1000kW DC powersupply - Help
- Thread starter h0tr0d
- Start date
Hillhater
100 TW
riba2233 said:
doubtful, since they are also a 2009 development and quoted at <2kW/kg energy density !
http://www.neces.com/18700-cylindrical-cell.htm
However..
http://www.racecar-engineering.com/articles/features/developing-mercedes-hpe-kers-batteries/
liveforphysics
100 TW
h0tr0d said:
If one wishes to make a 1MW BLDC controller, there are challenges ahead that make sourcing a 1MW power supply seem effortless.
After spending some time in this game, my advice would be to break it up into something like 10 controllers that each do 100kW. Or maybe buy 4 off-the-shelf P85 Tesla Model S controllers, and make sure to wind the motor to have 4 separate sets of phase leads exiting.
Or do away with the controller and use an SCRs to connect the motor directly to the pack and then use the same mechanical clutch as the ICE top-fuel cars to feed in torque to the wheels. At the end of the run the clutch disengages, unloads the motor and allows a contactor to safely disconnect the battery. An explosive fuse provides emergency disconnection under load. You might also want to consider an ejector seat.
Punx0r said:
I like the ejector seat idea, and I don't want to be too much of an efficiency zealot for no reason, as this idea would technically work, but the simple matter of rejecting the heat from the clutch and the fact that double the energy would be necessary for the trip are two reasons I don't like the SCR/clutch idea.
Leaving aside the clutch heat rejection issue for now, if you look at the SAFT battery discharge curve, staying as high up (most charged, less discharged) on the curve as possible gives the highest voltage and also the highest power density. Simple matters like this might require only 250kg of batteries versus 300kg. Now the real question, is can you get a controller(s) that weigh less than the saved weight (50kg). My gut says yes, particularly for pulsed automotive application, but no data or calculations to back that up.
Hillhater
100 TW
halcyon_m said:
How do you propose you can keep a high SOC ..and reduce the battery pack weight ?
The reverse would seem to be required....since that 400kg pack is going to be pretty much wasted after a few seconds. :?
liveforphysics said:
So that 6.0MW controller becomes 60 individual 100kW units !! 
..or a mere 40 Tesla controllers ! :?
I wonder what that would weigh ??
Hillhater said:
Using round figures, with 5.5MW needed at the top speed, this means that only 12-16MJ are needed, depending if the vehicle is running 1000' or 1/4 mile.
I know it's counter-intuitive, but due to the short duration (4.4-6 seconds) and the triangular power curve (if using a controller), the battery pack (using SAFT cells) has to be sized for power, and not energy. There will be 'fuel in the tank' of the batteries at the end of the strip, but all the batteries need to come along for the ride to meet the power figure.
This may not be the case with the 20kW/kg cells mentioned previously. I'm not sure if they are running at a C rate of 400 or higher, but for this application, anything less than a C-rate of 600 would be adequate energy-wise, not to mention the increased power density from not running the cell down completely when you need the power the most.
Hillhater
100 TW
Interesting point, even if there are a lot of unknowns still.
..but the fact remains, that unless you can put together a competitive drive package ( battery, controller, motor and ancilliaries) that weigh less than 250-300kg in total, ..then you are going to be at a disadvantage before you start.
..but the fact remains, that unless you can put together a competitive drive package ( battery, controller, motor and ancilliaries) that weigh less than 250-300kg in total, ..then you are going to be at a disadvantage before you start.
Considering that the ICE cars are traction, not power, limited an electric version is likely to be at a real disadvantage anyway. The nitromethane fuel is a lot lighter than equivalent batteries, the only saving comes from the electric motor potentially being lighter than the ICE.
Does anyone know the weight of a typical top fuel engine? They claim to make around 7.5MW. If it weighs as much a 200kg that's 37.5Kw/kg. I'm not sure how that compares to the power density of the best electric motors?
Good point on needing extra battery. It was a bit of a silly suggestion anyway I was surprised how relatively simple the mechanical clutch is on an ICE top-fueller. I was amazed how small it is considering the job it does. It doesn't actually lock up until something like the last second of the run!
Does anyone know the weight of a typical top fuel engine? They claim to make around 7.5MW. If it weighs as much a 200kg that's 37.5Kw/kg. I'm not sure how that compares to the power density of the best electric motors?
halcyon_m said:
Good point on needing extra battery. It was a bit of a silly suggestion anyway
I was surprised how relatively simple the mechanical clutch is on an ICE top-fueller. I was amazed how small it is considering the job it does. It doesn't actually lock up until something like the last second of the run!Hillhater
100 TW
Punx0r said:
230 kg complete with ancillaries.
wb9k
10 kW
riba2233 said:
That's the package they use. A123 is used in about 70% of F1 cars. Chemistry is more current now than what's shown in the link. Many holders of EV speed records also use A123. Ohio State's Buckeye Bullet uses A123 prismatics. Killacycle uses the 32113 cylindrical.
