Is a $1 Power Supply Any Good? + Wave Winners
Вставка
- Опубліковано 15 жов 2024
- 3 quick & easy power tests you can run with an oscilloscope: ripple, turn-on, and turn-off
Sign up now for the Wave giveaway: bit.ly/YTWave2020
Click to subscribe! ► bit.ly/KLabs_sub ◄
Wave 2020 overview: bit.ly/Wave2020...
Helpful Links:
Keysight Bench Facebook page:
/ keysightbench
Keysight RF Facebook page:
/ keysightrf
EEs Talk Tech Electrical Engineering podcast:
www.eestalktec...
/ keysightpodcasts
Twitter: @DanielBogdanoff:
/ danielbogdanoff
I recently came across some cheap boost converters on Amazon, these claim to take 5V in and boost it up from Vin to 25V. I’m a little skeptical, but I was curious how well they actually worked, so join me for a couple quick tests that I like to use to do a quick check of a supply, and we’ll put this thing through its paces.
First off, let’s measure the output ripple. You can do this manually, but I prefer to use the power app on the scope. To set it up, we probe the output and setup our measurement. We can get a better measurement with a better probe, but this standard passive probe will work fine. To set this up manually put your channel into AC coupling mode, zoom in, and make your measurements.
We have a micro USB input, so naturally we’ll use 5V as our input voltage, and our output voltage is clocking in around 20 V.
So, when we make the ripple measurement we see our peak-to-peak and RMS values on our 20V output.
I’m pretty happy with that, it’s just a couple percent. Ripple on a supply rail translates to noise, though, so for audio work or jitter sensitive designs this is probably not a good supply to use.
With my cursors I can see this is about an 350 kHz signal, but to double check it won’t hurt to turn on an FFT. I’ll do FFT, set my start and stop frequencies, and I like to use “max hold” to make things easier to see. Only use this for consistent signals - it’s not a good option for 1-off measurements.
I can see my peak, and a couple harmonics. I also see some 1/f action here, but my gut says this is also likely an artifact of this sawtooth shape. Notice how much this varies from cycle to cycle? That’s going to smear out the power over a range of frequencies, which is what we’re seeing here.
Let’s also take a closer look at the signal - do you see these little spikes here? If I measure the frequency of them with my cursors, I would bet this scope that it’s caused by the boost converter’s switching frequency. It looks like 1 MHz, but don’t be fooled. Let’s grab a few single captures and see what we get. It looks like these spikes happen every 350 kHz. Let’s see what we find if we pull the part number. It looks like there’s a 350 kHz internal PWM signal driven by an oscillator. There’s also a mention of a 1.2 MHz fixed operation mode - which is pretty easy to see on the scope as well.
Data sheets are great, but being able to verify it with the scope gives me a lot more confidence in this converter.
The last note on this measurement is that we’ve done it all with a simply resistive load, this performance will almost certainly change under a dynamic load, and you could test this with an electronic load or SMU.
So, we learned a lot from the ripple. The other two things I like to check are the turn-on and turn-off characteristics.
I’m going to use the scope app to measure these, but you can also do this manually by setting up a single-shot trigger
Let’s walk through the wizard and check our turn-on. The thing to watch out for when a power supply turns on is inrush current. When applying power right, the inductors and capacitors aren’t automatically at steady state, they have to charge up. This can pull a lot of current and do some damage - so it’s worth checking.
In this case, we see some interesting behavior. It boots up to one voltage level, sits there, and then moves up to the final output level. If we check the data sheet for the switching chip, we see that it actually has a built-in soft-start function. And it looks like it works! You’d still want to test this on various loads before moving on.
And last for today, the turn-off test. What you don’t want to happen when you turn a supply off is have weird power spikes or discharges into your system. You can set this up the same way as the turn-on test, but again I’m going to use the app.
And, a simple power off test shows things perform as expected.
Overall, this is a killer little supply and well worth the $1 just to have it on hand if I need something quick.
The power app also does a LOT more than just this, you can do a full suite of switching tests and all kinds of other measurements that are really tricky to make manually.
#PowerSupplyRipple #PowerSupplyTurnOnTest #PowerSupplyTurnOffTest #PowerSupplyTest #SMPSTesting #MeasureSMPS #switchmodepowersupply #oscilloscope #oscilloscopes #electronics #electricalengineering #computerengineering #SignalIntegrity #PowerIntegrity
That noise ain't pretty, but it'll get'er done! What's your go-to power chip?
Check out the resource libraries + Wave giveaway: bit.ly/YTWave2020
Cool... Wait?
XL6009 DC-DC adjustable boost module 5-32V in, 5-55V out with digital display :$3.22can. I plan to use it for first time getting fuller charge on my older e-bike battery this summer. Be nice to win contest to monitor it better.
The good old MC33063 is almost a universal choice. Widely available as various clones from China and easily salvageable from cheap consumer gear. Just don't forget some output protection and an input fuse in case it locks up and stops switching.
"...but that doesn't mean they are any good ... or are they?" Hey! Keysauce, Daniel here
vsause2
I am 50 miles away, and rapidly approaching
I love that keysight has a sales guy that actually knows what he's talking about. Usually large hardware manufacturers don't bother to do that. Very refreshing!
Thanks, I love my job and get to spend most of my time doing stuff like this!
@@KeysightLabs Hire me!
He's actually an EE IIRC. He just ended up in marketing.
@isohyde yup!
@@KeysightLabs wow, that was fast :) thanks for the awesome videos!
Other companies: upload ads on UA-cam
Keysight: what if we made actually good content
Big brain time
Why buy it for $1 when I can design and build it myself for $100 :-)
That's the spirit
My wife does not fully understand this yet but she is learning..
Excellent use of "automagically", sir!
"...things perform as expected, you'd still wanna go and check your currents before plugging this into a sensitive device..."
*Proccedes to plug it into a 3000x scope's usb port* :^)
"Do you impulse buy electronic components?" Oh yes :D
buys laser diode because laser diode...
This spoke to me so much lol
damn these sessions are really well done! It's not a typical corporate style video but its informational and not like an advertisement.
This guy ... these are advertisements done right! A little silly, entertaining, and genuinely informative. Doesn't feel like an ad at all, and now I want that oscilloscope!! Well done!! :-)
Thank you!
One thing I love about these scopes, is how smooth the interface works, its just soothing to watch ...
Have a few of those boost converters on hand. Only downfall was poor assembly on one I bought. Can't beat the price regardless. Great video🤟
I don't know if it's my go to but I like the PAM2320. Cheap, easy to assemble, doesn't need many external components, 3A output, doesn't get too hot, vin=vout, easy to adjust with a mcu dac over its range of 0.6v to 5v if you do some quick maths.
Keysight is doing a great job in educating people here in electronics
Thank you I needed that humor at the end "Your dc pain rail"
Congrats to the winners, keeping my fingers crossed for tomorrow.
Oh my gosh! I just bought some of these on Amazon a few weeks ago, and wanted to measure the ripple on with them, bout I didn’t have a scope, which is why I entered the giveaway. Lol.
Obviously, if I’m buying $1 boost converters, I can’t afford to buy my own scope. Lol
you actually can! just get something as barebones as a DSO138 for 20-30$ and it should allow you to do a lotta basic stuff. And then you can always grab a (less than) 200$ 100Mhz scope one day. 100Mhz is pretty good for most tasks.
Started watching this channel because of the giveaway but currently am enjoying it so much I just watch it for fun. Great content!
Heck, videos like this make me want to test my cheep parts before using them... and want an Oscope.
Give this man an award for becoming a DC power rail at the end.
I would measure the ripple under the normal load that this thing is gonna be used. It would be very different from an unloaded supply
This was all under load
@@KeysightLabs what was the load? The only thing that I saw connected to the output was the scope probe (I assume 1M ohm for X1 probe).
@@amirb715 They showed the triangle wave under 1MΩ and 1kΩ loads. The output voltage didn't show up on the scope but we have to assume it is being tested at 25V because that's what Daniel mentions at the start of his test video. So 25μA and 25mA respectively on the output.
The datasheets in the video were confusing. On one hand, they looked at the FP6288, but on the other hand, they looked at the B6289Y which is the same family but different technology.
The 6288 was only rated to 20V which would be impossible for the 25V test, so it must have been the B6289Y.
I just found your channel, and am checking out some of your videos. I'm happy, so far. I hope to learn how to do a bunch of stuff my dad spent decades doing nearly every day. 👋😊
Hey, guys, I know this is marketing but make no mistake, these are very interesting, illustrative and relevant videos you're making and I appreciate it. Thank you.
