John in CR
100 TW
The train example is about as conclusive as it gets.
Conclusive proof gearboxes are awesome.
- Thread starter Teh Stork
- Start date
I had a chance to drive a freight train
Into and out of the franzia winery
They take a long time to reach top speed
They are not in stop and go traffic
It doesnt matter
Cruising man
The main reason they are diesel elctric is a gearbox for diesel only would require too many gears
To stay in the powerband
While a electric has full torque from 0
I trust Ian Wright
If he designed his cargo trucks to use a two speed
There was good reason
The biggest mining trucks do use single speed
I guess the answer is
It all depends on design contraints
Speed range
Motor size
But it is an interesting thought experiment to think if there were none
maglev linear motor
Into and out of the franzia winery
They take a long time to reach top speed
They are not in stop and go traffic
It doesnt matter
Cruising man
The main reason they are diesel elctric is a gearbox for diesel only would require too many gears
To stay in the powerband
While a electric has full torque from 0
I trust Ian Wright
If he designed his cargo trucks to use a two speed
There was good reason
The biggest mining trucks do use single speed
I guess the answer is
It all depends on design contraints
Speed range
Motor size
But it is an interesting thought experiment to think if there were none
maglev linear motor
liveforphysics
100 TW
flathill said:
I like Ian Wright too. Wrightspeed is a local EV place for me.
However, his 2spd based drivetrain setup in the original Tesla was a reliability nightmare.
They dumped it for direct drive, and ended up with improved performance, range/economy, and became reliable.
I've driven a model S with performance package. It's a fixed single speed. You hit the throttle, it pins you back as you hear the little traction control sensors working to control wheel spin to a mild degree of slip (which IS where peak traction occurs, not when static like your highschool physics book would lie to you and say), and you still hear/feel it dialing torque back to the limit of the tires traction past 60mph, then it still just rockets you way past 100mph in a few more seconds. It's like the theoretical perfection in a drivetrain. No pauses. No interruptions in torque delivery (even if they are just for a short time, it's still a chassis upsetting interruption in torque delivery), and it's just dead smooth and consistent like some eerie force smashing you back and trying to rip up the asphalt as it catapults this giant 4000lbs sedan forward like a rocket with no signs of stopping till faster than I had road to go.
If you have to band-aid in the wrong motor for the job, a transmission can help. If you can design the motor for the application properly, IMHO, any transmission is just pure inefficiency, loss, weight, complexity, noise, and adding new failure modes.
In my own drag racing, I'm excited for making my own electric dragster. Is it because it environmentally makes sense? Hell no, electric dragsters will always be a greater environmental burden than gasoline dragsters, because gasoline dragsters simply don't use enough fuel for it to matter (I burned maybe 15gal in an entire season racing my drag car?). It's because I can finally dump having a damn transmission, which has cost me more events, and more money, and more time in replacing and fixing it's bullsh*t variety pack of failure modes, than the entire rest of the engine/car together has caused. I want to make an electric dragster JUST to not have a damn clutch and tranny to fix/replace anymore.
Rassy
1 MW
Flathill said:
Probably because it was designed to accelerate very slowly. I had a friend in high school who's dad was a freight engineer. He got to drive trains occassionly and one of the toughest parts was not to give it too much juice because it could spin the wheels and tear up the tracks.
motomoto
1 kW
- Joined
- Jun 28, 2010
- Messages
- 458
Ok, I hear all this talk about torque and how electric motors have maximum torque from zero RPM and therefore
a gearbox is never needed and is just a weight penalty and has frictional losses etc, etc.
What I fail to hear is that torque times rpm equals horsepower. When you are talking horspower and acceleration,
I would rev that motor where it makes the most horsepower and speed shift the gears appropriately to cause the
fastest acceleration. Using big heavy high service duty motors with one speed is one way to never have the fastest
combination but it is simple and it does last. I am more like have a motor that makes 4 hp per lb and reduce it
and gear it to win. It's torque x rpm and not just torque.
a gearbox is never needed and is just a weight penalty and has frictional losses etc, etc.
What I fail to hear is that torque times rpm equals horsepower. When you are talking horspower and acceleration,
I would rev that motor where it makes the most horsepower and speed shift the gears appropriately to cause the
fastest acceleration. Using big heavy high service duty motors with one speed is one way to never have the fastest
combination but it is simple and it does last. I am more like have a motor that makes 4 hp per lb and reduce it
and gear it to win. It's torque x rpm and not just torque.
