Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Buck with reduced radiated emissions

cupoftea

Full Member level 2
Joined
Jun 13, 2021
Messages
124
Helped
4
Reputation
8
Reaction score
8
Trophy points
18
Activity points
787
Hi,
We wish to do a Buck converter for 12v to 5v at 1A.
We will use a Synch Buck with external fets and bootstrap drive (most controllers on the market are synchronous). But we wont fit the low side fet...just leave it as a diode.
We will then increase the value of the upper series gate resistor to reduce dv/dt at turn on and reduce radiated emissions. Roughly what kind of reduction in radiated emissions will we see?
As you know, we can not fit both fets as when upper fet turn on is delayed, we would get shoot through if lower fet was also fitted.

Why doesnt the LTC1624 datasheet show such a series resistor being used?
 
Last edited:

mtwieg

Advanced Member level 5
Joined
Jan 20, 2011
Messages
3,602
Helped
1,270
Reputation
2,546
Reaction score
1,267
Trophy points
1,393
Activity points
27,113
As you know, we can not fit both fets as when upper fet turn on is delayed, we would get shoot through if lower fet was also fitted.
Delaying turn on will not cause shoot through. Delaying turn off can.
Why doesnt the LTC1624 datasheet show such a series resistor being used?
Probably because it's not necessary to pass EMC at this current/voltage level. Is there some reason you need super-low EMC?

Slowing down FET switching will only affect EMC above a certain frequency. For example if you reduce rise/fall time from 20ns to 10ns, then you could expect EMC above 16MHz to decrease by up to 6dB (in practice it will be less). But generally this is a last-resort method of reducing EMC, since it will also degrade efficiency.
chapter02-p3_img0006.ashx
 

cupoftea

Full Member level 2
Joined
Jun 13, 2021
Messages
124
Helped
4
Reputation
8
Reaction score
8
Trophy points
18
Activity points
787
Probably because it's not necessary to pass EMC at this current/voltage level. Is there some reason you need super-low EMC?
Thanks, this is in light of radiated emissions testing between 30MHz and 1GHz
 

dick_freebird

Advanced Member level 5
Joined
Mar 4, 2008
Messages
7,350
Helped
2,137
Reputation
4,279
Reaction score
2,001
Trophy points
1,393
Location
USA
Activity points
58,893
If you don't have a good low side switch, will your
bootstrap pump work properly? You'd be depending
on the inductor to sink the bootstrap current and
slew the HSS source node, not the LSS; at light load
might this make the bootstrap cap charge weakly /
inadequately?
 

mtwieg

Advanced Member level 5
Joined
Jan 20, 2011
Messages
3,602
Helped
1,270
Reputation
2,546
Reaction score
1,267
Trophy points
1,393
Activity points
27,113
If you don't have a good low side switch, will your
bootstrap pump work properly? You'd be depending
on the inductor to sink the bootstrap current and
slew the HSS source node, not the LSS; at light load
might this make the bootstrap cap charge weakly /
inadequately?
Good point, but the LTC1624 has an internal weak low side switch specifically for this reason. With other stepdown controllers this can be a problem though.
Thanks, i see your point.
Also, the following seems agood way of reducing radiated emissions in bucks...

Not a very compelling case for ferrite beads. They show that a simple resistor provides the same effect with negligible differences. In practice you would have to empirically determine the optimum components for such a method.

Also their measured waveforms don't make sense, especially figures 6/7. The disturbance on Vin and Vo actually comes before the switch node voltage. Indicates an issue with shoot through, or poor measurement technique.
 

cupoftea

Full Member level 2
Joined
Jun 13, 2021
Messages
124
Helped
4
Reputation
8
Reaction score
8
Trophy points
18
Activity points
787
....Thanks, and would you agree that its safer with ferrite beads than resistors, because you dont cause so much heating in the fet, because the ferrite bead generally just provides a little bit of ringing control, rather than the heavy damping of the switching node that can come from using a too high value resistor. It seems to me that even a low impedance ferrite bead can make an effect, and do no harm...so with radiated emissions testing being so expensive, it looks like this ferrite bead method is very worthwhile?...because every little improvement is definetely good in radiated emissions.
 

mtwieg

Advanced Member level 5
Joined
Jan 20, 2011
Messages
3,602
Helped
1,270
Reputation
2,546
Reaction score
1,267
Trophy points
1,393
Activity points
27,113
No, I don't agree with any of that, at least not in general. Sure I've used ferrite beads for EMC control (never in the manner specified by that Wurth app note), and I've seen that sometimes it's effective, sometimes not. Sometimes adding a ferrite bead actually drastically increases radiated EMI, and using a small resistor performed far batter. It's always somewhat trial and error.
 

cupoftea

Full Member level 2
Joined
Jun 13, 2021
Messages
124
Helped
4
Reputation
8
Reaction score
8
Trophy points
18
Activity points
787
Thanks, i am always a bit suspicious when they claim its just a very low R at low frequency , but a high R at high frequency...it must have some inductance.....i'd like to see a graph of complex impedance vs frequency, so we can see just how "resistive" they really are, or aren't
 

mtwieg

Advanced Member level 5
Joined
Jan 20, 2011
Messages
3,602
Helped
1,270
Reputation
2,546
Reaction score
1,267
Trophy points
1,393
Activity points
27,113
Most good manufacturers will specify the complex impedance vs frequency. For example, see figure 11 in that app note. Though keep in mind usually the plots are only given at one bias current (usually zero), but impedance can change drastically as the ferrite saturates.
 

LaTeX Commands Quick-Menu:

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top