I know I'm late to the ball game. I just started watching these daily about a week ago. I JUST realized that you have the calculator watch that I had in high school in the 80's. LOL NICE!!
The retro Casio watches are my jam
Ok, there are a TON of "how to use your scope" videos on youtube, which just shows you functionality examples. THIS video shows a real-world application to use a scope to gather valuable info about a circuit! Great stuff, do more!
Thank you!
Brave of you, to just put some stuff you got online into the usb port of your scope
I can only dream of being able to afford that scope, but it's still cool to see what it can do. I learned a few things from your commentary as well. Subscribed :)
"You can do this manually but I prefer to use the power app on the scope" like my trio cs2100 from 1980 has that option
Psst... He's trying to sell you a new scope 😂
I love those little chips, super useful to have around. Good to know that they are decent units, and you can't beat the price. Great analysis!
Had some bad experience with some of these DC-DC modules. Got a BB-Converter module set to reduce 5V to 3.3V. The module ended up boosting it to 30V before realizing 400ms later that its supposed to give out less than what is on the input.
Ouch! That'll fry everything in a hurry
Great video! Especially when testing components for which the typical buyers won't have the requisite instruments.
Ooo man I never thought of seeing a sales man pitching a scope in such a good way, having a hands on knowledge helps a lot
Thanks! I do have an EE degree, but get to spend a lot of my time putting stuff like this together.
@@KeysightLabs great, im still going trough my electronics engineering degree and enjoying it
Love the awesome background pieces.. apple IIgs, Retro HP scope, oldschool HP VOM
Capable scope with those built in apps. You guys have very nice products. Thanks for the demonstration and white papers.
Wooo, +10 to awesomeness for an Atari in the background. Congratulations to the winners.
- TIP Test for dynamic load conditions.
- TIP Never place a 10, 100nF sub ~470nF without a series resistor 10 or 100 ohm resistor. (seems to be common jump to conclusion measure to suppress noise)
Switching regulators/boost circuits has an interesting behaviour concerning "dynamic load conditions" and its behaviour with time, trade-offs comes down to two things.
- Fast transient response poor regulation (more noise, fast reaction speed).
- OR slower transient response, improved regulation. (less noise, slow reaction speed)
Trainsent speed (fast or slow) is obviously a critical parameter and application specific not saying there's a sliver bullet, good to understand these parameters.
An interesting fact, computer server power supplies are designed for fast transient speeds, at the expensive of high ripple (to prevent CPU brown-outs). That's why there are so many inductors everywhere.
Good tips!
Like that number of modules they're selling, estimates the probability of myself winning any measuring equipment from keysight. See you on next year on keysight wave's giveaway!
You are a cool guy and explain it fast and precise.
Learn new things from ya every day Daniel! Thank you.
Lol...love the test but that's my whole years income just for that DSO scope....I usually use a XL6009 buck/boost converter from china. It will do 3A and I will normally hand wind my own common mode chokes on the output to clean it up, probably have a dozen at any given time..
Been using the "SODIAL(R) 6pcs MT3608 DC 2A Step Up Power Booster Module 2v-24v Boost Converter for Arduino" six for "not much" set. Seems to work. But not tried with anything that might be impacted significantly by noise I have to say.
i modded one of these with some alligator clips onsilicone wires and a 3 digit, 8 segment voltage display for OTG powerr
I really like these kind of instructional videos from Keysight. They contain tons of useful information, straight to the point(s) and are a fun to watch. Keep up the good work!
Thanks! I use your power supply to USB board all the time!
You got my attention with potential free tools. However just watching your channel has taught me a lot. Thanks!
I enjoy your enthusiasm immesurably
One of the winners is Ian White, Dutch people will understand my pain
Holy cow, i just got my converter in the mail, and you do a video on it, sweeeeeeet!
Congratulations everyone! I'm totally not distracted anymore by the Atari, now all I can see is the classic Macintosh! What model is that? It looks a little like a II or an LC II? I like the 30cm PCB ruler!
It's a GS II, woz edition :)
@@KeysightLabs Oh wow! So cool. :) When that came out I was playing with ZX Spectrums.
The DC power rail in the end got me to laugh more than i want to admit.
