I'm currently researching building my own adjustable linear power supply, and I'd like to build in two analog gauges that show the voltage and the current being drawn from the supply.

It would have to be a high-impedance voltmeter, of course, to minimize loading. Not entirely sure how to implement an ammeter. The entire project will probably output 0 - +12V, and also have a separate -12V rail (both adjustable). I would probably implement manual switches on the gauges to switch between measuring different rails, as automatic switching/ranging is a project in itself.

Wondering if anyone had any experience with analog gauges and how to implement them.

 

I assume you referring to standard mechanical analog meters with a needle.

Voltmeters are simply connected across the power supply voltage outputs. Ammeters are connected in series with one of the outputs.

__________________
Carl
Curmudgeon Elektroniker

 

A low impedance voltmeter would work for this. It's reading an ideal voltage source.

 

Here's a good basic intro to analog meters.
And another, start reading on page 241

 

im looking to actually construct my own meters, i assume using opamps. i could just gut a cheap voltmeter, but the supply will be capable of several amps so i would like to at least build my own ammeter capable of handling that high of a current.

 

What kind of meter movements do you want to use (mfr, full-scale current, etc.)?

__________________
Ron

 

If you had actually read the references I provided, you would have learned that an ammeter capable of reading several Amps is actually a mV or mA meter movement with a resistive shunt across it. You would have learned that a voltmeter capable of reading several tens of volts is actually a mV or mA meter movement with a series resistor. No opAmps required!

 

What would be observed on an ammeter connected to just a coil of wire when a magnet falls through the coil?
The magnet is falling vertically through the coil, and the north pole of the magnet reaches the coil first. The ammeter is zero centered.

__________________
market samurai ~ marketsamurai ~ marketsamurai.com

 

The needle would jump. Why are you posting such a random and obvious question in someone elses thread?

__________________

 

i apologize for not reading before replying, but i replied from a mobile device, intending to steer the conversation. however I think it was misconstrued because it was a short reply.

anydangway, yes I am aware of how to construct a voltmeter/ammeter with series/parallel resistors. however, these are nowhere close to "ideal." information I am looking for is the basic operation of an "amplified voltmeter" as it is referred to in your linked reference (page 259 of the pdf, page 249 as labeled on the actual text scan.) However they merely mention its existence and not how it works.

the point is, a resistor and a gauge will work, but with a 1ma gauge and just a resistor, that 1ma is going to have to come from somewhere, and its going to come from your circuit, essentially changing the load. enter op-amps, stage right. they have a very high input impedance, and thus approach the "ideal" status.

seeing as I want to design an adjustable power supply, having something on there that sucks down current other than the circuit i am powering is not a good idea. it's also going to need to be manually switched so i dont explode the meter, which isnt very safe. what if i hooked something up wrong in the circuit im powering, and i accidentally dump a ton of power into my volt/ammeter?

I can at least imagine how an amplified voltmeter works, being that opamps have high imput impedance, but what about an "amplified ammeter?" I suppose that I could put something like a .01 or .001 Ω precision resistor in one of my output lines and stick a voltmeter across it, but is this how its done for "real" ammeters?

.01 would probably be acceptable, at a 10A load it would drop 100mV, which seems like a lot, but for something drawing 10A it probably doesnt need to be precise to 100mV. I wont be doing like, precision control systems on this power supply or anything. but i would like it to be robust and safe (safe as in not going to explode itself or circuits i hook up to it if i do it wrong. safe for my circuits, i could care less about my health =P) its a combination design-learning-experience for me as well as me getting something fairly useful at the end. so once im done with it i want it to be useful. just trying to apply some of the theory learned in class in a practical manner... having actual hands-on experience with things helps understanding a lot. and I figure a power supply has the potential for all sorts of applications. theres the actual power regulation, the metering of it, safety circuits, im sure i can incorporate things like comparators to give me warning lights (just throwing out ideas)... eventually i'd like to add a signal generator. probably a wien bridge oscillator core for sine wave production and then integrators and such to get square/triangle/sawtooth output. and of course amplification, which will be really fun. my own personal lab course that will teach me a lot about many different subjects I have studied.

sorry for the long winded reply. just looking for people who may have built their own supply with volt/ammeters in them that needed to be close to ideal, or otherwise have experience in that sort of thing.

 

by the way, sceadwian, where are you from in rochester? I go to RIT for electrical engineering, are you a student, or do you live/work in the area?


and that question he asked sounds suspiciously like it was taken directly from a homework assignment in basic circuit theory or physics... the correct answer probably includes a direction in which the needle jumps, i.e. towards negative or towards positive, but i think he should read his book, or at least google, and figure it out.

 

Apparently you want a meter that is not powered by the source you are measuring. Is this true?

__________________
Ron

 

The way "Real" precision power supplies tackle the problem of the voltage drop across the Ammeter shunt causing a voltage drop and a reduction in output voltage is to use "remote sensing", i. e. sampling the output voltage on the load side of the shunt. Likewise, the voltmeter is also connected on the load side...

 

Yes. since it's going to be as precise a supply as I can make it, I do not want my measurements to affect the supply's output. I want them as close to ideal as possible.

so there is still a shunt resistor for the ammeter, but if you set the supply's voltage to output, say 12V, when it is regulating for this, it will check to make sure that 12V appears at the load side of the shunt resistor?

makes sense.

I assume (in a commercially produced lab supply) that the shunt resistor would probably be switched for an auto-ranging meter, according to the output current and the defined tolerance spec of the supply.

So I suppose i shouldnt worry about the ammeter part of the project, as i'll just build two voltmeters and use one to measure across a good shunt resistor.

still wondering about how to go about constructing such a voltmeter, however.

 

The thing about using analog meters on a power supply is that the linearity of most meter movements is no better than 5%, their repeatability is like 2%, assuming that you calibrate them at full scale. They really suck if the reading is less than 25% of full scale, so that requires range switching.

If you want precision and repeatability, then use Digital Panel Meters. Besides, have you priced high-quality analog meters recently? You can usually buy several DPMs for what a good meter costs.