MC34063 layout advice

[djacobow]

I've made a PCB that has a MC34603 on it stepping down about 40V to 5V. It's my first project with a switching regulator on it. The regulator works fine, as near as I can tell, and I'm happy with the output ripple.

I'm spinning the board for other reasons and thought I'd really like to get the layout for the buck regulator "right". After looking at the advice in the datasheet and around the web, I'm having a hard time taking everyone's advice simultaneously. Keep the input and output current loops short is easy enough, but not every component can be "closest" to the regulator. I'm also unsure about whether there should be a ground flood fill and whether there should be ground plane on the back, and if so, whether they should be stitched up. I think this primarily will effect the current flow in the diode. I also don't have the right gear to see if this is emitting much EMI, so I have to rely on best practices. Also unsure if it's okay to have the ground fill go under the inductor -- though in this layout there is no way to hook up that diode and timing cap without at least a trace going under.

So, the layout is attached. I am showing two views, with the polygon connections and floods and without, so that you can see the component placement more clearly. But the fills are there on the real board.

In the picture. C3, C12, C14 are on the input. C17, C16, and C15 are on the output. D3 is the freewheeling diode. C13 is the timing cap. R24 and R25 are the feedback network. R12,R13,R14 make up the collecor resistance.

D4 and D5 are upstream of the regulator, rectifying an AC input.

I"m looking for advice. Does this seem reasonable?

[mariush]

Honestly, my first suggestion would be to use something better than MC34063, with more protection and with higher efficiency. Second would be to make an effort to get the input voltage under around 36v, making it possible to use a higher variety of switching regulators.

As for the layout, there's lots of things I would change.... here's just how I'd arrange the top at very least:



It's something I made quickly in paint, routes may not be correct (and at least one is missing between C13 and D3 and bottom left pin of the IC) , but you should get the idea.
 

[djacobow]

Honestly, my first suggestion would be to use something better than MC34063, with more protection and with higher efficiency. Second would be to make an effort to get the input voltage under around 36v, making it possible to use a higher variety of switching regulators.

As for the layout, there's lots of things I would change.... here's just how I'd arrange the top at very least:



It's something I made quickly in paint, routes may not be correct (and at least one is missing between C13 and D3 and bottom left pin of the IC) , but you should get the idea.

Thanks. The 40V is something I'm stuck with. 24vac. Those transformers run up to 28Vac unloaded, and that's where 40 comes from. I can string a bunch of diodes to get down to 36, bit that's a lot of diodes.

I will probably switch to a better regulator for another project, but considering that this one seems adequate for now, I'll stick with it for this one.

Your layout has some things I like, but the ground connections on the freewheeling diode and timing capacitor are absent (and completely blocked off by other traces) and so would need to come up from the bottom using vias. Is that ok? I dunno. That's why I'm here!

[mariush]

You have a 24v AC transformer which you say may go up to 28v AC at low current.
When you rectify that, you'll get a DC voltage with a peak of 28v x 1.414 = ~40v , MINUS the voltage drop on the two diodes in the bridge rectifier.  Depending on what bridge rectifier you have, that could amount to ~ 1.5v..2v so your peak will be at around 38v.
Remember this is peak voltage, you'll still need some capacitor to smooth things out ...
Anyway, even if you go with a peak voltage of 38v and you'd add let's say a couple diodes to drop to 36v, it would still be a risky idea to use an IC that accepts maximum 36v
You could add a resistor in front of the bridge rectifier as a sort of minimum load, let's say something dissipating half a watt or so, if the total power consumption is not a problem.

If the current is less than 1.2A, a nice and cheap (~0.65$ if you get 100, ~0.85$ in 10pcs)  I could recommend would be AOZ1282 : http://aosmd.com/res/data_sheets/AOZ1282CI.pdf
Absolute max ratings is 40v, the recommended max is 36v...  Runs at 450kHz so you can use smaller inductors and at the same time it's not so high that you have to be super strict about layout. You only need inductor and freewheeling diode, mosfet is integrated.

Yeah, in the layout I made in paint, i sort of imagined that you'd have all the right side as ground (a _| shape in a sense) and all the bottom layer could be ground as well (and use VIAs to connect the two) and maybe have a sort of star ground on bottom right corner.
Vias would be OK for the feedback resistors, but I'd like a thicker ground trace going to the chip itself. Some traces (like the one between C16 and C15) could go on the other side through vias breaking that bottom ground fill, if it would allow a thick ground trace to reach the IC.

[djacobow]

You have a 24v AC transformer which you say may go up to 28v AC at low current.
When you rectify that, you'll get a DC voltage with a peak of 28v x 1.414 = ~40v , MINUS the voltage drop on the two diodes in the bridge rectifier.  Depending on what bridge rectifier you have, that could amount to ~ 1.5v..2v so your peak will be at around 38v.

There is no bridge rectifier in this circuit because I need to drive triacs which will require that one terminal be at the same potential as "ground." So only a single diode rectifier. To get from 40V to 36V reliably requires 7 diode drops. That's a bunch of diodes. And, furthermore, the efficiency suffers because of the losses in all those diodes. It's nice to have a regulator that can just work from 40.

One part I am considering is switching to the LM2576HV, which can handle up to 45V. There is also a version that is fixed at 5V, so I saw two resistors. And it doesn't need a timing cap. So it is probably easier to use overall and a bit more modern with thermal and short-circuit protection. It is slow like the MC4603, though. 52 kHz, which is on the same order of the 4603.

If the current is less than 1.2A, a nice and cheap (~0.65$ if you get 100, ~0.85$ in 10pcs)  I could recommend would be AOZ1282 : http://aosmd.com/res/data_sheets/AOZ1282CI.pdf
Absolute max ratings is 40v, the recommended max is 36v...  Runs at 450kHz so you can use smaller inductors and at the same time it's not so high that you have to be super strict about layout. You only need inductor and freewheeling diode, mosfet is integrated.

That is appealing. I will check it out. I don't need nearly that much current. 500mA for this app will be fine.

Yeah, in the layout I made in paint, i sort of imagined that you'd have all the right side as ground (a _| shape in a sense) and all the bottom layer could be ground as well (and use VIAs to connect the two) and maybe have a sort of star ground on bottom right corner.
Vias would be OK for the feedback resistors, but I'd like a thicker ground trace going to the chip itself. Some traces (like the one between C16 and C15) could go on the other side through vias breaking that bottom ground fill, if it would allow a thick ground trace to reach the IC.

I'll work on that. Right now the back is a solid ground plane. There is other circuitry around this PSU, so though there is mostly ground flood around, it is interrupted and segmented here or there. There are lots of vias between the solid back and the front, but still...

[T3sl4co1l]

Layout? MC34063?  You're kidding, right?...

If there's ground plane under it, it really doesn't matter.  Without, I'd be a little concerned, but still not very much.  If you aren't seeing voltage spikes on the switching waveforms and input and output voltages, that's really all that matters.

Alternatives:
https://imgur.com/gallery/M1S0DbI

Tim

[djacobow]

Layout? MC34063?  You're kidding, right?...

If there's ground plane under it, it really doesn't matter.  Without, I'd be a little concerned, but still not very much.  If you aren't seeing voltage spikes on the switching waveforms and input and output voltages, that's really all that matters.

Alternatives:
https://imgur.com/gallery/M1S0DbI

Tim

This is my problem. There are so many options! Analysis paralysis. I just want to pick one or three that will be my go-to's basically forever. :-)

There is a solid ground plane on the whole back of the PCB and I'm not seeing voltage spikes. I am seeing ringing, but I'm not sure if this is just a measurement problem from my haphazard scope ground clip being attached to a USB connector several inches away.

[spec]

There is no bridge rectifier in this circuit because I need to drive triacs which will require that one terminal be at the same potential as "ground." So only a single diode rectifier.

You, could use a bridge rectifier if you drove the triacs via isolated couplers. That would keep the hash from the triacs away from your main circuit and allow a better raw power supply with half the ripple voltage and twice the ripple frequency which is easier to filter.