Converting DC Ammeter to AC Ammeter

[lionelsoni]

Wayne, I also doubt that core saturation is the culprit. Saturation occurs at the flux maxima, which happen at the axis crossings of the voltage waveform, not at the peaks. It is also not dependent on current, but on voltage. Or, more precisely, on the area under the half-cycle of the voltage waveform, which is why it increases both with increased voltage and with reduced frequency.

Is the waveform sinusoidal with no load? What kind of load were you using to draw the 1 ampere?

[Anonymous]

Wayne
You're right about the difficulty of easily finding inexpensive, general purpose AC Panel meters suitable for Toy Train layouts. I finally found a Trading Comapany source in New Zealand and recently bought 40 Analog Panel meters (20 Vac and 20 Aac) .

They sold out instantly last week to O-gaugers in pairs at my cost, including the sets I had intended for myself. End up costing the O-gauger about $35.50 per pair---Volt & Amp meter set: including meters, air freight to US, Customs clearance, foreign exchange bank's fee on NZ$/US$ conversion, padded mailer and postage to O-gauge recipent. (saved $3.65NZ$ per meter air freight by "piggybacking" on Trading Company instrument shipment from Taiwan to Auckland).

This was a one time "sample sale" enabled by the Trading Company "piggybacking". Not sure I can get more, I started out searching meters for my use, I don't intend to be in the meter business---order minimums are 500 per spec at Taiwanese Mfg Co, 100 per spec at the Trading Company which means 200 units(100 pairs) minimum per order. (There are meter suppliers in China, India and all the Indian sub-continent countries that advertise they can quickly deliver 1 million meters and have huge minimum order size).

Meters are 2-3/8" x 2-3/4" on the panel face with 1-15/16" dia barrel(need 2" panel hole) and rated at 2.5% accuracy. (My Wiggins clamp-on Aac from the shop verifies that rating), Voltmeter range: 0-30Vac. Ammeter range: 0-20Aac. Meters are assembled in Taiwan of Japanese parts and are the "moving magnet" or "moving iron" type. Similar appearing high quality "Iron Vane" type AC meters run about $85-95 per meter basic cost at Newark Electronics,etc.

Dewey Trogdon

[webenda]

The Mystery of the Waveform Solved

Why did the non-conduction time of the rectifier not appear on the load waveform?

Because I used a current calibrator that increases the voltage rapidly if no current is flowing, like during the non-conduction region of a diode. As a result, the time of zero conduction in the rectifier was nearly zero.

Here are waveforms for the circuit using my Lionel 1015 transformer as a current source.

Now the waveform across the load looks like I expected -- almost.

Now I am going nuts trying to figure out why my 1015 has such an ugly output waveform. I checked resistance of both coils (primary 20, secondary 0.6) and coils to ground (Infinity.) Could the core be saturating? Seems unlikely at only one amp.

Ignorance is bliss, knowledge can drive you nuts.

[lionelsoni]

I would have thought so too, Wayne. Perhaps the explanation is that the turn-on of a diode is not really abrupt as we often assume to simplify an analysis, but rather gradual. So the glitch might be there but feathered into the sine wave so smoothly that we can't see it. It looks like your oscillogram was taken at 10 volts rms. Maybe the glitch would reveal itself in a much-lower-voltage waveform.

[webenda]

DC AMMETER CONVERSION

Many of us prefer analog meters over digital meters for our control panels. Suitable AC meters seem to be in short supply. For this reason I purchased two Radio Shack DC analog meters, on closeout sale, planning to convert them to AC. Ezak's post, "Panel Meters" prompted me to show my conversion of the DC Voltmeter to AC. See:
<http://www.trains.com/community/forum/topic.asp?TOPIC_ID=36560>

The AC Ammeter was posing a challenge. I did not want to build an AC to DC instrument amplifier for it, because that would require an additional power source for the amplifier. The neat thing about analog meters is that they can be powered by the voltage or current that they are measuring. Then Lionelsoni suggested using a bridge rectifier wired to supply AC current to the load and DC current to the meter. Figure 1 shows how I imagined implementing the idea.

Figure 1

One of the problems when using half or full wave rectified AC on DC meters is that the meter needle tends to follow the ripple, evidenced by the needle vibrating at 60 or 120 Hz. Lionelsoni said, in part, "...the shunt might provide enough damping to make the meter readable." In other words, the shunt would dynamically brake the meter like dynamic brakes on a train engine brake the train. See <http://www.trains.com/Content/Dynamic/Articles/000/000/003/079uwrak.asp> for more information. Photo 1 shows my experiment.

Photo 1

I am happy to report that the meter needle is dead still.

Table 1 shows actual current, DC without rectifier, AC with rectifier, when meter is reading cardinal points.

Table 1
FS V = 0.09437 VDC
R = 0.09437 Ohm
0 - 1000 mA DC Ammeter
Meter____Actual__Actual
Reading_DC amp_AC amp
200_____186____209
400_____380____428
600_____583____654
800_____776____870
1000____971___1090

I had imagined a small glitch at the zero crossing point of the load waveform due to the nonconductive region in a diode's voltage curve. Puzzlingly, the voltage across the load looks like a smooth sine wave. [?]

Photo 2