Hosted by David L. Jones
And presented in Dave's unique non-scripted overly enthusiastic style!
-
EEVblog #90 – Linear and LDO regulators and Switch Mode Power Supply Tutorial
Posted on May 29th, 2010 46 commentsJust how different are linear, LDO and switching buck voltage regulators? You might be surprised!
It’s tutorial time.NOTE: There are many different methods of SMPS control, this is just one of the simple ones, go investigate!
EEVblog #90 - Linear and LDO regulators and Switch Mode Power Supply Tutorial [ 46.44 MB ] Download (3249)46 responses to “EEVblog #90 – Linear and LDO regulators and Switch Mode Power Supply Tutorial”
-
Hey Dave,
Great tutorial, liked it alot. I’ve got a question though, regarding output current from these kinds of regulators.
Does any of the regulators allow you to boost the current of your supply when dropping the voltage, kind of like what a transformer will do to a an alternating current?
-
@ Lasse
Yes, a switchmode will do that, but a linear will not.
But as always:
Power OUT = Power IN – LOSSES.
-
-
Hi Dave, thaks for the information.
I’m only a beginner, and I was just ordering a switch mode bench power suply, at the same time as I saw your video (what a coincidence).
http://www.elektor.com/magazines/2010/april/unilab.1285365.lynkx
This is a magazine with an article about a adjustable bench power suply, and thay sell the whole kit (without transformer) and display, just to learn more about the technique.
Thank you.
-
OH Dave, I wish I could get you addicted to the RC hobby lifestyle so you could start Video blogging about the electronics we see in the hobby.
Lots of fun stuff,
Voltage Regulators (staying on topic)
3 phase power supplies for Brushless DC motors
A range of RF toys
and More battery chargers than you have screwdrivers!
For instance, $105us + shipping gets you a flying wing that will do 95km/h with a 4 channel 2.4ghz transmitter – receiver, a 2.1Ah 11.1 Lipo battery that can sustain 40amp output, a 50 watt battery charger, and a ready to fly airframe with 2 control servos, a 3 phase power inverter with integrated 6v power supply for the on board electronics, and a brushless DC 3 phase power plant.
http://hobbycity.com/hobbyking/store/uh_viewItem.asp?idProduct=9742
http://hobbycity.com/hobbyking/store/uh_viewItem.asp?idProduct=8932
http://hobbycity.com/hobbyking/store/uh_viewItem.asp?idProduct=8338
http://hobbycity.com/hobbyking/store/uh_viewItem.asp?idProduct=11060
-
Oh, yes please! Explanation of BLDC motor control with sensorless feedback would probably take two sessions. The people into UAV hover platforms are tending to build them for the reason that standard RC ESC controllers just don’t react fast enough.
-
BAbbott May 30th, 2010 at 02:16
Dave is your garage/workshop air conditioned? This might be a tad naive but isn’t it intolerably hot in there? Here in the states I can’t stand the heat in my garage and I live in a slightly northern state.
-
@Babbott
No, the garage is not air conditioned.
Yes, it gets hot, and cold. Almost winter at the moment hence the jumper is now on. In summer you might see me sweating, and the camera sometimes overheats and switch off!
Not a problem, we Aussies just deal with it.
It’s not like it’s a tin shed. Terracotta roof tiles, insulation in the roof, and brick walls help the thermal inertia. But yeah, it can get over 40degC in summer occasionally.
-
-
Dave,
You’re a god! That was great! Keep doing these great tutorials and I also love your equipment reviews too. I learn so much more from your blog than I ever learned getting my BS at a University.
Todd -
Hi Dave,
Was wondering if you would ever do a blog on prototyping with BGAs and other pain in the ass packages. As more and more chips end up in these packages.
I noticed you have a xilinx BGA chip in your title graphic.
Did you ever attempt to solder this by hand? How did you go about prototyping with it?Great blog by the way.
-
I get the FPGA’s professionally loaded. Wouldn’t bother to do it myself unless desperate, too much mucking around and trial and error. It can be done and is supposedly not all that hard yourself once you perfect a technique.
The best answer is to avoid BGA’s at all costs!
-
-
Napalm May 30th, 2010 at 16:17
Nice tutorial Dave. As always to the point and I definitely learned allot from this. I would love to see more tutorials.
Thanks for the great blog.
-
Nice tutorial. I had a fear of switching mode regulators until I actually designed and built one. In the process of doing this, I came to exactly the same conclusions you did. I have not been afraid of using a switching regulators since.
-
Grapsus May 30th, 2010 at 22:49
Very interesting! The switching mode power supplies are failing all the time in the devices we use. For PCs it’s cheap and easy to replace, but for other devices, they never use the same form factor so the only way is to repair them at component level. Great tutorial!
-
Great intro to SMPS. This is one of your best blogs. Every project needs a power supply so it’s a valuable topic.
-
sneakypoo May 31st, 2010 at 01:33
I noticed you didn’t mention prices at all. From my limited research it seems like as soon as you step up to the switch mode regulators you also step up in price considerably. That might be another reason some of us hobbyists shy away from them.
-
Beauty… Awesome blog…
I know from different forums that battery driven microcontroller are usually powerd via switched power supplys – because linear ones are to inefficient.
Now I know why…
These are the kind of information hobby electronics like me need
Bit of theory, bit of practical usage, pros and cons…
Great -
Zaphod May 31st, 2010 at 21:32
This was a great episode. Thanks a lot. I think I will give the SMPS a try now!
-
Bearman May 31st, 2010 at 22:04
Hey Dave. Great site. Keep it up.
Speaking of power in equals power out… I have run accross sites on Youtube with folks trying to make “FREE POWER” devices. They are using what amounts to a perpetual motion device using motors and electronics to build what they call Bedini devices. What a hoot. All these poor folks out there are building all kinds of hand made motors and coils and magnetic who sa whats its to try to make this work. It cracks me up the way they flail away at tweeking these gadgets with obviously no scientific understanding of what they are doing
I bring this up because I would love to hear you explain how this is all, in your words “bullshit”. You have such an eloquent way of stating this in your past blogs on various topics. It really feel bad for these folks expending so much energy (no pun intended) in trying to make this crap work. I applaud their drive but they need to see the light. They need to see that they are not measuring power in and power out properly and have no clue as to all the loses in their systems via electrical resistance, heat and friction. Especially when they aren’t using prescision bearings and alligator clipped wiring and other poor electrical and mechanical practices.
Maybe they would be freed up to spend time and their drive towards realistic endeavors.
It would definitely be a good topic on proper engineering techniques and understanding the hidden obstacles of design.
View a few of these and you will get a kick out of them I’m sure.
What do you think?
Bearman
-
Kavee May 31st, 2010 at 23:55
I am enjoying the information presented, as much as the presentation style itself. Humm, I remember Steve (croc hunter), whom had the same enthusiastic verbal style, quite entertaining and informative. Keep doing the good work, cheers
-
@ Kavee
It’s funny that my NIDA coach is trying to beat that natural enthusiastic “Steve Irwin” style out of me!
-
-
Chris June 1st, 2010 at 10:57
Hey Dave,
Fantastic presentation, thank you. I’ve got a newbie question related to regulators:
With respect to the reference voltage used as one of the inputs to the op amp portion, how is that created/determined? It sounds like a catch-22 problem: in order to make a 7805 5v voltage regulator, you already need a regulated reference voltage.
Thanks!
-
@Chris
Look up bandgap voltage reference. They are easy to make on chips because the fixed voltage is a fundamental property of the silicon used.
-
-
Hey Dave. I love your blog. I’m thinking of switching to Foster’s in homage.
In all seriousness, keep up the good work.
-
Dave,
You should perhaps have mentioned that a big portion of switcher performance (wrt output noise) is layout related?
Nial.
-
@Nial
Yes. This was not mean as a proper SMPS tutorial, it was just to show the differences in operation between linear, LDO and switchmode.-
Jason June 3rd, 2010 at 14:50
I was hoping you would cover a bit more into the safe use of SMPS in noise-sensitive applications. One of the key reasons I’ve stayed away from these kind of regulators is that they seem to be quite noisy.
In fact at the moment I am designing a power supply for a UAV, and I am hesitant to use SMPS ICs because in testing they give off a lot of RF interference which will play havoc with any kind of wireless Command/Control communications. However, if I go with linear regulators I’m going to need significantly higher capacity batteries due to the terrible efficiency (in comparison with SMPS). If I go with bigger batteries I’ll have less weight capacity for a payload, or have to increase motor size, needing bigger batteries yadda yadda yadda.
Have you considered doing a tutorial on the EMC implications of switching regulators, and designing a board to try and reduce the amount of noise on your circuits?
-
Jason,
How did you test your design, and how good was your component layout?
I’ve had very good sucess with SMPSs but normally get my layouts reviewed by an FAE who knows what he’s talking about.
Typical advice is…
Forget the schematic. For a buck regulator place the input cap, output cap and big diode so their GND’s are as tightly together as possible. These are connected with a copper pour and connected to the GND plane with a couple of vias. Then place the rest of the componts round these. If the switcher device is sensibly designed you can get this tight up against the input cap and input to the inductor.The output of the switcher should flow through the pad of the output cap and the feedback for the sense resistors should be past this point. The sense resistors should be placed close to the switcher device with a long track back from the sense point if required.
The switching node (the switch out of the device, diode cathode and input to the inductor) should be a copper pour. A cut out in any planes under this node will reduce noise coupled to the planes.
This is pretty generic but it’s a flavour of the approach required for good results.
