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

Yes Porta, it does help. However before your client can either press the emergency button and comes to his senses, the unit has already switched the motor off. Even if he carries on pressing the reset button, the tests I have now done with various motors, he is never be able to damage your motor. What we can do is to see whether we cannot build a loop in the software that should he presses the reset button 5 times, it switches off for 10 minutes. Would that be a solution?. Let me see whether it takes not too many instructions/time up. Maybe we can make it as part of the first one second. The first one second is basically wasted time in anyway.
Bert

 

Thanks, Bert

PC

 

Hi Porta,

I would do it totally different. I would have the battery and then the toggle switch, then over-current sensing unit, then the motor control and then the motor. No emergency hit button. The client should be briefed, that when the blue LED starts flashing, he must immediately switch the system off and investigate why the motor is being labored. He further should be requested, that if the red flashing light has come on, he also immediately hit the main switch (it will reset the over current sensing unit) Although the over current unit has switched the motor off. He still must verify whether the propeller is clear from debris and free running.

I could make provision for a small buzzer output. I will attach a bad quality buzzer to your unit, but you need to replace it with a better quality one.

Porta, I know that customers could be doing stupid things, the unit will help them and you, to avoid very bad damage due to inexperience with hitting sand or collect debris. I have no solution to somebody who will not follow up your instructions.

I will make it the way the flow diagram is with the flashing LED's and buzzer output (5 Volt, max 20 mA). Are you happy with that?

Bert

 

I have been thinking about your real problem, whereby the propeller hit the soil or sand. The only way I can see it successful measured, is with a special clutch with springs and small magnets and ultra sensitive bipolar hall effect switches like I use for my brushless motors. This must be mounted close to the motor. I assume the motor is above the water level. If something hits the propeller, the spring will turn slightly, the magnet will be at a slightly different position and then the sensitive bipolar switches give a signal, which can be observed by a small micro and either switch the motor off, or sound the alarm. However that needs a lot of time and experiments. But it can be done nice and neatly. There are switches which measure the strength of the magnetism and others senses the polarity changes. The clutch can be very small, neatly over the propeller spring wire on a small pipe with springs. Bert

 

Hi Porta,
Here is a sketch in what I had in mind. A flans welded to the shaft, the shaft cut a little further up, with a pipe and another flans, which contains the holes for the magnets and the holes for the sensor. Any shift in the plates and the sensor will be covered and sent a signal to whatever electronics. You have only one IC socket pin, which could be modified to an digital input pin. This input could be used in the over current unit to stop the motor.
Bert

 

Thanks, Bert

 

Very clever solution Bert, but seems like a lot of work to build and with the fine tuning that would be required.

Below is another way I have used for the propeller hits on bottom where the prop simply slides and never breaks on the bottom or even rocks. The downside is the great inefficiency compared to RC props which greatly limits my range and speed. Broken RC props can cheaply and easily be replaced and it is easy to carry a handfull of the 2 blade type. But I have to spend time adapting them to a shear pin set up which doesn't work all that well....
PC


GO TO POST # 54 to see what prop strikes look like on this link: http://www.boatdesign.net/forums/inboards/prop-shaft-systems-24636-4.html

 

Hi Porta, can't you use a funnel around the prop? like I have made. photo's attached. Would that not be easier and you get more for less power, that is what the experts told me. Bert

 

Not really actual. You first mount the magnets or one magnet and then do the test how far the spring moves out and only then, you drill the hole for the sensor. Bert

 

Bert, thanks for the info and pics.

That setup might funnel more weeds in with the larger capture area and pinch them to the edges/side of duct, as I have no strut in front like you have- to block things away. Struts disturb the incoming water and drop efficiency from what I have read, and I need maximum efficiency. I need to retain the ability to remove the whole propulsion system in seconds, keep the weight low (3#) and propulsion system storage tiny, ie. what I have now. What you have may be similar to a kort nozzle which is discussed elsewhere on the boat list. It is supposed to be slightly more efficient at low speeds/high thrust but less efficient at high speeds, from what I recall.

PC

 

The client should be briefed, that when the blue LED starts flashing, he must immediately switch the system off and investigate why the motor is being labored. He further should be requested, that if the red flashing light has come on, he also immediately hit the main switch (it will reset the over current sensing unit) Although the over current unit has switched the motor off. He still must verify whether the propeller is clear from debris and free running.


PC: Must have overlooked something, what is the "main switch", do you mean my existing toggle, or is there a separate one on the sensing unit?





PC: What happens if the weeds float away from the prop, and the overload cause is cleared with the toggle in the ON position? Would flipping the "main" switch to reset suddenly cause full power to be applied after 1 second? Could it be arranged so that power will not be restored until the toggle has been cycled through the OFF position?

Thanks,

PC

 

Indeed it is your toggle switch. There will be no switch on the unit.

Your client should verify in anyway whether the prop is really clear from debris. But the unit will NOT switch off with slight overload. It will "loop" until the ADC noticed that the current is dropped again. Then the loop will go through the normal loop procedure and the blue flashing light should go off.

Here a very very quick sketch of a free slow running propeller ring, which freely runs around your axle. But because of the 3 mounting pieces, the water should prevent the ring from turning fast. It would be worthwhile to make something like that and judge whether it will solve your last problem. I always had a high marks for technical drawings, but have at present very little time to be a Rembrandt painter today!!

Bert.

Thanks,

PC[/QUOTE]

 

Hi E.K. You have to very patient with me. While doing testing with maximum and minimum currents with a brushless motor, it stalled and due to the force, it jerked, shortened the 24 Volt wires and one of the sensors got the full 24 Volt on the small output MOSfet. The stalling resulted in burning the battery wires to pieces. Although they are only 9 Amphour batteries, the current was horrendous. I am busy to put a new one in, but it is a hell of a job. Because I am using 4 x 110 Ampere MOSfets parallel x 6 legs, my controller and motor survived. Some photos of the works. But so far the testing is going well. I do the testing with my controller until the program is 100%. Only then I load it into the PIC12F675. Bert

 

Sounds exciting...! I have 12 AH AGM battery that is rated 18 mohms internal resistance...over 600 amps with a dead short until something burns up! It's rated for 250 amps for several seconds without damage. I've measured my lithium battery at about 5 mohms internal resistance...shorting it with heavy enough wire might cause an explosion!

It's good to see you are testing the failure modes of your design...!

 

Wooow.

The reason was that my expensive and very good quality Bourns potentiometer started to develop a cut internally. Needles to say that it suddenly jumped to full speed. With the above result. Tomorrow the 300 degrees Celsius glue will be hard and then I can assemble the motor again.

I have been able to make slides of the layout of both boards printout with a laser printer, thus I should be able to make now a decent pcb again.
Bert