If you’ve ever had a car with an electrical system problem you know how hard it can be to pin-point the source of your woes. Here’s a hackery solution that uses a diy PCB to monitor the current being drawn off of the alternator.The sensing is provided by an Allegro ACS758 integrated circuit. This chip measures current up to 150A and outputs an analog signal that can be measured by a microcontroller. In this case an AVR ATmega8 measures the signal and spits the info back to a PC via the serial port. This data can be graphed to help locate when too much current is being drawn for the battery to remain charged.
Check out that CNC milled PCB, what a beauty!
[Thanks Joshua via Elektronika]
I was just thinking about something like this a couple days ago. Would have been great to work out if it was the brushes or the regulator that was going. Awesome find guys.
*where’s the thumbs up emote…ha!*
Too bad this doesn’t give me any solution for finding a short which keeps burning out the fuse for the stop light in my wife’s Mazda 323.
What’s more bizarre about it is that the horn and the stop lights are on the same circuit, making me have to search the entire length of the car to find the short. Two days later and it’s still hiding from me.
OT… “Too bad this doesn’t give me any solution for finding a short which keeps burning out the fuse for the stop light in my wife’s Mazda 323.”
Check any flexible leads attached to the trunk / hatchback… where they bend is a common failure point for that kind of problem.
I hadn’t seen that Allegro part before. That’s really slick! Had I been trying to solve the problem, I probably would have ended up using a current transformer clipped over the lead. While “intrusive”, this seems to be a better solution. Another part to add to the toolkit.
I too am rather surprised the transmission is pulling 20 amps. I certainly don’t know much about all the transmissions that are out there now, but I thought the only current would be the TCC lock-up solenoid, and those are *maybe* 2 amps.
@Sam, there are two types of shorts. A resistive short, where there’s always current being pulled (you can check that by putting a sufficiently high rated amp meter in place of the fuse), and intermittent shorts. For the first type, an HP inductive current probe allows you to follow wires in a bundle, and determine if the current changes when it branches off to a device. For the intermittent short, you might consider putting an alarm of some kind that sounds when the fuse fails. Maybe tie it to an event, like brake lights on when turn signals are on, bumps, etc. Then start tracing the wiring harness looking for locations where it can short to ground. Whatever the case, tracing car wiring sucks.
@Sam
You could just abandon the horn and brake light circuits and run new wire. That may be faster than finding the short.
Stick a screwdriver in place of the fuse and follow the smoke.
@Ardu
You won’t have to follow anything — your car will probably be engulfed in flames shortly. Easy!
@Sam
Just dont do what we did once, wire one guys horn to his brake lights in the car park at work during lunch one day :)
A few feet of wire, 1 relay = hours of laughter
Thanks for all the suggestions, who knew we had a few car wiring geeks roaming Hackaday?
The whole problem has been frustrated by the fact that a rodent (probably a mouse) decided to make a nest in the engine and chewed up some of wiring to drag into its nest. I’ve had to clip and replace sections of chewed wires and everything seems in proper order. Disconnecting the bulkheads where I did the replacements didn’t help locate the short, however, so it’s elsewhere. On the other hand, running so much wiring has allowed me to find severed wires all over and I’ve got every light functioning in the head and tail lamp assemblies… except the stop light and horn.
Agent420, I’ll definitely try looking there. I’ve read that other common places are where bundles go through holes cut in metal (they come loose from their retaining wires and chafe). It makes practical sense to expect them there.
Jc, it’s definitely a resistive short. The fuse blows immediately when the stop light circuit is completed (pressing the brake). I’ve been using a continuity tester and disconnecting sections of the harness from one another (basically I ground one terminal of the multimeter and connect the other to the hot lead – the battery is disconnected of course). I almost wish that the wiring was broken up into shorter sections which could be disconnected, that would sure make this process a lot easier (and cheaper to just buy a new section of the wire harness instead of repairing broken ones).
Na, yeah, I’ve considered that and if it comes down to it that is exactly what I’ll do. It’s something I’d hate to do to the next owner of the vehicle, should we choose to sell it later on.
Ardu, I’ll pass on that. ;) The last thing I need is to fry a section of the harness when almost everything is functioning save the fuel gauge (on another circuit and it’s never worked) and the tail/horn assembly. This vehicle is a 1988 Mazda 323 turbo (there weren’t very many ever made) so finding a replacement harness wouldn’t be a fun task.
this chip ACS758 can sense a current from 50 to 200A.
could someone mention a similar chip capable to sense from 0 to about 25A?
Nice project — I’ve been thinking about doing something like this for a while, and this looks a lot more elegant than a current transformer would be.
@jc
Most modern transmissions also have shift solenoids that direct transmission fluid through the appropriate channels, allowing the transmission to be more precisely controlled by a computer. I don’t have any idea what kind of current they draw, but I would imagine they’re in the same ballpark as the TCC solenoid. A 4-speed transmission has 2 shift solenoids, and I guess that 5 and 6-speed transmissions have 3 (but don’t know for sure.)
@Sam
I agree with the other poster that said to run a new wire. If the fuse doesn’t blow until the brake is pressed, you should be able to locate the brakelight wire coming out of the relay and replace it from that point back.
But wait, the blinkers use the same bulb as the brakes in many cases. Do you have a 4 wire (tail, left, right, ground) or 5 wire (tail, brake, left, right, ground)? Places that install hitches (truck accessory dealers) are usually good sources of info on this topic.
That’s not how you hook up an ammeter. This will only indicate what the alternator is sending out. It does not provide the full picture – for example, current draw with the motor off, as it can only read what the alternator is producing. An automotive ammeter is properly connected to read all of the current in all systems except for what flows in the large cable between the battery and the starter. That is different than can be measured by the subject device.
@OTLabs:
I haven’t finished checking their whole line of products yet, but I see they do have smaller sensor chips. I’m looking at one from 5 to 50A right now.
http://www.allegromicro.com/en/Products/Part_Numbers/0712/0712.pdf
The picture shows his problem, he is using a re-manufactured alternator.
@KC9PFL
Remans parts are all you are going to get with older cars.
@OTLabs:
You may want to take a look at this part sold at Digikey. Looks like what you are looking for.
http://media.digikey.com/pdf/Data%20Sheets/LEM%20USA%20PDFs/lts25-np,e.pdf
@Sam
I’ve seen people install single filament bulbs into dual filament sockets before… check that, see if it’s not causing a direct sort when you step on the brakes.
JGN, “what the alternator is sending out” is what the truck is using, which was the point: the truck runs off of the alternator. The truck doesn’t run off of the batteries. The batteries are only to start the truck, and then the alternator charges them back up. You can actually remove the batteries on some vehicles while the engine is running.
Since the alternator and batteries are bonded, if you wanted to measure battery draw just move my device to the battery.
This sensor is actually better than an automotive ammeter, because it uses the hall effect to measure current. There is no current sense resistor, hence no voltage drop or power loss. And unlike a current transformer, the hall effect device can measure DC.
Can you say, clamp on fluke?
Eraser, I don’t believe a clamp-on fluke will measure the DC current from the alternator…
I believe they will.
http://www.myflukestore.com/p1252/fluke_337.php
Alright, but it does not graph the output like this guy did for only a fraction of the price. That fluke costs $300+.
No argument there :) Just that Flukes can do DC.
“That fluke costs $300+.”
Aren’t “Fluke” and “$300+” redundant…? :-|
@JB: thanks!
I have a High quality and cheap current monitor, if you need that plz call me from my home page