How to avoid triggering overcurrent protection

I have 4x motors that will be making lots of short burst movements in different directions. I'm driving these motors with some H-bridges which shut down when the source and sink reach 10A. This is fine - even though the motors sometimes demand more than that, the peak output current is 6A and that's been plenty for my needs.

However, I intend to power these motors over the 5v rail of a computer PSU, which provides 35A. Needless to say, this trips the overcurrent protection in the H-bridges, so the motors don't move at all.

I don't want to use a weaker power source since all 4 motors could be moving from a standstill at the same time.

So, I guess what I'm asking is how do I put a maximum of a 10A limit per H-bridge without dropping the voltage or using a different power source?

schmidtbag: I have 4x motors that will be making lots of short burst movements in different directions. I'm driving these motors with some H-bridges which shut down when the source and sink reach 10A. This is fine - even though the motors sometimes demand more than that, the peak output current is 6A and that's been plenty for my needs.

However, I intend to power these motors over the 5v rail of a computer PSU, which provides 35A. Needless to say, this trips the overcurrent protection in the H-bridges, so the motors don't move at all.

I don't want to use a weaker power source since all 4 motors could be moving from a standstill at the same time.

So, I guess what I'm asking is how do I put a maximum of a 10A limit per H-bridge without dropping the voltage or using a different power source?

I guess you will have to read the data sheets for the the H-bridges you are keeping a secret.

I guess the mystery motors are rated to run at 5 volts. So how does changing to a more capable power source cause problems?

But you never indicated you are stopping the motors before changing direction of rotation. That may be the source of your problems.

Paul

Ramp up the speed slowly. If the commanded speed is 70% then start at 10% and take 100-500ms to ramp up to 70%.

Also avoid sudden reversals.

Without knowing any details about your application nor parts, I can suggest you consider a two level current limit. This type of limiter is built into many controllers.

One current limiter could be at 15+ amps and act relatively fast.

The other limiter would be slower and limit at perhaps 8 amps.

You mention the multiple motors in a way that suggests you are looking for one current limiter for the whole application. I would suggest one limiter for each H-bridge.

@Paul It's not a secret, it just seems like irrelevant information. If you must know, it's a TLE 5206-2.

It doesn't [yet] matter if I'm changing direction, the motors don't even move from a standstill. I expect the motors to suddenly change direction but I need them to actually move first. However, if I swap out with a power source that doesn't exceed the amperage limits of the H-bridge, they will move just fine.

@MorganS The tricky part is if the H-bridges shut down due to overcurrent protection, they seem to stay shut down until they're power-cycled. It's kinda inconvenient to try and compensate for that when I could just simply limit the current.

@JohnRob I want to provide between 6-10A per motor, not 6-10A for everything; in other words, one limiter per H-bridge is exactly what I want to do. Having not much experience in electronics, I don't know a whole lot about how current limiters work. In particular, I don't know what it does when you exceed the specs, and I'm not entirely sure what to get. I saw something like this: https://www.digikey.com/product-detail/en/cantherm/MF72-003D13/317-1193-ND/1190968 which at least on paper sounds good, but I don't know if this actually does what I want.

A current limiter must reduce the voltage. A linear limiter such as a transistor will have to dissipate extra heat.

A switchmode limiter can be much more efficient: so efficient that it outputs more current than the input. Many switching power supply chips can be used for this but they are all surface-mount and must be used with a custom PCB.

If we must know? Yes, we must if you actually want useful help. If you don’t provide specifics for what you’re doing, the only help you can get is a generic guess. Is that what you want?

Okay, so you told which motor driver you’re using. That’s a good thing because now we can tell you that when the part shuts down, there is an output pin you can read to know that it faulted and that all that is required to restart the device is to change both inputs to zero. No power cycle required.

One look at the datasheet TLE5206 motor driver also tells us the 5206 has no built in ability to limit current. That means you either need a better driver chip as MorganS mentioned above or you need to control the current with your program and how you drive the 5206. How do you do that? MorganS told you in post #2.

If you don’t know how to do that, please read the sticky post here how to use this forum. Be sure to read and follow the instructions in topic 7 and post your code. Without seeing your code, we don’t know how you’re using TLE5206 which quite frankly, could be totally inappropriate for the part.

WattsThat: If we must know? Yes, we must if you actually want useful help. If you don’t provide specifics for what you’re doing, the only help you can get is a generic guess. Is that what you want?

Actually, yeah, that is kinda what I wanted.

One look at the datasheet TLE5206 motor driver also tells us the 5206 has no built in ability to limit current. That means you either need a better driver chip as MorganS mentioned above or you need to control the current with your program and how you drive the 5206. How do you do that? MorganS told you in post #2.

I never said it has an ability to limit current... I'm asking how to limit the current so it doesn't trigger the overcurrent shutdown trippoints mentioned on page 10. I'd much rather do that since it's a much simpler solution that I know for a fact works fine for my needs. I don't want to do MorganS's solution because as far as I'm concerned, it will make the motors accelerate slower than if I just avoid triggering the overcurrent protection via less supply amps, especially if I'm trying to rapidly change direction. Think of it like this: No matter what I do, the H-bridge's peak output is 6A. I can either stick with the full-blown 35A and slowly ramp up the speed (where I need to provide some margin of error slowing down the motor even further, so I don't accidentally trigger the OCP), or, I can just simply provide less than 10A per motor and go full speed where it has shown to work just fine. The latter is what I want to know how to do. Is it the "appropriate" solution? Probably not, but this is a one-of-a-kind personal project that I'm doing for fun so as long as I'm staying within the tolerances of the hardware, I don't care.

Without seeing your code, we don’t know how you’re using TLE5206 which quite frankly, could be totally inappropriate for the part.

Then why does it work perfectly fine when I use a variety of different lower amp power sources?

You cannot limit the amps and accelerate at the same speed using the same motors. Motors also cannot have instantaneous reverse acceleration because you need to allow some time for the magnetic field in the windings to dissipate.

The power supply is providing a maximum of 35 amps before it trips out or it’s voltsge droops (will will decrease the current for a given resistance). The motors draw the same current on a 35 amp power supply as they would on a 3500 amp power supply. Using an under-powered power supply would possibly limit the current, but will also likely cause other problems (the worst being death when the chinesium one you got from ebay bursts into a ball of flames devouring everything in its path)

You don’t “provide current” the motor “takes current” and all you can do is “limit current” with current limiting drivers or by lowering the output demand of the motor (by accelerating more slowly, which will reduce the amount of torque the motor needs to provide, and thusly its current requirements).

Show they datasheet of the motors, and from that you can determine how much current they require, or you can measure the current with a multimeter. The ntc thermistor may work, if the problem is the inrush current of the motors, but since we do not know what motors you are using, and their current requirements, it is impossible to say

This seems simple.

Under 10a power supplies will not trip controller because they don’t supply the motor with the stall current demand.

Over 10a power supplies will trip the controller because the motor is in a stalled state (O Rpm) and generally consuming 2-3x more current than running.

Test for this by manually spinning up the motor before powering it through the 35a supply.

Stating you “want” to provide 6-10a per motor using a controller that won’t control current and can’t switch past 10a makes me laugh inside..

What’s the data sheets on the motors tell you about nominal and stall current curve?

I never said it has an ability to limit current... I'm asking how to limit the current so it doesn't trigger the overcurrent shutdown trippoints mentioned on page 10. I'd much rather do that since it's a much simpler solution that I know for a fact works fine for my needs.

Here are several options, you can decide which is best for you since you apparently know some advanced physics about motor behavior that experienced people just don’t comprehend.

1) You put individual 6 amp supplies on each motor. 2) You design and build 6 amp current limiting circuits for each motor. 3) You buy new H bridges with internal current limiting.

Or, as a last resort,

4) You learn about the relationships between speed, torque and current in a dc motor. Once that is understood, proceed to learn about acceleration and deacceleration and their effect on current. Then you can modify your program and fix your problem.

You should also read this link, perhaps it will explain the answers you’re getting: The x-y problem

4) buy H bridges with enough current capacity and drive the nlmotors to their maximum performance easily.

Qdeathstar:
You cannot limit the amps and accelerate at the same speed using the same motors.

So whether I use a 6.5A source or a 9A source with the same voltage, that the motor won’t accelerate at the same speed, even though they’re both exceeding the peak draw current and below the overcurrent protection of the H-bridge?
If the answer to that is “no, they will accelerate at the same speed”, that’s exactly why I’m asking my question, and why I don’t want to slowly increase the motor speed.

Using an under-powered power supply would possibly limit the current, but will also likely cause other problems

That’s why I want to stick with the unit I have now with it’s abundant amperage but limit how much current goes to each H-bridge…

all you can do is “limit current” with current limiting drivers

That’s fine… that’s all I’m asking for.

Show they datasheet of the motors, and from that you can determine how much current they require, or you can measure the current with a multimeter. The ntc thermistor may work, if the problem is the inrush current of the motors, but since we do not know what motors you are using, and their current requirements, it is impossible to say

I don’t remember any specifics about the motors beyond being Mabuchi RS-555. There’s some variance between these RS-555 motors but it seems the highest stall current is a little past 9A.

Slumpert:
This seems simple.

Yes… I’m well aware of all that. That’s pretty much the basis behind my question.

Stating you “want” to provide 6-10a per motor using a controller that won’t control current and can’t switch past 10a makes me laugh inside…

What “controller”? The H-bridge? Because if you actually read my question (which it doesn’t seem you did) I want to prevent tripping the OCP of the H-bridge by limiting the amount of current it takes, but, without globally reducing the power supply itself.
If you’re going to be condescending, I don’t want an answer from you.

WattsThat:

  1. You put individual 6 amp supplies on each motor.
  2. You design and build 6 amp current limiting circuits for each motor.

