Energy to wattage ? / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Energy to wattage ?

1 2 next

Move Thread

LAN_403

Thomas F

Wed Jun 04 2008, 07:54AM

Registered Member #503 Joined: Sun Nov 19 2006, 06:37PM
Location:
Posts: 59

Hello all,

This is probably quite basic. I have a 10nF capacitor which is charged to 6kV. I need to discharge it and limit the maximum current to about 10mA ( ergo 600k resistor ).This current would be dropped across a mfr 0.5W resistor ( 1K) and the voltage ( 10V) would be used to trigger a comparator circuit.
The energy to be discharged would be given by 0.5*C*V^2. Is there anyway to calculate the wattage of the 600k resistor from this ? Or is just V * I = 60W ?
Any other factors to be considered ?
Can this impulse energy be dissipated in any other way ? A capacitor across the 1k ? Ceramic capacitors for kV seem to be more easily available than any other HV components these days.

Thomas

Steve Conner

Wed Jun 04 2008, 09:01AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Just divide the energy by the time the discharge takes. Energy/time = power.

If the discharge is repetitive, multiply the energy by the number of discharges per second.

Thomas F

Wed Jun 04 2008, 10:21AM

Registered Member #503 Joined: Sun Nov 19 2006, 06:37PM
Location:
Posts: 59

Hi Dr.Connor,

Thanks for the reply.

The energy is 0.5*(6000^2)*1E-8 = 0.18 Joules (units ??)
The time constant (R=500k,C=0.01mFd) works out to 5msec. Assuming 5 time constants for complete discharge,thats 25msec.
Power = (0.18/25)*1000 = 7.2 W ( still looks a bit high )
Suppose I have 10 discharges in a second how do I factor that in ?
AFAIK it cannot be 7.2*10 which works out to 72W which is more than the continuous rating.

Cheers

Thomas

Steve Conner

Wed Jun 04 2008, 12:58PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Thomas F wrote ...

AFAIK it cannot be 7.2*10 which works out to 72W which is more than the continuous rating.

No, it's 0.18 joules * 10 events per second = 1.8W

Thomas F

Wed Jun 04 2008, 04:55PM

Registered Member #503 Joined: Sun Nov 19 2006, 06:37PM
Location:
Posts: 59

Thanks for that.

Power = (0.18/25)*1000 = 7.2 W ( still looks a bit high )

So this would actually be 7.2E-006W/event then ?

Thomas

Bjørn

Wed Jun 04 2008, 08:22PM

Registered Member #27 Joined: Fri Feb 03 2006, 02:20AM
Location: Hyperborea
Posts: 2058

Power = (0.18/25)*1000 = 7.2 W ( still looks a bit high )

It looks fine, a tiny amount of energy discharged in a short time can deliver a lot of power.

It is 0.18 joule (Ws) for each event. 0.18 W for one second is the same energy as 7.2 W for 25 ms.

Thomas F

Thu Jun 05 2008, 05:35AM

Registered Member #503 Joined: Sun Nov 19 2006, 06:37PM
Location:
Posts: 59

Hi Bjorn ,

Thanks for that. Now the point is do I size the resistor for 7.2W or 1.8W or somewhere in between ?
The way I look at it I should estimate the maximum number of discharges / second i.e. 20 ( 25msec for charging / 25 msec for discharging).
So its 20 * 0.18 = 3.6W ?

Regards
Thomas

Bjørn

Thu Jun 05 2008, 06:03AM

Registered Member #27 Joined: Fri Feb 03 2006, 02:20AM
Location: Hyperborea
Posts: 2058

3.6 W is the simple answer. The complete answer is more complicated and depends on the type of resistor and the design of the circuit,

Some types of resistors have a very low thermal mass and may overheat in a very short time, those you should rate for the full pulse power. Homogeneous resistors like carbon cylinders will average out the power over several seconds and can within reason be rated for the average power.

If you rate the resistor at maximum power it will run hot and that is usually not a good idea when it comes to reliability.

Thomas F

Thu Jun 05 2008, 11:08AM

Registered Member #503 Joined: Sun Nov 19 2006, 06:37PM
Location:
Posts: 59

Thanks Bjorn,

Actually I am getting the resistors in a couple of days.It cost me a bomb ( $11 per resistor).The guy claims its made from some wire which is more expensive than gold. I have a sneaking feeling that I am getting ripped off , but its just 3 pieces now to get a proof of concept thingy.
This resistor is part of a divider which feeds a comparator input.

This was Steve Ward's idea of course,which worked great in a previous application,but in this one because of the capacitor ( energy ) at the kV output ,its blowing the sensing (1k) resistor.I am assuming this of course.

Thomas

Steve Conner

Thu Jun 05 2008, 11:24AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

I'd just have connected 30 ordinary 0.6w metal film resistors in series and got on with something else. More expensive than gold, meh.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob