Joined
·
255 Posts
The only people interested in extreme power output are the RC hobbyists, and one of a kind applications. EVs have too many cells in parallel that cell power output is not an issue at all.
Posts by Solarsail
1 - 19 of 125 Posts
Joined
·
255 Posts
But this is 20700 and not 21700.
Joined
·
255 Posts
Well, with the 100DL, the 580 kW load will be distributed over 85 cells. That is 17 A per cell or 5C. The heat generated with a 100 mOhm internal resistance is phenomenal. It will be about 30 W per cell, or 240,000 W for the pack. This is not coolable. The pack will glow after a minute! Even at 50 mOhm, the thermal energy produced is 120 kW (plus all the heat from the connections). This can be sustained for only a few seconds.
No wonder the 100DL wins the 0-60 but loses the 1/4 mile because it has to throttle itself back or you get destroyed cells. Also the lack of a 2nd gear in the Tesla causes the motor to operate in an inefficient region, draining more current.
The average user probably never exceeds 2C.
Joined
·
255 Posts
This must be a trick question?
The torque curve is pretty flat until it reaches a threshold speed whereby it declines by speed. Namely the motor efficiency drops rather steeply with higher speeds. The decline in torque is faster than the increase in RPM and thus power drops and efficiency decreases and motor thermal energy increases.So in order to have a faster car, the RPM must be dropped below the threshold, which means the gear should be increased. Thus we need at least two gears.
If you provide me the spec sheet for Tesla's motor, I will try to work it out.
Joined
·
255 Posts
I think what I am saying is that the Tesla motor efficiency drops at higher RPMs. Thus it would be beneficial if the reducing gearbox which I believe is reducing by a factor of 6 resulting in motor turning at 12,000 RPM at 95 mph - were to reduce to 3 over let's say 50 mph. Then there would be more power available to the wheels as opposed to heating the engine and increasing coil resistance and more work for the cooling system. Also much less noise and better comfort.
In the chart, look at 55 mph. R_Tq for the L is 255 (whatever unit) and at 75 mph 163 and at 90 mph is 130.
55*255 = 14,025
75*163 = 12,225
90*130 = 11,700
Whatever the unit may be. You can see the power to the wheels decreasing, even though power input is almost constant. Between 55 mph and 90 mph, power input drops 3.4% while power output drops 16.6%. I believe a 10% increase in power could be achieved at 95 mph if the gear ratio were dropped from 6 to 3, and the motor RPM halved.
This would also allow Tesla to increase the reduction at lower speeds, such as from 6 to 8 and get even better acceleration. Of course it should be the user's option to decide at what speed to change the gears - i.e. there should also be a manual mode.
In the chart, look at 55 mph. R_Tq for the L is 255 (whatever unit) and at 75 mph 163 and at 90 mph is 130.
55*255 = 14,025
75*163 = 12,225
90*130 = 11,700
Whatever the unit may be. You can see the power to the wheels decreasing, even though power input is almost constant. Between 55 mph and 90 mph, power input drops 3.4% while power output drops 16.6%. I believe a 10% increase in power could be achieved at 95 mph if the gear ratio were dropped from 6 to 3, and the motor RPM halved.
This would also allow Tesla to increase the reduction at lower speeds, such as from 6 to 8 and get even better acceleration. Of course it should be the user's option to decide at what speed to change the gears - i.e. there should also be a manual mode.
Joined
·
255 Posts
They are putting so much power that the cells are at the limit and can only be run for a few seconds before damage sets in, and they have to cut down on power. So that is how they lose the 1/4 mile sprint. With a higher drive reduction ratio, they can get more torque for less power at the start, and then at high speeds, less power wastage which means higher speeds. I think it was a very bad decision to drop the gearbox. A dual clutch gearbox with only two gears can be designed by their engineers in their sleep. I fail to see the challenge in shifting gears with a dual dry clutch. They will kill two birds with one shot. I am certain electric supercars will all be multi-gear.
