Nuts and Volts - February 2009

MAILBAG

Dear Russell,

In your column in the September 2008 issue (pages 26 and 27), you gave an incorrect explanation of the operation of transformers (and by implication, inductors). Your explanation is a widespread misconception about Faraday's Law. Faraday's Law is based on the phenomenon that a time-varying magnetic field produces an electric field. The EMF induced in an inductor or the secondary winding of a transformer is due to this electric field, not the magnetic field. This is proved by examining the operation of a toroidal transformer. Because of the essentially total cancellation of the magnetic field external to the primary winding due to its geometry, if the secondary winding is wound on top of the primary winding, it is not even in the magnetic field of the primary winding. Yet, the

replace a 100 watt bulb for about $100, which is cheaper than building one yourself.

I use 40 watt fluorescent tubes in my house; two of them give 10X the illumination of a 100 watt bulb, so I think that is the best solution although perhaps not the most efficient.

A CYCLING CIRCUIT

QI urgently need a circuit to cycle four pairs of thermocouple signals for a four cylinder air-cooled engine (CHT&EGT) at the cycle rate of 30 to 40 sec for the four cylinders. As each cylinder cycles on, I need it to also indicate the cylinder 1-2- 3-4 on a seven segment LED. A selective hold switch would also be helpful for extended viewing of any particular cylinder. The power source is 12 VDC.

The circuit doesn't need any signal conversion. Also, there isn't any common ground. This circuit would replace having to manually switch from cylinder to cylinder.

— Art Schwedler

AI remember answering this question about a year ago, but can't find it in any issue. Anyway, I got feedback that I could have done it simpler, so I will try to do that this time.

There are eight thermocouples (four cylinder and four exhaust) which are to be read in sequence: cyl1, exh1, cyl2, exh2, etc. You evidently already have the thermocouple circuit and display; all you need is an automatic switching circuit.

In Figure 7, I am using a Fluke model 80TK thermocouple interface but if you are using an instrumentation amplifier and cold junction compensation, you can use that instead. The input impedance of the 80TK is high enough that the resistance of the analog switch will not cause significant errors. I don't know much about thermocouples, but I believe that the negative lead has to

QUESTIONS & ANSWERS

toroidal transformer works.

— Michael S. La Moreaux

Thanks for giving me the opportunity to correct any misconception. You are correct that the flux does not have to "cut" the turns of the secondary, as I may have said, but the induced voltage is due to the time varying magnetic field. If it makes sense to you that the electromagnetic field extends beyond the magnetic flux field and thereby induces the secondary voltage, I won't argue with you. But, it makes more sense to me to think in terms of flux coupling the two coils of a transformer. Faraday's Law says that an EMF cannot exist without a time varying magnetic field, so it really doesn't matter which you assume to be the driving force, as long as you are consistent. I know how (in principle) to calculate the induced voltage from a time varying magnetic field; I don't know how to calculate the induced voltage from an EMF without going through the magnetic flux.

have the same interface as the positive. That means eight connections on the PCB; same as the positive. The schematic shows CD4051, one of eight multiplexers because it is available in thruhole plastic DIP; but 74LV4051 is a better device, available in surface-mount TSSOP. The program for the PIC16F627A controller is Figure 8. As usual, I can send you a programmed PIC16F627A for $5 if you don't have access to a compiler and programmer.