1000kW DC powersupply - Help

h0tr0d

1 kW
Joined
Apr 28, 2012
Messages
460
Hey guys!

A friend of mine is trying to build a 1000 kW controller for drag racing but he needs a 1000kW DC powersupply.

Any idea whats the cheapest way for that?

333 HP ESP120 3kW server PSU's together?
 
h0tr0d said:
Hey guys!

A friend of mine is trying to build a 1000 kW controller for drag racing but he needs a 1000kW DC powersupply.

Any idea whats the cheapest way for that?

333 HP ESP120 3kW server PSU's together?
I do this stuff for a living and most likley he dosen't need a 1000KW supply for most tests.
He should really think about his test methods before getting a power supply like that.
A 1MW 3 phase supply to the building to supply it isn't going to be cheap let alone the power supply itself.

A battery is probably unsafe for inital tests so some smaller current limited supply first.
There are ways and means of pretty much fully testing a controller without having the full power required available which are safer and more practable.
 
Ricky and HH,

You are both correct BUT, apparently, he wants a setup capable of 6 MW in future developments (he wants to beat 8000hp top fuel dragsters...) :shock:
 
I'd look into a railroad locomotive. Seriously :wink:

Or if he can live with short duration blasts at that power level, look at a bank of capacitors or batteries which is slowly charged from the grid or smaller generator and then discharged quickly at the high power. This will require some sophisticated power conditioning electronics as well.
 
h0tr0d said:
You are both correct BUT, apparently, he wants a setup capable of 6 MW in future developments (he wants to beat 8000hp top fuel dragsters...) :shock:
Hope he has an unlimited budget then :lol:.
At those power levels the health and safety also becomes a serious issue. Arc flash is bad.
 
Im sure there is a "formula" to work out what the limit is for an EV dragster is.
Its not just a matter of generating more power like the Fuelers do with chemistry and clever motor tuning etc...
..With electrics, you have to keep adding more battery and copper ( weight) and at some point the laws of diminishing returns start to hit home.
A battery pack capable of supplying 6 MWhrs for a few seconds would weigh well over 500kg alone, and motors with enough copper to convert those thousands of amps into torque/power, would also be no small weight.
So you start having to haul a ton of extra weight up to 300 mph...and stop it again !!
Anyone know what a current TF weighs ?
..Answer ..1000kg.. ( 2225 lbs minimum regulation, and most are close to it)
and their 500cu in motors weigh 230kg complete, ..so a rolling chassis weighs 770kg.
Add 6000kW of motors , controllers, and 500-600 kg of cells and i suspect you are now underpowered for the weight !!
Any one want to hazard a guess at the weight of a 6000kW motor ?

I hope that your friend has done his homework before he sinks any serious cash into this !

Meanwhile, for 1MW ..how about this..
https://www.youtube.com/watch?v=astZWo-NIP8
[youtube]astZWo-NIP8[/youtube]
 
Have you actually calculated the power required to compete with or beat a top fuel dragster? My suspicion is that it will be less than 2MW. If you figure acceleration at any given speed, its just physics. Obviously you have to assume some wind resistance at higher speeds. Now that I think about it, I have my doubts about being able to hit top speed. Thats where you're going to need all the power you can get.

If its just controller testing, he should think about using two identical controllers placed in a loop, where one is the controller and the other regenerates and circulates. This is dc motor control right?
 
AC system, guys.

6 MW is possible with 300kg of 20kW/kg batteries.
6 MW in 4 seconds seems doable with a pair of Tesla motors, CO2 cooled before each race...

Regarding power needed, I haven't done any calcs...
 
480 km/h top speed
1200 kg
4 seconds run

2,67 MW average @ wheels

If const torque
0 W @ 0 speed
5,333 MW @ topspeed

Right?
 
Agreed with Punx0r. Currently, the best batteries I can find are 90C cells from HK. http://www.hobbyking.com/hobbyking/...mate_7500mah_2S2P_90C_Hardcase_Lipo_Pack.html

At 90C, and somewhere around 2.8V per cell, that's 4050W and 350g for 11.5kW/kg, which would require 518kg of batteries to get 6MW, or 52% of the minimum weight.

I honestly was expecting this to be pretty much impossible given current battery technology, but... I'm afraid to say... might be possible.

This is, of course neglecting the weight of everything else. The biggest trick will be finding a type of machine that will be capable of such high peak power. I believe with this amount of pulse application, that anything involving magnets will likely demagnetize, so that suggests the variety of DC motors that don't have permanent magnets such as the drag motors that seem to be popular.

I could be wrong about the motor bit, but it's just out of my wheel-house to think of a motor going WAY WAY WAY WAY over it's steady-state rating for 5 seconds.
 
You must have missed the 130C Nanotech ?.....and i am sure I have seen cells with even higher C ratings. :shock:
But, even if they were real, available, without major sag, and even if you could find a 6MW motor (?) to make the 1200kg projected weight.....
....you still end up 200kg heavier than a Fueler.
Since the Fuelers have enough torque to lose traction pretty much at any point down the strip, power or torque is not the main issue,.....weight will be the deciding factor.
 