Congratulations you lucky winners
I want to know more about the Apple ][ gs and Atari in the background. I use my electronics gear almost exclusively to work on old computers and they can have some funky power supplies!
9 to 10Khz audio harmonics? That would sound just lovely as a power supply in a DJ turntable for playing vinyl.... Mention by example, my Stanton ST150 Mk1 has a high pitched oscillation which can be heard. But then again what can I expected for $599.00.
I can not afford an Oszi. I don't remember functions, algebra or analysis. I don't know how to work with an Oszi. But I like to see that somebody knows it. Subbed.
Great Video! I'm not an electrician 'cause I don't know how this stuff works, but it's still an interesting video. 👌
I had a small 3.3v circuit running off a similar buck-boost converter which was powered by an unregulated 18650 battery. For some reason when the battery dropped too far down (below 3v), the output voltage peaked 35v and killed all the attached circuits. I'm not sure why or whether that is standard for those regulators but that day was a sad day.
What you didn't measured is the ripple under small load and short circuit protection. They are usually terrible at this.
I'll have to make a part 2!
the one switching mode controller I always come back to is the humble MC34063 - I know there are much better and more modern versions, but...
Reading through the comments just to hear someone mention the good old MC34063. For the price and ease of use there is none better for simple, low power applications!
Soooo... when are you planning to show off that Nixie-tube? :)
As soon as I get it!
Seems to me a bit of additional filtering would solve your ripple problem. After all, most of these devices just use the chip OEM's test circuit for their design. Hardly optimised.
For sure! Definitely something that can be handled on the implementation side.
i need keysight oscilloscope. they are cool im loving them. but they aren't affordable for me 😑
There should be some seccond hand Keysight gear available for more affordable prices. The brand can be Agilent or HP but it is the same thing.
Wow that's a sophisticated scope.
My favourite boost converter is MIC2288
@Keysight Labs There's gotta be something wrong, I still haven't gotten my "You are a winner" email, is it cause I have a three word name?
I think it's because you're the only psycho
Keysight Labs Are you discriminating against psychos?
@@caddyguy5369 that depends, are you a psycho? I thought @TheOnlyPsycho was the only psycho...
Keysight Labs I'm leaning more towards socio.
@@caddyguy5369 oh thank goodness
@Keysight Labs should have a price for us that come here every day at 12!!!
hey.. how to become a winner of an oscilloscope?
Sign up at bit.ly/YTWave2020, the last drawing is tomorrow morning!
Is the winning process random? Or there is some kind of pattern /algorithm
too many people on this planet... third year participating Keysight Wave event... simply no luck :-((
What's the scoop on the ruler? I've not seen a foot long before and that one looks even longer like maybe a foot and a half.
That's actually two of them next to each other, they're about 6" each
@@KeysightLabs Thanks... I totally didn't notice and was thinking as a width reference like a rackmount case now that I look closer. Nice to see the width on the units are smaller too, didn't know that. Amazing how compact, full featured and then some!
I'm still in the world of advocating like the not crazies down the road from you guys that what the U.S., Canadian and other Diplomats, Staff, Family as well as others (including pets) dealt with at the Cuban and Chinese Embassy areas as well as deal with elsewhere... isn't the flue, fumigants or psychogenic. The issues are known surveillance and assault science/tech from signals that I am amazed how not well disclosed to the mainstream they are. Glad the Sept 1st 2019 CBS News 60 Minutes Reports and those interviewed came out with the truth finally... even though generically like the Associated Press "sonic recording". www.cbsnews.com/news/brain-trauma-suffered-by-u-s-diplomats-abroad-could-be-work-of-hostile-foreign-government-60-minutes-2019-09-01/
Would be nice to be able to go back to school and learn on modern equipment and not whatever I can find at surplus stores or dirt cheap elsewhere. Until then... if ever... fox hunting and RDF dreams to find the sources... along with generic recordings of signals not much better, though more of a frequency range, than the AP disclosed. Thanks again team!
One day I will have a oscilloscope like that. Amazing...😲
My favourite chip is the one that is the least expensive that day on the shenzhen market
Why are there 3 different frequencies when all tutorials say boost converter needs ..1?
A 2% noise at 10khz will be cleaned up by any decoupling caps on your device.
Otherwise a very nice commercial for the keysight scope.
