RESISTORS FOR CHARGING (DISCHARGING) LARGE CAPACITORS AT FAST RATE
Signification, Tue Mar 19 2019, 08:35AMI would like some expert opinions on choosing a proper resistor for discharging a large (uF)
capacitor at a fast rate. Here is a typical example:
The capacitor is rated at C=10,000uF and is charged to a voltage V=500v. The time constant
is only 1 second (5RC=5s). I plan on using a resistor R=100 ohm. The instant
R is connected across the fully charged C, a current I=5A is drawn. That is,
at t=0+, P=2500W is put on R. I integrated I vs T for the resistor over 5RC, and got,
an exponentially decaying function (over 0 to 5RC) a total charge q=5Q. At 5s, the power
dissipated by the resistor is only 1/10 W at 3.3v! I plan on using a 100W, 100 ohm
resistor. I have not had a chance to test any of this, but will soon...any opinions, or 'rule of thumb'
on choosing a resistor (ohms, watts), given the discharge time?
...adding (a bit later): Using the capacitor eq. Q=CV, I found that the charge on the
fully charged cap was 5Q, which agrees with the 5s discharge dq/dt=1A (should be delta's)
found by actual integrals over the 5 second integration period. I KNOW I don't need a 2500W
resistor (unless the discharge time was very, very long), and probably don't even need the 100W
resistor either (if not repeating this sequence at a high rate).
Re: RESISTORS FOR CHARGING (DISCHARGING) LARGE CAPACITORS AT FAST RATE
DerAlbi, Tue Mar 19 2019, 09:29AM
Complicated problem indeed.
First of all, you need to keep in mind that the energy dissipated will always be equal, no matter which resistance you choose. Capacitive capacitor charging is always exactly 50% efficient if you start at 0V. (If you start near 100% end voltage you reach near 100% efficiency. This is why charge pumps have good efficiency once they started up).
Since you always generate the same amount of heat in your resistor, you have to decide in what time frame you want the dissipation. Choosing a large RC-time constant will lower the resistor power requirement, choosing a short RC-time constant shifts the problem towards heat capacity energy absorbtion.
As a rule of thumb, THT resistors can be overloaded by a factor of 5 - 10 for up to 5 seconds; reducing the time, may increase power handling during the shorter time period (comparable to an I2t melting integral) however this time period has potentially lower limits as hotspots may form so that less and less heat capacity is active during the overload. So dont expect 1000x overloading potential for 1ms.
Googling for "power resistor overloading" may bring you to resistor manufacturers like Visay.
take a read:
DerAlbi, Tue Mar 19 2019, 09:29AM
Complicated problem indeed.
First of all, you need to keep in mind that the energy dissipated will always be equal, no matter which resistance you choose. Capacitive capacitor charging is always exactly 50% efficient if you start at 0V. (If you start near 100% end voltage you reach near 100% efficiency. This is why charge pumps have good efficiency once they started up).
Since you always generate the same amount of heat in your resistor, you have to decide in what time frame you want the dissipation. Choosing a large RC-time constant will lower the resistor power requirement, choosing a short RC-time constant shifts the problem towards heat capacity energy absorbtion.
As a rule of thumb, THT resistors can be overloaded by a factor of 5 - 10 for up to 5 seconds; reducing the time, may increase power handling during the shorter time period (comparable to an I2t melting integral) however this time period has potentially lower limits as hotspots may form so that less and less heat capacity is active during the overload. So dont expect 1000x overloading potential for 1ms.
Googling for "power resistor overloading" may bring you to resistor manufacturers like Visay.
take a read:
Print this page