I referenced both "Minimizing Buck-Boost (Inverting) Converter High-Frequency Switching Noise"(https://www.ti.com.cn/cn/lit/pdf/slva219) and "Working With Inverting Buck-Boost Converters"(https://www.ti.com/lit/an/snva856b/snva856b.pdf) articles.
But I don't know how to properly add an RC snubber circuit under Inverting Buck Boost application.
I learned about the location of adding the RC snubber circuit for Boost from article (https://www.ti.com.cn/cn/lit/pdf/zhca057).
I learned about the location of adding the RC snubber circuit for Buck from article (https://www.ti.com.cn/cn/lit/pdf/zhcaa52).
Which of the positions A, B, and C in the above figure is the correct placement position for the RC snubber circuit in the Inverted Buck Boost application?
Very much looking forward to and thank you for your response!
Hello!
We have received your case and the investigation will take some time. Thank you for your patience.
The RC snubber goes into position A.
Your value of RSNUB is what provides the damping, the CSNUB is a DC blocking components so as to not dissipate too much energy for the insertion of RSNUB.
The larger the CSNUB, the more power you will dissipate in CSNUB for a given frequency.
The power dissipated in RSNUCH for a given CSNUB/FSW is:
PRSNUB= CSNUB*VIN^2*FSW
The RC snubber is used to dampen the ringing cause by the parasitic Loop inductance and the capacitance at the switch node. RSNUB is selected as a optimized dampener, equal to SQRT(LPARA/CSWITCH). This is the RSENUCH value to use.
If we assume CSWITCH to be 10’s of pF’s of capacitance, the CSNUB will do well if its ~10X this, so I will say 100’pFs….
Note CSWITCH = COSS of MOSFET and parasitic C of switch node.
Now all you have to do is to select a RSNUB. Typically the parasitic L is low nH, say 1nH .
Lets just assume some numbers based of above assumptions.
SQRT (1nH/50pF) = 4.4ohm = RSNUB
Make CSNUB – 470pF. ~10 x the capacitance of CSWITCH
If switching at say 400kHz.and 12V in 27mW
At 2.2MHx will be = 150mW Approx..
If this does not snub well it typically means our estimation of CSWTICH and LPARA are a little off, usually lowering RSNUB helps this assumes LPARA is more than typical…
Hope this helps.