> Is it possible to make -10V from one or more PIC pins ? > The graphic LCDs I need to use for a project require > about -10V @ 5mA for Vee. > The -10V is fairly specific. Even slightly less negative > than that and the contrast is no good Single side capacitor pump will asymptote to about Vsupply + (Vswing x 2 Vdiode) x N for N stages. eg with 5V swing and Schottky diodes at about 0.3V unloaded drop you get about 1 9.4 2 13.8 By double siding / full wave driving the multiplier you approach (Vswing - Vd) x N 1 9.7 2 14.4 Loading will reduce this - usually noticeably, depending on design parameters. E&OE - brain tells me something may have been missed there but I think not (as, if I did, I'd change it ;-) ). So for a true 10V and 5V swing you don't get there either way with 1 stages= , so need 2 stages, so may as well use half wave as simpler. PIC pin, 4 diodes, 4 caps (incl output filter). Not expensive BUT tends to load nastily and be harder than inductive if mor= e than minor power is needed. Inductive scales easily to almost any Wattage as desired. ________ Using an inductor allows a far less "parts intense" solution than any other electronic alternative. Electromechanical may compete - eg pulsing a piezo and then disconnecting it. You MAY be able to capacitor couple a piezo and pulse drive it, thereby removing the need for DC decoupling during the "ringing" phase. PIC pin protection needs to be "well thought out" [tm]. Inductor shunning seems unwise, unless you are especially worried about EM= C aspects (into or out of inductor). A 'tiddly wee' inductor can be low size, low cost, visually almost indistinguishable from an R or C and no different to handle. These would do far more than asked for $NZ0.23 each /5 http://nz.element14.com/murata/lqh32cn101k23l/inductor-1210-case-100uh/dp/9= 522220 Complete basic voltage source then can be eg PIC pin, MOSFET, inductor, diode, output filter cap. I say MOSFET as it allows more cavalier / simple driving, if only just. But almost any old jellybean bipolar will do. PNP in this case in simplest topology. ____ Regulator: Regulation can be by known load, series regulation (std 3 terminal), zener (low precision shunt), or precision shunt. Of these the best good precision low cost solution is perhaps a TL431 and 2 resistors. Under 5 cents in volume in China, not too too many times more in ones in NZ. Farnellerates ... Wow! ... under $NZ0.20 for parts with IC in 1's (assumin= g you buy resistors sensibly). TL431, http://nz.element14.com/texas-instruments/tl431cdbvr/shunt-voltage-referenc= e/dp/1296031 C...suffix NOT expressly described in datasheet they point to. Probably a 2= % part. (A=3D0.5%, B=3D1%, "std" =3D 2%) http://www.farnell.com/datasheets/1305692.pdf Russell McMahon On 5 July 2011 18:10, IVP wrote: > Hi all, > > is it possible to make -10V from one or more PIC pins ? The graphic > LCDs I need to use for a project require about -10V @ 5mA for Vee. > The -10V is fairly specific. Even slightly less negative than that and th= e > contrast is no good > > I've looked around and found switched capacitor and capacitor-diode > designs for positive doublers, positive triplers but not greater than uni= ty > inverters. Some of those designs I've used before and found them quite > adequate. A Schottky diode-capacitor inverter is pretty close to -5V > but I think a tripler would be better than a doubler so I can adjust down > to -10V > > For the time being I'm using an SMPS IC to make it but would like to > know if it's possible with discretes, and without an inductor if that's > possible. Unless the omission of an inductor means the inclusion of a > heap of other components > > TIA > > Joe > -- > 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 .