Man oh man. Where do I start? Ok let's start here. The model s can not accelerate any faster from ~ 40mph and up because it is battery limited and no amount of gear changes will improve that. Simply add a better battery for drage racing and it will accelerate from 40-155mph faster.
Conclusive proof gearboxes are awesome.
- Thread starter Teh Stork
- Start date
speedmd
10 MW
For normal driving and normal driving speed range, the gear box idea is of limited value if the motor is sized correctly. We went through most all of this in the field weakening thread. Unless you have a rock crawler that you want to also drive at Le Mans it makes no sense. You need to able to use the crazy amount of torque developed with the low gearing. Tractors are another possible example if they are also used at highway speeds along with pulling stumps.
motomoto
1 kW
- Joined
- Jun 28, 2010
- Messages
- 458
I think a lot of the conflict of ideas is that everyone has a different viewpoint. I am a motocross guy who lives to climb impossible
hills. Someone else is thinking what is needed on asphalt on the flat in a car. Totally different applications.
I would try to do what Luke speaks of but do whatever it takes to meet my needs. Not everyone can design a motor for a
specific application. We have to use what is out there and band aid the setup however needed.
hills. Someone else is thinking what is needed on asphalt on the flat in a car. Totally different applications.
I would try to do what Luke speaks of but do whatever it takes to meet my needs. Not everyone can design a motor for a
specific application. We have to use what is out there and band aid the setup however needed.
nutnspecial
10 MW
MD and Arlo, it might be important to first clarify the stated goal. I didn't mean to impress a goal of improved acceleration within the initial 1:1 ratio. . . . that already kicks fastest supercar ass priced 3x, up to ~100mph . . .
After maybe 80mph/ .125mi tesla begins to come undone, and it just gets worse.
This is to another 700hp car that uses 'gearbox', and has that highly inferior IC torque curve. I do assume people commenting observed some drags and compared vehicles.
Seems like exactly a case for considering at least one additional TALLER gear. Sure, consider the best combo of more volts and amps in leu of gear #2, and consider bigger better motors and controllers too.
After maybe 80mph/ .125mi tesla begins to come undone, and it just gets worse.
This is to another 700hp car that uses 'gearbox', and has that highly inferior IC torque curve. I do assume people commenting observed some drags and compared vehicles.
Seems like exactly a case for considering at least one additional TALLER gear. Sure, consider the best combo of more volts and amps in leu of gear #2, and consider bigger better motors and controllers too.
DanGT86
100 kW
If you added that extra gear to the tesla to increase its top speed then you will have reduced total system efficiency like they keep saying. If you increased the size of the motor you have increased the cost. The scenario with the tesla is no different than adding a gearbox to a 5kw motorcycle to get trials use then hwy use for the way home. It would work and save my tiny motor from melting but would not be be more efficient than improving the motor or controller.
What the no gearbox guys keep saying is that if Tesla wanted to get higher top speed they would be better to change the motor design in a way that provided a higher top speed and then hit it with higher amps at low speed to get the acceleration back.
Luke said it far back in this thread that I hope I'm paraphrasing correctly: we need to ditch the misconception that hitting a high kv motor with huge phase amps at low rpm is bad for efficiency.
The reason they tesla doesn't retool fro high speed is because the money and design time is not worth it to provide a feature that customers don't commonly use. In this case that feature is high top speed.
Apparently we have all been conditioned to think a certain way because of the old days of very limited motor choices and weak controllers. We continued to think gearing is the secret because its easier for a guy at home with a sprocket catalogue to band-aid a shortcoming than it is to design a new motor. Doesn't mean the "design a different motor" camp is wrong.
What the no gearbox guys keep saying is that if Tesla wanted to get higher top speed they would be better to change the motor design in a way that provided a higher top speed and then hit it with higher amps at low speed to get the acceleration back.
Luke said it far back in this thread that I hope I'm paraphrasing correctly: we need to ditch the misconception that hitting a high kv motor with huge phase amps at low rpm is bad for efficiency.
The reason they tesla doesn't retool fro high speed is because the money and design time is not worth it to provide a feature that customers don't commonly use. In this case that feature is high top speed.