Are you serious right now? This is what I’ve been asking how to do the entire time. There’s no “advanced physics” that I know, but people like you just keep wasting time with pedantic and condescending remarks. Drop the ego you suspect I have, re-read the last sentence in my original post, and you’ll realize the only thing I was asking how to do were the options you provided (mostly #2).
I prefer to stick with the computer PSU I have since 35A is plenty if I limit 6A per motor. But, I don’t know how to to do that, hence me being here…

  1. You buy new H bridges with internal current limiting.

If that was the best option to me, I wouldn’t be here. But since the H-bridges are known to do what I want without overheating, I don’t have a compelling reason to get rid of them.

Drive the bridges properly with speed ramps and you will have no problem. This will accelerate faster than methods 1 and 2.

3 and 4 seem too expensive for your budget?

MorganS: Drive the bridges properly with speed ramps and you will have no problem. This will accelerate faster than methods 1 and 2.

3 and 4 seem too expensive for your budget?

Pardon my ignorance, but I don't understand how a speed ramp will be faster if 6A is the peak output current. If I want to give some margin of error to avoid tripping the OCP, to my understanding, that means I'm going to be outputting less than 6A, in which case wouldn't that be slower than if I just supplied a constant 6A (via a current limiter)? Not sure if this means much, but the motors reach the desired speed in a fraction of a second.

As for option 3 (controller with current limiter), to me, it makes more sense to just stick with what I have and simply buy some current limiters. They're not expensive and it'd accomplish the same goal. I'm just not totally sure what it is I need. Honestly, before I posted my question, I wasn't aware the thermistor limiters existed. I don't want option 4 (a different/better motor controller) because I'm using the 5v 35A power source for more than just the motors. So, if I basically give the motors unrestricted access and they all power up at the same time, that may cause problems elsewhere. In other words, option 4 will fix one problem and cause others.

The current isn’t related to top end speed, it is related to torque. Starting from zero requires a lot of torque because there is no momentum. If you ramp the speed up more slowly you build up momentum which will reduce the current needed to achieve incremental increases in speed. So by ramping up slowly, you use less current but can achieve the same top end speed.

However, we don’t know what you are doing with your motors, so you might need the torque.

If you limit the current you limit the torque. There is no way around that fact.

So whether I use a 6.5A source or a 9A source with the same voltage, that the motor won't accelerate at the same speed,

It depends on the current requirements of your motor and how the specific supply you are using responds to being overloaded, if it is overloaded.

even though they're both exceeding the peak draw current and below the overcurrent protection of the H-bridge?

We don’t know the current requirements of the motor. If they both have the ability to supply the demand from the motor, then they are not limiting the current

and why I don't want to slowly increase the motor speed.

If you limit the current you limit the torque. There is no way around that fact.

If you limit the torque you limit the acceleration (slowly increase the motor speed). There is no way around that fact.

Qdeathstar: If you limit the current you limit the torque. There is no way around that fact.

I know this. I've stated several times I'm ok with that.

It depends on the current requirements of your motor and how the specific supply you are using responds to being overloaded, if it is overloaded.

Well considering the context, the supply isn't overloaded and the motor is reaching its stall current (which is more than 6A, and to remind you, that's the peak output current of the H-bridge). So - does that mean a 6.5A source will accelerate slower than a 9A source?

EDIT: I guess another way of looking at my question: is the 6A peak output current of the H-bridge a recommendation or is it literal? Because I don't understand how the motor is supposed to get more than 6A if the H-bridge says it can't provide more than that. So, if it can't get more than 6A, what difference does it make if the power source provides more than that? To me, the only thing that matters is if the power source provides less than 10A, since that's what triggers the OCP.

We don’t know the current requirements of the motor. If they both have the ability to supply the demand from the motor, then they are not limiting the current

In regards to my question, what difference does it make if the motor is only going to get a peak of 6A no matter what?

If you limit the torque you limit the acceleration (slowly increase the motor speed). There is no way around that fact.

Again, I don't care. I said from my original post that this was fine.

The chip you are using has an [u]under[/u] voltage lockout. You can read the the error bits to determine this.

Using a computer PSU at 5V to drive DC motors at high amps will likely drop that 5v to set that lockout.

Have you just tried to hook up the motors directly to the PSU without the H-bridge?

Those PSU supplies are good for steppers (the 12 v rail for a 3D printer) but I would not use it for driving 4 DC motors.

Just my opinion.

The H bridge will trip out if it the motor draws more than six amps (as you have experienced) or the voltage will drop, or the h bridge will become damaged if the motor pulls more than 6 amps.

So - does that mean a 6.5A source will accelerate slower than a 9A source?

It depends on the current requirements of your motor. If your motor only needs 5 amps, acceleration will be the same, but if your motor needs 10 amps, the current and voltage applied to the motor may not even be enough to turn it.

I checked the datasheet…

VUV (on) is 5.3 to 6.0V. The chip should not even switch itself on with only 5V power supply.

VUV (off) is nominally 4.7V but could be higher or lower. So a tiny sag in input voltage can trip it off.

If you want to continue using 5V then pick a different driver. These are not intended to run on 5V. Very few motor drivers are.