Amp-meter from solenoid

[jrz126]

I've been curious to see how many amps my audio amp is drawing in my car. approx. 13.8VDC 50-100A
At my place of employment, I noticed that they use some rather large solenoinds to operate some of the switches. They have a 1 to 2" opening, so I can easily get my 4 gauge wire through there.

Is it possible to use this as a poor-man's ampmeter?

 

[stevez]

The current input to and output amplitude from your amp will be changing in response to the audio input. The conductor that carries that current will have a magnetic field that is changing as well and would induce a voltage/current in the solenoid coil if placed within. How much voltage/current would be generated would have to be determined experimentally and you'd have to calibrate it against some known point of reference. The problem I see - the input current is DC and while constantly changing it might not represent the total input power. Example: with a steady state input tone it seems likely that the DC input to the amp will be constant (assuming that the audio is isolated from the DC power input by the power supply) and with a constant current the magnetic field won't be changing so it won't induce a current/voltage in the solenoid.

You might make a series resistor with a length of wire. The ohm value of the resistor would be very low - the wire gage would have to tolerate the current though it may get warm/hot to the touch. The voltage drop across this resistor would be very low but might give you useable information that better represents the current flow. Note that the power cables themselves are likely to have some voltage drop that could be sensed however the challenge presented by long sensing leads might be significant.

Sounds like a good project.

 

[bmcculla]

If you wrap one loop through the solenoid one way for the + lead and loop the other way for the - lead you would get a nice differential current transformer.

I don't think you have to worry about any DC. I doubt the amp would deliver any DC as it would be a waste of power. Getting the whole bandwidth of the signal through the transformer might be difficult though. It sounds like its worth a try though.

 

[Roff]

If you wrap one loop through the solenoid one way for the + lead and loop the other way for the - lead you would get a nice differential current transformer.

I don't think you have to worry about any DC. I doubt the amp would deliver any DC as it would be a waste of power. Getting the whole bandwidth of the signal through the transformer might be difficult though. It sounds like its worth a try though.

He doesn't want to know output power - he is interested in measuring supply current.

 

[Nigel Goodwin]

Many years ago in Practical Electronics (the magazine now known as EPE) they did a series of car projects using the LM3914 LED meter chips.

One project, which has always stuck in my mind - because it was so obvious and simple - was an ammeter. Rather than try and add something in series, with it's attendent losses, they used an opamp to amplify the small voltage drop across the battery earth strap. All the cars electrical current passes through this earth strap, and by biasing the meter in the centre, they were able to produce a charge and discharge display, with the centre zero.

Obviously you don't know the resistance of the strap, but you can easily calibrate the circuit by altering the gain of the opamp to give a reading with a known load - such as your headlights!.

I once knocked one together, fed from a 9V battery, using a multimeter for the display, in order to test an alternator - the original was powered from the car battery of course.

Such a simple idea, and so easy to do - it still impresses me more than 20 years on :lol:

 

[jrz126]

Input current is what I'm looking for.

I know it will change depending on the audio input, but I'd just like to get a general idea of the current draw.

I'm planning (some day soon I hope :wink: ) to get a microcontroller, 4x4 keypad, and 2x16 lcd in there. Then I will have all sorts of options for projects (Roof leds, digital heater controls, MPG calculator (will keep a log of MPG too),umm...I'm sure by the time i get the micro in there I will have thought up several more)...

I was just wondering if this would work or not so that if the oprotunity comes up for me to get one of these coils, I know I will have a use for it.

 

[panic mode]

**broken link removed**

 

[Oznog]

I made a project like that, but never got it to totally work.

The ground strap is not a good choice, because the resistance varies a lot with temperature. Temp varies a lot with current as well as engine compartment temp. You would need to modify it with 4 points too or all you'll be reading is unpredictable contact drop. Also, the voltage really is just too low to measure accurately.

A shunt is a good choice. Some of the surplus catalogs carry a 200 amp shunt. Since modern alternators can put out well over 100 amps, this is a good idea. The Hall effect devices have a bit of trouble if you exceed the rated current, it can permanently magnetize them (causing an offset error), and their offset error is usually large anyways. So using a 100 amp Hall device on a system which might be able to exceed 100 amps (even briefly) is a no-no.

I used a 24 bit Linear ADC with no preamp. This is handy since they can measure 0.3 v above Vdd, so if you hook Vdd to 12v, then use a -5v reg to get 7v and use that for the ADC ground you can measure directly. The ADC output then has to be optically coupled to ground reference. If you used an op amp, you'd probably need to generate a positive voltage above 12v since op amps can't read voltages above (or usually near) Vdd. Note- it would also have to be an extremely low offset op amp, a chopper like LTC1013.

The Linear ADC was disappointing, I could not make it live up to its specified offset and offset thermal drift spec. Also the noise was pretty high, and it's not fast enough to take a crapload of readings for averaging.

 

[TheOne]

LEM have nice products. Used them for quite a few applications in the past.

**broken link removed**

 

[Nigel Goodwin]

I made a project like that, but never got it to totally work.

I threw one together in a few minutes, and it worked perfectly, I only used a 741 for the opamp.

It might only be a very small voltage drop, but that's what the opamp is for - and you can even measure the starter current!.

 

[Roff]

<snip> The Hall effect devices have a bit of trouble if you exceed the rated current, it can permanently magnetize them <snip>

I have not heard of this, nor could I find reference to it. Is this from personal experience? Can you cite literature to confirm this?
Edit:
I thought that you were referring to Hall Effect devices in general, but after looking at **broken link removed**, I'm assuming that you are referring to residual magnetism in the core. Below is an excerpt from the datasheet:

Magnetic offset error (VOM): The magnetic offset is due to the residual magnetism (remnant field) of the core material. The magnetic
offset error is highest when the magnetic circuit has been saturated, usually when the device has been subjected to a full-scale or highcurrent
overload condition. The magnetic offset is largely dependent on the material used as a flux concentrator. The larger magnetic
offsets are observed at the lower operating temperatures.

With a 100 amp excursion, the worst-case offset is 0.8 amps - not too bad if you are looking at the big picture. On the other hand, if you're looking for the reason your battery is going dead every night, it's worthless. I wonder if the residual magnetism would be much worse after an 800 amp excursion (which the part can tolerate briefly).
Of course, you could build in an offset adjustment that would be enabled only when a known zero current condition existed.