Apparently we have all been conditioned to think a certain way because of the old days of very limited motor choices and weak controllers. We continued to think gearing is the secret because its easier for a guy at home with a sprocket catalogue to band-aid a shortcoming than it is to design a new motor. Doesn't mean the "design a different motor" camp is wrong.
nutnspecial
10 MW
nutspecial said:
Initial post and 2 linked drags. . . .Followed by LFP post of:
Which makes no sense to me. Volt and/or amp limitation doesn't dictate the useful possibility of a single additional TALLER gear for this case afaik.
Also, I fail to see how going '1spd improved range, improved 0-60mph, AND got higher top speed'.
Edit// Ok thanks Dan, your above reply gives me something to grab onto.
DanGT86
100 kW
nutspecial said:
Glad it helped. I am seriously just starting to believe it myself. There are some pretty impressive resumes arguing against gearboxes so I am inclined to believe them. I have also noticed the high powered single speed builds all seem to be so wheelie prone that low end torque doesn't seem to be a problem. I think with the proper suspension the deathbike would handle a trials stage or hill climb without melting.
I think we owe the lack of gearboxes to the unbelievably flat torque curve and efficiency curve of electric motors in general. Area under the curve is really what you are after in a variable speed vehicle. Like motomoto just said, sometimes we just have to work with whats out there or what we got. I think there will continue to be really nice builds with gearboxes that meet the builder's needs. We just need to be careful not to credit the band-aid as scientific evidence.
Nutspecial. You are failing to realize that the motor and controller are working as a buck converter. As the speed increases the demand from the battery increases to keep the same acceleration. So if you gear the Tesla different it will not accelerate any faster it is battery limited.
I have dyno graphs from my wife's zero I built her. It accelerates hard from 0-170kmph and it has no transmission and that's a higher top speed then we want/need. At 0 its very low current from the battery and the battery current increases on a strait line to the top speed. Phase (motor current stays the same). The motor current is what determines the torque which is also the acceleration. So as you increase the rpm the battery current and there for power increaseses if you want a Tesla to accelerate faster at higher speeds like 60mph and above YOU NEED A MORE POWERFUL BATTERY! I will post graphs later and maybe help you understand.
nutnspecial
10 MW
Agree
Agree
This would be opposite. It would be gearing UP a 700hp motor with plenty of capacity to do so. Keep initial 1:1 that does great, and add 50-100% taller 2nd, possibly better competing at top speeds, reasonable weight, cost, launch, cargo capacity, and eco friendliness over similar results of larger batteries and controllers and motors.
After racing mclaren, pit the two teslas against each other and compare the 7 criteria.
Agree
Agree (When overvolting, overamping can offset. Provided motor(s)/battery/controller(s) is/are sufficient?)
Agree
I understand that, but consider it ill advised to rely on in objective argument at much lower levels.
It so happens my inital 'larger idea' is neither pro mechanical nor pro E advantage.
DanGT86
100 kW
For what its worth, the tesla in the tesla vs. Ferrari video is 1000lbs heavier and has 40 less hp. According to ET calculators the Ferrari should beat the tesla by roughly a second in the 1/4 mile. Drag racing involves power to weight. Torque isn't really a factor. No gearing in the world would change the weight or power of that tesla so no transmission would have changed that race. Tesla did not get overtaken by lack of gearing. The fact that the Tesla started off ahead is likely do to traction at the start. If both cars were gas powered with the same power to weight it would have gone exactly the same way.
DanGT86
100 kW
In the early days of EV drag racing. Cars like the white zombie used to cross the finish line at relatively low top speeds compared to the ET. The hp vs. weight predictions were a little off compared to gas racecars. Due to controller motor and battery tech not being good enough they would gear for lower speed to keep from melting components and then plateau before the finish line. The lack of high powered EV components produced a peaky power output so direct drive EV racecars would suffer from lack of gearing. Fast forward to some better tech and extremely high amp capable controllers and the power is now the same all the way down the track. If you look at the current fastest EV drag cars they run almost exactly the same speed and ET as there gas counterparts of equal weight and power. This means they are able to apply full rated power the whole time with less or no gear changes compared to the gas cars that need gear changes to stay in the power band.
