Jye Smith (the lead ExpressLRS developer) has done some testing with TVS diodes to see if they provide any benefit when soldered to the ESC LiPo pads along with a low ESR capacitor. In a nutshell a low ESR capacitor on it's own is all that is really required. Adding TVS diodes just adds extra weight (and sometimes pose a challenge where to put them on a quad) with no real benefit. This is also backed up by two reputable FPV community members (Ryan Harrell and Wezley Varty).

Source (Facebook): https://www.facebook.com/groups/FPVLIFE/...6832922178

Jye Smith Wrote:Soooo... I was thinking about adding a TVS diode (or 3) to my ESCs for extra protection.  It would be super easy to solder at least 1 on to the vbat pads along with the cap.  But before going ahead I figured some testing is in order!  I wanted to see these 'spikes' and how well the TVS diode performed.

I did a heap of bench runs simulating motor acceleration and more importantly braking.  Along with simulating prop wash, and finally prop strikes a.k.a crashing.

In the end I couldn't find any benefit to adding a TVS diode after the addition of a cap.

Below is a link to some of my testing and results.  If anyone has data showing the benefits Id love to see it.

https://docs.google.com/presentation/d/1...PbA7gcnB6U

Ryan Harrell Wrote:This is pretty much what I've been saying from the beginning on TVS diodes. They don't hurt, but for our application the cap is actually more effective. Possibly the TVS diodes may make a difference on big 9" and 10" cinelifters? But for 5" class drones the cap by itself removes the spikes effectively enough that the TVS diodes don't make a difference. Unless maybe folks aren't installing their caps in a useful way (cough cough, at the XT60, cough)

Wezley Varty Wrote:To add some more supporting data to this...

Here's some testing I did a few weeks back on the bench with an FC + ESC and a single motor with a 5.1" prop. Pretty much the same results as Jye Smith.

I have TVS diodes on my 9" only because it came with the FC+ESC stack and I had the space on the quad to do so.

Does it make a difference? In some extreme case perhaps.

FETtec for example markets it as a Spike Absorber and not to reduce noise and so they always recommend using it with a suitable low ESR capacitor.

(12-Apr-2022, 04:44 AM)kafie1980 Wrote: FETtec for example markets it as a Spike Absorber and not to reduce noise and so they always recommend using it with a suitable low ESR capacitor.

It seems that FETtec ESCs need them because their ESCs are a pile of crap and if you don't stick on one of their spike absorbers you will have a burnt out carcass of a quad before you've even completed the maiden flight  That is probably the only reason that the Spike Absorber even exists. They should just integrate the TVS diodes into their ESCs and be done with it.

I always put lowESR, even on 30g 2.5". Even a tiny 220uF is better than nothing.

I have a whole pile of <20A AIO ESC in my graveyard drawer.

One big ESC purchase deciding factor for me is if I have the MOSFET in stock.
All the MAMBA, XRotor and JHEMCU ESC I have use same MOSFET, and are easy to repair.

If the ESCs are the things making the noise, how would this prevent that? That's like sitting in a room with an air raid siren and putting sound insulation in the walls to protect your hearing. I find it amusing when FCs put them at the power leads to defend against all that noise generated by the lipo.............