Proper design of the H-bridge. Your design is inadequate, because you do not completely understand either the design theory or the theory of how brushed DC motors operate.
Buy a professionally designed H-bridge capable of 100 A at 50V, and it should be fine in that application.
If it worked well then you would not keep destroying your hardware. Therefore it doesn't work well.
Still I am not the only one telling you that your design is inadequate, you are rejecting others saying the same thing. Why ask a question when you don't want an answer?
You don't know enough to design a viable H-bridge.
As you choose to ignore me I will choose to ignore you. The thread will be muted and so I will not receive any more notifications from the thread, so I will not be able to respond.
Good luck with your project and goodbye.
I'll admit I did not read every post. However looking at the circuit I would guess the current limit circuit is unidirectional. When the motor causes a reverse current to flow the current limiting circuit thinks this is all good because it will be below the threshold of the current preset.
One MOSFET? Two? All four?
How fast do your MOSFETs switch? timeConstant = 10Kohm*87nF=870us
Are you preventing shoot through in code? How much time does the code take to switch the MOSFETS? Are you doing PWM?
Where's the code?
Thank you very much for your reply and interest. However, I think you are interpreting my post differently than I intended.
I am not good at English and am using a translator, so I think my 'question' is not conveyed well.
I would like a solution for the linked content.
I am curious about other solutions and methods.
Thank you very much for your interest.
I am asking because I am really curious. I think your answer will be of great help to me.
Here is your statement.
'Proper design of the H-bridge. Your design is inadequate, because you do not completely understand either the design theory or the theory of how brushed DC motors operate.'
Thank you for your sincere answer.
The reason is that
(1) There has never been a problem when electrically driven. (For a long time)
(2) However, I actually saw problems with 'FET' and 'DIODE' when external force was applied manually. I thought that 'Back EMF' was the cause. So, I asked for advice on how to solve 'Back EMF'.
(3) However, I am not good at English, so the translation is not good.
(4) Are you saying that the cause is still 'Shoot through'?
Is it correct to say that the answer to my question about the cause of the failure is 'Shoot through'?
I do not know much, but
Isn't 'Shoot through' the cause of problems when electrically driven without external force? I'm writing this because I don't know much, so please explain it in detail.
And I really, sincerely never ignored your opinion. It's just because I don't know much. I may have misunderstood.
I wonder if it's right to see 'Shoot through' as the cause.
So all four FETs burn out. And you are using pwm of maybe 980hz (510us pulse for50%duty cycle pulse on a system with a time constant of maybe timeConstant = 10Kohm*87nF=870us.
I’m no expert but I don’t think your h bridge is designed to be fast enough to use pwm.
Please post your code. Or at least a minimum reproducible example.
I think I said it wrong.
If you are interested in learning the basics of electronics and electrical circuits, I strongly recommend EE courses at a technical university. It takes several years of study and practice to learn the material you will need to be successful with this project.
Thanks for the recommendation.
Did I answer incorrectly or make a mistake in my design? If you could point it out, it would be a great help to me.
Many, as forum members have tried to explain, but you don't seem to understand the posts.
Thank you for your reply.
In conclusion, are you saying that I did not understand what the other members said?
If that is true, I will reflect on it.
And sorry. I don't know if I misunderstood, but I think you thought there was a difference between what I wrote and what the 'forum members' answered.
If it was my ignorance that made me misunderstand, I'm really sorry.
I will read the reply again.
Thanks for the recommendation of 'SN754410'.
However, the motor capacity is large.
MOTOR : 24V, 400W
You are right.
The current is measured through one resistor.
Is there a way to measure the current in the other direction?
Which FET dies? Is it always the same FET? Does a different FET die depending on which direction the motor is energized and which direction it gets pushed? Which way is "Foreward"? You have two schematics, #11 with Q1-Q4 feeding motor+ and motor-, and #12 with IC7 and IC9 feeding MOT_A and MOT_B. It makes confusing and hard to communicate.
Do you switch directions often? How do you switch the directions? How big of a dead time is there between turning off Q3 and on Q4? Are all 4 pins on PWM or only Qxx and Qyy? How is it energized when it is stopped? PWM?
How much force/torque does it take to blow a MOSFET? Does it get much hotter as it blows?
I’m not at all an expert on this, but for detailed advice on your redesigning something, I’d think one would need detailed answers to questions like those.
Thank you for your reply. I appreciate your sincere advice.
I'm really sorry, but I may have misunderstood,
but it's unfair.
I said that it breaks in two cases.
Looking at the above, it's not that it breaks when you simply drive it electrically, but that it breaks due to an external force, and that's what I asked about.
It's not that it breaks when you drive the motor electrically.
Looking at your question, isn't it a question about the motor breaking when you drive it electrically?
Am I misunderstanding? That's because I really don't understand.
I'm saying we can't really understand well enough to give you advice from your descriptions. For instance, by "in the stationary state" is the power off? Or do you have a couple of the MOSFETS engaged in a braking? Partial braking with PWM? I really do not know enough to tell you the right fixes in any of those cases, but they seem like they might be different failure modes and perhaps require different fixes.
The only one I was pointing out above was that the capacitance of the gate, taken with the resistance gives a time constant that's about 1ms long, which to me means you ought to switch with much more than than 1ms in the code to avoid shoot-through. And that use with a 490 or 980Hz PWM might incompletely saturate the MOSFET. I don't know enough to say that fixing those issues with smaller resistors would solve your problems, but those look like potential problems that may exist haven't been eliminated by your answers.
Maybe your total system as implemented can only handle 1/10th of the expected 400W load without dying, and normal operation only needs 39W unless you stall the motor.
I'm quite sure that is the case.
In fact, I'm surprised that the circuit worked at all. The MOSFETs are rated for only 40 A, they are not logic level and don't have the appropriate gate drivers, while the motor stall current could be 80 A or higher.
I haven't seen an answer to post #6.