So if there is a drastic power decrease in the tesla's dyno curve that occurs before the vehicles top speed then yes maybe a transmission would help. But again, that's assuming they wouldn't change another factor like motor design battery choice or limiting as Arlo is probably screaming right now.
If you can achieve the max power of your system throughout the required speed range without melting it you just don't need a transmission with changeable gearing.
So if there is a drastic power decrease in the tesla's dyno curve that occurs before the vehicles top speed then yes maybe a transmission would help. But again, that's assuming they wouldn't change another factor like motor design battery choice or limiting as Arlo is probably screaming right now.
If you can achieve the max power of your system throughout the required speed range without melting it you just don't need a transmission with changeable gearing.
nutspecial said:
Increase the cost VS what? If you add a transmission you increase the cost as well. What do you think its cheaper a slightly bigger motor with a bit more copper and steal or a whole transmission and all the cost associated with it.
DanGT86 said:
NOPE wrong again. After Tesla first announced the P85D and P90D with Ludicrist upgrades they had listed them at 762hp but that is false and they had to relist the power figures at 532hp for the P90DL and just a bit less for the P85DL
https://www.teslamotors.com/en_CA/models
That's 198hp less basically 200hp less
http://auto.ferrari.com/en_EN/sports-cars-models/car-range/f12-berlinetta/#specifications
It is battery limited. You cant gain any more acceleration below ~30mph with a tesla because it is traction limited then above 30mph it becomes battery limited. Because in order to keep acceleration that high you need to increase battery current (power) as you increase speed.
Here is the first run with the Split phase Zero 75-7 I put together running 2 size 4 sevcons for a drag bike I'm helping build.
This runs no transmission just a sprocket from the motor through a chain to the rear sprocket on the rear wheel.
And here is a Run from my wifes bike. Which runs a belt form the front sprocket to the sprocket on the rear wheel. Both of these accelerate very hard and evenly from 0-top speed obviously the one with more power and 2 controllers accelerates harder
But notice at the speed increases the current from the battery increases. I will try to find one that shows the torque later as well. But the torque is a strait line from 0-just before top speed with the same torque number the whole time.
This runs no transmission just a sprocket from the motor through a chain to the rear sprocket on the rear wheel.
And here is a Run from my wifes bike. Which runs a belt form the front sprocket to the sprocket on the rear wheel. Both of these accelerate very hard and evenly from 0-top speed obviously the one with more power and 2 controllers accelerates harder
But notice at the speed increases the current from the battery increases. I will try to find one that shows the torque later as well. But the torque is a strait line from 0-just before top speed with the same torque number the whole time.
Attachments
macribs
10 MW
- Joined
- Jul 22, 2014
- Messages
- 3,702
Arlo1 said:
IIRC you seem to have double the power on that motor then original? From 40 up to 82 hp. Brutal.
You are running two controllers on that motor, did you rewind the motor to be a 6 phase motor?
I don't understand how you can run two controllers and double the output power?
liveforphysics
100 TW
Since there seems to be confusion on the topic, model S does spin it's tires until the speed it hits it's battery current limit.
All a transmission can add to that is weight and loss.
Ultimately the EV powertrains roll is to convert stored electro-chemical potential energy into a tractive effort.
The parts of the drivetrain that accomplish this is the copper and iron and magnets. The more copper you have, the less resistive loss for the conversion to occur. The more iron you have, the lower the specific flux density and lower rate of hysteresis and lower eddy potential.
This is why big motors going slow have the potential to be most efficient, as well as make the conversion from the energy stored in your pack to your intended tractive force for the application with no further additions of losses or parasitic weight and moving parts to fail.
All a transmission can add to that is weight and loss.
Ultimately the EV powertrains roll is to convert stored electro-chemical potential energy into a tractive effort.
The parts of the drivetrain that accomplish this is the copper and iron and magnets. The more copper you have, the less resistive loss for the conversion to occur. The more iron you have, the lower the specific flux density and lower rate of hysteresis and lower eddy potential.
