sinewave form on the oscilliscope i heard a loud "screeee"ing noise of ultra high frequency and it made me sick lol
This is all total nonsense! Why not put the capacitor on the output to smooth it out into a useful DC voltage and stop messing around with worthless saw tooth signals?
Never in the past human history have folks waited so eagerly for a capacitor!
I would swap the optocoupler with more high frequency ones, so that the waveform is more like a square wave. This reduces losses inside the MOSFET, thus gain up the efficiency quite a bit!
so interesting, I want to see that with the capacitor, can not wait!!
I would most definitely trust the readings on the oscilloscope than the cryobi dmm
But, why not just add the cap and do it in this video.
Because that's enough for today.
Could the bump (similar efficiency reading) in the output be because that is near the natural frequency of the circuit?
Julian .. Small problem solution - add a 'fine tune' pot in series with the adjuster pot. Also wondering if the 50Hz environment is affecting the coil. I like this series.
I enjoyed this Julian. Going through each components and finding little efficiency gains could be time consuming but ultimately satisfying. I wonder how high you can get the average efficiency? I should organise a sweepstake!
Did you ever do the vid with the output cap? I can't see one.
Would like to see an efficiency versus output current chart at fixed input and output voltages. That's what they seem to give in data sheets anyway. Would give you an idea of how it stacks up versus commercial ones.
what would be the efficiency if you were to wrap a wire around the inductor and then short the end together so you basically have a shorted turn.
30% for 11v - will the same numbers hold if it was from 24 to 13? or are we also expecting 30% drop from 24v to 1v? would be interesting to see if it would make sense to have two bucks if you do a large drop... say if you were going from 24v to 1v would it make sense to do 24v to 10v and then down to 1v
@Julian Ilett, If you are ever bored, it would be neat to see you do a similar test on the QD188 to see how efficient it is, especially at the 5v usb outlet since it uses 4s2p setup.
Also, I freaking love your vids mate. Always look forward to the next one that comes out.
* grabs a plate of rice, sits back & enjoys the video *
:)
could this be easily made to cc CV buck converter.? and what would the max output current with 22-28 v in and 12.8v out? and if I made this would it be efficient enough for daily use. in a solar powered lighting system. lights 12 v. solar and batts 24 v. thanks
How did you decide on a "reasonable" switching frequency?
Do you actually like that sound CRT TVs used to make, or is it rather meant as a joke to annoy someone else?
My question is if i have a source 10v and i put two stepp down to 5v in series and on the other end again in parallel so i converting 5v to 5v on each steppdown so a have a eficiency of 100%??
Did the capacitor video ever get posted? I just watched the buck converter series but don't see it... Maybe I'm just blind.
The efficiency of a boost-buck converter will be pretty sad.
We'll wait with abated breath for that episode, and the follow up WITH the capacitor.
Surely a 10-Turn pot would help here? :-)
Can you please run the same tests at much lower switching frequency, such as 100 Hz? Semiconductor switching losses will be affecting measured efficiency. And what percentage of the losses are also the result of inductor core losses? Your meters will also be introducing measurement errors, because of the non-sinusoidal waveforms. Finally, can you check the currents by using the same meter on input, then output? That would eliminate the differences in readings between two meters, neither of whose errors are exactly linear wrt current levels. Thanks.
make a video on how to improve this efficiency!
I can't see the point of this. Without a capacitor after the inductor, all the measurements from your watt meter and multimeter are dependant on how they treat discontinuous signals viz are we talking true rms averaging or what. I have never seen a SMPS that doesnt fit a small cap to reconstitute a true dc output value.
I want to find out which components affect efficiency. After the capacitor tests, I'll change the Shottky diode for a regular silicon diode to see what effect that has. Then I'll replace the diode with a MOSFET and attempt to make a synchronous rectification buck converter.
Julian Ilett HURRY UP!!!!!!
thank you
Julian is not making a production SMPS here more an experiment to build one from first principles. Besides I bet a lot of Chinese switchers have no qualms omitting that output cap.
What is your resting heart rate.
it does not increase the current, must be the inductor and freq used. The L used is commonly 150khz on most converters that increase the current when bucking
It would have been helpful at the start to mention something about the load. If one is needed, if the value is important etc
It is the opinion of some excellent viewers who feel that you have let us all down (a wee bit). Capacitors?
Interesting topic!
Julian don't so be happ because your measurement is not correct. This is not efficiency vs. output voltage, but you also measured it with different output power. So please repeat this video with 4 W load on all output voltage.
Or to be fair with a constant load current (say, 1A) otherwise you will be drawing 4A @ 1V wich would be very very inefficient. DC-DC converters are rated at max input or load current, wich can be exceded before reaching max power
Can anyone provide me with the name/link of this device at 3:21 of this video, the title is given below
"Aukey PB-T11 30,000mAh Quick Charge Lightning Power Bank"
in which he said I've got a quick charge 2.0 compatible test unit.
That device (and another like it) will be appearing again soon.
Julian Ilett Hey Julian can you recommend a RELIABLE 18650 battery brand? Also where I can buy them?
@Julian Thank you. I like your videos.
hell sir!....nicely done
No output capacitor!?
As the voltage output goes down, the current wants to go up, so you should match the impedance to conserve the power input/output.
Think of class D amplifiers.
Thanks Julian!
Would like to see your videos more often :)
And would like to see VOCODER !! :)
I can clearly hear the 15KHz :P
It's been a while since I've heard frequencies that high :/
Bloody hell that frequency gets annoying
Scope is not for measuring precise voltages
Maybe a True RMS Multimeter could help out. Maybe someone could lend you two Gossen Metrahit Energy Multimeters?
Of course the more work the buck converter needs to do, the more energy is wasted. It's thermodynamics.
The Muppet 2 project maybe this MMPT solar panel bucket bust converter with higher then the 12v volt possibilities looks to be something to watch, also because my solar panels are going to 68V open and, there are not many thing out there that goes so high and when I want to charge a 48 Li-ion (13*4.2 = 54.6V float) battery there are almost nothing useful, so in My case it will be very useful,
I don't know about you other people ?? What do you think ?
cds.linear.com/docs/en/datasheet/8490fa.pdf
Best Jørgen Thomsen Denmark DK
You don't really need to waste time attempting exact integer voltages, just note the values and the power ratios and plot what you see. It'll still be a straight(ish) line. Also, whenever I see 0.69, I think "Hmm. That's Ln(2) again. I wonder ..."
You stepped power down with voltage, how bout you try to keep power level constant while stepping voltage down (one step at a time), changing two values at the same time does not paint the same picture...
I spot Krispy Kreme, yum.
My poor ears!!!!!! as soon as i saw a sinewave form on the oscilliscope i heard a loud "screeee"ing noise of ultra high frequency and it made me sick lol
Good afternoon
well, if you only have a simple light bulb on the output you don't even need a buck converter, you can just use PWM
This is all total nonsense! Why not put the capacitor on the output to smooth it out into a useful DC voltage and stop messing around with worthless saw tooth signals?
Nice video, but please plot white line on black background.
Why should he?
What are you trying to do don't understand what you are doing??
Hello random people in the comment section, but no one will notice me anyway :(
Did I hear someone speak? No it was the sound of tumbleweed drifting across my room...
ThePoramaTV
hello. how you are?
Random noise gets random replies...
oi oi wee man.....
Hello