Other Parts Discussed in Thread: LM5160, TIDA-01418
Hi, We use LM5160 for EV compressor controller's low voltage power supply solution.
The solution is very similar to TI's TIDA-01418 demo file.
For EV compressor application, we have High voltage 400V and low voltage 15V/driver(AUIRS21814) and 3.3V/MCU(F280049C)。
We use one point connection for HV side and low side gnd, and we use isolated probe to test the 3.3V voltage.
There are some issue:
1. 3.3V is not smooth, there is spike for every switch frequency period, however ,if I add a 2.2nf Y-capacitor between 12V's GND and3.3V'GND,there is almost no spike for 3.3V.
2. When compressor working, there is a big sipke in some switching point ,more than 7.0V without Y capacitor, almost 4.9V with 2.2nF Y-cacapcitor.
3. We use a typical -fly-back transformer, with 20uH of primary side,and I meausured the leakage inductance with neary 1uH@1kHz.
However the volatage is not as the turn ratio shows.
I first set the primary voltage to 8V, while secondary side goes to 17.5V and 7.1V, so I have to decrease the primary voltage setting to 7.1V ,get 15.4V and 5.1V output.
So
1. Any suggestion to improve the voltage spike of 3.3V output ? This may damage the MCU since the MCU's max VDDIO is 4.6V.
2. The Y capacitor between primary and secondary ground is important in the LM5160' design? or with good layout, it can be removed?
3. The primary voltage is right,but the sencondary's wrong voltage because of the transformer or low current consumption?(12V 0.05A input,with compressor off,12V 0.065A,with compressor on)
in the TIDA-01418‘s sch, the primary inductance is 25uH,the leakage indutance will be the reason?
Regards
Arrow