This is why big motors going slow have the potential to be most efficient, as well as make the conversion from the energy stored in your pack to your intended tractive force for the application with no further additions of losses or parasitic weight and moving parts to fail.
DanGT86
100 kW
Very tidy explanation LFP.
Wow. Didnt realize it was that low. Either way that further emphasizes the point that the Tesla was going to get killed in that race. It didnt lay down at speed due to lack of transmission. It just got beat by a more powerful lighter car and no gearing in the world can magically create power to change that for the tesla.
Arlo1 said:
Wow. Didnt realize it was that low. Either way that further emphasizes the point that the Tesla was going to get killed in that race. It didnt lay down at speed due to lack of transmission. It just got beat by a more powerful lighter car and no gearing in the world can magically create power to change that for the tesla.
liveforphysics said:toolman2 said:-This all reminds me of debating with Luke maby 5 or 7 years back when he wanted to use 100-80 hobby king motors at 15-20 000rpm and use something like 6 tooth sprockets, ceramic bearings and wrap the can in carbon fiber to stop it exploding, he refused to accept the existence of eddy current losses that i had actually measured, that might limit the useful rpm. Those days when you didn't listen were special, don't you think Luke?
I was so confused, and you and Miles and Bigmoose and many other were offering excellent advice I was foolishly ignoring.
I've had a dyno and track educate me the hard way since then, and now I appreciate core losses and how to make good use of iron and copper.
I also was once heavily bias to axial flux motors, but now I respect inrunner, outrunner, axial all as equals just each with different packaging options.
Good of you to admit that Luke.
My concern now though is (like Arlo said re: huge going nowhere thread) that there has been a lot of confusion created for the the average reader.
-ie, one could say this:
'After all any frictional loss and built in failure modes from a using a loaded scrubbing tire and multiple bearings that are ultimately not giving us the direct forward force that we desire is a poor design choice that is ultimately doomed to fail into chunks of broken rubber and metal'.
This is more of a philosophical or futuristic outlook, and therefore cant really be wrong,
but when taken to the next logical conclusion has us all skimming along on magnets with the conductors in the ground in an apparently perfect world. (that i dont think many of us would actually like)
What initially has the appearance of being open minded and forward thinking, is actually just ONE very narrow and limited possibility, that if taken on by folks here that you have a substantial effect on (who are likely not motor designers!) results mostly in stagnation (analysis/design paralysis) whilst waiting for availability rather than doing what we humans tend to want to do -deal with, experiment and enjoy what IS available to them, and currently gives the best performance and efficiency.
liveforphysics said:
Ok, in an attempt to clear this up and confuse people less: That being the case its clear then that the Tesla has zero need for a gearbox then, ie it can wheelspin up to the battery current (and therefore power input) limit and thats it.
But heres the critical bit Luke and we need the answer to contain a yes or no:
Is it correct to say that if it was to be highly loaded at relatively low speed, like say a race that is a decent grade uphill, OR say a much higher speed race with lower load, like down a long straight downhill. -That there is likely to be a benefit to either speed or efficiency by changing the drive ratio that it currently has.
liveforphysics
100 TW
toolman2 said:liveforphysics said:toolman2 said:-This all reminds me of debating with Luke maby 5 or 7 years back when he wanted to use 100-80 hobby king motors at 15-20 000rpm and use something like 6 tooth sprockets, ceramic bearings and wrap the can in carbon fiber to stop it exploding, he refused to accept the existence of eddy current losses that i had actually measured, that might limit the useful rpm. Those days when you didn't listen were special, don't you think Luke?
I was so confused, and you and Miles and Bigmoose and many other were offering excellent advice I was foolishly ignoring.
I've had a dyno and track educate me the hard way since then, and now I appreciate core losses and how to make good use of iron and copper.
I also was once heavily bias to axial flux motors, but now I respect inrunner, outrunner, axial all as equals just each with different packaging options.
Good of you to admit that Luke.
My concern now though is (like Arlo said re: huge going nowhere thread) that there has been a lot of confusion created for the the average reader.
