DeWalt XR Brushless Design Secrets
Vložit
- čas přidán 24. 07. 2024
- Teardown and study of an "XR" class cordless tool from Dewalt. The motor driver, hall effect sensor and micro controller are torn down to the silicon die level. A look at the engineering that drives the impressive performance of these tools!
More details at my blog: electronupdate.blogspot.com/2...
Correction on video: Power transistors are FETs.
good source of data on Hall effect sensors: phys.libretexts.org/Bookshelv... - Věda a technologie
Woah! This electric handtool has more computing power, than my fist 3 computers together… 😮 I like your analysis and your style - thank you!
The three phase inverter module seems to be very similar to this one: Fairchild FSB44104A
Good to see new content. Hope more of your electronic bits fail!
oh, I hope not…. that was an expensive bit of kit to replace on short notice 😀
Love your videos. Good solid deep dive reverse engineering. Need more channels like yours. I can't get enough! Thanks!
Awesome video! Speciality chips on boards from many of these manufacturers these days for sure!
Outstanding!
Why did they use IGBT and not MOSFET?
I thought IGBT is used in high voltages, where MOSFET can't be used.
Yeah its quite odd, the lowest volt IGBT I saw from onsemi are 300V, high for a 20V tool. The Vce loss on those would be massive at 2V+. You can find a similar module with FETs, FTCO3V455A1 (rated 40V, so this module is either 40V or 60V I guess), though it doesnt have the ceramic block.
Also, the trenches he shows in the video look similar to some vertical MOS die photos I see online.
But, I don't have anywhere near the knowledge of electronupdate.
IGBT's are standard drivers for 3PH motor drives (BLDC and VFD) due to their extremely high power capability, at very high speeds, with very low gate drive currents, and they do it all very efficiently.
Could be MOSFET, as the massive gate array is indicative of it, and for the current of a 20V tool an IGBT power loss is very high, as the saturation voltage is high, but a MOSFET can approach a wire resistance wise. For high voltage the IGBT wins for the resiliance of the integrated PNP transistor over the MOSFET, though the drive is a small FET that only has to handle the base saturation current, so for a 20A load probably only 5A, easy enough to do, while the rugged transistor does the work. Also the turn off is a lot less of a problem, as the miller capacitance means there is a lot less charge induced into the gate you have to remove.
@@StreuB1 Yes, standard for 3ph motors that run at 240V+, not
Yes that is almost certainly not IGBT, probably MOSFETS.
Nice details.
Thanks!
keep going made video sirr i love this.from Indonesia always waiting you updated 🙏👍
I wonder if their electronics are shared over their product range. Would make sense, as they essentially are very similar in functionality.
Very likely, same basic functions, only you need to very a few parameters like current and ramp rates, plus a desired speed at various points of the range. Would say that probably half of that flash is used to store look up tables, as that is a lot easier to do than a multiply, and for a motor driver quicker, as you only need one table and simply use an offset to get each phase drive out of it. They likely use the PWM on the drivers to ramp the current up per coil per cycle, using the look up table to get it, as that is probably non linear, so instead of having to do motor coefficients and multiply to get the right one they simply took a few, did a sweep, and got an average for the lot, and put in the table. motor driver the top of the die has both a boost circuit and a push pull driver, to enable them to use the same type of IGBT drivers, fits in with having the 3 diode blocks on the ceramic, there as both overvoltage clamps and also to provide this boost voltage. Probably part of the trench drivers is actually a capacitor, in 3D to get more capacitance per unit area, exactly like the trench MOSFETS do to get more gate area.
These motors rock. They can break your arm.
I don't think the hall effects sensor would be based othe ancient LM 108
Op amp. These are very common and probably have special chip design for this purpose. Also in order to get reasonable hal voltage the hal element must be made of semiconductor, not metal PCB-traces.
7:22 when you said read out the data, do you mean visually? Because I thought that flash memory couldn't be observed like that.
electronically…. looks like one could drop microprobes down.
Editing note : volume is super low on the video
yep
The audio is terrible! At first I thought it was a bot, but tbem I realised it was just some really strange distortion.
Others reckon ir has low volume, but I didn't feel that... onlt sone very strange modulation or 3D encoded audio or something? 😮
it was recorded by the multi tool itself