Electric motor technology, design and discussion.

[Xoon]

I created this topic for sharing, learning and discussing electric motors and their associated systems. 

 

First thing I want to discuss is cooling systems for EV electric motors. To get the most power out of the motor you can simply overvolt the motor, since more current=more power, this however decreases the lifespan of the motor. The reason why is because electric motors have magnet wire isolation rate to last 20 000 hours at their rated temperature.  Let's take the cheapest type, which is usually rated at 150 degrees Celsius.  Let's say it operates at 150 degrees, and you overvolt it and increase the temperature to 160 degrees, you have now halved the life of the motor to 10 000 hours. Increase it to 170 degrees and it halves again to 5 000 hours, and this continues to the motor burns. 

 

How do you counter this? 

 

You can wind the motor with thicker wires, which creates less resistance, and less heat, but this also hurts the power density of the motor, since thinner wires can be more compact. Thinner wires also increase the amount of turns you can do per pole, increasing the power.

 

You can have the case be a heatsink, though this gets to ridicules levels quickly, making the motor extremely heavy. We have for example a 5hp motor at 140kg, a complete no-go for EVs. 

 

You can use a higher grade isolation, being rated as high as 500 degrees Celsius in some cases. The issue here? Higher cost and a thicker wire for the same resistance. It also has some implications. If you are using a PM motor you have magnets inside, which demagnetize at a certain temperature. You want PMs for compact motors, and the stronger the better. Neodymium magnets are the strongest, but they also demagnetize at 60-80 degrees Celsius, which means that having higher grade wire does not really do anything. You can of course use weaker ferrite magnets that work up to 250 degrees C, but then you lose power density. 

 

If you are a incredibly rich person that wants to make his own toy you can use sliver, which weights less, has less resistance and conducts heat better, at about 77 times the price, compared to copper.

 

Now lastly the best solution in my opinion is cooling. This can be done in two different ways, air cooling and liquid cooling. Air cooling requires the motor to have a lot of surface area, be open and therefor not water or dust proof and to have a powerful fan blowing over it, a bigger fan, means a more cool motor. This however, is about as effective as in cars.

 

Now lets get to the point. Liquid cooling. There are several approaches, we can give the motor a water jacket, and simply pump the water through it. We can have channels than run through the stator and rotor. Or we can (simplified) fill the motor with transmission oil, like they did in the Toyata Prius. Here the rotor circulates the oil as far as I have understood.

 

 

The method I prefer is using channels that run through the stator, closely touching the wingdings. This is simply because copper is good at conducting heat, while steel is not. Though, I have no idea to circulate cooling fluid through the rotor.

 

 

Which brings me to my question. How do you circulate the coolant? 
I was thinking about putting a pump right on the axle.  This seemed like the good idea, until I realized that the electric motor also has to spin in the other direction, which would make all the coolant flow in the other direction too. I guess you can use a valve system to make sure it always flows one direction, but then you need a bidirectional pump which is not as efficient. 
Alternatively, you can use this:

 

 

Does anyone have a better idea of how to solve this? And how to cool the rotor? 

 

Also, would it be possible to use a vortex tube to cool the air before it passes through the radiator? 

 

[Xlucine]

1) Is it really a terrible thing if the coolant flow reverses while you're backing into a parking space? If you're going 70 down the highway in reverse I can see it being an issue, but at low speed for a short time the motor should manage

2) Put another motor in to drive the coolant pump - AFAIK this is what is done for aircon systems. Assuming constant efficiency for the electric motors (i.e. adding the pump load to the big motor is just as efficient as the smaller motor), it'd probably be better for energy economy as you can control the coolant pump speed independently (and turn it off when the motor is cool)

[Meplat]

Board ate my post. 

[Toxn]

My only contribution to the field of electric motors was the suggestion that you put a lot of DC through a bar spinning between two circular plates after getting it up to speed to prevent arc welding.

 

Unfortunately; my plans for a simple railgun electric motor have not been picked up by major electronics companies as of yet.

 

Edit: note that, for the purposes of cooling, my idea nicely dodges around the issue entirely by having the plates ablate into hot plasma after a short while.

[Xoon]

On 15.9.2017 at 11:02 PM, Xlucine said:

1) Is it really a terrible thing if the coolant flow reverses while you're backing into a parking space? If you're going 70 down the highway in reverse I can see it being an issue, but at low speed for a short time the motor should manage

2) Put another motor in to drive the coolant pump - AFAIK this is what is done for aircon systems. Assuming constant efficiency for the electric motors (i.e. adding the pump load to the big motor is just as efficient as the smaller motor), it'd probably be better for energy economy as you can control the coolant pump speed independently (and turn it off when the motor is cool)

1. What I am afraid of is a hot spot in the coolant circuit going back into the motor and demagnetizing the permanent magnets. 

 

2. This is probably the best solution, though a bit heavier. 

 

 

[Xlucine]

Maniacs want to sink charge into lithium batteries 5 times faster than current safe limits:

Quote

New sensor tech for commercial Lithium-ion batteries finds they can be charged 5 times faster

 

Researchers at WMG at the University of Warwick have developed a new direct, precise test of Lithium-ion batteries’ internal temperatures and their electrodes potentials and found that the batteries can be safely charged up to five times faster than the current recommended charging limits. The new technology works in-situ during a battery’s normal operation without impeding its performance and it has been tested on standard commercially available batteries. Such new technology will enable advances in battery materials science, flexible battery charging rates, thermal and electrical engineering of new battery materials/technology and it has the potential to help the design of energy storage systems for high performance applications such as motor racing and grid balancing.

https://warwick.ac.uk/newsandevents/pressreleases/new_sensor_tech/

 

1) It'll be hilarious if after all the money that's been spent on faster charging batteries, it turns out that all you need to get similar charging rates to gasoline is just a better sensor

2) This has the potential to sink an ungodly amount of power into the new tesla lorry. ~5 MW in a charger is absurd