How to create a 3000 Ampere peak current ?

[walkura]

Let me start with the history of the (intriguing) problem at hand .
At my work we build battery chargers for cars .
Some have a so-called startaid which means that you can hook it up to youre out of juice battery and start with the charger as aid .
Ahum *supposedly* a 35 Ampere charger could deliver a 300 Ampere peak (given that the output voltage drops to 7.5 instead of 15 volt)
(i don't believe a word of those specs to be honest)
But here it comes somebody called over a startingunit for a MIG jet .
28 volts on the output with the capability to sustain 3000 Amperes for 1 second .
Curious (and mildly excited) i tryed to puzzle and calculate .
Simple calculated it means U*I*T = 84000 joules .
Even with 3 transformers of 28 volt 100 amperes on 3 fases i hardly scratch the surfaces .
28*100*3 i come to 8400 joule .
When i calculate that according to the same *supposedly* ratio as the 35 Ampere charger then i'm still 12000 joules short .
I must confess it made me curious and reminded me of this picture .
http://www.tesladownunder.com/HVEasterExplosion1000.jpg
(easterbunny meets 3.5 KJ)
I'm aware that its not audio related but it keeps me puzzled .
Maybe some of you here had any experiences with huge peak Amperages .
For now it just eludes me how it can be done


Have a nice eve Walkura

[Frank Berry]

Add a BIG bank of electrolytic capacitors to the output of the power supply and really HEAVY GAUGE cables.

Of course, it would take some time to charge the capacitor bank through current limiting resistors. When the capacitors charged to some voltage near the output voltage of the supply, a contactor would pull and bypass the current limiting resistors.

Still .... 28 volts at 3000 amps. The conductors would have to be REALLY big.

What kind of a connector would you use to pass 3000 amps?

[rikkitikkitavi]

simple:

the transformer is rectified. this is then feed to a battery, probably lead-acid parallell to the output.

the battery is designed to provide huge currents for short durations.

a normal car battery with a capacity of 65Ah can provide 65*3600 = 234000A for one second in theory, in practice the current is strongly reduced due to internal resistance to a few hundred amps.

[mzzj]

Originally posted by Frank Berry
Add a BIG bank of electrolytic capacitors to the output of the power supply and really HEAVY GAUGE cables.

Of course, it would take some time to charge the capacitor bank through current limiting resistors. When the capacitors charged to some voltage near the output voltage of the supply, a contactor would pull and bypass the current limiting resistors.

Still .... 28 volts at 3000 amps. The conductors would have to be REALLY big.

What kind of a connector would you use to pass 3000 amps?

For 1 second discharge batteries are more cost-effective than caps. Lead-acid AGM's or A123-systems Li-Ions for example. 11-series 30-parallei(total 330) battery pack of ANR26650M1 cells would be capable of supplying over 3kA for 10secs minimum and battery pack only weights around 20kg !


Cables... 70mm2 cable drops 0.8v/meter at 3kA so it would be good enough for cables shorter than 4 meters or so. ( 1 sec, not continuous) Same goes with connectors, for example welding cable connector DIX 70/120 should handle 3kA for short perioids easily.

check out those busbars on power stage:
20khz PWM @2kA, 24V

[walkura]

Thanks for youre reply's so far .
I calculated the option of capacitors by the following formula joules = C*V2 / 2 with 28 volt @ 1 Farad i get just 392 Joules or 28 coulombs.
(i left school almost 25 years ago so forgive me when my formula's are a lil rusty)
Which in practice would mean that even when you amount a square meter of capacitors you don't even get close .
The option of battery's i did consider but didnt calculate thorough .
About 10 years ago we did a test with peak Amperage of Nickel Iron Battery's (now known as Nickel hydride) we measured over 200 Amperes out of the AA model .
My doubts in this setup is the how to switch it .
You cant create anything with *hot* cables (afterall this sort of peakpowers make big holes in youre platework when you drop a cable)
For rectifying i was thinking to use freerunning synchronous rectification & switch with n-mosfets like irfba1405p.
http://www.irf.com/product-info/data...irfba1405p.pdf
For now thats as far as my plans were .
Its a bit the season for batterychargers so i didnt have the time to do more or less anything .

Have a nice day Walkura.

[peranders]

Walkura, have you seen my thread about kA?
Failure in a huge capacitor board

I'm working right now with lot's of amperes