At 05:31 PM 3/15/2010, ivp wrote: >I've been looking around for a sealed lead charger. A piece of >equipment I'm building has to be run off a battery, 6V or 12V, >and I prefer to use an SLA. > >I figure I could make a constant current supply from an LM317 >(or is that just current limited ?) and a constant voltage supply from >a 7805 or 7812 baseline I can offer some suggestions but need a tiny bit more information: what is the source of the charge current? First off - I should mention that most of my experience is with small Sealed Lead Acid batteries - 1.2AH through 35AH, both 6V & 12V. Although the majority of my designs are used under garden-variety conditions, two of my designs (so far) are used in (and certified for) hazardous locations - one for Class 1 Div 1, the other for Class 1 Div 2. In general terms, I usually charge batteries to their equalization voltage, then drop to float mode when the charge current falls below some low value. If the charging supply is intermittent, (plug-in when charge needed), I always start up in equalization mode. If the charging supply is continuous, I switch into equalization mode on a periodic basis (once per week typically). Although I have used LM317-type regulators for some of my chargers (and still do), I find them inefficient in that they require a supply voltage at least 2.5V higher than the highest battery voltage + drop across the blocking diode. That's OK if you already have that supply voltage available (charging a 6V SLA from a 12V automotive supply, for example). If I need to supply the charger power supply, I'll either use some form of MOSFET output device or, sometimes, modulate the output voltage of a switch-mode buck-convertor. Chargers other than solar-powered tend to be linear supplies, although I have done a couple of switch-mode chargers. If I have enough voltage headroom available, the pass device is often one or more LM317 devices - I treat them as high-gain transistors. If I don't have the voltage headroom available, I tend to use P-channel MOSFETs as pass devices. Current sensing is most often one or more Zetex ZXCT1008 and suitable shunt resistor(s). I do two distinct types of solar-powered chargers. The easiest solar charger is what I call a bang-bang controller, where I use a MOSFET to put a short across the solar panel when the battery voltage is above its desired level. You pick the frequency of the on-off switch to suit your needs - I've gone as slow as sampling once per two seconds (0.5Hz) to as fast as 10ms. I don't notice much performance difference as the switching speed changes. Bang-bang chargers are suited for larger solar panels and battery-banks - anything above 60W solar panel size to perhaps 200W. I'd probably go back to series pass devices for systems with more than 200W of solar panel but haven't had to do anything that large yet. I generally size the solar panel such that its highest current output (full sun in mid-summer) is less than the maximum that the battery bank can accept. That means that I don't have to monitor charge current, only voltage. The other solar charger is a linear charger done with MOSFETs. I like to use these on smaller systems, where currents are low enough that heat dissipation is not an issue. One recent design has been successfully in use for 1.5 years so far (two winters in Arctic conditions). I'm quite proud of that one - quiescent current drain with no solar input is a few microamps. The entire design is analog (quad op-amp) but the PIC controller sets whether the charger is in equalization or float mode (one logic line). Finally, I'll mention something that not enough people consider: if you have a product that is only ever charged from an automobile's cigarette-lighter plug, do consider using a 12V SLA. The charge circuit is essentially a plug and something to protect against shorts: the charge voltage you get from an automobile charging system is almost exactly what a 12V SLA requires. The only downside is that the charge voltage is temperature-compensated for the battery sitting under the hood as opposed to the temperature inside the automobile's interior. In practice, I find that is not much of a problem at all. Hope this helps! dwayne -- Dwayne Reid Trinity Electronics Systems Ltd Edmonton, AB, CANADA (780) 489-3199 voice (780) 487-6397 fax www.trinity-electronics.com Custom Electronics Design and Manufacturing -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist