2
\$\begingroup\$

Trying to refresh my knowledge I was design and construct a fixed unregulated power supply 12V/5A using parts that I have around. Parts are 150VA/12V toroid, an 25A rectifier bridge in KBPC case and 20 electrolytic capacitors 1000μF/35V each. Trying to calculate a plate heatsink for the rectifier, I test my construction with an ohmic load (a tungsten lamp rated 55W/12V) and monitoring the rectifier case temperature enter image description here I find the temperature rise abnormal and I cut the power around 100oC, then I remove the capacitor bank, and get temperature again enter image description here

This is the graph enter image description here

I can not explain the reason of this difference in temperature rise. I used another rectifier but no deference even using a 35A one.
The maximum current during charging period is calculated about 33A but it lasts just 1,8ms and with a duty cycle 0.22 the total dissipation is about 6.6W in total. In no load condition, there is no consumption by the capacitors.

\$\endgroup\$
1
\$\begingroup\$

Its going to come downto powerflow and how "DC" your DClink is.

Without the cap the 55W bulb will be provided with ABS(Sin) voltage

enter image description here

This provides an RMS voltage of Vpk/1.414 and thus the power reduced. the 55W/12V bulb is an equiv of: 2.618R with an XFMR giving out 14.1pk voltage => 9.97Vrms => 37.98W

Now when you add the capacitors, you are "smoothing" the DClink enter image description here This increases the power supplied to the resistive load as it is now a DClevel. 20mF worth or capacitance should provide a DC output. 14.1V on the DClink => 75.939W Increased power, increased heating.

\$\endgroup\$
  • \$\begingroup\$ Yes indeed this is correct. However my question was about the power that this specific device can handle (i.e 25A) versus the requested current by the load (6~7A). A deep study in the manufacturers data and recalculations indicates that even for 7A rms (causing a total 6~7W dissipation on the bridge), the case temperature in free air rises to about 115oC which is "normal", since junction temperature maintains to about 135oC. So to use a KBPC rectifier in free-air, should be consider very carefully. Is it trough that this package case it is designed to use mainly in motorcycles? \$\endgroup\$ – GR Tech Oct 31 '13 at 22:00
0
\$\begingroup\$

I don’t know if it is a coincidence, but when we draw a reverse derating curve (in red) this gives the case temperature rice in free air as a function of current. If derating curve slope considered as 1/Rθjc, the reverse is a plot of Rθjc. Is this correct?

enter image description here

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.