Hillhater
100 TW
Unless you can put that 6000kW into a pair of hub motors, you are going to need a transmission of some form.?halcyon_m said:
Ohbse
10 kW
They don't have a transmission. Direct driven 12" rear end with hydraulic multi plate clutch that's computer controlled. Over the course of the run RPM actually drops as the clutch is released before climbing again at the end. Clutch plates last one run. Fuel flow is primary means of metering out power, this increases a bit in the final 300' to get to the peak 10,000hp/12,000 ft/lb figure due to aerodynamic loads.
To say these things are traction limited is technically true, but still ridiculous as they can pull 5.6g from launch. If they accelerate for a mere 60' and shut off the engine they will coast through the traps with a high 8 second pass.
Fuel load is 55KG, engine + dressings is 230kg, fuel tank/lines etc probably another 20KG - call it 300kg all up as a ballpark. No weight to be saved beyond the engine as you would still require clutch/driveshaft/diff etc
To say these things are traction limited is technically true, but still ridiculous as they can pull 5.6g from launch. If they accelerate for a mere 60' and shut off the engine they will coast through the traps with a high 8 second pass.
Fuel load is 55KG, engine + dressings is 230kg, fuel tank/lines etc probably another 20KG - call it 300kg all up as a ballpark. No weight to be saved beyond the engine as you would still require clutch/driveshaft/diff etc
Ohbse
10 kW
halcyon_m said:
Totally possible - however then you have issues with reduction. There's no way you can run without some form of reduction considering the tire size of 36" tall which grows over the run to over 44". You're talking about delivering peak torque through a virtually stalled rotor if driven 1:1, at this power level that would result in atomised motor I suspect.
The clutch plays possibly the most critical role in the dragster - it allows smooth delivery of near instantaneous savage torque. The exact same issue would exist, probably to an even greater degree with electric motivation. Instant torque breaks things, it shocks tires into slip and can dramatically increase your ET. The race is won or lost (by substantial margin) in the first 60', it's very easy to get it wrong and I don't think you'd have much hope of getting a direct driven electric drivetrain to cope nearly as well as a very highly evolved clutch - at least not at the ludicrous power levels we're talking about.
Top fuelers hit over 120mph in 18 metres or a little over 6 rotations of the tires.
Ohbse said:
Maybe you're not familiar with motor controllers these days, but current regulation is torque regulation, and with a properly designed motor controller, you can regulate torque 16000 times a second with ease, which is likely more than a hydraulic clutch actuator response rate. So... there's an edge over ICE tech right there. Not saying it's a win on this whole electric dragster discussion, that's just a forte of motor controllers. So you can have instantaneous torque to have constant acceleration, or a ramp in acceleration to have a constant value of jerk, or a third order spline to control jounce to a constant value... or anything in between. To automate it to limit slip would be traction control though, so I'm not sure how that would bode in the rules. Having the throttle response characteristic to be consistent with the examples I gave would not be traction control, but it would limit peak loadings on mechanical components and may help reduce weight elsewhere in the frame.
Ohbse
10 kW
Oh I have no doubt that there's better ways to do it than the clutch method employed by the current rule set of the class (which bans all computer control I've just realised). I guess my point is that it would be a lot more achievable to get something electric going quick by running with the clutch, it's well understood, there's a lot of experience around and it still employs a useful function in shock isolation (accelerating 2300lbs to 60mph in 8 feet is definitely what I would consider shock loading)
Back to more interesting things, like what motor/motors would actually look like to hit the torque numbers required!
Back to more interesting things, like what motor/motors would actually look like to hit the torque numbers required!
Hillhater
100 TW
As halcon_m said,.. a well designed controller(s) would eliminate the need for a clutch pack...but also as you say, it seems unlikely that you could design a realistic high power EV drive that doesnt require rpm reduction to the wheels
Hi wb9k, ..are any of those packs rated at 20kW/kg ?
wb9k said:
Hi wb9k, ..are any of those packs rated at 20kW/kg ?
Hillhater said:
I didn't say there wouldn't be a gear reduction. What I did say was that the motor(s) could be mounted longitudinally to avoid the need for a right-angle pinion gear and the losses that are associated with them. If there was room enough, a planetary set would be optimal, but it's not likely that there would be enough room without a secondary belt.
From my calculation (5.6G, 100kG load, 36" diameter tire at launch), that would require 25,000N-m at the wheels, 12,500 for each.
That would be 12 YASA 750's per wheel, geared 1.4:1 to reach a top speed of 300MPH, if you were going AC permanent magnet. I'm sure we can find a higher power density motor to use particularly for 4 seconds.
Some of the performance figures for the top fuelers quoted in this thread are mind-bending Building such a vehicle propelled by any means clearly requires a huge amount of technical skill and money. Lots of money.
I think whatever motor(s) are chosen would be pre-chilled, perhaps with liquid nitrogen, and massively over-driven.
Is a permanent magnet synchronous motor necessarily the best choice?
Building such a vehicle propelled by any means clearly requires a huge amount of technical skill and money. Lots of money.I think whatever motor(s) are chosen would be pre-chilled, perhaps with liquid nitrogen, and massively over-driven.
Is a permanent magnet synchronous motor necessarily the best choice?
Similar threads
