H-Bridge shoot-through

[Diver300]

I put together an H-Bridge using BTS 443 P smart switches as the high side drivers and FDMS7682-d MOSFETs as the low side drivers. The schematic is attached. All the control and the feedback come from a microcontroller running on 3 V.

The H-bridge works and I can run the motor forwards and backward.

The problem is when I tried to run PWM. The BTS 443 P are not very fast, so the idea was to leave them on, and operate the MOSFETs from the PWM signal. When the MOSFETs turn on, the BTS 443 P that is turned off lets some current through for a few microseconds.

So if I am running forwards, the forward BTS 443 P is on all the time and the reverse BTS 443 P is off all the time. For PWM, the forward MOSFSET turn on and off at 20 kHz. The problem is that each time the forward MOSFET turns on, the reverse BTS 443 P lets current through for a short time. It is quite a lot of current, in excess of 15 A, and for several microseconds. That becomes a problem when the switching is happening at 20 kHz.

Is there is an alternative high side driver that wouldn't do this, or is there some other way of preventing shoot-through?

Is there something in the data sheet for the BTS 443 P that should have told me that this would be a problem?

 

[shortbus=]

Not sure if this will be helpful.

On the first page under "applications" It states "no PWM with inductive loads" As a DC motor is both resistive and inductive, that may be part of the problem.

On page 9 it shows/illustrates the problem internally when driving an inductive load.

 

[Diver300]

I wasn't trying to use the BTS 443 P to switch the motor with PWM. That was done by the MOSFETs. The voltage would have been clamped to near the supply rails.

I was getting shoot-through when the motor was replaced with a 1 kOhm resistor. It seemed to be the rapid drop in output voltage which turned on the BTS 443 P

 

[danadak]

Dead time/band in PWM implementation, where MOSFET drive does not overlap
for a small time, see PH1 and PH2 in PWM setup window below. This chip is a SOC,
many more resources available onchip for V and I measurement, general logic thru
high level functions. See right hand window for resources used/left.....

Regards, Dana.

 

[Diver300]

Danadak, I was not driving the reverse high side driver or reverse MOSFET at all.

The two states were:-
1) Forward high-side driver on, everything else off
2) Forward high-side driver on, forward MOSFET on, everything else off.

The reverse high-side driver was never turned on, but for some reason, it let some current through just after the forward mosfet turned on.

 

[rjenkinsgb]

I'm guessing it's triggering a part of the driver protection circuitry - the data mentions "Fast demagnetisation of inductive loads".

The device may be trying to damp what it sees as a fast negative spike from an external source, as it would get get with an inductive load it was driving being switched off.

They are possibly too "smart" for that application & the designers may never have envisaged another switching device being in the same circuit.

You could try adding snubbers across them, to reduce the dV/dt rate it sees? eg. 10 Ohms and 0.1uF?

 

[Diver300]

I'm guessing it's triggering a part of the driver protection circuitry - the data mentions "Fast demagnetisation of inductive loads".

The device may be trying to damp what it sees as a fast negative spike from an external source, as it would get get with an inductive load it was driving being switched off.

They are possibly too "smart" for that application & the designers may never have envisaged another switching device being in the same circuit.

You could try adding snubbers across them, to reduce the dV/dt rate it sees? eg. 10 Ohms and 0.1uF?

I think that it might be something like that, but I'm not sure if it's a deliberate protection device.

Since I posted this thread, I've been looking at the datasheet and possible alternatives. There is no ground connection on the BTS 443. When it is off, the supply, the logic input and the output are all at 12 V. The supply and output are drain and source of an N-channel MOSFET. I assume that the gate must be at 12 V because most of the rest of the IC is at that voltage. When the output is rapidly pulled down to ground, that is the source of the MOSFET. Any capacitance in the drive circuit will tend to hold the gate higher than ground, and it will take time for the gate voltage to also get to ground voltage. During that time, the MOSFET is at least partially on.

I will look at a snubber, but I suspect that it will end up dissipating as much power as the shoot-through is.

I think that I need a high-side driver that has a ground connection.

 

[Diver300]

I've found that the BTS50060-1TEA has a ground connection and the same package.

Unfortunately, it's a completely different pin-out, so it will be difficult to adapt even to a circuit board for a proof of concept.

Also, Farnell have none available for nearly a year.......

 

[crutschow]

The momentary shoot-through of the off MOSFET may be due to its drain-source capacitance.

 

[danadak]

I've found that the BTS50060-1TEA has a ground connection and the same package.

Unfortunately, it's a completely different pin-out, so it will be difficult to adapt even to a circuit board for a proof of concept.

Also, Farnell have none available for nearly a year.......

Try getting a sample directly from manufacturer. Also try contacting their FAE
and/or sales rep to get a sample.

Regards, Dana.

 

[Diver300]

The momentary shoot-through of the off MOSFET may be due to its drain-source capacitance.

Have you got any idea how I would find out the drain-source capacitance?

I would have assumed it would be too small to make a difference. I think that the drain-gate capacitance would be more important as it would tend to turn the MOSFET on.

 

[crutschow]

Have you got any idea how I would find out the drain-source capacitance?

Look at the data sheet output capacitance (example below):

 

[rjenkinsgb]

I think that I need a high-side driver that has a ground connection.

Do you have a 1K to ground / negative on the IS pin? No idea if that will have any effect, but it gives another part of the circuit a ground reference.

 

[Diver300]

Look at the data sheet output capacitance (example below):

View attachment 132410

The output capacitance doesn't seem to be quoted for the BTS443P. However, if I average the capacitance at 5000 pF, charging and discharging that to 12 V at 20 kHz would only give just over 1 mA average, and I am seeing several amps.

I've not got an oscilloscope handy but I'll post the waveform of the current pulse when I can.

 

[Diver300]

Do you have a 1K to ground / negative on the IS pin? No idea if that will have any effect, but it gives another part of the circuit a ground reference.

I've got a 1.3 k resistor to ground. I don't think that the BTS443 can use that connection as a ground reference because it would lead to a false current reading.

That is why I intend to try one with a separate ground connection.

 

[danadak]

How/where are you measuring the transient current ?

Regards, Dana.

 

[Diver300]

How/where are you measuring the transient current ?

Regards, Dana.

I've been measuring the waveform using the current monitor signal from the BTS443, and the average current is shown on the display of the bench power supply that is running the system.

With the motor removed, there is no shoot-through and the average current is <100 mA for the whole circuit. With the motor replaced with a 1 kOhm resistor, the average current is more than 2 A and the circuit gets hot.

 

[gophert]

Are you certain the P-mos is connected the right way around. Drain to motor, Source to Vdd. A common error for first time P-MOS users.

 

[danadak]

I've been measuring the waveform using the current monitor signal from the BTS443, and the average current is shown on the display of the bench power supply that is running the system.

With the motor removed, there is no shoot-through and the average current is <100 mA for the whole circuit. With the motor replaced with a 1 kOhm resistor, the average current is more than 2 A and the circuit gets hot.

I am sure you have followed the constraints here -

Plus I see its got its own signal path response -

Although its response looks like classic one pole amplifier response.....so no explanation for
transient peaking here......is there a spice model for this part ?


Regards, Dana.

 

[Diver300]

Are you certain the P-mos is connected the right way around. Drain to motor, Source to Vdd. A common error for first time P-MOS users.

I'm not using P-MOSFETS. The high-side drivers are the "smart" devices that contain an N-MOSFETs and a charge pump. The low side are simple N-MOSFETS.

The circuit works fine when PWM isn't used. If there is any shoot-through when starting the motor, it only happens for a few microseconds and it wouldn't be noticed.