If it is true that the 100L is drawing 580 kW - I believe this is for the motor and for all base systems and cooling, including line losses - and assuming an internal cell resistance of 60 mohm (which increases with cycle life), the current drawn from the pack is 1,475 A and heat generated by the pack, not including line thermals, is 130 kW, or about 16 W per cell. Each cell is drained at 17A which is 5C - way above the spec (2C Panasonic, 3C Tesla) I believe - unless they have a new chemistry, which they have not shared. So when Musk says "will share patents", he means, I will not patent, so I won't have to share. Or more likely, they are just overheating the cells as it is doubtful that they can collect 130 kW of heat. Hence the very short duration of maximum acceleration.
With a gearbox, many of these issues can be mitigated.
Joined
·
255 Posts
It should be noted that according to the chart, the vehicle is drawing almost 40 kW at standstill. Compare this to a Leaf's 0.5 kW base load. I would guess that this may be the result of entering Ludicrous mode and the cooling system is feverishly pre-engaged to cool down the systems and the pack, prior to the sprint.
Front Ludicrous:
55 mph 140 = 7700
90 mph 93 = 8370
So there has been a gain of 8.7%. Obviously the power is being shifted to the front. But not as much as the rear is losing power. Also note the base load of 40 kW. So that would mean that the loss is actually larger as a percentage of power consumed by the motors.
That is a good point that power may be transferred to the front wheels as it speeds up. That may explain the loss seen at high speeds in the rear train. In fact the chart indicates that the front motor is getting more efficient at higher speeds!
Front Ludicrous:
55 mph 140 = 7700
90 mph 93 = 8370
So there has been a gain of 8.7%. Obviously the power is being shifted to the front. But not as much as the rear is losing power. Also note the base load of 40 kW. So that would mean that the loss is actually larger as a percentage of power consumed by the motors.
Joined
·
255 Posts
dcb - I think the calculation for ICE horsepower is somewhat different than pure electric and is bloated. I have seen elsewhere that 28 HP of ICE power is equated to only 15 kW electric motor (about 20 HP). This may explain the high 588 kW figure of the dyno -- which does not include the base load which is at least 40 kW. With a 50 kW base load, it is hard to believe that 638 kW is being consumed.
The Tesla already has a reduction gear of 9.73:1. They can have two sets of these with a dual dry clutch system where only one is engaged at a time. There will be absolutely no loss in efficiency. (A clutch has other uses such as in towing or cruising.) Why not have a quicker car with more torque and a quieter and faster car at high speeds? All you need is a dual parallel reduction drive. I would choose 13:1 and 6:1 where the gear changeover is programmable (higher for sports mode).
The Tesla already has a reduction gear of 9.73:1. They can have two sets of these with a dual dry clutch system where only one is engaged at a time. There will be absolutely no loss in efficiency. (A clutch has other uses such as in towing or cruising.) Why not have a quicker car with more torque and a quieter and faster car at high speeds? All you need is a dual parallel reduction drive. I would choose 13:1 and 6:1 where the gear changeover is programmable (higher for sports mode).
Joined
·
255 Posts
Wow - at this ratio, the motor RPM is about 21,000 at 95 mph assuming 245/45R19 tires. Does this reduction include the differential? At 21,000 RPM you are smoking a lot of things, and the noise is certainly not so pleasant (if you can hear that frequency!). This is not a good idea and makes the drivetrain costlier than it should be.
Dual (or even triple) 'parallel' reduction is the way to go. Should make the gearheads happy.
Yes, the cells are the limit at 5C -- see my calculations in a post above.
Joined
·
255 Posts
So are you saying that there is no reduced performance in 1/4 mile? I find it mind-boggling that they would pre-heat the cells.
At 1,500A, the pack is generating 135 kW. Do you know the cooling capacity of the system for the battery pack? It would have to be at least 100 kW. So that may explain where the missing power is going. I tend to think that the limiting factor will be the cooling of the battery pack, and that should start kicking in after maybe 30 seconds. Has anyone seen the temperature chart (for the inside of the battery
) at these rates?
Joined
·
255 Posts
Ahh ... thanks for the correction. That is a lot better.
My point is that the dyno, made for the sports and muscle car market, will tend to overrate things. I have seen that in a slightly different context, that power by ICE is discounted when compared to electric power. By as much as a factor of 1.4. See for example Torqeedo.com. So power is not the same if marketing says so.
