## Sunday, February 2, 2014

### I hope you find this blog enjoyable and helpful. Support it with a small donation.

I'm working on something that requires me to measure high currents.  I don't need to be exact, I just need to know how many Amps are running through a winch so that cabling and relays can be sized.  It wasn't until this project came along that I realised I had no way to measure current over a couple of Amps.  I tried to measure the resistance of a wire and measure the voltage drop across it, but it was too clumsy and error prone.  An old automotive current meter I had should have done the job, but it seems to be dead.  The best and least frustrating solution was to do things properly and build a tool for the job.  A quick trip to Jaycar for a couple of parts and I was on my way.
 Current-Shunt connected to multimeter
The set-up is really simple, it consists of a current shunt and some cabling to connect it to a load and a multimeter.  If you're unfamiliar with the concept of a current shunt here's a quick recap.  When you need to measure a current the easiest way to do it is to run the current through a resistor of known value and then measure the voltage drop across the resistor.  This resistor is called a current sense resistor or a current shunt.  Ohm's law is then used to calculate the current.  When you need to measure large currents however you need a special low value resistor.  That's where specially constructed current shunts come into play.  The one I bought is rated for 200 Amp and at this current has a voltage drop of 50mV across it.  This means it has a resistance of 250 micro Ohms (50 mV / 200 Amp).  It also means that at full current it dissipates 10 Watt, something to keep in mind.

To measure the current of a load, the shunt is inserted in series via the alligator clips and the voltage across the shunt is measured on the multimeter.  Every amp thought the shunt will cause a 250 micro volt drop across the shunt.

 Current Shunt
To connect the shunt to the multimeter, the tips were cut off a cheap set of multimeter probes and eye terminals were crimped onto the leads.  You can make your own, but this way you get a nice set of moulded banana plugs on the ends of the leads.  The size of these leads doesn't matter because no current should flow through them, they are only used to measure the voltage across the current shunt.  To connect the shunt to the load you need some serious cabling.  The current I am trying to measure isn't 200 Amps, but since the shunt is rated for 200 Amp I might as well size all parts for that current in case I need it in the future.  This means that the cabling has to be quite thick.  Trying to find the cable thickness needed for this is difficult.  Everyone has a different answer, the cross sectional area also depends on the the accepted temperature rise, the insulation, and where the cables are located.  For my situation, I bought the largest crimp terminals that would fit on the shunt, and by coincidence these lugs were the exact size of some old welding cable I had.  The cable has a cross sectional area of approximately 25 square mm.  I would've preferred something a little thicker, but this should do the job adequately.

 Car battery clips
It worked out nicely that the largest battery terminals from Jaycar were also rated for 200 Amps.  They also make things look a little more professional.
 Battery clip teeth
One side of the teeth in the battery clip can be taken out by removing a screw.  The cable can then be crimped into it.
 Teeth soldered to cable
Although the cable is crimped into the fitting, I wasn't entirely convinced it would hold, so it was soldered into place.  It's not the greatest job in the world, I don't have a soldering iron that could put out that much power, as most of the heat will be drawn down the large copper cable and dissipated into the environment.  What I do have however is a small blow torch for browning Crème brûlée that was just the right size.  It's a balancing act to get the right amount of heat into the copper and not melt the cable insulation, but I think I got the hang of it.

I've tested it by running 4 amps from a power supply though it.  The reading was close to 1 mV.  A better multimeter would make a difference, but this is precise enough to get me in the ball park.  All up I'm quite pleased with the result.  I have a useful tool that will come in handy in the future.  A pre-amplifier across the shunt would be a nice addition, kind of like a big brother to the eevblog ucurrent.