How to make an over-current fuse/switch to avoid damage to a brushless motor

EK, we can solve all problems, as long we understand them. I understood for myself, my motor can handle 90 Ampere for 10 minutes. Thus my conclusion is, if I set it for 70 or 80 ampere, it does not matter, whether my 24 Volt drops to 15 Volt or go to 32 Volt or whatever, because a stalling of my motor at 24 Volt means some hundreds of ampere, probably close to the 2000 Ampere. My DC resistance is very low. Thus by 12 or 15 Volt or below 24 Volt or 32 Volt, the 70 or 80 Ampere still cut the system off.

Now in your case. The unit can go down to 6 Volt and still works, but as long your motor by stalling still will reach the 30 Ampere.

However by greater than 12 Volt, I may have to place a 12 Volt zener over the DC - DC converter to limit the risk of damaging the DC - DC converter. The IC can handle up to 30 Volt. If I now put 24 Volt on it , your 30 Ampere cut off may be too low and should be 50 Ampere, which is also fine for 12 Volt. Why? I thought somewhere you mentioned that the motor can handle 60 Ampere.

Alternative we have to measure the input voltage and do the testing of the over voltage in stages. But that we can do only in a 8 pin PIC12F675 with 4 Mhz internal. With 20 Mhz you need either a more pin IC or go to P Mosfets. With the consequence that the IC fails or is removed or has oxidation's in the pins, will no longer be "fail safe" . It is your call, in the meantime I have to find a solution to my real problem in how I can see what is going wrong in the registers of the PIC12F675 and why.

I assume now that I have lost the opportunity to help you in this summer time, I am so sorry. We will get it all right with time. Will you be happy with my explanation?

Bert

 

Just to elaborate on the above. We can make it that it will work from 7 Volt to 30 Volt. But I think you will have to consider a cut-off current of much more than 30 Ampere, probably 40 or 50 Ampere. It does not matter, because whether it stall at 12 Volt or stalls at 24 Volt, your stalling current will be in the hundreds/thousands without the protection of 50 Ampere in anyway.
I think that if your motor stalls, it will take in anyway a few hundred microseconds from running to dead stop, just because of your mechanical flexing and therefore to cut the 50 Ampere within 20 microseconds will not harm your motor. If the Microchip IRFP064N can handle each 390 Ampere for 1 microsecond with a 25 Micron golden wire (you have 2 parallel), I am convinced that the copper coil wire of your motor, which is something like 4 or 5 mm (+/- 18 mm2) can handle a few thousand Ampere for a few microsecond also. Thus cutting the current at 50 Ampere, approx up to 20 uSecond, I cannot see any damage what so ever. Will that put you at ease? Bert

 

If you take 2000 Ampere for 2 seconds, if you do not have a protection and your battery drops to a few volt, let say 4 Volt, you have 2000 x 4 x 2 seconds = 16 000 joule, that is nearly more than a lightning strike. Your motor is a goner. But if it is only 20 uSeconds, it will be 0.16 joule. Your motor does not even get warmer. The above should put you really at ease. Bert

 

OK Bert, that's great then. My mind is at ease!

My motor has low IR so it will easily reach 30 amps on a stall. It will probably do that even with 1 volt applied!

I do think the design needs to handle a wider range of voltages or will probably be blown up at some point due to dumb operator error (ie me). 6v to 30 volt sounds great.

Given a nominal design set point of 30 amps, how do I calculate what the trip point will be for any given actual battery voltage? You mentioned 50 amps might be necessary but I'm getting lost with all the numbers. Conceptually I prefer a trip point of 30 amps but I realize it will change if the supply voltage changes...however I don't understand how to calculate the new number in order to verify it will be suitable (or not).

And yes, we have all winter to finish things. But I am not concerned since I expect to have many more summers!

 

Good, I am relieved.

yes, indeed, 21 milliOhm and 70 Ampere.

I have never met a dumb Canadian, thus I have faith in you. I will try for 7 to 30 Volt and can only discuss this with you when I know all the consequences after testing.

