Part 2: Test the analog features of the TPSM84A21 EVM
To be Tested Features of TPSM84A21:
- Output voltage ripple measurements (with and without external capacitors)
- Tests at high output current with external load
- The Vin, Vout and PGOOD start up waveforms
- The Vin, Vout and PGOOD shutdown waveforms
Features of TPSM84A21 IC:
- Ultra-Fast Load Step Response
- Efficiencies up to 88%
- 1% Output Voltage Accuracy
- 4 MHz Switching Frequency
- Synchronizes to an External Clock
- Power Good Output
- Pre-Bias Output Start-Up
- Programmable Under-Voltage Lockout (UVLO)
Output voltage ripple measurements (without external capacitors)
Using the reference below Understanding, Measuring, and Reducing Output Voltage Ripple
Probe Sensitivity – Using 10X vs 1X Probes: When trying to measure signals in the sub ~10mV range, using a 10x probe is not appropriate. The sensitivity of the probe is not high enough to have a clean signal reading. A simple 1x probe should be used for measurements under 10mV.
Measurement Conditions in the oscilloscope: 1x probe settings, Noise Reject ON, LPF ON, No o/p load, BW limit set 20MHz, MEMDepth 70Mpoints, AC Coupling, 1Mohm input impedance. Ch1: Vin Yellow Ch2: Vout Blue, Measurement cable from the EVM to the scope is a RF cable 1x with BNC.
The Vout ripple is also load dependent but here the measurements where done without external load
Data-sheet Preview
Without Filtering Capacitors at the input and output For Vin = 7.969V
Without Filtering Capacitors at the input and output For Vin = 10V
Without Filtering Capacitors at the input and output For Vin = 14V
One can conclude from these results that there is a slow rise in the output ripple from 5 mVpp to 8mVpp. But still it is low in comparison to other fast switching DC/DC converter.
Output voltage ripple measurements (with external capacitors)
Measurement Conditions in the oscilloscope: Noise Reject ON, LPF ON, No o/p load, BW limit set 20MHz, MEMDepth 70Mpoints, AC Coupling, 1Mohm input impedance. Ch1: Vout Yellow After filtering capacitors Ch2: Vout Blue measured before filtering capacitors.
Tests at High output current with external load: Very short cables were use to connect from Rload to the DMM.
The measurements were also verified using shunt resistance of 0.01 ohm at the oscilloscope.
| Vin | Vout Measured | RLoad cement Resistor + Burden resistance of Fluke 179 | Output Load current calculated | Output Load current Measured |
|---|---|---|---|---|
| @8V,10V,14V | 1.2V | 0.5ohm +0.037ohm = 0.537 ohm | 2.234 Amp | 2.263 Amp |
| @8V,10V,14V | 1.1V | 0.5ohm +0.037ohm = 0.537 ohm | 2.0484 Amp | 2.055 Amp |
| @8V,10V,14V | 1V | 0.5ohm +0.037ohm = 0.537 ohm | 1.862 Amp | 1.891 Amp |
| @8V,10V,14V | 0.8V | 0.5ohm +0.037ohm = 0.537 ohm | 1.489 Amp | 1.502 Amp |
| @8V,10V,14V | 0.6V | 0.5ohm +0.037ohm = 0.537 ohm | 1.1173 Amp | 1.128 Amp |
| @8V,10V,14V | 0.500V | 0.5ohm +0.037ohm = 0.537 ohm | 0.9310 Amp | 0.955 Amp |
The device closely delivers the required output current although one can see a decline as the current requirement increases. The measured output current remains stable for Vin voltages from 8V to 14V.
Rload = 0.1 Ohm
| Vin | Vout + Burden voltage 37mV/A | RLoad 21W Resistor + 0.037ohm | Output Load current calculated | Output Load current Measured |
|---|---|---|---|---|
| @8V,10V,14V | 1.2V | 0.1 ohm + 0.037 ohm = 0.137ohm | 8.759 Amp | 8.351 Amp |
| @8V,10V,14V | 1.1V | 0.1 ohm + 0.037 ohm = 0.137ohm | 8.029 Amp | 7.61 Amp |
| @8V,10V,14V | 1.0V | 0.1 ohm + 0.037 ohm = 0.137ohm | 7.299 Amp | 7.1 Amp |
| @8V,10V,14V | 0.8V | 0.1 ohm + 0.037 ohm = 0.137ohm | 5.8394 Amp | 5.584 Amp |
| @8V,10V,14V | 0.6V | 0.1 ohm + 0.037 ohm = 0.137ohm | 4.379 Amp | 4.211 Amp |
| @8V,10V,14V | 0.5V | 0.1 ohm + 0.037 ohm = 0.137ohm | 3.649 Amp | 3.565 Amp |
In both the tests, the resistor became very hot very quickly. I had to turn off the setup after sometime to avoid any melting situation. Even with the 21Watt 0.1 ohm the device turns quite hot. At high current and for low Vout the efficiency of the converter is as mentioned in the datasheet. The measured output current remains very stable for Vin voltages from 8V to 14V.
Another important conclusion can be derived here that the difference between the calculated load current and actual measured load current also increases drastically to around 300 mA when loads start drawing current higher than 7.5Amp.
Although the deviation of Measured current@load from Calculated current value@load for 1Amp to 7Amp was only in the order of 20mA which can be understood as drawn by additional cable resistance.
The Vin, Vout and PGOOD Start up waveforms
Setup: Vin = 10V, Vout = 500mv Iout = 1A, Rload = 0.5ohm
The Vin and Vout rise times are comparable to those in the datahseet. Vin@6ms; Vout@4ms; PGood has a steep slope. One can see here that the PGOOD signal in ON only when the Vout reaches its stable output voltage~99%. Whereas Vout start rising when reached around 7V. And Vout has a faster rise time.
The Vin, Vout and PGOOD shutdown waveforms
Setup: Vin = 10V, Vout = 500mv Iout = 1A, Rload = 0.5ohm
The shutdown waveforms were rather difficult to catch as they should be measured in the 200ms/div window and not in 200us/div(as in the datasheet)
Vout and Pgood have a sharp falling edge but the Vin signal takes around 2 sec to fully get to the base level.
Short Conclusion: The TPSM84A21 is good device with fast output rising voltages. It is able to drive higher currents consistently even with as low as 8Vin and 1.2Vout without any issues. But it falls short in achieving a higher efficiency at high load currents at low input voltage. The voltage output ripple is well under control (less than 20mV) even in high current range which is pretty good.
References:
Power Tips: Measuring Vout Ripple in DC/DC Converters
Power Tips: Simply estimate load transient response
Understanding, Measuring, and Reducing Output Voltage Ripple
Line and Load Transient Testing for Power Supplies
TI SWIFT
Power Module EVM Review: Part 2 Test Setup Description
TI SWIFTPower Module EVM Review: Part 4 Dynamically adjust the output voltage using external DAC
Go Back To: Road Test Review of TPSM84A21 Power Module
Top Comments