Author Options:

Will adding leds to my injector circuit damage my computer in my truck? Answered

The injectors are 12 volt, 12 ohm, about one amp. I would be wiring the leds in parallel. The lights work, I'm just worried about some unforseen error.



Best Answer 8 years ago

I would be hesitant about adding anything the a circuit that gets feedback to the computer.  It will introduce errors and an extra current draw that the computer is not expecting and your mileage may vary.

The leds are 'noid lites', show my injectors firing. At idle they blink, at speed they glow normal, and at acceleration they glow bright. The circuit is common hot, with the computer controlling the ground. I wired the leds in parallel with the injectors, common hot. The leds have resistors.
A week later the truck dies on the highway. I found a bad connection between the ignition switch and the spark coil modulator. The truck is a 1990 Ford Ranger, no distributor.
The spark circuit and the injector circuit are not interconnected, but the computer runs both.
I don't know if the connection problem was related to putting diodes in the injector circuit, or just normal for a twenty year old truck.
I tried to post images, but the files must have been too big. What is a good file size for image downloading? My avatar went real fast.
I want to thank everyone for the answers, all were very good.

.  What do you mean by "bad connection between the ignition switch and the spark coil modulator"? Did the wire burn up? Did a connector come apart? Corrosion?
.  If the wire burned up, you may have fried the ignition system, but maybe not the computer.
.  If a connector came apart, plug it back in and see what happens.
.  Corrosion? Clean it up and apply a VERY small amount of bulb grease when putting back together.

It was light corrosion in the connector near to the one to the injectors. Separate connectors, nice o-ring units for four or more conductors. Found it by checking with my schematic to see where the power went to the plugs. I had power everywhere but the modulator, so I started unwrapping. Tracked back from the modulator to this connector, which had power in but not out. Yay!
Just wiped it off a little, and the truck has been running fine, and I swear my voltage gauge on the dash is showing a higher reading than before, too!
Just to be safe, I unplugged the noid lites, they were just an experiment. I thought they were cool, but nobody else seemed impressed.

This is normal corrosion for an older vehicle, especially if the connection is subject to the elements. I'd suggest that you spritz the connection with a silicon based wire protectant...maybe all of them since you found one already...Folks often don't realize that moisture eventually ruins the connections and can be prevented by periodic maintenance by lubing them with a sealant.In a pinch, even WD-40 will do, although I don't recommend using it in general for the purpose.

There's so many products out there, which do you like best? WD40 is ok for some stuff, but I've heard it is damaging to printed circuit boards! Something about it penetrating the board and changing values.
On the same truck, I have to replace the exhaust donut, and the nuts on the flange are twenty years stuck. My mechanic recommends brake fluid, what do you think?

Take a quick trip to the autoparts store and tell them you want "wire drier" or similar (if they don;'t know what you're talking about, let them know that you want a spray that can be used on connectors to reduce corrosion) I can't think of the actual name right now and am too lazy to do the research.

But yes, DW-40 is an "in a pinch" solution, since it's not the very best thing, though it will work ... in a pinch.

Personally, when it comes to frozen exhaust clamps, I say torch 'em and replace with new. They usually end up sheering off anyway. If they're part of the vehicle, rather than a U-clamp that can easily be replaced, heating the nut to near red hot, then allowing it to cool enough to get the wrench on and give it a twist, rapping it with a hammer following heat/cool, and lubing with penetrating oil are about the only ways I know to break the rust weld. I'm personally a fan of doing it in that order for seriously rusted nut/bolts. Also helps to clean up the exposed end of the bolt thread with a stiff wire brush to remove as much of the accumulated rust prior to trying to turn the nut. You can use almost any torch to heat up an exhaust system nut.

