Dear Tom and all the others that offered advice Thank you for all the advice. Your idea about the way to read the pressure transducer looks very interesting. My current prototipe Altimeter/Variometer uses a MPX5100A pressure transducer with a 13 bit A/D. It can display altitude to a 1 meter resolution with great accuracy. The advantage of the MPX5100A is that it is already temperature compensated and linearized. I have gone into great pain to implement lookup tables to make this instrument as accurate as possible, and do not really want to change that part of the design at this stage. I'll keep your advice in mind for a next generation model. All I want to change at this stage is to use four penlight batteries instead of a 9V battery. You say that it should not be necessary to use the voltage step up converter. I assume then that the discharge curve of the four penlights in series would drop very slowly initially from 6V down to 5V, and once it goes below 5V it means that the batteries are almost discharged anyway and could be thrown away? Thanks once again to you and all the other people that offered advice. Rgds Werner > Reply-to: "Tom Rogers" > From: "Tom Rogers" > To: > Subject: Re: PIC circuit running on penlight batteries > Date: Thu, 11 Dec 1997 10:27:15 -0500 > Werner: > > Don't use that a/d. Use one with an internal 2.5 V reference, then do the > pressure measurement ratiometrically. Essentially the reference is used to > excite the pressure transducer. If your ptrans isn't the bridge type, get > one that is: it'll be cheaper without the amplifier and its error. Forget > the regulator and let the PIC run from 4 alkaline cells. If you must > regulate, do it with a low dropout regulator, preferably driven directly > from the PIC output pin. Arrange for the external circuitry to draw > negligible power when operating. If more power is needed, use a high side > switch or a p-channel logic Hexfet. > > Check Microchip for an app note on powering directly from the output pins. > See IR for Hexfets. See Analog Devices for information about ratiometric > data conversion; check the AD7710 and its peers for examples. Learn about > the strain bridge in the ptrans. And always do the math: look at all the > data sheets until you understand them and can calculate (or estimate) the > power situation for all operating conditions. > > You can have a PIC driven circuit with multiple LCD displays and peripherals > operate several hours per day for the shelf life of a set of AA cells. > Choose the right PIC and you can self monitor the battery condition. > > --Tom Rogers VP-R&D Time Tech Inc. > > -----Original Message----- > From: Werner Terreblanche > To: PICLIST@MITVMA.MIT.EDU > Date: Thursday, December 11, 1997 2:16 AM > Subject: PIC circuit running on penlight batteries > > > I'm currently designing a product that will have a PIC running of > four 1.5V penlight cells. There is also a 12bit A/D and pressure sensor > on the circuit which both demands very accurate 5V supply. > > This sounds like a fairly simple problem to solve, but I would still > be interested to hear how other people would tackle the same problem. > The full requirements to be met are: > > 1. Stable 5V supply over the entire battery life > 2. Software ON/OFF control of the supply voltage > 3. Automatic wake up of the micro every 10 minutes or so. > > I was planning to use a MAX860 to double the voltage and then just > use a low dropout fixed 5V regulator like the LM2936-5 to regulate it > down to 5V. My circuit only draws about 20mA, so the efficiency of > the MAX860 should be more than 90%. The MAX860 also provides a > shutdown function, which should make it possible for the micro to shut > down its own supply. > > How do the rest of you guys out there that makes battery operated > equipment ensure a solid 5V over the entire life of the batteries? > Is this proposed method that I want to use an acceptable way of doing > this, or is there a better way to skin the cat? > > Another question : What is the best way to switch the 5V supply to > other parts of the circuit. I suppose it would be a MOSFET, but > what I really want to know is the part number of a good general > purpose MOSFET that is widely available. Every time I try to order > one from my databook, I get told that I need to buy a minimum of > 10 000 and that it would take 8 weeks to deliver. This is very > frustrating when you only need a couple for prototyping. Where do > you guys get your MOSFETS from? (Especially you fellow South > Africans) > > > Rgds > Werner > FOR SALE: Parachute. Only used once, never opened, small stain. > > -- > Werner Terreblanche http://users.iafrica.com/w/we/wernerte/index.htm > wterreb@plessey.co.za (work) OR wernerte@iafrica.com (home) > Plessey SA, PO Box 30451,Tokai 7966, Cape Town, South Africa > or at home : Suite 251, PostNet X5061, Stellenbosch, 7599 > Tel +27 21 7102251 Fax +27 21 7102886 Home +27 21 8872196 > ------------------------------------------------------------ > > > FOR SALE: Parachute. Only used once, never opened, small stain. -- Werner Terreblanche http://users.iafrica.com/w/we/wernerte/index.htm wterreb@plessey.co.za (work) OR wernerte@iafrica.com (home) Plessey SA, PO Box 30451,Tokai 7966, Cape Town, South Africa or at home : Suite 251, PostNet X5061, Stellenbosch, 7599 Tel +27 21 7102251 Fax +27 21 7102886 Home +27 21 8872196 ------------------------------------------------------------