Racing and the humble E-bike

recumpence

1 GW
Joined
Apr 19, 2008
Messages
5,304
Location
On Earth right now. That can change at any time, t
Hey Guys,

I have been working with a few guys lately regarding racing applications of various kinds. Also, there are a few threads running about racing setups and motor heat issues and I wanted to address that in a dedicated thread.

DISCLAIMER:

I am not a racer. I have rarely competed in any organized event. I am sure Luke and others will chime in with track experience. However, I want to layout the basics of racing setups and a few principals to get us off on the right foot.

We are all used to building commuter setups. Commuters differ from racing vehicles primarily in duty cycle as well as comfort. What I mean by that is;

Duty Cycle (this is also addressed as continuous versus peak power output). The duty cycle is the ratio of time a system can be hammered versus the time it must rest. For example, a refrigerator or a commuter car has a 100% duty cycle. It can run all the time. A circular saw may have a 20% duty cycle. So, for every 2 minutes it runs, the saw should sit for 8 minutes to cool. Now, for racing, we need to take the burst rating of a motor and somehow make that its continuous rating (turn the 20% duty cycle rating and trick it into a 100% duty cycle). (Oh, I am speaking about track racing here, not drag racing or hill climbing. Those can be accomplished with a high burst rating or low duty cycle rating alone). The primary issue for our bikes in racing applications is, obviously, cooling. Remember, for general riding, we run our motors with no additional cooling what-so-ever. That is fine for low output. But, for high output, active cooling is needed.

As for comfort, in racing, very little seat padding is needed, and we do not care about noise or a somewhat harsh throttle and brake setup (within reason), or looks, for that matter. What matters in racing is speed and durability, PERIOD!

To make our humble little motors to survive in a racing application (full throttle, full brakes, full throttle, full brakes, repeatedly) we need to remember a couple things;

#1 High RPM with low gearing is our friend. Low KV motors are fine to reduce heat and increase efficiency. But, for maximum power, you gotta spin it fast and gear it down. For example, my 4 turn 3220 was dynoed in delta and wye. The numbers are very telling. In wye the motor peaks at 95% efficiency. However, in Delta, it only peaks at 92% efficient. Yet, the Wye motor peaks its efficiency at only 2.3kw, whereas the Delta motor sees over 90% efficiency from 2.6kw all the way up to 9kw! The Delta motor has nearly 4ft lbs of torque at 9kw and 14k RPM. So, geared down for typical track riding, would net 50ft lbs at the rear wheel!

#2 Cooling. We need some sort of additional cooling. As for the Astro motors I am familiar with, heat sinks work well. However, it is good to pull even more heat from the motors. I am working on an integrated solution for that right now. Remember, for general riding, we hit the throttle once in a while and cruise the rest of the time. In racing, we are on the throttle, and off the throttle, then on, then off again. That is where the heat really builds up. Heat sinks with fans is the minimum needed to make a dent in cooling in racing conditions. Other solutions are even better. :wink:

#3 Overbuild it! For general commuting, we just need to make sure the bike can hold us and get us to our destination with charge to spare. In racing, you need to make sure the system can be beat to death and survive. That is a far different situation.

We are asking a lot out of our street systems. However, they CAN be made to survive. We just need to think outside the commuter box. What seems like stupid overkill on the street, may be barely adequate on the track. Even brakes are an issue (if we can get the motors to survive). I am working on a braking solution right now as well.

Racing is a matter of finding the absolute limit of a given component or group of components, then running that component or system right up to 99% of that number so it just barely survives the entire race. That requires testing or overbuilding to the point that you know there is a large margin to work with. However, that margin usually in expressed in added weight. That weight can slow you down.

At any rate, I am not a racer. But, these are some general guidlines to help out. I could post 10 times more information. But, for now, these are some things to chew on. I am sure others will chime in soon.

Maybe we need a racing section? Much of what I advise people on regarding their systems would be totally different if they were asking about a racing setup versus and commuter. Heck, I am building a racing trike for a customer right now that is going in a far different direction than any street commuter...........

Matt
 
Great topic Matt, these brushless motors can produce outrageous burst power levels yet wouldn't survive unless a 10 sec drag race.

Yet, the Wye motor peaks its efficiency at only 2.3kw, whereas the Delta motor sees over 90% efficiency from 2.6kw all the way up to 9kw! The Delta motor has nearly 4ft lbs of torque at 9kw and 14k RPM. So, geared down for typical track riding, would net 50ft lbs at the rear wheel!

I want to understand the RC motors and try to study the topic but can't seem to find answers, honest, I read this forum alot and try hard to relate info, yet am learning ever so slowly (the Drive Calculator software is driving me batty) So, above;

Q1: if a 4 turn produces 50ft lb, what would an 8 turn produce with similiar rear wheel speeds? (even a little more/less or alot more/less response would help)
Q2: For continous running, once heat sets in (lap3), are the 2 motor performance limits much different, or does heat even out final power output?
Q3: how does limited voltage (12S), such as with CC HV160 controller affect hi/lo Kv or Delta/Wye considerations? (an 8T in Delta would be pretty hi-Kv, yet would it handle similar amps to a 4T in Wye)

You may not race, but I look to folks such as you for answers. Even though I have clear stable goals, I don't know enough to plan my next build.

