MOSFET overvoltage problem

Thread Starter

Barta

Joined Dec 22, 2020
3
I am developing project of MOSFET switch that is going to be used as power switch device in application with 12S li ion battery. To keep power wastes in the minimum level, I have to provide the lowest possible RDSon value. In that application, the maximum voltage of battery will be at the level of 50.4V, so I've decided to use 60V MOSFETs. I really dont want to use 80V ones, as we know, the higher VDS breakdown voltage is - the more RDSon the MOSFET will have. But I am not sure, if this MOSFET will sustain any unpredicted, induced peaks of voltage (there is also BLDC motor on that power rail).

There may be situation, where this device have to get disconnected while full load due to "hard fault error" like overcurrent threshold exceeding.
1. Am I able to specify the value of overvoltage peaks from BLDC motor while this situation?
2. Is the BLDC motor driver protected from that kind of problem - disconnecting power in full load?
3. Am I able to protect this 60V MOSFETs from unpredicted, induced overvoltage breakdown peaks?

Pseude electrical schematic:
1608637822297.png
 

LowQCab

Joined Nov 6, 2012
4,023
It sounds like you are building a Large Drone. (why are you making us guess ???)
You should have "Bulk-Capacitors" EVERYWHERE that you can physically fit one,
then don't worry about Voltage Spikes.
You may also want to add Ceramic Capacitors to the inputs of the ESCs,
(Electronic Speed Controllers), (DC to 3-phase, Brushless Motor Controller),
to absorb RF Hash from the high speed switching FETs in the ESCs.
These Capacitors may also slightly increase performance and efficiency.
DO NOT install any Capacitor in between any ESC and a Motor,
twisted wires and Ferrite Cores around the twisted group of Motor wires are OK.
DO NOT install separate Ferrite Cores on individual Motor wires.
Mount the ESCs as close to the Motors as is practical.
The shorter the wires are between the ESCs and Motors, the less RF Noise will be radiated into the Air.
If you are building a Long-Range Drone, remove any Heat-Shrink-Tubing from the ESCs
and use other means of insulating them from Short-Circuits, they need cooling air-flow.
Both sides of the Board radiate heat.

ALL WIRING, between the ESCs and the PDB (Power Distribution Board),
and the Battery Wires to the PDB, should all be twisted to reduce RFI radiation,
this includes ALL low-voltage control wiring, everything must be twisted.
Adding Ferrite Cores everywhere will also help to quiet things down.
These precautions may also increase your RC (Radio Control), and Video, SNRs (Signal to Noise Ratios),
which may increase their useful distance range.

If you are still concerned, double the FET Voltage Rating, and double the number of FETs.
Better yet, just use a single HUGE FET like this one ........
75V 400A IXFN520N075T2
It's also much easier to mount than 5 smaller FETs, and has Screw-Terminals on top.
The only time the Battery should see more than about ~100 Amps
is during a Full Throttle "Punch-Out", (going straight up),
and this can't happen for more than ~2 or ~3 seconds max.,
or you may loose your Drone permanently.
( you DO have RTH (Return to Home) properly programmed in the "I-Nav" software right ???).

You will also need a "Smoke-Stopper" Light-Bulb, (usually ~7.5W, 120V), connected across the FET,
that will smoothly charge those 12 to 15 Bulk Capacitors that are soldered to everything,
BEFORE the FET is quickly ramped-up to full On, (~15Vgs).
This bulb will warn you of any possible Short-Circuits by
continuing to produce a warm glow after connecting the Battery,
(anything longer than about ~1-second may indicate a problem),
and will reduce Arcing-Damage to your Battery Connectors.
.
.Push Button 2 Flat .png
 

shortbus

Joined Sep 30, 2009
10,045
If you are still concerned, double the FET Voltage Rating, and double the number of FETs.
Better yet, just use a single HUGE FET like this one ........
75V 400A IXFN520N075T2
A man after my own heart. I have preached the Isotop use and am usually shot down. I can't for the life of me figure out why people try to manage many small mosfets together when one single one will be much easier in the circuit. And a plus side of the Isotop is no insulation needed for the heat sink mounting.
 

Thread Starter

Barta

Joined Dec 22, 2020
3
Application with IXFN520N075T2 would provide like 76W heat at 200A, assuming that RDS(on) does not increase in function of temperature. In my application - 5 MOSFETs in parallel provide 1.5W per MOSFET of wastes (also 200A current), so it will be almost 70W less. The efficiency in my application is crucial.
 

shortbus

Joined Sep 30, 2009
10,045
Application with IXFN520N075T2
I'd think that was just a part to show what he meant. There are many, many others with low Rds numbers. You aren't taking into count the problems people have with using multiple mosfets in their motor controls, it's one of the biggest questions people coming here ask about, how to solve the problem of burning one or more of the multiple mosfets in use.
 

LowQCab

Joined Nov 6, 2012
4,023
Maybe he read the Spec Sheet wrong, and his Motor Specs wrong ........
That's 0.0019 Ohms Rds On.
200A X 50V = 10,000 Watts, what planet do you live on ??
Where do you get 200A and 76 Watts of Dissipation from ???
Do you seriously think you will ever pull 200A through a single 10-gauge wire with no losses ???
That's 50 Amps per Motor at 50 Volts, = 2500 Watts of Heat per Motor, are you kidding ????
The Motors will catch on FIRE at that kind of Dissipation.

The Calculations for the Drone I am planning out go like this ........
12S, 50V, 4- 3S 12,000 mAh Batteries in series.
The Batteries are rated at "100C", which equals ~1,200 Peak Amps, but this is nothing but advertising hype.
Yeah, they might put out 1,200A for a few seconds if you short the output leads together.
On with the Specs ......
The ESCs (Electronic Speed Controllers) are by ADP, and are rated for 120A,
and are the smallest ones rated for 50V (12S).
The Motors are by "T-Motor"....... "T-Motor Antigravity 6007 KV160".
Using one of T-Motor's recommended Carbon Fiber Props,
( sized for maximum Hovering efficiency, Hover Time is close to 1- hour ),
and the 100% maximum Power at 50V only draws 15 Amps, or 75 Watts,
while producing 10.6 pounds of Static Thrust.
Times 6 Motors = 90 Amps, and, 63 POUNDS OF THRUST, with an all-up Drone weight of 16 pounds.
That means that I have on-tap almost 4-Gs of vertical acceleration, which is absolutely SICK Performance.
Going with more powerful Motors would just be silly.

So, figure it out ....... Lets call it 100A max for easy Math,
50V / 100A = 0.5 Ohms total Load Resistance,
The Huge FET I suggested has 0.0019 Ohms Rds-On,
0.0019 Ohms X 100A = 0.19V Voltage drop across the FET,
0.19V / .0019 Ohms = 100A, so everything squares-up,
0.19V X 100A = 19 Watts of Heat Dissipation from the FET,
meaning that you can basically run it with NO Heat Sink at all,
since the percentage of time at 100A is likely to be near zero,
( Hovering Current is ~13 Amps Total ),
and for no longer than 2 to 3 seconds worst-case.
( that's 2 to 3 seconds at over ~100 MPH !!!! )

Now, are you trying to tell us that you are building a Drone WITH TWICE AS MUCH POWER ????
( even then, we're talking about ~40W max dissipation, not 68W ??? )
If you're building an all-out Dragster/Hot-Rod Drone,
what do you need with a fancy convenience item like a Battery Switch,
that just adds weight and complexity ???

I guess you just made a mistake in your Math calculations.
 
Last edited:

shortbus

Joined Sep 30, 2009
10,045
Maybe he read the Spec Sheet wrong
That was my first thought too. But since he didn't specify what mosfets he was using I couldn't compare yours and his. I still say a single mosfet is the way to go, if he really wants his project to get off the ground(pun intended).
 

LowQCab

Joined Nov 6, 2012
4,023
BTW,
I just reread this whole thread to make sure that I didn't miss anything.
Barta expressed concerns about Short-Circuit Protection.

Part of this is handled by the ESCs, (if you use the same ESCs that I am going to use).
From the Spec-Sheet ........
---------------------------------------------------------------------------------------------------------------------
Protection Mechanisms,
3 stage bus current limiting,
Over-voltage protection,
Current Sense methodology:
Impedance controlled GND Plane Phase Current limiting,
Over-temp protection.

The ESC will limit current to
60A when below 30% throttle,
120A when between 30-60% throttle, and
200A when over 60% throttle.
The ESC will reduce Regenerative Braking Response when it detects a voltage rise on the bus whilst braking.
Motor phase current is limited to 200Amps.

This applies to ......... "APD F3-120" ESC's from "Advanced Power Drives" Australia.
------------------------------------------------------------------------------------------------------------------------

As far as shorts created by wires being damaged in a Crash .......
The Batteries should be mounted in a manner that
causes them to be ejected from the Drone in the case of a Crash.
This will also instantly disconnect all power to the Drone's wiring upon impact.
This has the bonus of limiting damage to the Drone caused by the inertia of the weight of the Batteries.
( the Batteries may be as much as ~50% of the total weight of the Drone )

There is also a need for careful lay-out/routing of wires to insure that they are kept out of "harms-way",
.
Protective coverings on wires were they may be subjected to sharp edges, or crushing of the insulation,
.
Physical barrier protections provided for the Power Distribution Board (PDB) and its soldered wiring connections,
.
Insulation ( Spaghetti ) sleeves on all added-on Bulk Capacitors so that their leads will not be exposed,
.
Solid physical mounting/protection for all added bulk Capacitors to prevent the Capacitor's leads from
being damaged, or physically separated from, the body of each Bulk Capacitor,
.
Adequate "slack" or "loops" in all wiring runs to prevent
the application of any stress, stretching, or excessive bending, to the wires,
under any condition, including possible structural failures in the framework of the Drone.
.
In addition to the above precautions,
all wiring, and all individual electrical components should be rated for at least
twice their expected operating parameters, in both Voltage and Current,
with particular attention paid to adequate Heat Dissipation.

