[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Ammeters and Voltmeters



Original poster: David Speck <Dave@xxxxxxxxxxxxxxxx>

Gerry,

If you are talking about op amps and diodes, then you should look up the circuit for a precision rectifier, or absolute value circuit. These circuits use a pair of diodes ingeniously connected with an op amp in such a way that the op amp completely compensates for the diode's forward drop. Only problem with these circuits is isolating them from the high potentials used in TC work, but you can use a 1:1, or different rating scaling transformer to provide isolation between your measured circuit and the op amp stuff. Since very little current is drawn, the transformers maintain pretty linear operation. For commercial power monitoring circuits, the standard way of measuring AC currents is the use of a current transformer. The usual standard is to scale what ever your max desired power current will be to 5 amps. Most commercial switchboard meter movements have a full scale deflection for 5 amps AC, and the faceplate is labeled for the scaled current. Thus, if you want to measure a circuit with a 100 amp max current, you buy a 100:5 amp current transformer, and connect it to a panel meter with a 5 amp full scale deflection, but a scale that reads from o 0 100 amps. Such meters and current transformers are readily available on eBay. You can also buy prepackaged current and voltage transducers that very nicely take a defined range of AC volts or currents and convert them to a pretty DC current or voltage ready to feed to your analog panel meter. They are usually not too expensive if you are patient on eBay, and save having to reinvent the wheel. One transducer variant fits an old standard, the 4 - 20 mA current loop, where 4 mA represents "0", and 20 mA represents full scale. Less than 4 mA indicates a failure of the signal system. Many digital process control meters, and even a few analog moving coil meters are built for the 4-20 mA standard. Nice thing about this standard is that the reading is independent of the wire length or resistance between the transducer and your display meter, and there is a fail safe indication of your circuit is open -- the meter is displaying something way less than zero. MPJA.com has some nice current transducers for $12 each that have switch selectable input ranges of 5, 10, or 50 amps and convert that to a 4 - 20 mA current loop. They are even split core units, so you can place them around the measured circuit without having to open the wire to pass it though the measuring core. Even though digital panel meters look sexy and op amp circuits look neat in the lab setting, the electrical environment of a running TC is so full of electrical noise that most of the TCers I know take the easy way out and stick with good old fashioned moving coil analog meters, and avoid the fancy electronics. Diodes can rectify all sorts of spurious RF signals, and even a well shielded enclosure will not guarantee that your solid state circuits will give reliable indications when the TC is firing. Therefore, IMHO, I'd suggest looking at hamfests or on eBay for analog meters scaled for the measurements you want to make.

HTH,
Dave

Tesla list wrote:
Original poster: "Gerry  Reynolds" <gerryreynolds@xxxxxxxxxxxxx>

Hi Ed,

What you say sounds like a problem. One would need to drop more voltage than 50mv to account for the diode drop (that I forgot about) like you say. I suppose one could put an opamp in the meter to boost the voltage taken from the sensing resister before rectifying to make the diode drop less significant. But the diode drop would still cause an offset error.

Gerry R.