Saturday, February 8, 2020

Another in process ESC project with Infineon 6EDL04 gate driver.

This is still more under construction than the LC 2, but it's in the pipe as well for the next big PCB batch. Working name ACE-NX.


LFPAK56 power stage. 25x46mm 2 layer board, same dimensions and a lot of structural similarities to the LC 2.

Infineon 6EDL04 gate driver (6EDL04N02PRXUMA1) - Datasheet. Cool, huh? The overcurrent trip will be an interesting feature to hook up and try out sometime later as well, could help to increase the bulletproofness further, but this one doesn't, the idea here is straightforward and simple.

Power supply is usual trusty ST L5150BN LDO for the 5V and AP3012 boost off the 5V rail for the +12V gate drive supply. The SOT-223 linear and then an AP3012 feeding the gate driver is a standard setup to get a stable gate drive rail, a lot of BLHeli_Whatever things with Fortior drivers have something to that effect.

I haven't seen anyone else make use of these Infineon EiceDriver chips yet. I have seen plenty of IR2101s and similar, a few FAN7888 projects on RCgroups, and a ton of Chinese drone stuff with Fortior FD6288. The 6EDL04 caught my attention for being a smallish TSSOP package and having integrated bootstrap diodes.


The Fairchild FAN7888 and Fortior FD6288 are also among the candidates leading up to me picking the Infineon, but both require external diodes, the Fortior is pure Chinesium (I literally cannot find a datasheet that isn't in Chinese) and not the easiest to get shipped quickly and reliably in the US and the Fairchild is a large SOIC package. There was a promising MPS MP6531 found by RCgroups user AlkaM which goes a step further and includes its own buck/boost gate drive supply onboard, but that one has availability issues and is also an exposed-pad package which is alright but a tad annoying.

The Infineon and AP3012 almost package better than the discrete drive setup! Killed quite a few components and traces. If I wind up liking the results of these, discrete drive may be on the way out of my designs.

Pinouts on this thing are bs_nfet except that obviously, the high side drive is NOT inverted like it is with the discrete drivers. A board definition is going to need to be created for that, probably called ace.hex. Perhaps later on it will be prudent to swap some pins around in the board def for better routing between ICs if the 6EDL04 setup works nicely.

This will obviously be able to run some extremely beefy mosfets that have larger gate charges with the added gate drive grunt. PSMN0R9 is what I have in mind.

No comments:

Post a Comment