And finally,
use only the Motor Manufacturer's recommended Propellers matched to the particular Motors
that you have selected for your project, to avoid possible Over-Heating of the Motors.
A Motor with Props that are too large in Diameter, or that have too much Pitch-Angle for its rated output,
can cause Insulation Failure of the Motor Windings,
which could result in a FIRE, and/or, Short-Circuiting of the Battery, which could also cause a FIRE.
.
.
FET Symbols 2 .jpg
 
Maybe he read the Spec Sheet wrong, and his Motor Specs wrong ........
That's 0.0019 Ohms Rds On.
200A X 50V = 10,000 Watts, what planet do you live on ??
Where do you get 200A and 76 Watts of Dissipation from ???
Do you seriously think you will ever pull 200A through a single 10-gauge wire with no losses ???
That's 50 Amps per Motor at 50 Volts, = 2500 Watts of Heat per Motor, are you kidding ????
The Motors will catch on FIRE at that kind of Dissipation.

The Calculations for the Drone I am planning out go like this ........
12S, 50V, 4- 3S 12,000 mAh Batteries in series.
The Batteries are rated at "100C", which equals ~1,200 Peak Amps, but this is nothing but advertising hype.
Yeah, they might put out 1,200A for a few seconds if you short the output leads together.
On with the Specs ......
The ESCs (Electronic Speed Controllers) are by ADP, and are rated for 120A,
and are the smallest ones rated for 50V (12S).
The Motors are by "T-Motor"....... "T-Motor Antigravity 6007 KV160".
Using one of T-Motor's recommended Carbon Fiber Props,
( sized for maximum Hovering efficiency, Hover Time is close to 1- hour ),
and the 100% maximum Power at 50V only draws 15 Amps, or 75 Watts,
while producing 10.6 pounds of Static Thrust.
Times 6 Motors = 90 Amps, and, 63 POUNDS OF THRUST, with an all-up Drone weight of 16 pounds.
That means that I have on-tap almost 4-Gs of vertical acceleration, which is absolutely SICK Performance.
Going with more powerful Motors would just be silly.

So, figure it out ....... Lets call it 100A max for easy Math,
50V / 100A = 0.5 Ohms total Load Resistance,
The Huge FET I suggested has 0.0019 Ohms Rds-On,
0.0019 Ohms X 100A = 0.19V Voltage drop across the FET,
0.19V / .0019 Ohms = 100A, so everything squares-up,
0.19V X 100A = 19 Watts of Heat Dissipation from the FET,
meaning that you can basically run it with NO Heat Sink at all,
since the percentage of time at 100A is likely to be near zero,
( Hovering Current is ~13 Amps Total ),
and for no longer than 2 to 3 seconds worst-case.
( that's 2 to 3 seconds at over ~100 MPH !!!! )

Now, are you trying to tell us that you are building a Drone WITH TWICE AS MUCH POWER ????
( even then, we're talking about ~40W max dissipation, not 68W ??? )
If you're building an all-out Dragster/Hot-Rod Drone,
what do you need with a fancy convenience item like a Battery Switch,
that just adds weight and complexity ???

I guess you just made a mistake in your Math calculations.
Thank You for your time and contributions to this forum as they provide some comment sense review of calculations made in planning a project which are easily overlooked and put things in their proper perspective. It is great to have an educated man giving his experience and knowledge to lend proper perspective to how to plan a project and how to reevaluate both goals and objectives and the proper way of achieving our goals. Again Thank You
 

LowQCab

Joined Nov 6, 2012
4,023
Thank You for the Complement,
I've just got way too much time on my hands,
and only another ~20 years or so to do a complete Data-Dump wherever it might do some good,
without getting myself "disappeared" for my efforts by some corrupt monopolistic multi-national corporation.
Too bad I don't get paid for it.
When we change over to the "Quantum Financial System" and go back to proper "Common-Law Courts",
maybe I'll get an opportunity to build my Zero-Reciprocating-Parts, ~90% efficient, mostly made out of Plastic, Diesel Engine,
(which also makes sick amounts of close to zero-pollution POWER by running on ~80% water, with no Radiator).
I can dream .......
I ain't a genius fur nuthin'.
.
.
 
Top