See, that gets back to economics, especially where weight is concerned.
If electric motors aren't expensive, and weight isn't a big deal, then there's no need to have the smaller motor. Just get a huge motor that will perform as a car engine would need to in first gear, but ALL the time :lol:
The gearbox ultimately does lower the efficiency of any motor by just adding friction to it. Cost is also added, in terms of the gearbox.. it really seems like the best use of your $ to just have a big motor!
The reason that gasoline cars have always had gearboxes is that the engine itself has higher mechanical friction and higher pumping losses, the larger it is. So you need a transmission to keep the RPM as low as possible and the load the highest to get the best efficiency. Seriously - drive your car one gear down ( 4th instead of 5th, or 5th instead of 6th ) on the highway and see what kind of fuel economy you get
But a geared system does offer you advantages in weight, but not economics, and not in efficiency compared to the big honkin' motor. Because you always have more mechanical friction to deal with.
If electric motors aren't expensive, and weight isn't a big deal, then there's no need to have the smaller motor. Just get a huge motor that will perform as a car engine would need to in first gear, but ALL the time :lol:
The gearbox ultimately does lower the efficiency of any motor by just adding friction to it. Cost is also added, in terms of the gearbox.. it really seems like the best use of your $ to just have a big motor!
The reason that gasoline cars have always had gearboxes is that the engine itself has higher mechanical friction and higher pumping losses, the larger it is. So you need a transmission to keep the RPM as low as possible and the load the highest to get the best efficiency. Seriously - drive your car one gear down ( 4th instead of 5th, or 5th instead of 6th ) on the highway and see what kind of fuel economy you get
But a geared system does offer you advantages in weight, but not economics, and not in efficiency compared to the big honkin' motor. Because you always have more mechanical friction to deal with.
AussieJester
1 TW
I just think they sound cool shifting 
KiM
KiM
liveforphysics said:
Totally agree see what I posted way earlier in the other thread
"Also gearboxes dont hold up well to high torque edrive
See tesla 2 speed development saga on the roadster
They tried 4 suppliers
All sheared gears
They could have made it work with enough money and an in house design
Like the wrightspeed gtd
But they just oversized the controller and found it was faster up to the designed top speed
And more efficient real world because an oversized controller has less resistance
More switching devices in parallel"
Biff
100 W
liveforphysics said:
It is true that given a motor and a desired torque, and to use the least amount of power to achieve that torque is to do at low RPM, since RPM * torque is power.
That is however a very different situation than how to use your power most efficiently. Quite often increasing speed and reducing torque increases efficiency. If you are working in the range of a motor where hysterisis losses are much more significant than eddy current losses, your core loss will change linearly with speed. At the same time, your copper loss will change with the square of torque , since current is directly proportional to torque, and power is related to the square of current.
so if you are operating at 60W core loss, and 100W copper loss, in one gear, then change to a gear with 1/2 the ratio (motor operating 2x the speed 1/2 the torque, same output power), you should change to 120W core loss and 10W copper loss for a total loss of 130W vs the initial situation where you had 160W of loss, save yourself 30Watts.
What people underestimate is the in-efficiency of a gearbox, and they overestimate how much they can actually improve efficiency by "working in the sweet spot" of the electric motor. Lets continue with the scenario I outlined in my last paragrah, asuming the motor is 90% efficient in the first scenario.
if the loss of the motor is 160W and it is 90% effiency, it would be putting out about 1.6kW. When re-geared, it would still be putting out 1.6kW of power and have only 130W of loss , or about 92% efficient. Only 2% better, in a pretty extreme example where copper loss was significantly higher than core loss to begin with (which gives the gearbox its biggest advantage). You would be very lucky to find a gearbox that would have less than 2% loss, and most gearboxes would give you that loss all the time, so not only would you not be better off in that situation, you would be sacrificing that energy all the time to the gearbox when your motor would be operating at peak efficicney without the gearbox.
There are some smart gearboxes that have one fixed gear, and one reduced gear, such as the SMESH, which is just a planetary gear with a way to lock it into a 1:1. http://www.smesh.eu/ This would allow you to get lots of torque at low speed without sacrificing efficiency at high speed. It certainly adds complexity, and a bit of weight, but is pretty good.
As gwhy mentioned, if you want to get the best acceleration and you have a low peak power from your motor/controller setup, a well designed and tuned CVT is the way to go, as it gives you the ability to apply the peak amount of power your motor can deliver to the rear wheel over the greatest range of speeds.
The other thing a gearbox can do for you is make things possible. If you have a motor that can't produce enough torque to actually accomplish a task, adding a gearbox can allow you to produce the required torque at the wheel.
I guess this back to the beginning of the post:
Can a gearbox reduce the time required to accelerate your electric vehicle from a stop to your vehicles top speed? yes, but it takes a good design, I would bet that most gearboxes actually slow you down, due to the time required to shift.
Can a gearbox increase your system efficiency? very doubtful. only in very specific situations where the gearbox is carefully considered would this actually be true. If it does make it more efficient, it probably won't be by very much.
Can a gearbox make you accelerate faster off the line? it better, or you are doing something wrong.
Can you achieve similar acceleration off the line by increasing the motor / controller size rather than adding a gearbox? Probably, and by doing it you will likely see other benefits in performance at the same time, such as overall efficiency gains, or better peak performance.
There is definitely a place for a gearbox in electric vehicles, but I would bet that a lot of the time people don't consider their options carefully enough, and overestimate the advantages of adding a gearbox over finding larger motor.
How many of the electric motorcycles in the Isle of Man TT had transmissions?
-ryan
Another strange way to think of it is an indirect analogy
Are you designing for sensitivity (output per volt)
Or efficiency (output per watt)
It all depends on the moving mass (cargo + vehicle)
And load (impedance) matching
And how often you are slowing down and speeding up
Hz
Are you designing for sensitivity (output per volt)
Or efficiency (output per watt)
It all depends on the moving mass (cargo + vehicle)
And load (impedance) matching
And how often you are slowing down and speeding up
Hz
John in CR
100 TW
Is there anything to gain by using a manual clutch for launches in order to get the motor rpms maybe even just before launch. It would act like a CVT just for the first revolution or two of the tire? I see three side benefits, reducing shock load on the system, a cheap way to get precise throttle feel, and moving some of the waste heat during launch out of the motor... but is there maybe some overall efficiency gain there too by getting the motor rpms higher earlier than they'd be with a direct drive?
Lebowski
10 MW
A bit like a torque converter with lockup capability.... i like that ideaJohn in CR said:
I wish I could take more time to answer more properly, but here goes.
- Induction motors came up again, these are not similar to a PMM.
- To run higher phase currents you need more stator material. This leads to higher hysteretic losses. To reduce copper losses you need more copper. You completely overlook this fact.
I've got my own prototype that is somewhat similar to that Ian Wright gearbox. No clutch - speed controll of motor inbetween gears match motor and gear speed. I believe that the most powerful drive solution (over a wide range) can be realized by a hi-rpm PMSM with a lightweight gearbox.
From an economy perspective motor you need less magnets, less stator iron, less copper and "less" power electronics by adding a gearbox. Somewhere this balances, I'm sure.
- Induction motors came up again, these are not similar to a PMM.
- To run higher phase currents you need more stator material. This leads to higher hysteretic losses. To reduce copper losses you need more copper. You completely overlook this fact.
I've got my own prototype that is somewhat similar to that Ian Wright gearbox. No clutch - speed controll of motor inbetween gears match motor and gear speed. I believe that the most powerful drive solution (over a wide range) can be realized by a hi-rpm PMSM with a lightweight gearbox.
From an economy perspective motor you need less magnets, less stator iron, less copper and "less" power electronics by adding a gearbox. Somewhere this balances, I'm sure.
JennyB
1 kW
Miles said:
That makes sense for that application. I find that with really steep hills or strong headwinds I have to work hard to assist the motor at 8-10 mph, because the slower I go, the less power I have available. From playing around with similar motors on Justin's simulator I discover that gearing down by that ration the power available would be similar (so I wouldn't climb any faster), but the motor would be in a range where it was operating twice as efficiently, and where power is still increasing as the revs drop.
John in CR
100 TW
JennyB said:
Jenny,
You're seeing the direct result of typical DD hubmotors being such compromised systems, because they are simply geared too steeply with normal size bike wheels. That forces them to be underpowered for their weight. Gear down with a smaller tire and volt up to get back your desired top speed and overall performance increases dramatically.