For those who aren't familiar with bench-testing high power electronics, you can reduce the cost of a supply by just supplying losses and circulating the rest. I did that with one of my previous jobs with this unit http://www.amsc.com/pdf/PM3000W_DataSheet.pdf

Here's a diagram as to how it gets connected. The tricky part is control. Power flow can be in either direction.
 
Saft VL5U, 5Ah can be discharged at 400C continuous. That's 2000A.

Thanks for the idea halcyon!
 
I need some of those cells...

6MW for 6 seconds = 8333Wh

The saft cells return 42Wh/kg at 100C and 36Wh/kg at 200C. At 400C let's assume that continues linearly to 24Wh/kg.

8333 / 24 = 347kg

Add some reserve capacity, casing and interconnects...

Even done from a peak power perspective, at 400C the cells deliver 19kW/kg.

6000kW/19 = 316kg.
 
Punx0r, you gotta remember that at low speed, the full power won't be required, so peak power is 6MW, but average is around 3MW (triangular power profile).

Does that help? I'm sure it helps with thermal issues on motors (that we have yet to get to)
 
Punx0r said:
Even done from a peak power perspective, at 400C the cells deliver 19kW/kg.

6000kW/19 = 316kg.
:?: :?: 19kW/kg ..data source ?
The ARL test report states the power density at 400C to be 14kW/kg.
so 6.0MW would need 6000/14 = 428 kg of cells.
Ref http://www.google.com.au/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0CB8QFjAA&url=http%3A%2F%2Fwww.dtic.mil%2Fcgi-bin%2FGetTRDoc%3FAD%3DADA494956&ei=ZHs0VLe-BpKuogTk3IGoAg&usg=AFQjCNGf1lF-pfrSNyBet6MZpPMOs3oEQA&sig2=S4iC0_-8XPEC1m-mHsCidw&bvm=bv.76943099,d.cGU

Without even thinking in detail about motor or controller sizes and weight , i would guess you will be 500kg over weight compared to the Fueler.
..and that will kill any chance of being competitive !
 
Maybe there is an assumption that it is above 20C? The single value of power density is a little silly considering it depends on temperature and state of charge, but it's been shown that higher temperatures decrease chemical impedance.
 
Beating a top fuel dragster over 1000' in reality is going to be very, very challenging. These days they run more like 10,000 HP.

Beating it in theory is possible, battery power is most definitely not the biggest challenge. I suspect the power stage, converting all that current into motion without letting out the magic smoke will be difficult. Also a substantial challenge - safety... how does one adequately fuse something that's essentially going to be straight up shorted out anyway? Any accidental discharges can result in some pretty big clouds of sprayed molten copper and arc flash is pretty nasty.

Best suggestion would be to take the very highly refined technology in existing dragsters and just change the source of power, don't try to reinvent the wheel... rather than designing a power stage capable of varying power, just run with the existing hydraulic clutch setup that delivers varying power levels over the course of the run.

If anybody is interested in the incredible tech in top fuel/funny cars have a look at this: http://www.motoiq.com/MagazineArticles/ID/3099/Nerds-Eye-View--Inside-10000-horsepower.aspx
 
Ohbse said:
Beating a top fuel dragster over 1000' in reality is going to be very, very challenging.

I don't think anyone said this would be easy, or even plausible at the end of the day. Entertaining to think about.


Ohbse said:
a substantial challenge - safety... how does one adequately fuse something that's essentially going to be straight up shorted out anyway? Any accidental discharges can result in some pretty big clouds of sprayed molten copper and arc flash is pretty nasty.

Would you qualify this as less or more dangerous than TF dragsters through the various years in the past where many engines and parts exploded? I agree that arc flash is pretty nasty, but so can a rogue blower.

Thanks for the link though, entertaining for sure. Efficient? not in the least.
 
halcyon_m said:
Maybe there is an assumption that it is above 20C? The single value of power density is a little silly considering it depends on temperature and state of charge, but it's been shown that higher temperatures decrease chemical impedance.
The ARL tests do have data for power density at 60deg C, but only for 200C and below discharges.
At 200C (1000 A) the Power density increased by 0.8 kW/kg (8.7 to 9.5 kW/kg) .. so you might expect a similar improvement at 400C..maybe ?.. still a ways off the 20kW/kg figure,..but damm good by any standards !
Who is to say that 400C is the limit with these cells ?
They tested a 4800 amp pulse discharge for a fraction of a second, so maybe 2500 A could be possible for a few (2-3) seconds , which would give the 20kW/kg result ??

I wonder what cells the F1 teams use in their ERS battery packs, ?
..they need all the C rate they can get !
I believe Ferrari use Saft cells, so maybe they have a more developed version of this one.
 
Very true. That report is from 2009 - who knows what has been developed since then?

Hillhater said:
The ARL test report states the power density at 400C to be 14kW/kg.

Well spotted. Looking again I can see that Table 2 clearly states 14kW/kg. I think I must have wrongly extrapolated 19Kw from elsewhere - it was late last night ;)
 
Back
Top