Why is there no on/off switch integrated in a VESC

@Maxid have you seen the group buy @longhairedboy is doing for the new 100a maytech vesc? It has a built in switch like you’re talking about.

As an EE you know why you want the switch to be a separate part, especially if the switch isn’t designed very well and will fail very soon, going to permanent ON mode. It would make more sense to integrate the switch into a BMS.

1 Like

…intergated antoispark from the vesc? Would that work idk just throwing pennies…

You just seem not very constructive. Bashing the idea seems weird - there are ESCs out there already that have this functionality. So it must work somehow. Why can’t the VESC do it? You also never answered the inrush question.

It’s definably possible, proper gate protection and thermal protection is a must have. These are the two main reason why “anti spark” switches blow

1 Like

You mean having charged caps on your inlet and then flipping the switch? Ofcourse things will happen. Sure, caps draw current to preserve charge in an attempt to keep voltage. But its essensially a big ball of energy waiting for something to happen. The AS increases the “turn on time” by slowly turning on the FETs. You cant accomplish this with the powermosfets controlling the motor. You would need separate FETs if you must have it integrated on the same card. On the subject of power electronics especially the driving of the FETs for motor control, its a very critical design aspect, using them for a combined purpose will only give you a headache.

I stll don’t get it. Why would I need FETs to turn on slowly? I don’t need AS protection - this is not what this is about. My question is about having an on/off switch on the VESC directly.

The FETs on the AS turn on slowly, when they’re “off” (Vgs < Vgth) the FETs has high impedance(Mega Ohms). By slowly increasing the Vgs the impedance falls, enabling the current to flow after awhile they’re fully open(mOhms). Simply doing Off->On will produce the in-rush current we’re so desperatly trying to avoid.

If you want a yes or no answer. Yes, it is possible to intergate a functioning switch on the card. But you shouldnt use the Switching FETs alone for that purpose.

If the mosfets wouldn’t turn on slowly capacitors will consume over 500A current and blow the mosfets :wink: that’s is a problem of most anti spark switches big capacitive loads :slight_smile: that’s why you have sparks when you plug in because that large current inrush into vesc capacitors without any limitations…

1 Like

I don’t want to use any FETs. Keep the capacitor connected to the battery for all I care --> no inrush.

@Kug3lis AS is not part of my question/this topic. I know how it works and I would like to not need one anymore.

Then capacitors are useless because the additional tracks switches and etc will destroy their ESR capabilities :wink:

Why do hobby ESCs have this functionality then? :thinking: If the capacitors would be useless I guess the Chinese manufacturers would remove them to save cost - but they don’t. Hobby ESCs also don’t look like they have additional FETs built-in just for the switch. How are Roxxy ESCs able to do it for example?

@linsus I also found this statement from mosfet - High-side RC ESC switch - Electrical Engineering Stack Exchange

This is a fundamentally bad idea. Electronic switches always have loss, and ESCs already put two in series with the motor, which have to be carefully chosen to minimize this. You now want to add a third, and of course circuit topology means you need to do it on the more difficult high side where P channel is weak and N channel means complexity. Instead, look to improving the control signal to the existing FETs. It usually comes from a microcontroller and typically is only enabled by an arming sequence. Replacement firmware is available for common implementations.

I know too little about the electronic details but isn’t he saying that you actually should do it the way i proposed?

Yeah, sort of. I get what hes coming from. With the added FET switch and losses etc, you just have to weight pros and cons. My guess is that even with that solution implemented he’d have a highcurrent on/off switch to still be able to completly separate from the source.

On the VESC the gate driver is the DRV chip. Now, I wont say I’m familiar with how the switching is controlled from the MCU. I try to stay far away from code. But hes proposing to do the turning on as a separate sequence through the MCU. Which will require you to do some software magic.

If you google inrusher limiters they’re pretty much in any application that draws a high current, often as a separate unit. If you’ve spent top bucks on some nice components you’d want to protect them for some pennies extra.

If your really passionate about the subject go to vedders forums and ask, I’m sure you’ll be provided with a comprehensive enough answer.

Regarding the Caps, they need to be as close to the ESC as possible for best effect. One of thier benefits is keeping voltagespikes as smooth as possible during operation. Guess you’d call it decoupling somehwhat.

I have one e-board(HW 4.10) with just a simple mechanical switch, no AS. Works alright. you can even hear the “pop” from the current when turning on, but if/when the FETs finally blows, I’ll only have myself to blame.

It’s hard to argue with person with little electrical engineering background (didn’t wanted to be rude).

Regarding the post you found that’s why I implement N channel high side switch which here at least no one is selling :slight_smile:

Capacitors used and placed close to switching circuit is that because of their low ESR they can provide current spike faster than battery would which makes everything in waveform department much nicer you can find many videos on youtube about esc and capacitors :slight_smile:

1 Like

Not the new one from @Kug3lis it’s 300A continuous and it’s overbuilt for a reason.

because the BMS is a better place for it if you simply can’t stand having a discreet unit.

because the switch will only turn off the ESC and none of the other stuff you’ve plugged into your pack via BEC/SBEC/UBEC unless that’s included with the ESC.

integrated power switches in ESCs are a great solution for minimal LiPo builds, however.

1 Like

He said himself it can do 50A and needs active cooling for higher current.

Edit: checked the thread again - it’s 100A

Lol, I said without radiator at 50A constant mosfet is warm, and yes for higher CONSTANT currents you need active cooling, because if you knew a bit about electricity you would calculate how much energy is dissipated as heat, you think at 100A there will be like simple part and it will do it without any cooling you are really wrong :smiley:

Point me where I said its only 100A? :smiley: I have personally run over 300A using my lipos without any problem just needed to use a big as fan to cool it down :wink: But for esk8 loads which currents are pulses this active cooling is not required if you not going to pull 100A for like 1min non stop

1 Like

Woah - all I said was that you need active cooling for higher currents. That’s what you say in that thread. Not sure what your post is trying to tell me here. If you want to hear that you have the best AS available - you probably do. But I am not going to spend 70+€ for an AS on a second hand Lou board.

higher constant currents, I think