Excited to hear your cool Astro plan; couldn't put together an RC build with out all your prepared packages.
Will aka SoSauty
 
All good points. I have raced many things in my life from dirt bikes, to jet skis, to cars. HEAT is the biggest issue in each one, either you are trying to get enough, or trying to keep it to a minimum. For this application it is obviously keep it down and the best thing I can think of (for the motors at least) is liquid cooling. There is a thread here about a small motor with a water cooled mounting base that caught my attention. I think this TYPE of configuration would be well suited to our bikes. You need to get a nice compact water chamber that will effectively pull heat from the hottest part of the motor, the windings. Then you need to figure out how and where to mount a heat exchanger/tank for the water or coolant. The last part is some kind of simple pump. I would do a mechanical one simply because electricity is at a premium and I think tapping off the chain or something would be pretty easy and barely make any noticeable extra consumption. The real challenge of this is making it all compact and fit cleanly on the bike, and that is where the time will be consumed in the designing process....

That being said, multiple gear ratios will GREATLY extend the heat tolerance of any motor that is currently running a single speed due to the lower loads it will see on takeoff and in the corners where not as much top speed is needed but the power to accelerate out of them IS....
 
I used to machine liquid cooled car audio amplifier housings (15 years ago). I used a simple 12v DC pump designed for a pop-up camper and an aluminum tranny oil cooler as a radiator. I could not believe the difference it made!

I am working on a new drive system that pulls heat from the motor far better than my current V4 unit. This is not an update, it is a completely "From Scratch" new system.

I have considered liquid cooling. I may do some research in that direction. However, just being able to pull directly off the can, rather than through the faceplate only will make a big difference.

Matt
 
Great topic Matt. I am happy to see you get involved in it. Those of us that run an Astro know the feel of the raw power. When you have that kind of power it is only natural to want to put it to test in race situations. We all know that the Asto 3220 makes the same max available power no matter what wind it is, you just have to adjust the amps and volts to get the same numbers. However if you limit yourself to the 48 volts of an HV160 then it is obvious that you need the faster wind to get the maximum power out of it. In my head faster winds means higher amps and higher amps means more heat.

When speaking with Astro Bob, he told me that the Wye wind would always be slightly more efficient than a delta wind, and the Dyno bears that out. I am sure that if you could up the voltage on the Wye wind it would make the same max power as the delta on the dyno.

I have learned to try and always oversize what I need to get a reliable running electric motion. If I think I need 2000 watts, I buy batteries, controllers and motors rated for 4000 watts, then run them at only 2000 watts and I usually get good reliability out of them, maybe in race situations we need to up the ratio to 4:1

On a related note, Matt, how are those oversized Astro motors coming along? I think they would be the solution we need!
 
recumpence said:
#1 High RPM with low gearing is our friend. Low KV motors are fine to reduce heat and increase efficiency. But, for maximum power, you gotta spin it fast and gear it down.

I don't think this is so cut and dried. There are different advantages to both high and low Kv setups. The Joby motors are an example of how the dimensions of a large diameter low Kv motor can be advantageous for passive heat dissipation.

I'd want to do the power density comparison at the final output shaft with any reduction apparatus included in size and weight calculations.
 
I'm looking toward a rebuild and waiting ever so patiently!

just being able to pull directly off the can, rather than through the faceplate only will make a big difference.
Ummm hmmm. Agree, put some soft aluminum around the can and heat really flows away/power stays :p

Rather pleased with what the 3210 can put out. I never surpased 31mph on Willow Springs track last week, thought I was dialed in for 31mph, yet due to higher than expected load voltage, large tires, and single speed; found I was geared for 33mph, alot of topend torque went to waste. Think what the 'lil Astro could do with a properly geared 3 speed :idea: I didn't so much as have pedal 1st gear working!
 
gogo said:
recumpence said:
#1 High RPM with low gearing is our friend. Low KV motors are fine to reduce heat and increase efficiency. But, for maximum power, you gotta spin it fast and gear it down.

I don't think this is so cut and dried. There are different advantages to both high and low Kv setups. The Joby motors are an example of how the dimensions of a large diameter low Kv motor can be advantageous for passive heat dissipation.

I'd want to do the power density comparison at the final output shaft with any reduction apparatus included in size and weight calculations.

Even a Joby motor will put more power to the ground if spun up faster, then geared down. That is why F1 cars run such insanely high RPM.

I am talking absolute power to the wheel/s here.

Matt
 
A motors physical flux gap dimensions and the materials flux saturation density determine the torque its capable of reaching.

From there, the power is entirely proportional to the RPM you can get the motor to spin.

