More kV = More power? What are the trade-offs?

[Miles]

All teams obviously want to win and are eager to have the fastest machine. Everyone concludes: go for the highest no-load rpm motor. Is this the best strategy? (considering the racing conditions where one of the main failure is an overheated motor or battery)

Interesting...

If all the bikes are, effectively, geared the same.... Assuming all the windings are identical and disregarding control discrepancies, the motor with the highest no load speed will have the lowest continuous torque capability.

I guess we need more detailed information about the race to guess the best strategy, though....

 

[liveforphysics]

I was reading about the agni mods being done for the TTXGP.

They wanted as close of flux gap as possible, and as high of KV as they could get. I think the hot ticket was doing some precision post-build machinging on the rotor flux faces, and shaving the faces of the case and installing approprate shims to get the flux gap as close to zero as possible. This of course lowers the KV, and I think they did wider brush contact areas on the rotor by 1-2 winds in the advanced direction to get the KV to jump back up again.

Unlike an RC motor or something though, your rotor is made of copper... so you don't really have the RPM stability/capability needed to get much of an advantage over another racer by jacking the KV up.

 

[rolf_w]

If all the bikes are, effectively, geared the same... disregarding control discrepancies...

the racing rules:

• battery peak voltage is 59V
  controller peak capacity 450A , set max. current 320A
  batteries mustn't be swapped, they have to be charged between the races (e.g. supermotard up to 8 races a day)
  brush plate mustn't be adjusted during the race (available are so called 'booster brush plates' which allow the rotation of the brush plate whilst driving; here a video demonstrating the booster)
  any gearing ratio is allowed (Quantya uses two stages, see e.g. here)

of course:
those teams with low Kv motors change to smaller sprocket to gain max. speed.
before the race, all teams try to guess the proper setting (advance) for the brush plate, in order to have best overall performance

are these the best strategies?

All teams have strategies/components to overcome the thermal problems of the motor (i.e. cooler, heat pipes, ...). The causes for overheating are many e.g. without any mods the individual brushes reach very different temperatures due to different Ri (production differences and bad matching); the new high-rpm Agni motor sparks under load and >4000rpm... etc.

advancing the brushes results in higher max. speed but also in more heat. All the teams are very scared about this. (what is the predictable explanation? higher Kv = higher losses?)

In space vector control of PM sync motors field weakening is a technique to increase Kv (lower magn. flux) - does advancing of the commutation pattern do the same?

 

[johnrobholmes]

Kigher KV = higher amp draw = more heat


Interesting rules. That is a very low voltage restriction.

 

[liveforphysics]

When pack voltage and controller current is pre-determined...

If you put 10kw into a 93% efficient motor, you've got 700watts of heat if the motor is 10kv or 500kv...

Kigher KV = higher amp draw = more heat


Interesting rules. That is a very low voltage restriction.

 

[johnrobholmes]

You are assuming they ride the amp limits 100% of the time. Modifying a motor to have a higher KV will increase the average amp draw, all else the same and assuming there is ANY loads below the regulated threshold.

 

[rolf_w]

...Modifying a motor to have a higher KV...

As explained earlier: we don't modify the motors - the Kv of the same motor type varies up to 10% due to variation of the PM's B-field (we assume that's production tolerance). Thus the bike manufacturer & racing teams select the highest no-load-rpm motors to be mounted in the racing bikes.

Given the racing rules (max current and voltage limits), the possibility to set the brush advance and the gear ratio and always running the motors at their thermal limits, a highest Kv might not be the best strategy to select the motor. A lower Kv and shorter gear ratio might produce less heat at the average speed? (of course we don't run the machines all the time at 20kW during a race - the average power is more around 10kW)

To win a race is thus a very delicate procedure: balance the risk of an overheated motor/battery with highest max. speed vs. highest torque; battery voltage/ energy balance over an entire racing day. The pilots have to tune their driving style and tactics according these electromechanical realities...

 

[Miles]

Given all of the above, I think choosing the motor with the lowest Kv makes more sense.....

 

[liveforphysics]

Given all of the above, I think choosing the motor with the lowest Kv makes more sense.....

With the fixed voltage and controller currents, I was thinking the same thing.

 

[johnrobholmes]

Whether the motor has a timing change or just a construction difference to get a different KV, it doesn't really matter. "Modification" wasn't meant to imply you are hacking them up.


I agree, since the controllers are amp limited it will probably suit you best to use the lower KV motors assuming you can hit the amp limits with it.

 

[swbluto]

I agree, since the controllers are amp limited it will probably suit you best to use the lower KV motors assuming you can hit the amp limits with it.

Yeah, I was thinking that was the assumption in the previous posts. Otherwise, I was thinking, BLASPHEMY! With a sufficiently low kV motor, you could cause the motor current to go below the rated current limit with the motor's increased resistance, so there'd be less power going out, so there'd be some optimum value that's at neither "extreme".

 

[swbluto]

I think I'm starting to find more tradeoffs with high kV. A higher electrical RPM reduces the commutation period which reduces the amount of time for "phase current" to ramp up to the level predicted by phase resistance; with a sufficiently high L/R ratio, this can significantly reduce the predicted average phase current and thus torque. I had this nasty surprise when I was trying to predict performance with a 9c motor - I was pretty ecstatic thinking I could extract 30 mph up a hill with a 9C, but I double checked with ebikes.ca's model that includes inductance, and it was a pretty nasty shock. The predicted speed fell down to ~22 mph, which is a pretty big difference. It looks like I'm looking back at my RC motor, though I still don't think that'll be ideal for a cross-country trip due to legality and reliability.

However, for most RC motors, I *think* the L/R ratio is pretty small so this might be negligible. Someone needs to measure L or find the L/R time constant.