Re: Protecting a DC motor
Bookmark this page:
 Yahoo!
 Google
 Windows Live
 del.icio.us
 digg
 Netscape
|
|
||||||||||||||||||||||||||||
|
Posted by on August 7, 2008, 2:32 pm
Please log in for more thread options often. But when it does, it blows the controller in less than a second. I assume when the motor stalls, it behaves more or less like a short circuit placing quite a few amps down the MOSFET (IRL7843 with an rds of 5 milliohms) which ends up with a crack across it. The MOSFET also has a source to gnd series 0.01ohm 6W current sensing resistor, which also blows up. (100 amps at 0.01r is 100W) The system is designed to operate at 10-20A nominal, so the controller is OK under these conditions, the jamming is an unexpected situation. Ideally, i'd like a fast "intelligent" fuse that can trip during these unexpected situations. I don't have access to the controller's internal circuit, I know it has some form of current sensing (it has tripped before) but somehow it is not catching these fast high-current conditions. Any ideas? | ||||||||||||||||||||||||||||
|
Posted by Jon Slaughter on August 7, 2008, 3:05 pm
Please log in for more thread options >>
>> >> >> >> Does it only blow when turning on? or because the motor is stalling >> somehow? >> Or is it more "random"? > Only when the motor stalls (mechanical jam) which doesn't happen very
> often. But when it does, it blows the controller in less than a > second. > Ok, its cause the maximum current occurs durring stall(at least with most motors). > I assume when the motor stalls, it behaves more or less like a short
> circuit placing quite a few amps down the MOSFET (IRL7843 with an rds > of 5 milliohms) which ends up with a crack across it. The MOSFET also > has a source to gnd series 0.01ohm 6W current sensing resistor, which > also blows up. (100 amps at 0.01r is 100W) Well, it behaves as a resistor. Once the motor starts turning a back emf is generated which reduces the current. >
> The system is designed to operate at 10-20A nominal, so the controller > is OK under these conditions, the jamming is an unexpected situation. > Ideally, i'd like a fast "intelligent" fuse that can trip during these > unexpected situations. > > I don't have access to the controller's internal circuit, I know it > has some form of current sensing (it has tripped before) but somehow > it is not catching these fast high-current conditions. > > Any ideas? > Well, it really depends on what you need. If the stall is only temporary and the motor and "restart" somehow then there are many options. If you have to manually remove the jam then there are a few other options. But realize that the mosefets blowing is a good thing else your motor could end up getting fried. Some possible options: 1. Use a circuit breaker. 2. Use a thermal switch. (kinda an automatic circuit breaker but probably not really a smart way to do it... but cheap and easy) 3. Use a smart current limiting technique like current folding. This may or may not be easy to do for you depending on your electronic skills or ability to find and integrate such a device into your system. 4. A fuse(if you like wasting fuses). 5. Current limiting resistor(like a lightbulb). Not the best method as it wastes power and might cause the motor to run slower but might introduce just enough resistance to limit the stall current so the mosfets don't blow. (I don't recommend this one as its really just very dumb current limiting but then again might work fine for your needs) 3 Is the best and can be accomplished relatively easy. What you do is have a small very precise resistor in the path. like a 0.010Ohm resistor, say. Since you know the voltage is 24V then the current through the resistor will give a voltage on it of I*0.010. The power dissipated in the resistor is very small and has little impact on the circuit. (for about 30A it will dissipate 9W) You then sense the voltage drop across it which you can then compute the current. When the current passes a threshold(say using a comparator) it will trigger a relay or mosfet to turn off the power. (or even switch in a high resistance light that could be used as a malfunction indicator) It can be accomplished with just a few components like are resistor, comparator and transistor. Better techniques would use current folding which actually reduce the current but require a more components and knowledge. I'm sure theres other techniques out there too but you really need to decide what you want to do and what you need to do for your specific application. | ||||||||||||||||||||||||||||
|
Posted by Paul Hovnanian P.E. on August 7, 2008, 4:02 pm
Please log in for more thread options
Jon Slaughter wrote: >
> >>
> >> > >> > >> > >> Does it only blow when turning on? or because the motor is stalling > >> somehow? > >> Or is it more "random"? > > Only when the motor stalls (mechanical jam) which doesn't happen very
> > often. But when it does, it blows the controller in less than a > > second. > > >
> Ok, its cause the maximum current occurs durring stall(at least with most > motors). > > > I assume when the motor stalls, it behaves more or less like a short
> > circuit placing quite a few amps down the MOSFET (IRL7843 with an rds > > of 5 milliohms) which ends up with a crack across it. The MOSFET also > > has a source to gnd series 0.01ohm 6W current sensing resistor, which > > also blows up. (100 amps at 0.01r is 100W) >
> Well, it behaves as a resistor. Once the motor starts turning a back emf is > generated which reduces the current. > > >
> > The system is designed to operate at 10-20A nominal, so the controller > > is OK under these conditions, the jamming is an unexpected situation. > > Ideally, i'd like a fast "intelligent" fuse that can trip during these > > unexpected situations. > > > > I don't have access to the controller's internal circuit, I know it > > has some form of current sensing (it has tripped before) but somehow > > it is not catching these fast high-current conditions. > > > > Any ideas? > > >