-ie, one could say this:
'After all any frictional loss and built in failure modes from a using a loaded scrubbing tire and multiple bearings that are ultimately not giving us the direct forward force that we desire is a poor design choice that is ultimately doomed to fail into chunks of broken rubber and metal'.
This is more of a philosophical or futuristic outlook, and therefore cant really be wrong,
but when taken to the next logical conclusion has us all skimming along on magnets with the conductors in the ground in an apparently perfect world. (that i dont think many of us would actually like)
What initially has the appearance of being open minded and forward thinking, is actually just ONE very narrow and limited possibility, that if taken on by folks here that you have a substantial effect on (who are likely not motor designers!) results mostly in stagnation (analysis/design paralysis) whilst waiting for availability rather than doing what we humans tend to want to do -deal with, experiment and enjoy what IS available to them, and currently gives the best performance and efficiency.
I see powertrains with a slightly different perspective, but maybe that's because my own life does involve creating scratch built power train solutions rather than implementing existing off the shelf parts.
A friend I gave my old Honda Insight to recently got it running as a pure EV from an old Zero motor. It uses the OEM Honda Insight transmission to enable the swap to work, and without that transmission the vehicle would have to pick being able to climb hills or go on the highway but not both. In this example of using something designed to move a motorcycle to move a 10x heavier car around, the transmission is a critical element to make the project possible, and without it he wouldn't have an EV to be driving today. From a pragmatic perspective for DIY'ers making things work with whatever they've got access to use for parts, I whole heartedly agree with you that it's good to just get more functional EV's in the world through any functional electro-mechanical mechanisms.
Yet, if we are going to mature EV designs to shine with the most performance and least wasted energy and most robust topologies, it's going to take some folks not settling for piecing together off-the-shelf options.
It's the paradox of total contentment in what can be done today being the conclusion of innovation and evolution, yet total discontentment of what can be done today causing stagnation. Somewhere between contentment/discontentment there is a reasonable path forward for folks who want to create the future of EVs.
I am super thankful for your amazing work with so many awesome EV's and the education you've helped share with so many Toolman!
ATB,
-Luke
liveforphysics
100 TW
toolman2 said:But heres the critical bit Luke and we need the answer to contain a yes or no:
Is it correct to say that if it was to be highly loaded at relatively low speed, like say a race that is a decent grade uphill, OR say a much higher speed race with lower load, like down a long straight downhill. -That there is likely to be a benefit to either speed or efficiency by changing the drive ratio that it currently has.
I see where you're going there and I'm glad you brought it up. To compensate for the big motor at high RPMs and light loads, Tesla used an induction motor so the core losses are highly variable and load dependent rather than PM system that flips the orientation of the iron around every time the magnet moves over the tooth. To further alleviate the wide performance envelope losses, by using a large induction motor in the rear and a small induction motor in the front, when they want to cruise down the long straight downhill in your example they completely stop powering the rear motor, and it's core-loss reduces to windage and friction alone, and however much iron core loss is in the stator and rotor becomes mute (they call it "torque-sleep mode").
This is a prime example of getting to have your high power acceleration cake, and still get your watt-sipping cruise cake, and also a prime example of what can happen when folks aren't satisfied with the status-quo powertrain limitations and apply themselves towards creating some better options.
nutnspecial
10 MW
Hmm, yes/no? ^^^ :wink:
_____________________
From earlier posts of dan/arlo/luke:
So 1000lb heavier and 200hp underpower, the S technically performs even more admirably vs mclaren, though it will be perhaps harder to improve perfomance competitive to mclaren in a balanced manner. Perhaps we could agree that 'battery= drastically weak link in E' already, and we'll need a fair amount more I assume.
So within reasonable comparable cost and weight, make the car compete as best with mclaren as possilble. Assuming the drags were after any limiting (active limiting or just specs 40% off or confused), we have 532hp and 1000+lbs doing fairly well in competition as seen in the drag.
How much MORE do we need to gain maximum shaft output of motors, and how much difference will it make in weight, cost, and performance vs mclaren? Will motors and controllers need upgraded? Just how identical could they become?
