Olin is assuming that you connect the battery to the collector of the BJT, control voltage to the base, and the emitter to ground through a sense resistor. The error comes from the fact that whatever current the base terminal draws will ALSO flow through the emitter. In other words, Ie=3DIc+Ib. The only way you could really avoid this (and still use a BJT) is to either use a BJT with a sufficiently-high current gain that the base current is negligible OR put the sense resistor in the collector lead. However, the latter option will require a differential amplifier to measure the sense resistor since it is no-longer ground referenced. Sean On Fri, Apr 15, 2011 at 10:09 AM, V G wrote: > On Fri, Apr 15, 2011 at 9:56 AM, Olin Lathrop = wrote: > >> =A0That depends on how much error you can tolerate. =A0You can probably = get >> transistors with a gain of 30 in that range, especially since you don't >> need >> high collector voltage ability. =A0If one part in 30 is good enough, the= n >> that's fine. =A0If not, you can use a smaller transistor around the big = one >> to >> make a darlington, which then will have a combined gain that is the prod= uct >> of the two individual gains. =A0If the second transistor has a gain of a= t >> least 50 (easy to get), then the darlington gain would be 1500. >> >> You could also use a FET between collector and base to drive a big fat T= O-3 >> power transistor, like the ancient but brick-outhouse 2N3055. =A0That >> channels >> the base current to where it becomes part of the overall current, so the= re >> is not error, and a gain as low as 10 or 15 from the big transistor is >> fine. >> > > I'm not understanding where the error comes from, or what exactly the err= or > is. > > Also, I'm not sure if I need put some sort of a resistive load on the BJT= , > or if I can just connect it to ground. > > Why would I have an error if I can control the BJT by sensing the current > through a current sense resistor? (something like the top few here: > http://dipmicro.com/store/index.php?act=3DviewCat&catId=3D529) > > I'm going to make some simple circuits on LTSice and try them out. > -- > 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 .