Re: 2003 Tiburon - no spark

it is the 2.7L v6 engine. I have a multimeter and can do the basics, but i am not the advanced in testing. The fuse pops about 3-4 seconds after i turn the ignition key to on position. I can easily unplug all of the sesors except maybe the canister close valve. Can i just plug them back in one at a time until it pops?

3-4 seconds means no "hard" short. You can try that, but without quite a bit of electrical troubleshooting experience, you can get in over your head pretty quickly.

I don't know what the schematics show for this circuit Bob, but he'd likely be ok at least trying this approach - assuming that the circuit does really consist of the components the OP first indicated.

You're right that 3-4 seconds implies a short with a resistance and not a direct short., and that kind of troubleshooting with a meter can get confusing for folks who are not a little experienced. Depending upon how the meter is set, it can potentially either look like a direct short (leading to replacing the wrong component in vain), or not even show the short (leading to the obvious oversight).

I think in this case - again, assuming the circuit is as described, just go ahead as the OP suggests and disconnect the loads and selectively re-connect them. The caution with this technique is to make sure to inspect the wiring as well as the components.

Over his head pretty quickly??? Hell - every time I get into troubleshooting automobile wiring/electronics I make the same statement - "I hate working on car wiring". But... I do it, and I've been doing it a long time, so I understand this stuff pretty well, but I still hate it!

What concerns me is given that there's a delay before the fuse blows, and wiring at 12 volts typically can't fail in a way that gives a high resistance short that's hard enough that it blows a fuse. If it were a high resistance short, he'd be dissipating about 240 watts (~12 volts at ~20 amps to blow a 10 amp fuse in a short but not instant time) at that high resistance point - much smoke would be present. The other possibility is a motion induced hard short to ground. If he in fact has something that is driven - a valve, etc - that has a short, and there's a delay before that device is actuated, he runs the possibility of damaging the driver. I mean, how many times can an output take a short before failing. To try to explain to use a current limiting device in place of a fuse for troubleshooting - a lamp, etc.- and what the results should be is really difficult.

I disagree. If it's a 10A circuit the normal load would be ~12/10=1.2 ohm. It's going to look like a short to anything but a specialty resistance measuring circuit. I doubt it's really that low and it is not that low per sensor so measuring may locate the bad sensor. I would expect something like a 50% safety factor on the fuse so I'd expect an agregate load of ~2.4 ohms. Each individual load will be over that. Possibly well over that. If the circuit wiring is properly sized it will not smoke with that load. I have seen connectors with an ohm or more in a bad connection.

Polarity probably will also be an issue. You are going to need to use the positive voltage resistance lead on the power lead to the sensor. If you are not sure what one that is use a second meter set to measure voltage and measure the voltage accross the probes of your resistance meter.

Once You know the polarity probe the leads to the sensors with any ground lead connected to the chassis. I bet several will be over 100 ohms so you can discount them for a first pass. If one is clearly below 2 or 3 ohms it is probably the problem. If all of them look OK then it is probably a function of the sensor activating and drawing too much current. The easiest way to find that will be to disconnect all of them then reconnect one at a time. You can either wait for the fuse to blow or connect an ammeter accross a blown fuse. Monitor the current draw as each sensor is connected. If I was doing this much that is what I would start with. ;-)

If the fuse is blowing with the key on but not cranking the engine I bet you will find the answer fairly easily. You probaly have a blown semiconductor junction. It still has resistance, as does the packaging leads, but it's well below what it should be.

nothermark

I'm not talking about burning the wire. A properly sized fuse will open prior to the wiring heating significantly. What I am concerned with - if it is a wiring issue - is that the point where the wire is being grounded, and whatever the path to ground is is dropping the entire battery voltage over what is likely a small area. According to the fusing tables, you need 15 amps of current to blow a 10 amp fuse in 4 seconds. So, the entire dissipation over the wire and the short to ground is 12v * 15a = 180 watts.

15 amps would be a total resistance to ground of .8 ohms. If he looks at O2 sensors, they start out as fairly low resistance, so that may be a little misleading.

If he's got a digital meter, most of them will not forward bias any diodes present in things like position sensors, so he won't even see them. They are not likely culprits anyway.

If he puts an ammeter (like the one in his meter) across the open fuse, he'll either cook his meter, or blow the current fuse in it (if there is one). Something that works well - if you don't have access to a current limited power supply - is to use a high current lamp - like one or both filaments in a 60 watt headlamp - in place of the fuse. Low current flow = no heating of filament, so little drop across it. Short to ground = light lights, and current is limited to about 5 amps for one filament, or about 10 amps for 2 in parallel. In either case, you can start disconnecting (or connecting) things and watch continuously for what changes.

