Do you know a way to prevent short circuits in a h-bridge?

[Lewis Mojica]

Hello!

I'm designing a circuit to control a dc motor, now i'm working on the h-bridge and i made the following schematic:



Then i realize that if both buttons are pressed at the same time it would produce a short circuit; every transistor would be in saturation so, vcc would be connected to ground from q1 to q4 and from q2 to q3. Then i made this schematic:



In the schematic above, the logics gates prevent the short circuit, so when both button are pressed the logics gates do not active any transistor. I also added diodes to each transistor to protect then from reverse voltage from the inductive load.

Now, i'm wondering, are there a best way to protect this from circuit short circuit?

[Neilm]

HAve a look at H-bridge driver chips. They include the bootstrap you will need to correctly drive the upper FETs (I wouldn't use transistors) and will include shorting and shoot through protection.

[tecman]

Many H-Bridge drivers put a small dead time between high and low devices to eliminate cross coduction.

paul;

[Cyberdragon]

The problem with driver chips is that they are normally digital or PWM controlled and aren't meant for something this simple. The logic chips would be far simpler than getting a driver chip to work.

The type of transistors would be determined by the application. If it's a small battery-powered thing, bipolars would probably be fine. FETs would be suitable if the motor had significant load on it.

The simplest solution, depending on the size of the device and power requirements, would just be two SPDT buttons or relays wired like a two-switch light circuit in a house. So each button reverses the polarity of each side, and if you pushed both they'd just be at the same polarity again.

[jmelson]

You cannot connect the bases of the high side transistors to the bases of the low-side transistors.  They need separate drives, as their emitters are at different voltages.

Jon

[KhronX]

Well, not if you want / need to use all NPN / N-channel transistors. If the high-side ones are PNP / P-channel, then a common drive signal with the low-side one is fine.

You cannot connect the bases of the high side transistors to the bases of the low-side transistors.  They need separate drives, as their emitters are at different voltages.

Jon

[rstofer]

H-bridges don't really work like your schematic.  There will be a time where one transistor is turning off while another is turning on.  Both are transitioning and there will be a period where both are partially on.  This will usually blow both transistors/MOSFETs.

[Lewis Mojica]

H-bridges don't really work like your schematic.  There will be a time where one transistor is turning off while another is turning on.  Both are transitioning and there will be a period where both are partially on.  This will usually blow both transistors/MOSFETs.

I don't see how can be a period where a couple of transistor will be partially on, i mean, the transistors are controlled for the logic gate which only will output 1 or 0, saturation or cut-off.

[Lewis Mojica]

The problem with driver chips is that they are normally digital or PWM controlled and aren't meant for something this simple. The logic chips would be far simpler than getting a driver chip to work.

The type of transistors would be determined by the application. If it's a small battery-powered thing, bipolars would probably be fine. FETs would be suitable if the motor had significant load on it.

The simplest solution, depending on the size of the device and power requirements, would just be two SPDT buttons or relays wired like a two-switch light circuit in a house. So each button reverses the polarity of each side, and if you pushed both they'd just be at the same polarity again.

Sadly i'm not able to use SPDT buttons, this schematic is part of a circuit that will replace another circuit which is broken, and for several reasons it is better to replace it. The device in which i will out the new circuit use simple push button (open or close). I also avoid to use relays in cases that isn't strictly necessary to improve the useful life of the circuit.

I'll make another schematic with mostfet.

[BNElecEng]

MOSFETs are not ideal devices. They take a finite amount of time to change between the ON and OFF states. Dead-time is used in gate driver circuits to take this into account, by adding a small time delay to the switching logic. If you haven't heard of dead-time before, I'd suggest a quick google of the topic. It's a very practical concept that gets used in many applications.

[mikerj]

The circuit is doomed as drawn. You have no current limiting into the low side transistors bases, so the output of your logic gates will be clamped at around 0.7v, which means you'll have nothing coming out of the emitter followers on the high side.  An emitter follower isn't ideal anyhow, unless you don't mind the additional voltage drop and higher dissipation?

[C]

Lewis Mojica

First all logic takes time to change states and actually has three states for standard logic, 1,0 & changing.

Second there are many H-Bridge chips that have simple to use inputs. Some just have two direction inputs. A lot are simple direction and on inputs. The PWM is just used to control ON to gain speed control.