Ok, I hit the surplus place this morning, got a 10A analog ammeter for $10 and came up with 4@0.75r resistors with no wattage rating, but they're big old canister things about 5/8 * 3/4 * 3" that the gentleman very kindly gave me for $2.95 ea. I hooked them up in parallel. I had some IRF1010's in stock 3 of which I bolted to an old CPU heat sink and stuck a computer fan on it. I was originally planning on running the FETS in the linear region but I couldn't get a smooth enough ramp up via the potentiometer, so I went to PWM and a small microcontroller, but probably could have done it with a 555. It works quite well though even with the big heat sink the fets get surprisingly hot when I'm running it right at 10A (the resistors get pretty warm). But I cooked it for quite a while and nothing exploded, so I think we're ok. I also happened to have an ACS711 kicking around that I'd been meaning to play with so I stuck that on and fed it back to the micro-controller as an analog input since as the fets heat up the current definitely creeps. I'm not sure I'll go so far as to implement a PID controller but I'll definitely code in a little correction. But on the whole for a Saturday afternoon not too bad. Thank you for your help. DougM On Sat, Oct 1, 2011 at 12:35 PM, Dwayne Reid wrote= : > At 07:23 AM 10/1/2011, Electron wrote: > >>I think it's easy at surplus locations to find batches of power resistors= .. >>E.g. I got one (almost for free) containing one thousand 5W 100ohm resist= ors. >>I still haven't found the time, but I'd like to make a programmable sink >>with them, some mosfets and a PIC. > > Metal Oxide 5W resistors work out nicely - they range from completely > non-inductive to somewhat (barely) inductive. > > I tend to choose lots of higher-value resistors, then put them in > parallel as needed. =A0Putting several of a resistor in series, then > putting that series string in parallel with a smaller number of > resistors limits the power dissipation to the weakest element, which > can be as low as a single resistor. > > For example, choosing 100R resistors would limit my load rating to > about 22Vdc - not high enough for me. =A0A better choice would be > 470R. =A0Or, in the case given by Electron (above), make each resistor > out of several 100R resistors in series. =A05- 100R resistors in series > would make 500R @ 25W. =A0That's good up to about 113V continuous - > usable for most anything that I might need. > > Using all of those resistors in that fashion (200 strings of 5) would > give you almost a full 8 bits of resolution, all the way from about > 2R5 through 500R with a maximum working voltage of about 113V > (continuous) and a maximum power rating of 5kW at the lowest > resistance setting. > > Eminently reasonable, actually. > > One of the many advantages of choosing Metal Oxide resistors is their > extremely high tolerance to severe overload without failing or > changing resistance. =A0I've run them red hot for several hours just to > see how they behaved and was pleasantly surprised at their stability. > > dwayne > > -- > Dwayne Reid =A0 > Trinity Electronics Systems Ltd =A0 =A0Edmonton, AB, CANADA > (780) 489-3199=A0voice =A0 =A0 =A0 =A0 =A0(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 > --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .