Long battery leads and inductance

I'm building a Bormatec Maja. Like many UAVs it's a pusher with the motor at the back and I want to put the battery right at the front for balance. I Googled for some ready made battery extension leads and didn't find any. But I did find this discussion about the problems caused by longer battery leads.

www.rcgroups.com/forums/showthread.php?t=952523

I'm not an experienced UAV pilot, but I'm a bit surprised that I've not heard of this before now as I would imagine many UAVs have the battery at the front and motor at/near the back.

The Maja fueslage is 1.2m long. That mean's I've got a distance of 80-100 cm (30-40 inches) to cover.

I could bring the ESC forward a bit by extending the leads between the ESC and motor a little, but I can't do much as there is only really room for the ESC at the back. Or can I move it forward into the main fuselage compartment (within 10cm of the APM 2.5)?

I'm expecting to have to install something like this

http://castlecreations.com/products/cc-cap-pack.html

I want to ask how Maja, and other pusher UAV pilots manage this?

The problem really lies in the "coil of wire" like effects (parasitic inductance) on the battery lines and the way Metal Oxide Semiconductor Field Effect Transistor speed control circuits are designed to work. MOSFETs in this application are used as an all-or nothing switch that turns on and off very quickly. The unfortunate thing is that high currents develop strong magnetic fields, and when the current is interrupted suddenly by the MOSFET doing what it's supposed to do, very high transient overvoltages can be generated by the collapsing fields around the stray inductance of the power wiring.

MOSFETs don't like being constantly "kicked in the ass" like this (really referred to as avalanche voltages) at levels higher than their maximum ratings. One way to solve the problem is by giving the sudden voltage spike energy somewhere else to go, and this is what the capacitors are for. Another method is to tie a zener diode between the drain (assuming n-channel) and gate, to literally turn the MOSFET back on for a bit to dissipate the transient (if you don't understand what I just said - you're a more and bigger capacitors person).

The whole notion of a "sinusoidal" MOSFET switching circuit sets off my hogwash detector. If an ESC doesn't need a big capacitor (you can't make it too big electrically), then the latter approach is probably what they're depending on. Biasing MOSFETs in the active region (as in what applying a sinusoidal signal to the gate would do) and asking them to handle a high drain current usually guarantees nearly instant vaporization of the device.

I have a rather large run in my, ahem, model of around 15 feet at the extreme, and found a 300,000 uf capacitor quells the effects of the inductance rather well (yes, that's .3 farads folks). So yes, you can't go too large, and there's the added benefit of vastly increased power supply voltage stability in the opposite direction (less "sagging" due to transient battery impedance).

Replies

Brad Hughey January 24, 2013 at 6:14pm

The problem really lies in the "coil of wire" like effects (parasitic inductance) on the battery lines and the way Metal Oxide Semiconductor Field Effect Transistor speed control circuits are designed to work. MOSFETs in this application are used as an all-or nothing switch that turns on and off very quickly. The unfortunate thing is that high currents develop strong magnetic fields, and when the current is interrupted suddenly by the MOSFET doing what it's supposed to do, very high transient overvoltages can be generated by the collapsing fields around the stray inductance of the power wiring.

MOSFETs don't like being constantly "kicked in the ass" like this (really referred to as avalanche voltages) at levels higher than their maximum ratings. One way to solve the problem is by giving the sudden voltage spike energy somewhere else to go, and this is what the capacitors are for. Another method is to tie a zener diode between the drain (assuming n-channel) and gate, to literally turn the MOSFET back on for a bit to dissipate the transient (if you don't understand what I just said - you're a more and bigger capacitors person).

The whole notion of a "sinusoidal" MOSFET switching circuit sets off my hogwash detector. If an ESC doesn't need a big capacitor (you can't make it too big electrically), then the latter approach is probably what they're depending on. Biasing MOSFETs in the active region (as in what applying a sinusoidal signal to the gate would do) and asking them to handle a high drain current usually guarantees nearly instant vaporization of the device.

I have a rather large run in my, ahem, model of around 15 feet at the extreme, and found a 300,000 uf capacitor quells the effects of the inductance rather well (yes, that's .3 farads folks). So yes, you can't go too large, and there's the added benefit of vastly increased power supply voltage stability in the opposite direction (less "sagging" due to transient battery impedance).
Joshua Ott January 24, 2013 at 2:26pm

I'm familiar with the thread you linked to, and this seems to be a legitimate concern. The cap pack is a reasonable solution (cost, weight, and size). The flexibility in battery placement is worth taking on another component like this, in my opinion.

This reply was deleted.

Discussions

Long battery leads and inductance

Replies

Activity