I don't even care if it can't 'beat' mclaren, what is acheived is the most comparable and competing ev vs mclaren. *cont below
So for this argument to define a cutoff in gearing benefit from hamster power to EV (or presently beyond),
All perfectly accurate afaict, and good valid info, but my dumbass fails to see technical connection or any explanation to why in EV
*cont- so take that best competing version of a 1spd S and copy it, adding then (at least) the 2nd spd. Could that help with any highly disproportional weakpoints remaining to mclaren, giving a more well rounded package in performance, cost, weight, handling, etc?
Then design and run custom 1spd vs custom 2spd comparing the above 4 criteria, and also comparing and considering the potential between the 2 ev's of trading some 'efficiency' for better weight, cost, range. . . and even some 'greenness' when trading steel vs lithium and/or copper.
_____________________
From earlier posts of dan/arlo/luke:
I agree with this, and that whole post.
Yes I am.
This is true. With a STRONG relation to human power, ICE, or hamsters in a wheel. 'all eletrons man.
Yes
Yes, motors are rated combined @ 692, but *battery limited* to 532
So 1000lb heavier and 200hp underpower, the S technically performs even more admirably vs mclaren, though it will be perhaps harder to improve perfomance competitive to mclaren in a balanced manner. Perhaps we could agree that 'battery= drastically weak link in E' already, and we'll need a fair amount more I assume.
So within reasonable comparable cost and weight, make the car compete as best with mclaren as possilble. Assuming the drags were after any limiting (active limiting or just specs 40% off or confused), we have 532hp and 1000+lbs doing fairly well in competition as seen in the drag.
How much MORE do we need to gain maximum shaft output of motors, and how much difference will it make in weight, cost, and performance vs mclaren? Will motors and controllers need upgraded? Just how identical could they become?
I don't even care if it can't 'beat' mclaren, what is acheived is the most comparable and competing ev vs mclaren. *cont below
So for this argument to define a cutoff in gearing benefit from hamster power to EV (or presently beyond),
^ my observation baseline you can help refine if you like. Hopefully it's a solid start before further arguing definition of specific values.
Yeeess. . . . for a single speed. . . . how does that help your argument?
Right, it rotates it's tires until current limit. . . . and even after letting off. Saw alot of rotation going into speed in insane mode, but the tires hooked up all the way. Was there something I missed?
Yes, basic statement. In fact 'electro chemical' could perhaps be broken down further? *Perhaps there are similarities to human and animal, and even ice.
Yes. . .
Yes. . .
All perfectly accurate afaict, and good valid info, but my dumbass fails to see technical connection or any explanation to why in EV
???
*cont- so take that best competing version of a 1spd S and copy it, adding then (at least) the 2nd spd. Could that help with any highly disproportional weakpoints remaining to mclaren, giving a more well rounded package in performance, cost, weight, handling, etc?
Then design and run custom 1spd vs custom 2spd comparing the above 4 criteria, and also comparing and considering the potential between the 2 ev's of trading some 'efficiency' for better weight, cost, range. . . and even some 'greenness' when trading steel vs lithium and/or copper.
liveforphysics
100 TW
When you pin it in a p90d, it rides the traction control limiting wheel torque as you feel and hear this neat little high frequency chirping of the tires breaking loose, then torque being dialed back, then tires breaking loose and torque being dialed back.
Hillhater
100 TW
I hate to sound like a parrot, but the thread seems to have settled for the status quo that practical EV's ....( ignoring bikes for now please) , require a reducttion drive ( transmission) if not a multi ratio transmission.
We have not yet explained why this is so.
if Motors can be made with sufficient torque for DD drive ( not necessarrily "in wheel" motors,) what are the reasons no one has yet produced a car using that configuration ?
I would not expect Tesla etc to reengineer thier drivetrains now, but why didnt they, or GM, etc, go DD originally ?
Is it just cost or is there some other technical barrier yet to be overcome ?
We have not yet explained why this is so.
if Motors can be made with sufficient torque for DD drive ( not necessarrily "in wheel" motors,) what are the reasons no one has yet produced a car using that configuration ?
I would not expect Tesla etc to reengineer thier drivetrains now, but why didnt they, or GM, etc, go DD originally ?
Is it just cost or is there some other technical barrier yet to be overcome ?