> Well, it really depends on what you need. If the stall is only temporary and > the motor and "restart" somehow then there are many options. > > If you have to manually remove the jam then there are a few other options. > > But realize that the mosefets blowing is a good thing else your motor could > end up getting fried. > > Some possible options: > > 1. Use a circuit breaker. > > 2. Use a thermal switch. (kinda an automatic circuit breaker but probably > not really a smart way to do it... but cheap and easy) > > 3. Use a smart current limiting technique like current folding. This may or > may not be easy to do for you depending on your electronic skills or ability > to find and integrate such a device into your system. > > 4. A fuse(if you like wasting fuses). > > 5. Current limiting resistor(like a lightbulb). Not the best method as it > wastes power and might cause the motor to run slower but might introduce > just enough resistance to limit the stall current so the mosfets don't blow. > (I don't recommend this one as its really just very dumb current limiting > but then again might work fine for your needs) > > 3 Is the best and can be accomplished relatively easy. What you do is have a > small very precise resistor in the path. > > like a 0.010Ohm resistor, say. Since you know the voltage is 24V then the > current through the resistor will give a voltage on it of I*0.010. The power > dissipated in the resistor is very small and has little impact on the > circuit. (for about 30A it will dissipate 9W) > > You then sense the voltage drop across it which you can then compute the > current. When the current passes a threshold(say using a comparator) it will > trigger a relay or mosfet to turn off the power. (or even switch in a high > resistance light that could be used as a malfunction indicator) > > It can be accomplished with just a few components like are resistor, > comparator and transistor. > > Better techniques would use current folding which actually reduce the > current but require a more components and knowledge. > > I'm sure theres other techniques out there too but you really need to decide > what you want to do and what you need to do for your specific application. The OP should get some data from the motor manufacturer about damage due to overloads over time. Any scheme is going to have to allow for the motor's starting characteristics. The protection must allow the 80A, 100 ms starting current, but operate at 80A for durations beyond that. Circuit breakers or thermal devices with the proper I-T curves can be matched to the motor's characteristics. It would also be possible to design a foldback circuit that would allow the starting surge but then step up its sensitivity to just above the normal full load operating current. -- Paul Hovnanian mailto:Paul@Hovnanian.com ------------------------------------------------------------------ Why are so many towns named after water towers? | ||||||||||||||||||||||||||||
|
Posted by Jon Slaughter on August 7, 2008, 6:06 pm
Please log in for more thread options
> The OP should get some data from the motor manufacturer about damage due
> to overloads over time. > > Any scheme is going to have to allow for the motor's starting > characteristics. The protection must allow the 80A, 100 ms starting > current, but operate at 80A for durations beyond that. No, not necesssarily. The 80A is full torque. If you don't need full torque at startup then you don't need the full current. In fact chances are he only needs the amount of current that will overcome friction. (even if it rotates at 1rph...) This should be quite low. Of course you don't want to wait for ever for the motor to get up to full speed so it really depends on what the motor is driving. It's easy to implement anyways. A simple resistor-cap delay that prevents the current limiting device from triggering. > Circuit breakers or thermal devices with the proper I-T curves can be
> matched to the motor's characteristics. It would also be possible to > design a foldback circuit that would allow the starting surge but then > step up its sensitivity to just above the normal full load operating > current. > It's all going to depend on the specific problem that is causing it to jam and the type of solution that they want. | ||||||||||||||||||||||||||||
|
Posted by Paul Hovnanian P.E. on August 11, 2008, 7:37 pm
Please log in for more thread options
Jon Slaughter wrote: >
> > The OP should get some data from the motor manufacturer about damage due
> > to overloads over time. > > > > Any scheme is going to have to allow for the motor's starting > > characteristics. The protection must allow the 80A, 100 ms starting > > current, but operate at 80A for durations beyond that. >
> No, not necesssarily. The 80A is full torque. If you don't need full torque > at startup then you don't need the full current. In fact chances are he only > needs the amount of current that will overcome friction. (even if it rotates > at 1rph...) This should be quite low. Of course you don't want to wait for > ever for the motor to get up to full speed so it really depends on what the > motor is driving. > > It's easy to implement anyways. A simple resistor-cap delay that prevents > the current limiting device from triggering. > > > Circuit breakers or thermal devices with the proper I-T curves can be
> > matched to the motor's characteristics. It would also be possible to > > design a foldback circuit that would allow the starting surge but then > > step up its sensitivity to just above the normal full load operating > > current. > > >
> It's all going to depend on the specific problem that is causing it to jam > and the type of solution that they want. This would be the way to go if the OP could dig into the drive controller. My impression was that the solution had to be external. It might be possible to add current limiting in series with the existing controller, but that may introduce some contention between the two of them. Adding a simple thermal overload that is essentially a closed circuit until it trips open would be the most straightforward. -- Paul Hovnanian paul@hovnanian.com ----------------------------------------------------------------------- Have gnu, will travel. | ||||||||||||||||||||||||||||
>
>
>
> Does it only blow when turning on? or because the motor is stalling somehow?
> Or is it more "random"?