Ok, I unplugged all the sensors except the the canister close valve by the fuel tank and canister purge valve under the throttle body. I turned on the key and measured the current without the fuse in and it read 12.8 Then as soon as i put the fuse in it popped. I should mention the the car was hit from behind recently enought to damage the rear bumper and taillight. Have to leave town tomorrow but will try some of the things suggested.

Thanks for all your help.

I am thinking pinched wire around the purge valve only because of the accident. It looks like you have about 1 ohm so it could be a bad connection limiting the current. If you are getting 12.8 A I would be tempted to put in a 20 A fuse and see where things were gettting warm. I'm not sure that is a good answer to do alone as I would be concerned about burning something up. If you can have someone turn the key on for 10 or 15 seconds while you hold the wires to the sensors in question you may find one heating noticeably. Just watch for wires getting hot enough to melt insulation.

The other thing that might happen is that you blow the sensor open and stop drawing current. If it was a bad power transistor the initial failure would be the reverse biased junction shorting so the total wiring resistance becomes the current limit. The internal leads in the package are usually the next thing to fuse but they usually blow open. An open circuit should let the diagnostic routines find the bad sensor.

I think he indicated the problem circuit does not include the O2 sensor. I'm not too worried about the power draw as long as he does not jumper the fuse with a large wire. As you pointed out, and he measured, the current is limited to around 12 - 15 A so On times of 10 seconds or so won't do more than run down the battery or blow the bad sensor open. Even melting something is not all bad as it points out the problem. The only thing I would feel bad about there is one wire in a large bundle and it took out the bundle.

I was thinking more about a reverse polarity diode accross the sensor. I've seen them on a lot of automotive comm gear as the power from the alternator is pretty dirty. Most meters will detect that. That's how we check transistors. ;-)

Not really. It's already taking several seconds to blow the fuse so we know the current is limited. A 20 or 50 A current meter like he needs is pretty rugged. What I think I advised is that he pull all the sensor wires and put them back as he watched the meter. I assumed he would turn off the power or pull off the lead if he pegged the meter. Perhaps I assume too much?

Your way is more elegant, just takes more fussing around. If I was doing much trouble shooting I would probably build a lamp bank. I was giving him a one shot quicky approach.

Which is my concern. At 15 amps, something is getting hot. I would prefer not to torch something in the harness.

It really depends on the meter. I typically use Fluke 73s. In ohms mode, they don't "see" diodes real well. That is done intentionally. They do have a mode for looking at semiconductor junctions.

We really don't know what kind of meter he has. The less expensive meters have a fuse on the low current path, and nothing but a current shunt on the high current path. Cat. III / IV meters like the ones I use are actually fused at relatively low values - 440 ma, and 11 amps for the high current port. For high current DC I use either an external shunt, or a DC clamp on. For AC, I normally use a clamp on. The DC systems I work on can have available source currents as high as 40,000 amps at voltages from 12 on up to 480 volts. I also work on switching equipment and control systems for voltages as high as 230,000 volts at lots of current. So, maybe you can understand why I'm not real big on the keep feeding it fuses technique.

In automotive, it might work without major issues, but it may also eventually damage something besides what is already broken. feeding a line with a current limited source - either a fancy power supply, or a lamp as a ballast resistor - is a pretty effective technique. When you've isolated it to one line, and it's a wiring issue, you can feed that line from either end, and determine which end the fault to ground is closer to, and maybe even estimate relative distance. I work on a wide variety of DC powered devices that communicate with serial data, tones, voltages, and currents. The things I recommended against are just things I cannot do because I have to be very careful not to cause additional damage, as repeatedly re-fusing a known failed circuit - even in a car - can do.

I'm disputing that your suggestion will work. It just makes me cringe when you suggested shorting across a blown fuse with an ammeter when we are not sure whether that is fused, or will just melt. Also, the assumption that he'll get his meter off it fast enough is another variable.

Uhhhh..... That would be...... I'm NOT disputing..... Trigger finger a little quick on the send button...

I'm just getting caught up on this thread and have read the conversations between you and northmark. Good stuff. I have to say - I have used the technique of throwing a bulb in the circuit as a current limiter. It's a great technique for more than one reason. Besides providing the current limiting benefit, it also gives a pretty visible, easy to "read" indication, which is often times a big plus when working on car wiring. I wish I had thought a little further down the troubleshooting process to have recommended that Bob.