True! If i were to implement this into a design, I'd probably figure out the ripple at my specific load ranges and pick caps accordingly.
@@KeysightLabs sorry I did not notice it was an add from you, I just thought it was a UA-camr checking a power supply. Your scope is very impressive with its touch screen display and performance but maybe a little bet too expensive for me. You have a lot of products reasonably priced for the pro, maybe you should develop some low cost front ends, display less Bluetooth or usb interfaced designed toward the amateurs the market is large and if you amortize the R and D cost on a large market it would be profitable for all of us. I'm thinking about a sub $100 100Mhz display less scope. Can you do it s the question?
@@johnkubik8559 That would be awesome! I've been wanting an open-source type front end for a while
Cool vid... Thanks, Now I just need the scope to make those measurements :-) BTW: Your DC power rail imitation would be improved if you drank a lot of coffee to put some jitter into the "signal"
Good call!
Stop teasing with that ruler man! So sad the supplies were already gone when I tried to get one :/ Thanks for the giveaways and tips though :D
thank mr keysight
Why use a fancy oscilloscope when I can jam it against my tongue and gauge the ripple by the amount of pain?
You do you, dude
It seems they're useful and that's for sure !!
Well, actually not that bad as I initially thought. But of course there's still the old wisdom: You always get what you pay for! ;)
Great impression! :D
AMS1117 is a way to go
Gg for the winners
Great video! Nice impression lol
So what could you do to filter out that noise?
Filtering/bypass capacitors would be the main thing
Everything went above my head.
we can decrease the ripple by using L-C filter or capacitance doubler circuit .
Simply adding a capacitor directly across the output would do wonders. May be all that's needed unless the load is quite high.
@@rich1051414
Yes increasing the capacitance will decreasing the ripple but not so much cause the output already has a big cap.
@@mohammedjawadforelectronic9190 0:57 I am not seeing a massive capacitor. I see a small ceramic on there.
@@rich1051414
That's right its a small one , due to high frequency , but I was mean using LC filter is better than C filter
@@mohammedjawadforelectronic9190 Oh for sure. I was simply stating doing nothing more that throwing a large value electrolytic across the output would immediately massively improve performance. Yeah, a proper LC filter or even a RC filter would be much better, of course
Oscilloscopes are so cool
Nice oscilloscope, I never used any oscilloscope worth more than 1k USD, hopefully u guys can make it cheaper.
Engineer's sense of humor :)
1:00 How does it not create shortcut???
It is very advanced. I love it😍
Every video makes me want to have one, if I don't win anything (which will probably happen haha), would be great to get some discounts to buy one.
Most of them are noisy especially if you use them with a sensitive circuits that needs a stable Power supply. Anyhow, hope to be lucky next time.
A bad DC power rail is like being on magic mushrooms.... very mellow for a little while then... "OMG the spiders, spiders EVERYWHERE!!!", then mellow for another bit, then "OMG why is your nose SOOOO big??!"... mellow... "I can't stop breathing!!" ...mellow... on second thought a DC power rail isn't like magic mushrooms. lol (btw I'm depressed I haven't yet won the scope of my dreams).
Lol, I'll take your word for it
😂😂 so funny
Nice job
If I could have a scope as the one in the video.
I will very appreciate that.
Geat video thanks for posting
If I had a Real Scope I could do that. Hint Hint. LOL I use Boost and Buck converters all the time. They do the job and are CHEAP. Auto Buck Boost are the bomb for a varying voltage source like solar or wind turbines.
Testing the PSU without load will cause discontinous mode and a very high ripple at a low frequency. Loading it with e.g. 20% of rated current should give a much lower ripple output. Even a beautifull scope cannot fix a bad test setup...
This was all tested under a load
@@KeysightLabs I dont think the 1kohm load was enough to get it out of the pulse skipping mode for very light loads. The 9.1kHz frequency is probably a pulse train of 1.2MHz switching. You should see the frequency get faster as you load it more then eventually itll exit pulse skipping mode and youll see a much lower ripple as it continually switches at 1.2 MHz. You should definitely test it at a heavier load, say 1A, and see the performance then! 😊
It's not how well they work that matters but how LONG they work well.
That's a great point, I should throw one in our cycling thermal chamber.
I get feeling he actually enjoys doing what he does
I do!