Nial.
-
-
-
-
As an electronics newbie, I’ve found SMPS problematic – not the theory, but the component-selection side!
I wanted a straight-forward step-down for some 5V logic, so I looked around and decided the LM2574 ‘Simple switcher’ would be just that. Choosing the inductor then became more painful…
The data sheet’s inductor selection guide recommended a 470uH inductor for my load. The data sheet recommended specific inductors, and I went with that as I have no idea how to select an inductor otherwise. The available selection vary massively in resonant frequency and other characteristics beyond the basic inductance, and I have no idea how important they are!
The recommended inductor was a) giant b) rather expensive. So I’ve built myself a simple test SMPS as proof-of-concept, but it’s really not anything I’d ever want to use for a real project.
Any hints on selecting components for small, cheap SMPS?
-
sgf,
Have a look at Nat Semi’s webench, it’s pretty good for characterising designs, component selections etc.
If you plug in your input voltages, output voltage and required current it’ll do the design for you and propose components. You can select alternatives and it’ll re-characterise the results after the changes.
Note that good SMPS performance is very dependant on optimal layout. Get some advice here if you’re unsure.
Nial.
-
Ah, excellent!
That site looks a lot better than taking a random walk through the data sheets on a distributor’s website, trying to piece together something.
Not knowing what parts would make decent substitutes was a big hindrance to me, but between the alternatives list and the models/graphs/etc., I think I should be much more confident next time. Thanks.
As for the layout, I’m sufficiently newbie that I’m still mostly using breadboard!
-
-
Have you seen the new LMZ line? They have the inductor integrated. I’m thinking of ordering the demo board
http://www.national.com/store/view_item/index.html?nsid=LMZ12003DEMO
to power an USB hub (>=2 A). The only hassle: you have to replace 3 SMD resistors to get 5 V out…
Or does something speak against this idea?-
The only thing with the LMZs is that they’re still quite expensive ($7 each). You also have to pay attention to the layout/grounding of the converter and the placement two caps for best results!
When the cost comes down a bit they’ll be ideal!
Nial.
-
-
-
Thanks once again! I’ve been mucking around with solar power from 12Vdc to 5Vdc for various microcontroller items and rather than contribute to the heat death of the universe by burning energy in a pass transisitor, I’ve been starting to use switch mode regulators. It takes too long to charge a battery in the Pacific Northwest to only burn it off uselessly.
-
You are correct. Most beginners don’t understand. This tutorial will really help them feel more comfortable and willing to work with switchers.
Thank You,
GuruSantiagoWant to learn more about electronics?
The GuruSantiago can help. Checkout his videos here:
http://www.youtube.com/user/ElectronicsIsFun
And follow him on twitter @ElectronicsFun -
Hey Dave,
I love your videos. I’m a EE student and got a late start into electronics at 16-17 but of course in high school I never knew anyone that was into this stuff as much as I am. Even in university I might know one guy that’s sort of into the electronics side of EE, the rest seem to care more about power transmission so it’s refreshing to see these helpful videos from a guy that’s just as excited about how cool circuits are as I am.
I’m actually working on designing my own power supply and I was wondering if you could point me in the right direction for finding either an adjustable boost converter or an adjustable buck-boost converter
that can take my 24V@6.5A DC power supply and give me ~1-50V, 100W output. The ICs digikey/mouser has to offer (that i can find) seem to be rubbish and I might just make my own if I can’t find one. -
You’ve missed one important detail….
In case of a LDO regulator, the reference voltage[REF] must be applied on inverting(–)
input of the error amplifier because the PNP transistor inverts the phase by 180°, thus the overall feedback should be 180°(negative feedback). Of course, the middle point of the resistive divider should be tied to noninverting(+) input.No offense meant.
Mihai
-
kyndal February 19th, 2012 at 05:36
how do you organize your parts bin?+
keeping anti static in mindtook me all night to find a NE555 and an LM358 in my baggies….
how much should i worry about static
for “normal components” like logic chips, opamps, ne555s etc..for micros and stuff i use proper static protection..
There is a whole blog for ya !
love the blog. keep it up -
Dear friends,
Please tell me Where are LDO and SMPS are used in comparsion manner..?
-
Nathan Dugan April 27th, 2013 at 11:52
Greetings from Carolina! I’m bored to death at work so I decided to browse your site on my iphone during lunch break. I enjoy the knowledge you present here and can’t wait to take a look when I get home. I’m shocked at how quick your blog loaded on my phone .. I’m not even using WIFI, just 3G .. Anyways, good site!Arlington Residential Commercial Roofing, 5902 Willow Branch, Arlington, TX 76017 – (817) 606-8773
Leave a reply
-
Follow the EEVblog
Use coupon code “EEVblogViewer” for 25% off!
Lasse Dalegaard May 29th, 2010 at 23:23