That is the easiest from all problems. If you motor can handle 70 Ampere for a long period of time (refer to your thread No 19), I would not set the trip point at 30 Ampere, but higher like 50 Ampere. The rise time of the stalling situation will not be in nano seconds but in a few hundred microseconds, but most likely in 1 or 2 milliseconds, thus if the PIC12F675 reacts within 20 uSeconds, you will not exceed the 50 Ampere by more than a fraction. If you set it too low, you may get false triggering. Special when you suddenly need to increase the speed to maximum, when you use 24 Volt batteries, probably not with 12 Volt batteries. But it is your call.

You state 21 mOhm i.e at 12 Volt and that means 571 ampere stall current. At 24 Volt it means 1142 Ampere. However it will only be that sort of current, when there is no more any inductance and that only happens when the motor has stopped totally. But you have to stop the mass of the motor + propeller + whatever is the obstruction. Between over current of > 20 Ampere and stalling you have a rise time, due to the fact that the motor needs time to stop, even if this is in microseconds, it will never be faster than the time it takes to measure the over current i.e. 20 uSeconds. Play it safe and take 50 Ampere, you have then still 20 Ampere until you reach your maximum allowable current of 70 Ampere forever or for a long time.

Thank you, I am pleased that we have the time to put it properly together. I like to go to the ADO Elephant National Park end of next week for a week during my birthday.
Bert

 

OK, understood. I'm still trying to understand one more thing. Given the existing design for a 30 amp trip point (I assume at 12v?), how do I know what current it will trip at when using different supply voltages? What is the formula for calculating that?

I'm trying to understand how the unit will behave when I apply different supply voltages (either due to different batteries or due to voltage sag as the batteries drain etc). I thought you were saying the set point would change as the voltage changes but perhaps I misunderstood you on that. Does it remain at 30 amp regardless of the supply voltage?

 

Good question.
The PIC12F675 works from 3 Volt to 5.5 Volt, but it has a 5 Volt regulator feeding it and keeping it at constant 5 Volt with an input voltage of 7 - 30 Volt. Actual 30,5 Volt, because we are using a polarity diode to prevent damage by accidental reversing the plus and minus. THUS !! the 30 Ampere is fixed by the from 7 to 30,5 Volt, by the microprocessor. Most likely also lower than 7 volt, because the IC works from 3 to 5.5 Volt and if the regulator drops below, the 5 volt, it is no longer regulated and it drops to below 5 Volt. Thus any voltage between 3 and 30,5 Volt ) ** you only have a cut-off of +/- 30 Ampere.

That is the first issue. The second issue is that your normal average current will be by 12 Volt maybe??? 10 Ampere by the PWM half open and probably 20 Ampere when the PWM is fully open. But that is at 12 Volt. By 24 Volt you will have the current doubled i.e. 40 Ampere by full throttle. And by 30,5 Volt most likely 47 - 48 Ampere.

Conclusion: With a 24 Volt battery fully charged to 28.8 Volt or higher, you will not be able to use the PWM fully and full throttle, because you will exceed your 30 Ampere cut off as set by you. If you use 50 Ampere, you are safe for the 70 Ampere maximum your motor can handle and you are able to use your system up to 24 Volt.

The microprocessor + fixed shunt resistors dictates the 30 or 40 or 50 Ampere, regardless of the supply voltage. It is your call whether you have to post the IC back to me to make changes, if you discover that the 30 Ampere is too low.

HOWEVER your consumption, (usage current) by the motor will change by the different sagging and battery voltages. Because the motor is not Voltage regulated. Happy???

Is it already snowing in your part of the world?

Bert

 

EK, are you able to measure the current at 12 Volt with the prop in the water? You only have to double it for the 24 Volt usage + proportional for over the 24 Volt. Bert

 

Thanks, that helps a lot.

It is 20 deg C here and sunny today. We just went for a nice day trip in the kayak. About 25 km. Winter is coming but even so we rarely get snow (lots of rain), and if so it doesn't last long.

I looked at my battery current draw in typical scenarios (4s Lipo battery). At 5kph slow cruise current draw is only about 2 amps. At 6.2 kph normal cruise speed the current is around 3.5 amps. My maximum viable speed is 9kph and is around 10 amps. Peak speed before something breaks is just over 10kph and the shaft starts to buckle...even this is only 14 amps.

10kph is quite fast in a kayak.

So I think I'm happy with 30 amp trip point. In fact perhaps I should go lower but have it in the back of my mind to strengthen the shaft to allow running at 10kph so the 30 amp trip point is probably good.

 

I'll just jump in here on some things you may have already considered. You could use your present fuse in series with the electronic one and add a by pass switch to give you more flexibility on break current and also a back up mode arrangement. For running at 10 kph and above, you might consider a longer separate quick release 0.25" shaft/prop which could take that load without buckling. That separate shaft would also serve for an emergency "limp home" mode should the 3/16" shaft fail.

I've been experimenting with a dozen or so breakers on my various outings. I have found a couple that do a fairly good job under real time conditions of stall, and also with labored run with weed tangles. May come in handy since I have 6 micro boats to consider. I live in South Texas, and can go out all year, even in Winter- unless it's raining, windy, or have a rare freeze (no rare snows for several years). It's 33 deg C and sunny outdoors today.

PC

 

No fun to kayak in heavy rain. Hope the nice weather will hold.

That changes the scenario. The 14 Ampere is an average and looking at the oscilloscope, at the 10 Ampere, the spikes are approx double. We do realize that because of the PWM, the current is an average, while the ADC will trip at all the spikes when higher than the set limit. I may agree now, that maybe we should lower the trip voltage to 24 or 25 Ampere.

What about when you will use a 24 Volt battery? Or will you never use a 24 Volt battery? It makes my life easier when only the system will work from 7 Volt to 15 Volt.
Bert

 

Good point.

The reason why I tackled EK's problems first, because I thought it would be the easiest with only 1 tripping point. Also I have to be honest, for weeks, actual more than a month, I have not been able to get my platform working on which I test the brushless motors. Today I have been able to find the 3 faults. Nothing more difficult than to have multiple faults in a system, as it let me run around in circles. Very unusual, but one optocoupler went on the brink, this together with me using a too low wattage globe/bulb in serial to avoid any serious damage during testing of the circuits. Plus, what is very unusual, resistors which suppose to be 15 Ohm (brown green black) were 75 Ohm purple green black, which looked the same, have been mixed by the supplier in their bin. This is not the first time, that I received the wrong items and have to be double careful in the future. As a result I have been changing the software, as I thought it was my software what was not working properly anymore. That let me even more run in circles.

Porta, I will do my best to get your unit going. With the brushless motor test platform now working again, I am now able easier to simulate on the ICD 3 debugger and programmer your circuit with a PIC18F2431, and when everything is working as it should, transfer the program to the PIC12F675. I have to mention that I will be at the ADO Elephant National Park as from next week to celebrate my birthday. (36 degrees Celsius next week Sunday !!) But I will do my best to get some test results before that. . Bert

 

Regarding the 30 amp trip point: It is interesting that you are seeing 2x voltage peaks at the battery. Is this normal? To me it suggests the input capacitors on the ESC are not doing their job. Or possibly your test leads have unusually high inductance. In any case, I think 30 amp trip point seem about right and could be a bit lower if that made things easier.

Regarding battery voltage: I'm using 4s 16Ah Lipo batteries. Fully charged they are 4.2v per cell = 16.8 volts plus some kind of safety margin. I have legacy lead acid AGM batteries I use also.

If I go to the 5/16" 17-7 SS shaft and run significantly higher continuous power I may need higher voltage with a reduction drive but if that causes complexities then it isn't necessary ( I may never get to it anyway and can deal with it separately if it ever happens ).

PC...your suggestions are good as always. Thanks for the ideas.

 

No not at the battery, over the shunt resistor. That is logical, because you have a pwm working for you and that means pulsed current = pulsed voltage (milliVolts) , with an average reading like you had and therefore the peak has to be higher. Your battery voltage, I did not measure high voltage peaks.

Porta has no such problem as he uses a resistor and does the speed differently, therefore it will be easier (I think) to get to the correct values for him.

not a problem, but at 16.8 Volt fully charged battery and using your pwm at more than 1/2 position, closer to full, I may have to retract my 25A and go back to 30 Ampere.

What means substantial higher voltages????? , 2 x 16.8 = 33,6 Volt, yes that needs a switch to compensate for going over the 30.5 Volt. We tackle that problem at a later stage.

I agree, Porta has always some unusual solutions.
Bert