JennyB
1 kW
John in CR said:
Actually, this is an limited-amp gearmotor system.
I understand that for any system there is an amp limit, set either by the controller, or by the motor becoming saturated and converting more and more of the supplied current into heat. With a smaller wheel the motor revs faster for a given road speed, and therefore can deliver more power for the same amps. What I've got works well for me (most of the time), because if the road is in any way favourable I don't use any power at all. My typical usage is now 7.5-8 watt hours/mile and falling. 8)
A two-speed as Miles suggests might be even better, but I suspect what would better suit many here would be the reverse - a direct drive for pootling round at a 20-25 mph cruising speed and a 1.7:1 or so GTFOOTW overdrive. Any motor big enough to pull that would have no trouble climbing just about anything at cruising speed in direct!
This can be a big range of losses. As you make more power IE motorcycle or EV you will need a transmission which can handle a lot more power causing it to have a lot higher losses. But if you use a 500w ebike with a derailer then it will be next to nothing.Teh Stork said:
This chart is quite wrong unless you are running a very low power bike where you can suddenly engauge the next gear without damage.Teh Stork said:
You need to add losses of drag to the line with the transmission. So in first and second the lines will be a bit lower and in third it will be below the non tranny bike.
As well you need the shift points to drop into the negatives during shifting.
My argument was a choice between bigger motor or adding a transmission to get the same end weight will have no power advantage im not talking about de-railers im talking about decent power levels 20kw+ electric motorcycle and EV teritory.
Teh Stork you should make a new graph showing motor A then a bigger motor (motor B) then motor A with a transmission added and make sure to add losses for drag and weight.
John in CR said:
Ha, i just lower my phase amp to battery amp ratio to get that effect.
Makes the torque of the motor far more linear. Much easier to control at low RPM, and yeah, less waste heat too.
On high power setups, it's the only way to fly imho.
John in CR
100 TW
Arlo1 said:
Paying more for a controller than for a motor simply isn't happening, not when I can get the power delivered for about 20% of the price. If control is so good, where are the videos of high powered rigs on one wheel, which is relatively easy to do with an ICE. Did I miss these videos, or do they not exist because control is still lacking?
Yes, I need torque throttle, but at a cheap price.
Other than the weight, what's wrong with the clutch idea?
liveforphysics
100 TW
Let me tell you John.
Having the exact amount of torque on tap to the wheel that you want, with no backlash and no mechanical engagement jerks and jolts and clanking bullshit parts and things to manipulate with your fingers being not wrapped firmly around the bar and throttle is NOT the path forward for EVs, and the ability to not need any of that added bullshit, and instead just deliver exactly the torque you want the instant you want it, in the precise amount you wanted by only twisting your wrist IS one of the biggest beautiful attributes of the EV riding experience crushing gas bikes.
I should add, you don't get that from anything speed control based or battery current limited based control. You get that from the controllers that are made to do it, and sadly, so far I think that list is very short right now. Sevcon or lebowskis is about all ive ever seen on a bicycle.
Having the exact amount of torque on tap to the wheel that you want, with no backlash and no mechanical engagement jerks and jolts and clanking bullshit parts and things to manipulate with your fingers being not wrapped firmly around the bar and throttle is NOT the path forward for EVs, and the ability to not need any of that added bullshit, and instead just deliver exactly the torque you want the instant you want it, in the precise amount you wanted by only twisting your wrist IS one of the biggest beautiful attributes of the EV riding experience crushing gas bikes.
I should add, you don't get that from anything speed control based or battery current limited based control. You get that from the controllers that are made to do it, and sadly, so far I think that list is very short right now. Sevcon or lebowskis is about all ive ever seen on a bicycle.
This guy made his own torque based dc controller for you agni guys
We should pester him to open source it
Im sure he has made no money on the last off it
Given how overpriced it is
http://www.altmann.haan.de/riding_on_electrons/default.htm
On a real motorcycle (or car), the throttle setting is equal to a force (a torque) and not a speed
We should pester him to open source it
Im sure he has made no money on the last off it
Given how overpriced it is
http://www.altmann.haan.de/riding_on_electrons/default.htm
On a real motorcycle (or car), the throttle setting is equal to a force (a torque) and not a speed
