wdp@dukee.egr.duke.edu (Dev Palmer) (07/12/90)
Date: 12 Jul 90 Mr. Streeter: Since you are dropping 7 V across the regulator, you are generating quite a bit of heat and the regulators may shut down or at least begin to shut down. I had a circuit where the regulator was pulsing on and off. It looked like low voltage to a multimeter, but an oscilloscope revealed the thermal cycling. If you really want 3.00 A (giving 15.00 W), you really should heed collie's suggestion to use a TO-3 package, but you should use the LM350K adjusted to 5 V and mounted on a BIG heat sink (heck, mount it on the firewall!). That should give you all the juice you need. Buy it from DIGI-KEY (1-800-DIGI-KEY, part number LM350K $5.56) or your favorite supplier. Also, check National Semiconductor's Voltage Regulator Handbook. The 1982 edition has a section called Automotive Applications which should be very informative. Good Luck! Dev Palmer Electrical Engineering Duke University (not DIGI-KEY or National Semiconductor) Durham, NC 27707 (919) 660-5282 wdp@dukee.egr.duke.edu
john@frog.UUCP (John Woods) (07/14/90)
In article <987@cameron.egr.duke.edu>, wdp@dukee.egr.duke.edu (Dev Palmer) writes: > Since you are dropping 7 V across the regulator, you are generating quite a > bit of heat and the regulators may shut down or at least begin to shut down. > If you really want 3.00 A (giving 15.00 W),... use the LM350K adjusted to > 5 V and mounted on a BIG heat sink This would be my recommendation, too, but another thing to try would be a resistor in series between the battery and regulator. Set it to drop 4-5 volts at maximum current (1.3 ohms or so at 3A), and it will dissipate half the power for you, keeping the regulator(s) much cooler. Keep the resistors away from the electronics, and everything runs cooler. I use a similar trick on this computer (frog) to run a 12V disk drive from a 24V supply. If your heat sinking is really effective, you can get by with the cheaper LM350T (plastic TO220) but in frog's case, it helps that there is a fan 1/2 inch from the heatsink! The LM350 is a great part. The 5A and 10A regulators are so much more expensive that for just screwing around it makes more sense to wire a pass transistor by hand, but the LM350 is cheap enough that it's better to use it instead. -- John Woods, Charles River Data Systems, Framingham MA, (508) 626-1101 ...!decvax!frog!john, john@frog.UUCP, ...!mit-eddie!jfw, jfw@eddie.mit.edu
streeter@theory.lcs.mit.edu (Kenneth B. Streeter) (07/24/90)
Thanks for all of the suggestions that I've received for the 12V to 5V conversion in a car. What I've done is replace the 7805H (220 package) regulators with 7805KC (TO3 package) regulators to drive the logic circuitry. The logic draws relatively little current (each regulator needs to push about 150 mA) and the pair of regulators are probably actually overkill for this. To drive the current-hogging LEDs, I've used a LM323K (TO3 package) with a 3" x 5" finned heat sink. In this way, separate regulators drive the logic and the LEDs, so any power cycling that may occur due to the LEDs making the regulator too hot affects only the LEDs (a momentary loss of power to the LEDs doesn't hurt anything, whereas a momentary loss of power to the registers, etc, would lose the state information.) What I've done is basically the "simple" fix -- better, more robust suggestions were given, but I wanted to use the parts that I could find at local second-hand electronics stores and flea markets. I'd like to thank everybody that gave suggestions, especially: Tom Bruhns (tomb@hplsla.hp.com) Jeff DePolo (depolo@eniac.seas.upenn.edu) Mark Hoy (mark@retix.retix.com) Cary Logan (logan@titan.tsd.arlut.utexas.edu) Stephen MacMinn (macminn@powertool.crd.ge.com) Dana Myers (dana@devnet.la.locus.com) Phil Ngai (phil@brahms.amd.com) Steve Nuchia (steve@eddie.mit.edu) Dev Palmer (wdp@dukee.egr.duke.edu) Ahmed Shamim (ahmed@nicsa.ircc.ohio-state.edu) Norm Strong (strong@tc.fluke.com) David Thomas (bbx!yenta!dt@unmvax.cs.unm.edu) John Woods (john@frog.UUCP) -- Kenneth B. Streeter | ARPA: streeter@im.lcs.mit.edu MIT LCS, Room NE43-350 | UUCP: ...!uunet!im.lcs.mit.edu!streeter 545 Technology Square | (617) 253-2614 (work) Cambridge, MA 02139 | (617) 225-2249 (home)