Joined
·
255 Posts
major, thanks for the charts - they are very helpful.
Does the inefficiency also reduce the torque, or just the output power?
The first chart, running at 4000 rpm with 90 Nm of torque requires about 38 kW. Let's say this is an electric Smart car. With a tire circumference of 1.1 m and a reduction ratio of 7:1, you get 37 kph. The efficiency is 92%. At 35 kW power, you will be accelerating quite a bit. So you accelerate to 74 kph and the rpm is 8000. Your efficiency is about 93%. Now you are approaching 111 kph which is 12,000 rpm. From the chart you see that the efficiency is quickly being pushed down to 80, 70, and then 60%. The faster you go, efficiency drops dramatically.
What is causing this sudden drop in efficiency?
It goes to show that depending on the motor and load conditions, the motor can become so inefficient that if we just increase the power without accelerating, we will actually be using less power, per unit of work.
BTW. my calculation for RPM was amiss. Karter2 has the correct speed.
Does the inefficiency also reduce the torque, or just the output power?
The first chart, running at 4000 rpm with 90 Nm of torque requires about 38 kW. Let's say this is an electric Smart car. With a tire circumference of 1.1 m and a reduction ratio of 7:1, you get 37 kph. The efficiency is 92%. At 35 kW power, you will be accelerating quite a bit. So you accelerate to 74 kph and the rpm is 8000. Your efficiency is about 93%. Now you are approaching 111 kph which is 12,000 rpm. From the chart you see that the efficiency is quickly being pushed down to 80, 70, and then 60%. The faster you go, efficiency drops dramatically.
What is causing this sudden drop in efficiency?
It goes to show that depending on the motor and load conditions, the motor can become so inefficient that if we just increase the power without accelerating, we will actually be using less power, per unit of work.
BTW. my calculation for RPM was amiss. Karter2 has the correct speed.
Joined
·
255 Posts
Tesla could use some "real" good news. I would think even if they had improved energy density by 20%, it would be huge news. I am afraid we may have hit a hard ceiling. The 3.4Ah came out 3 or 4 years ago and there has not been any density improvement by Panasonic (or Tesla) since then.
Joined
·
255 Posts
The 3.4Ah capacity is for a 18650. Now since the 2170 is 1.466x the volume of the 18650, one would expect the proportional capacity for the 2170 to be 4.98Ah. But if the capacity is confirmed at 4.6Ah, then that is a step backward! (Note that the 'chemically usable' volume proportion of the 2170 is actually higher - about 1.5x, which implies a capacity of 5.1Ah.)
The T3 LR pack is estimated at 78 kWh nominal (and 74 kWh usable). 78 / 4416 / 3.6 = 4.9Ah -- which I believe is more likely (than 4.6Ah).
Joined
·
255 Posts
One theory is that there is no gear-changing transmission - just each of the two rear motor having a different reduction ratio - one for low speed high torque, and the other for high speed. Of course with a differential.
I think even at the predicted capacity of 4.9Ah for a 2170 cell (which would be the same energy density as the 18650 cell), it should be possible to place the 2nd 100 kWh in the R2. Also note that for best performance, you do not want to have 200 kWh capacity, but rather 100 kWh capacity to save on weight - as long as the 100 kWh battery can produce the required power.
But if you had to fit 200 kWh with today's cells, there is room in the frunk, and also the trunk, and also double layering under the rear passenger seats. It won't be pretty, but I think it is doable.
I think even at the predicted capacity of 4.9Ah for a 2170 cell (which would be the same energy density as the 18650 cell), it should be possible to place the 2nd 100 kWh in the R2. Also note that for best performance, you do not want to have 200 kWh capacity, but rather 100 kWh capacity to save on weight - as long as the 100 kWh battery can produce the required power.
But if you had to fit 200 kWh with today's cells, there is room in the frunk, and also the trunk, and also double layering under the rear passenger seats. It won't be pretty, but I think it is doable.
Joined
·
255 Posts
Yes, except that all three motors will be powered simultaneously when high torque is required. The rear two motors having different reduction ratios, will provide best torques at different vehicle speeds. There will be considerable overlap between the two rear motors. The front one will be mid-range, while the two rear motors will supply low-speed and high-speed torque and power, due to their different reduction ratios. Both will drive the differential.
With a 2-speed gearbox, it is possible to eliminate the rear differential, and obtain even more torque because both motors will avoid the high-speed inefficient region. But then two gearboxes would be required. So I am a bit skeptical that they would take this approach. If they were going to use just one gearbox in the rear, then why have two motors? Conceptually a 2-speed gearbox using a planetary gearset is simple. But if Tesla is already having difficulty with the reduction gearbox, then woe be on them to attempt a 2-speed box.
I wonder if it is beneficial to mechanically disengage the high-reduction ratio motor at speeds of let's say above 150 mph. This would allow the two rear reduction ratios to diverge. Assume in the current Model S, the ratio is 10. In the R2, they could be 14 and 7. But if the high-ratio motor could be disengaged, then they could take it to 16 and 6, and cover a wider operating range with more torque at the low end.
With a 2-speed gearbox, it is possible to eliminate the rear differential, and obtain even more torque because both motors will avoid the high-speed inefficient region. But then two gearboxes would be required. So I am a bit skeptical that they would take this approach. If they were going to use just one gearbox in the rear, then why have two motors? Conceptually a 2-speed gearbox using a planetary gearset is simple. But if Tesla is already having difficulty with the reduction gearbox, then woe be on them to attempt a 2-speed box.
I wonder if it is beneficial to mechanically disengage the high-reduction ratio motor at speeds of let's say above 150 mph. This would allow the two rear reduction ratios to diverge. Assume in the current Model S, the ratio is 10. In the R2, they could be 14 and 7. But if the high-ratio motor could be disengaged, then they could take it to 16 and 6, and cover a wider operating range with more torque at the low end.
Joined
·
255 Posts
So if we eliminate the possibility of two 2-speed gearboxes, then in order to have two rear motors, the following architectures remain:
a) Different reduction ratios + rear (and front) differential. The Model 3 reduction ratio is close to 10. So for the R2, it would be 14 on one motor and 7 on the other motor. A further twist here is that if the 14 motor could be disengaged by a clutch, then the ratios could be wider such as 16 and 6. The front ratio would remain at 10.
b) Another possibility is just one 2-speed gearbox in the rear for one of the motors. One motor runs at ratio 14 the other at 7 but also has the benefit of the 2-speed gearbox and then can also run at 14, for low-speed high acceleration. I don't think this would be a good design.
a) Different reduction ratios + rear (and front) differential. The Model 3 reduction ratio is close to 10. So for the R2, it would be 14 on one motor and 7 on the other motor. A further twist here is that if the 14 motor could be disengaged by a clutch, then the ratios could be wider such as 16 and 6. The front ratio would remain at 10.
b) Another possibility is just one 2-speed gearbox in the rear for one of the motors. One motor runs at ratio 14 the other at 7 but also has the benefit of the 2-speed gearbox and then can also run at 14, for low-speed high acceleration. I don't think this would be a good design.
Joined
·
255 Posts
I think this would be something like two identical rear ratios at let's say 14 and a front at 7 - for a 2:1 ratio. The problem is that this will not take you up to 250 mph. The rear motors cannot deliver the power at those speeds, and the front is underpowered and too high RPM for 250 mph is required - the spread of the ratios is not wide enough. But I like the idea that at high speeds, power is mostly delivered by the front, and I think that will make for better high speed driving. So the ratios I propose are 18 for RM1 (rear motor 1), 10 for RM2, 6 for FM, for a 3:1 ratio of the reduction gears.
0 - 60 mph, 100% power on all three motors, torque ratio (RM1, RM2, FM): 60%, 35%, 5%
60 - 150 mph, power is cut to RM1, 100% to RM2 and FM. Torque ratio: 25%, 55%, 20%
150 - 250 mph, power is cut to RM1, RM2, and 100% to FM. Torque ratio: 0%, 30%, 70%
Electrowrks, what is the effect of switching from delta to wye in the stator configuration of a PM motor?
1 - 19 of 125 Posts
