Further to my comments below (I did some searching and reading) you might try a PTC thermistor (spelled properly this time) like the ones in this datasheet: http://www.vishay.com/docs/29086/29086.pdf. They've got a bunch more types of high-current surge protect PTC's that might be a better match. The larger ones in this datasheet have a cool resistance of a couple of ohms and a hot resistance of 10^3 to 10^6 ohms. They can handle peak currents of 10-13 amps and I think can heat up in 1-2 seconds.=20 If one of these were in parallel with your current voltage dropping resistors they would provide a significant starting voltage/current for the motor for a second or two, then the dropping resistors would hold it at 12V while the motor runs.=20 Kind of like a light bulb but more durable. -----Original Message----- From: piclist-bounces@mit.edu [mailto:piclist-bounces@mit.edu] On Behalf Of Fred Langley Sent: Tuesday, July 26, 2011 6:57 PM To: 'Microcontroller discussion list - Public.' Subject: RE: [EE] Quick and dirty Power Supply 48->12V What you need is a thermister with a very positive temperature coefficient.= =20 A light bulb is an example of such a thing.=20 Put three 12V automotive brake light bulbs in series with the motor. They flow about 1A at 12V. The motor would then get 48V as a starting surge for = a fraction of a second and settle down to 12V once the bulbs are heated up. Its quick and dirty and possibly fragile and un-reliable if the bulbs burn out, but these bulbs last many years of applying brakes in a vibrating car, so maybe it would be OK.=20 I think there are more solid state sorts of thermisters that might work. I think they use something similar on some power window motors so you do not burn out the motor if you keep ten switch pressed.=20 -----Original Message----- From: piclist-bounces@mit.edu [mailto:piclist-bounces@mit.edu] On Behalf Of Denny Esterline Sent: Monday, July 25, 2011 10:02 AM To: Microcontroller discussion list - Public. Subject: [EE] Quick and dirty Power Supply 48->12V Ok, so I have product that has a small 12V motor running off a 48V bus. It's exceptionally low duty cycle, about 5 seconds a day. And low power, about 500mA. Given that, I had been using a couple resistors and a zener diode to power it off the 48V bus. It's a little less than elegant, but it'= s been working fine and fit within my space constraints. Recently I was asked to build a variation of it that needed a larger motor. I'll spare you the gritty details, but eventually I found a suitable motor and shoehorned it into position. It was also rated 12V so I gave no thought to powering it. Now the face-palm moment - it needs more current (duh). The kicker is it needs at least 2A to start and drops back down to about 1A when running. Attempting to alter the resistor values to accommodate the 1A run current leaves it with a very anemic startup (won't start under load) and adjusting the resistors to allow 2A for startup results in excessive speed or very unhappy zeners. The simple fix would be bigger zeners, but I can't seem to find anything bigger than 5 watt rated parts. The elegant fix would be to lay out a small buck regulator, time is not my friend. This needs to ship before I could have boards made. The quick fix would be an off-the-shelf DC-DC converter. I haven't been abl= e to find any smaller than about 2"x2"x1" thick and they all seem to need to be PCB mounted, see above. So, I'm turning to the list for out-of-the-box suggestions?? Thanks -Denny --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .