I'm reviewing an evaluation board for the TPS54A20 DC/DC converter from TI. This switcher is specific for low voltage designs. The output range is 0.5 - 2 V. That's a very narrow range. In that range it can deliver 10 A, with a typical input of 12 V.
Let's dig a little deeper into the "Two-phase, Synchronous Series Capacitor Buck Converter" design. |
Series Capacitor Step-Down Converter
The Buck converter uses a series capacitor to help stepping down the input voltage.
In a previous blog, I asked if anyone could explain the operation of the circuit better than me. jc2048 did that in a comment to that blog:
When Q1a is on, current flows through the capacitor and coil La from Vin - charge accumulates on the capacitor, the coil establishes a magnetic field, and the load is powered. When Q1a turns off, and Q2a turns on, the coil La keeps the current to the load going, gradually depleting its magnetic field (because that's what coils do). At the same time, Q1b turns on and the energy in the capacitor (notice that the negative of the capacitor is now connected to ground by Q2a) is used as the power source for the other phase, with Lb establishing its field and driving current into the load. After a time, Q1b is turned off and Q2b is turned on, and Lb keeps the current going by retrieving energy from its field.
The steady state average (there has to be some ripple) voltage on the cap is half the supply because during one phase it hangs from the supply and for the other it sits on ground. If the supply voltage to the coils is to be the same for both phases, that can only happen if the capacitor voltage splits the supply in two (the voltage driving La is the supply minus the capacitor voltage, the voltage driving Lb is the capacitor voltage).
The measurements are fully in line with that explanation. On the scope capture below, you can see that:
This capture is taken in steady state, input Vin = 12V, with the probes 1 and 2 of the scope over the series capacitor Ct.
I have to measure the voltage over Ct that way because neither of the cap's connections is at ground level and I don't have a differential probe. So I measure voltage at each side, and will use my scope's Math function to calculate and show the difference between these channels. That difference is the effective voltage over that cap. |
You can see that the left side of the cap (yellow trace) swings between 6V and 12V (Vin/2 and Vin).
The right side (blue trace) between 0V and 6V (Gnd and Vin/2).
The effective signal over Ct is the purple trace, at just above 6V.
The white horizontal line on the trace, at 6V, is a cursor. I put it just below the purple trace (to avoid hiding it).
So the purple trace is just a little above 6V.
When zooming out on the result (trace above), I measured 6.04V.
The application settings used:
Vin: 12V
Vout: 1.2V
Load: 3Ω
Switching frequency per phase: 2MHz
This measurement is with a load way below the reasonable load point of the circuit. It starts to get interesting at the 80% efficiency point, where the load draws 3A.
(I'm on the red line, with Vin == 12V).
I'll need a load that is 1.2V / 3A = 0.4Ω for that. It has to be able to dissipate 1.2V * 3A = 3.6W.
At that point, the traces over the series capacitor will also become more interesting, because we'll get real pull towards ground and Vin, instead of the ripple we get at the current low load.
With the 400mA load that I'm using, my efficiency measurements matches relatively well with the graph (my intention wasn't to measure efficiency- I did the calculation just for fun) .
I draw 0.4A at the output, and the source delivers 0.07A.
| U (V) | I (A) | P (W) | Efficiency (%) | |
| in | 12 | 0,07 | 0,84 | |
| out | 1,2 | 0,4 | 0,48 | 57,14% |
As bye bye for this blog, a trace taken with Vin = 10V, but with the signals spread vertically on the scope to see each one properly.
| Related Blog |
|---|
| Low Voltage Step-Down Converter TPS54A20 - First Check |
| Low Voltage Step-Down Converter TPS54A20 - Series Capacitor |
Top Comments