Pulse Toaster running off battery

[cupoftea]

Hi,
We are doing a "pulse toaster" which gives a stream of high current pulses, 40ms long, which pulse into the element of the toaster.

Pulses are 300W "high".
Element is 0.41 Ohms.
Average power is only 50W. (so T = 240ms and Pulses are 40ms long)

The battery is 11V1 lithium (3 lithium in series) The battery (needing to be small) overheats due to the rms of the current pulses.
As such, we wish to look into having a capacitor bank outside the battery, and so the pulses get drawn from that instead.

.....But Do you agree, this is pretty well a non-starter......there's no way we could have a reasonable size capacitor bank to stop the pulses from being drawn from the battery anyway?
Basically, do you agree, we need a Boost converter to boost up to a high voltage capacitor bank.....then use a Sync Buck to buck down to the element for the pulses?

We need a solution which is small.

 

[barry]

If my math is right, you’re asking for a lot of energy (12 Joules). That would be like a 0.2 Farad capacitor, assuming all the energy comes from the cap.

 

[betwixt]

The capacitor is only going to level the peaks, it won't reduce the overall consumption so the power if anything drawn from the battery is going to be higher. Are you sure the overheating is caused by pulsing the current and not the average load? In my thinking, the heat produced by the heavy load is also pulsed so it has a cool down period between them, the average heating will not be improved.

Brian.

 

[crutschow]

The capacitor is only going to level the peaks, it won't reduce the overall consumption so the power if anything drawn from the battery is going to be higher.

Not necessarily.
The high peak current of the short pulse will have a higher RMS current value for a given average current, which will thus cause greater heating and loss in the battery's series internal resistance.

 

[cupoftea]

Yes, i see your points...i think the only way is to use a Boost converter off the battery, and have it drawing the average power level from the battery in constant current..then draw the pulses from the output of the booster.
Would you agree.?....perhaps making the element resistance higher aswell

 

[BradtheRad]

By adding an LC input filter, you draw smooth current from the power supply. (Namely 9 Amperes continually rather than 39A pulses.)

Instead of a boost converter, consider a buck-boost. The coil generates current pulses briefly to the output stage.

Notice if you can install 13 V worth of batteries, it provides sufficient power to the load and no need to boost anything. Then you can omit L2 D1 C2.

 

[cupoftea]

Notice if you can install 13 V worth of batteries

Thanks, but sorry we cant. We are using the biggest battery pack we can manage for.
I wasnt sure why a Buckboost was better than a booster?
I appreciate initial inrush is lower with a booster though, so thankyou.

 

[crutschow]

A large capacitor across the battery will significantly reduce the large peak battery current and the consequent high RMS dissipation, so it seems that would be a good and easy solution.

 

[cupoftea]

.....Thanks, as Barry appears to imply above, the capacitor would have to be very big to make the drawn current pulses be diverted to the capacitor, rather than being drawn from the battery......its 300W at duty of 0.17 and the pulses are 40ms long..... and i cant see the battery current becoming a nice constant , steady DC , due to simply putting a capacitor outside the battery.

 

[BradtheRad]

I wasnt sure why a Buckboost was better than a booster?

Yes, a boost converter appears equally suitable. It helps to add a switching device at the output.
The load still gets 300W pulses (53W average).
Components do not need to be as massive.

The converter draws 5.5A average from the batteries.
The LC input smoothing filter isn't absolutely necessary but it can be included if desired.

 

[SWRmeter]

Hi,
We are doing a "pulse toaster" which gives a stream of high current pulses, 40ms long, which pulse into the element of the toaster.

Pulses are 300W "high".
Element is 0.41 Ohms.
Average power is only 50W. (so T = 240ms and Pulses are 40ms long)

The battery is 11V1 lithium (3 lithium in series) The battery (needing to be small) overheats due to the rms of the current pulses.
As such, we wish to look into having a capacitor bank outside the battery, and so the pulses get drawn from that instead.

.....But Do you agree, this is pretty well a non-starter......there's no way we could have a reasonable size capacitor bank to stop the pulses from being drawn from the battery anyway?
Basically, do you agree, we need a Boost converter to boost up to a high voltage capacitor bank.....then use a Sync Buck to buck down to the element for the pulses?

We need a solution which is small.

You need LiPo 3S batteries of higher current capability, or try use 3SxC LiPo batteries with a higher rating.
Adding electronics is a waste of time where the battery itself can do the job.