What diodes should I use?

[olof102]

Help me!!

I'm currently building this circuit **broken link removed**
A H-bridge using the L298 IC.

But I'm stuck now, in the text it says: "D1 to D4 are flyback diodes. Their forward current should correspond with the the expected load."

What does it mean? I use a 12V to run a 100W DC motor. It draws about 2A constantly.

Help me pick some diodes!!

 

[Boncuk]

You could use SB360(60V/3A) or SB560(60V/5A) Schottky-diodes.

 

[olof102]

Thank you!
What does the diode in this circuit do? Stop current from flowing back into the IC? Am i right?

 

[Diver300]

No. The motor is inductive, so you can't stop the current in the motor very quickly. When the IC turns off, the diodes give the motor current somewhere to go. Without them, you would get some very large voltage spikes each time the motor turned off.

 

[Boncuk]

No. The motor is inductive, so you can't stop the current in the motor very quickly. When the IC turns off, the diodes give the motor current somewhere to go. Without them, you would get some very large voltage spikes each time the motor turned off.

Objections!

Turning off the motor power is achieved rapidly.

The back EMF is the problem to deal with.

Back EMF voltage can reach 6 to 8 times the supply voltage level, e.g. supply voltage 12V, back EMF = 72/96V.

The diode must be able to reduce that back EMF to a level which won't "kill" the switching transistor.

Schottky-diodes work perfect for spike suppression when rated for the coil nominal power.

Fast recovery diodes work perfectly as well.

Here are some types to use: MR85X, where MR851 is 100V/3A, MR852 is 200V/3A, MR854 is 400V/3A and MR856 is 600V/3A. The strongest fast recovery diode is the RGP30M (1000V/3A)

The diodes must be able to terminate a relatively high voltage (hence high current) within a very short time repeatedly.

Boncuk

 

[crutschow]

Fast diodes are not particularly required for the suppression of inductive kick. Fast diodes are rated for fast recovery turn-off. For suppressing an inductor, you just need a diode that turns on rapidly, and all diodes do that. So don't worry about the speed of the diode here. A Schottky diode does have a lower forward drop but that is likely not important for inductive suppression.

 

[Vizier87]

agreed with carl.

 

[Hero999]

If you're not worried about speed, then why not just rely on the MOSFET internal diodes on a MOS bridge?

I know that this isn't a MOS bridge, I'm just asking.

 

[crutschow]

If you're not worried about speed, then why not just rely on the MOSFET internal diodes on a MOS bridge?

I know that this isn't a MOS bridge, I'm just asking.

They are connected in the wrong location. The diodes need to connected across the inductance, not across the transistor.

 

[mneary]

I mentioned in OP's other thread that 1N400x are also adequate. Don't know why this came up again.

 

[Hero999]

They are connected in the wrong location. The diodes need to connected across the inductance, not across the transistor.

No they're connected in the correct direction for a bridge - in reverse parallel with the transistors.

 

[Boncuk]

Fast diodes are not particularly required for the suppression of inductive kick. Fast diodes are rated for fast recovery turn-off. For suppressing an inductor, you just need a diode that turns on rapidly, and all diodes do that. So don't worry about the speed of the diode here. A Schottky diode does have a lower forward drop but that is likely not important for inductive suppression.

Hi Carl,

I agree with you partly.

A Schottky diode has a lower forward drop than a silicium type diode. Lower voltage drop at high current means the diode has to dissipate less power than a silicium type diode.

The speed of the diode determines if the back EMF is reduced fast enough in order not to damage the driver transistor.

Most data sheets suggest Schottky diodes for recovery - certainly not without reason.

Regards

Hans

 

[crutschow]

A Schottky diode has a lower forward drop than a silicium type diode. Lower voltage drop at high current means the diode has to dissipate less power than a silicium type diode.

In most inductive suppression applications the diode current flows for a very short time, thus the diode power dissipation is low for any diode.

The speed of the diode determines if the back EMF is reduced fast enough in order not to damage the driver transistor.

As I noted it's the forward turn-on speed that's important and all diodes are fast at turn-on.

Most data sheets suggest Schottky diodes for recovery - certainly not without reason.

Certainly you want to use Schottky diodes for the flyback in switching power supplies to improve efficiency. For spike suppression in inductive circuits, I don't see the need.

 

[mneary]

Carl is right, reverse recovery time isn't important if the forward conduction ends before a reverse bias is applied. Therefore diode speed isn't important in most 'catch diode' applications. Speed is only important if forward current is still flowing when the voltage reverses (as in many switching PS).

Reverse recovery of diodes is misunderstood by most engineers I know. Most of them don't need this information, so that's sort of OK. But that means that most engineers will give the wrong answer when asked.