This thread has been locked.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
Dear sir/madame,
We have some problems with using the LM5155 chip as a SEPIC converter on our PCB. With this converter we'd like to variate the rpm of a DC motor using a 10k potentiometer connected externally to P1 and P2 (see figure). The converter is part of a new product of our company, however the MOSFET used is getting too hot within a minute and during usage the circuit draws more current that the output motor is pulling (in unloaded state). We expect that there is a design flaw, but we do not see the reason behind failure. The requirements of the buck boost is: input voltage: 24v, output voltage, 2-32v, unloaded output current: 0.05A loaded output current 1.1A. De switching frequency chosen is 440kHz. A figure of the circuit is inserted in this message, as is the Bill of Materials for the components used. Did we design something wrong?
Kind regards,
Robin
Hi Robin,
Thanks for reaching out with your question and for using the LM5155.
The schematic appears to be wired correctly for a SEPIC configuration. Is there really only on capacitor on the input and output? Typically there is a bank of capacitors (ceramic + electrolytic) to create a capacitor bank. The ceramics will help reduce high di/dt current loops when doing the layout.
What condition is the MOSFET getting hot? Is the out voltage at 32V?
Thanks,
Garrett
Hi Garrett,
Thanks for your help. Yes there is only one capacitor for the input and one for the output. The capacitance however is relatively high for a ceramic capacitor (the output voltage ripple may be relatively high since a DC motor is controlled with it), would increasing the capacitance help or would increasing the number of capacitors help? For the latter one, why would that be the case, the ESR isn't too high according to www.ti.com/.../snva168e.pdf The MOSFET is getting hot for all output voltages, however much quicker when applying 32v. Even without a motor/load connected to it, it then takes about 5 seconds to reach 50˚C and rapidly increasing further. Would a board layout help in identifying the problem?
Thanks,
Robin
Hi Robin,
Typically having more capacitors of different types will help keep the output capacitors from heating up. Ceramics are good for conducting high RMS ripple currents and electrolytic capacitors are good for bulk capacitance. But this probably won't have a large affect on the MOSFET temperature.
What is the maximum temperature the MOSFET gets to? Also can you send the layout over? It is a possibility that the copper around the MOSFET is not large enough to help keep it cool.
Thanks,
Garrett
Hi Garrett,
The maximum temperature of the MOSFET will be reached eventually, such that it will fail. We have measured 150˚C. We are also thinking that the problem might arise from an in-proper ground plane design (see figure), could this effect be so catastrophic? The lay-out of the buck-boost is shown in a separate figure.
Hi Robin,
I just realized that the MOSFET that was selected is only a 2mm X 2mm package and the thermal resistance is very large. 153C/W. So if 1 watt of power occurs in the MOSFET (switching losses plus conduction losses) then the junction temperature of the MOSFET will be 153C higher than ambient. This will probably damage the device.
I suggest switching to a MOSFET like the CSD18534Q5A. This is a 5mm x 6mm MOSFET with similar parameters and a much lower thermal resistance. (50C/W).
Thanks,
Garrett