.  Glad to hear you got it back to running. :)

Thanks, what's a fellow to do when you have a ton of dead truck in your driveway! The mechanic I called was butchering the wiring looking for power. He was using a probe, and breaking strands right next to a plug-in at a coil. At least do such damage far away enough from a molded connector for a field repair! Now I have to worry about that connection failing. I have my eyes on a junker just like mine, will grab computer, module, connectors, and coils.

You can post just about any resolution using the original query or in a reply, although the snapshot shown in the post will automatically be parsed to a smaller version. I tend to stay near 1kx1k just to be nice.

Thanks, I looked everywhere, could not find file size limits for uploads. My service here is very slow, less than one m/sec upload. Text goes quickly, but images can be slow.

Well now, that all depends on the LED in question. An led which only draws 10mA or there-abouts should not make an appreciable difference, but I would not say so for any larger currents.

I would consider adding a pigtailed 12V-ready (embedded resistor) to ease the installation and increase durability.

I cannot comment on the pulse characteristics of the leds either, so if you're planning to use them for a diagnostic tool (which to me seems to only logical use since they would otherwise flash much too often to be seen as "flashing"), then you might want to consider looking only at peaks of the waveform a phototransistor captures. Also must be careful to avoid missing jitter and perhaps avoid using a schmitt trigger, since that might remove real data, even if it makes the output *look cleaner...

good luck!

The safest thing to do would be to put a voltage follower on the injector, then drive the LED or whatever.

True that, but he wants to connect the leds directly to the load...I find too many people here don't actually answer the questions asked, and instead tell others how to do it just like they would (or like an instructable they read would do it) rather than answering the question asked...**unless there is a hazard or something too dumb for words...

(I'd actually tap the clocking signal with a nand gate or similar to buffer (follow) the signal to minimize parasitic effects, but he asked for tap off the load...)

(but all of this still begs the question of why the author would seek to slave an led off the injector signal, load or otherwise, and imo, an author *should say as much up front to receive a true "best" answer. Otherwise, those of us who can actually give them the help they desire are left hobbled with a partial picture and have to guess the underlying use (so as to give the best possible advise)....wastes our time and possibly the author's as well. (and why I suggested that he *may be thinking to use this as a tach signal, since it's not like anyone human will be able to see the flashing, except ~possbily at idle...)

> except ~possbily at idle
.  An injector fires at half the engine RPM. That means the LED would be flashing at 400Hz or more at idle. Much too fast to perceive.
.  I'm guessing that the OP is wanting some blinking lights - which just ain't gonna work without some kind of freq divider and/or pulse extender.

700 rpm = ~12 rotations per second. Methinks your calculation is off a bit.

Comes out to 6 flashes per second, which very well might be perceptible, although it may also just look like a dimmed led.

.  Accckkk! I'm giving up math - RPM -> Hz throws me for a loop.

lol...no need for shame. I had already considered it, then forgot what I had been thinking...read your note and doubted myself, until I redid the math and remembered that I had already run the numbers (was going to post the logic the first time ...not sure why I didn't)...so no bad.

In any case, even 6 or 7 HZ is enough to make for just dimming rather than flashes, if the circuit doesn't have a nice sharp square wave once it hits the led (parasitic capacitance...idk).

.  Even if it blinks at ~1Hz at idle and ~5Hz at WOT, I'm inclined to agree with Re-design - it's likely to interfere with the computer and shouldn't be done.
.  Judging by cb92's Feb 15, 2010. 3:27 PM post in this topic, he's already fried something.

Ideally yes, but I doubt an led loading the circuit 10mA will make any real difference. As far as I can tell, the problem had nothing to do with the led, but originated in faulty wiring.

For heavy-ish usage devices, I don't see any problem with adding one low-power LED in the way you describe (plus suitable resistor).
What do you want to do this for and what's the engine's rev-range?


Sean has a lot of useful info - you can read a few mA off of the ground side of the injector without worry.  I would highly recommend installing an opto-isolator on the circuit to ensure you draw the bare minimum current from the system, then use the isolator to trigger a higher load circuit.