One thing that comes to mind is that it may rely on a certain range of battery internal impedance to work properly. Either it wants to pull high current pulses from the battery and can't do that from your wall-wart, OR, the wall-wart voltage stays high at higher currents than the batteries do and there is no current limiting action that you would have with the batteries. For example, I've seen LED flashlights which have no current limiting other than the battery internal resistance. If you run them on a PS that can source more current they blow out the LEDs. Sean On Wed, Jan 23, 2013 at 10:31 AM, William Bulley wrote: > I got tired of buying new AA alkaline batteries and later tossing them... > > I have a cloth tapestry, about 24 inches high by 36 inches wide, which > shows a sylvan cottage scene complete with creek, pine trees, snowman, > and snow. Within layers of cloth, this tapestry features colorful LED > powered enhancements (i.e., the snowman's nose is lit, the roof line > of the cottage is outlined in multicolored glory, etc.). Some of these > points of light are solid, while many others blink. > > Inside, the technology consists of a battery pack that holds three AA > cells and has an on off switch. The electronics are powered by these > batteries. I assume the electronics consists of some silicon logic to > blink the points of light (those that blink), while another part of the > circuit routes voltage to the remaining LED emitters. > > When batteries are inserted, and the switch turned "on", the display > lights up, some blinking, some solid - very colorful and delightful. :-) > > A matrix of fibre optic strands carry the LED emitters output to all > the various points of interest on this tapestry. All this works fine. > > One other feature: once powered "on", the display electronics "times out" > after six hours, and the tapestry goes dark until the slide-switch is > moved from "on" to "off" and then to "on" again. All this works fine. > > Like I said, the thing eats batteries like the "Cookie Monster"... :-( > > After literally a couple of days, the brightness dims, and within a few > more days the batteries are drained and must be replaced. This is not a > good design, IMHO. :-( > > It occurred to me that a low voltage "wall wart" that produces 4.5 Volts > DC might be employed to replace the three AA cells in the power pack. I > purchased an inexpensive 4.5 VDC wall wart from DigiKey. This unit has > a switching (not linear) design, but as measured with a DC multimeter, > it produces a stable 4.5 VDC exactly. Maybe I should look closer at the > output with an oscilloscope, but I have not as yet. > > After snipping off the connector at the end of the cable, I soldered the > positive lead to the "+" terminal of the battery power pack (all batterie= s > removed), and the negative lead to the "-" terminal of the power pack. A= t > this point all that remained was to plug the wall wart into the mains out= let > and slide the power switch to "on". Voila! I had a mains operated displ= ay > and would never again need to purchase AA alkaline batteries for it. :-) > > Then some time later, maybe a day or two, maybe less, I noticed that the > display was dark. "Ahh" says I, "that six hour timer has caused this." > I then moved the slide-switch from "on" to "off" and then to "on" again. > > NOTHING! ZIP, ZERO, NADA! :-( > > At first I thought of a loose or broken connection or solder joint, but > I doubted that. Then I thought maybe the wall wart had died due to poor > Q/A or other component failure. But no, neither problem was at fault. > > I unplugged the wall wart, and left the power slide-switch in the "off" > position. I tried my initial test after thinking about the problem for > a few days. Lo and behold, the darn thing worked just fine! Well, not > quite, for what I had missed on the first go around was the frequency > of the flicker (blinking LEDs) was more rapid than under battery power. > > And worse still, the six hour shut down timer now seems to last maybe a > minute or two at best. How can this be? I am supplying 4.5 VDC, there > are no components to heat up or cool down. There are undoubtedly some > capacitors/condensers involved in the logic circuit and in the wall wart > but 4.5 VDC should be 4.5 VDC no matter how it is delivered. I have yet > to destroy the tapestry to inspect the logic circuit, but I thought I'd > first ask the list for wisdom or any explanation of this odd behaviour. > > This "enhanced" design has a serious failure mode - when it works at all, > and that seems to be unpredictable - it "runs" at an accelerated rate. > > Sometimes it doesn't light up at all! :-( > > How can this be? > > Regards, > > web... > > -- > William Bulley Email: web@umich.edu > > 72 characters width template ----------------------------------------->| > -- > 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 .