An astro at 5krpm is only capable of aproximately half the power of an astro at 10krpm.
 
Just like gas motors, you have a given torque per cubic inch potential, once that is reached, you have no choice but to spin it faster and gear down if you need more power...
 
The power path is going to include at least one opportunity for reduction unless its some kind of hubmotor, but how much reduction is too much (weight, complexity, etc.)?

Can the Joby/Astro Kv analogy be applied to 100K+ RPM motors? What are the practical limitations, variables, and break even points? I'm just trying to think in concepts using extremes.
 
Well, as RPM increases the parasitic losses inside the motor will tend to increase. Windage, hysteresis loss in the stator, bearing friction, etc. Also, tolerances would become more and more important to control balance and vibration. So while power continues to increase with RPM it will gradually get more difficult and more expensive to realize. Actually, it's probably more accurate to say that power density increases with RPM, since you could get the more power at the same RPM by increasing the motor size as well. An F1 engine produces a few times the power of an average car engine in the same displacement, but they also have a very short lifespan and probably cost as much as your house. Thus, the tradeoff.

Heat would also be an issue, obviously. Even if you can maintain the same efficiency the absolute losses will increase as power output increases. The higher the power density the more challenging the heat is to dissipate.
 
gogo said:
What are the practical limitations, variables, and break even points? I'm just trying to think in concepts using extremes.
Eddy current losses in the stator laminations go up as the square of the shaft speed so, beyond a certain point, they will dominate.
 
rhitee05 said:
Well, as RPM increases the parasitic losses inside the motor will tend to increase. Windage, hysteresis loss in the stator, bearing friction, etc. Also, tolerances would become more and more important to control balance and vibration. So while power continues to increase with RPM it will gradually get more difficult and more expensive to realize. Actually, it's probably more accurate to say that power density increases with RPM, since you could get the more power at the same RPM by increasing the motor size as well. An F1 engine produces a few times the power of an average car engine in the same displacement, but they also have a very short lifespan and probably cost as much as your house. Thus, the tradeoff.

Heat would also be an issue, obviously. Even if you can maintain the same efficiency the absolute losses will increase as power output increases. The higher the power density the more challenging the heat is to dissipate.

That is exactly why I setup systems for people (and advise those who are not buying anything from me) based on their intended use and needs/wants. If an individual says they do not mind a little more noise, and higher wear, but really want performance, I run the KV higher and gear it down. If he wants absolute reliability and is willing to sacrifice performance, I run the KV lower and gear up.

I rarely ever advise running up near 15krpm (with an Astro) unless racing is the application. But, even 12krpm, is low enough to provide reasonable reliability and still very good power when geared down. This is a great compromise between power and reliability. Running below 7krpm is nice and quiet, efficient (at low power levels) and VERY reliable. But, you will sacrifice power.

Matt
 
Ok so we all agree that we need the efficiency of our high rpm motors, but how do we actually make them more reliable? Why don't we have a little brainstorming session with new ideas??
 
I had an fan idea for out runners. It's a Squirrel Cage type fan that bolts to the end of the motor can and runs over and outside of the motor can. Below are some examples of the kind of fans I am talking about. The flange face is drilled to attach to the place where a propeller would go in a Turnagy motor or it's bored out to shaft size. Has to be made to run at 6,000rpm or more...I was not able to find one that had a flat flange and was the right size to just fit over the motor can... What do you guys think?

19285k54p1s.png17750a_235955.jpgcagecwLg.jpg
 
Genious idea there! If you spin it the right way, itwould pull the air away from the motor! Ill have to look into that with my 12 FET coming soon!
 
Whiplash said:
Genious idea there! If you spin it the right way, itwould pull the air away from the motor! Ill have to look into that with my 12 FET coming soon!
Well, thank you!

Now that I am thinking more about it one could make a casting and bore it out to fit the outside of the motor can and screw it to the can using the existing threads used to hold the motor can to the end flange and bearing flange. There could be a right and left hand version so it would cover either rotation.
 
Racing ebike.

Massive front brake?
Check.

Custom super strong CroMo frame?
Check.

60kw capable battery?
Check.

400amp 120v Kelly?
Check.

15kw brushless motor?
Check.

Drive system made from #428 motocross bike chain and sprockets?
Check.


Very sticky tires made for leaning way over?
Check.


goldenon.jpg
 
liveforphysics said:
A motors physical flux gap dimensions and the materials flux saturation density determine the torque its capable of reaching.

From there, the power is entirely proportional to the RPM you can get the motor to spin.

An astro at 5krpm is only capable of aproximately half the power of an astro at 10krpm.

That's true if we're talking about the short term peak.

For continuous running, the limit is usually set by the heat dissipation.

It's a moot point which condition we should be looking at for racing. Is 10 minutes a short term peak, or long enough to melt a motor?

+1 for a racing section, but we need to decide whether it's e-bike racing or all e-racing, and we do have to guard against proliferation of sections.

Nick
 
Back
Top