DIY: How to test for spark and to troubleshoot spark problems

[RonJ@HT]

A very common cause for no-start engine problems is weak or no spark at the spark plugs. I created this thread to describe a relatively simple yet effective method to...

Check spark

1) Unplug all four fuel injector clips to avoid flooding the cylinders with gas during the procedure. (Thanks goes to Sonia's Driver for this tip.)

2) Buy a spare NGK spark plug for your engine (inexpensive $2-3) and gap it properly.

3) Remove the plug wire from one spark plug and insert the spare plug into the end.

4) While firmly grounding the threads of the spark plug against a clean metallic area of the valve cover, have a buddy crank the engine. If you are concerned about getting shocked, you can wear a rubber glove on the hand holding the spark plug, though this is really unnecessary if you take care to keep fingers or other body parts away from the spark plug tip.

5) With the engine cranking, you should see bright white spark (with only a faint bluish tinge) rather than weak orange, yellow, or blue spark at the tip of the plug.

6) Repeat procedure above for the remaining three spark plugs.

------------------------------------------------------------------------------------------------------------------------------

Outcomes and troubleshooting

A) Some or all plugs have no spark or weak orange, yellow, or blue spark

Check the plug wires, distributor cap, and rotor. See FAQs sticky for diagrams of component parts.




B) All plugs have weak orange, yellow, or blue spark

Check whether the coil in the distributor is bad. The service manual or FAQs sticky has a diagram that describes resistance measurements to test the coil. Here are some tips for doing the coil test:
------------------------------------------------------------

Resistance testing the coil

Remove coil from the distributor for tests. Make measurements inside a room at about 70F. Note that Ohm scale recommendations below are irrelevant if you are using an digital multimeter that auto-ranges.

Primary coil resistance
The spec reading for the primary coil reading is very low resistance, so you need to use the lowest Ohm scale (e.g., 20 Ohms) for testing.

Steps:
1) Set multimeter to the lowest Ohm scale.
2) Touch the two meter probes together and note the reading. This reading represents the internal resistance of your multimeter.
3) Measure the primary coil resistance and note it.
4) Subtract the meter internal resistance reading from the primary coil resistance reading to obtain the actual primary coil resistance.

Secondary coil resistance
This is much easier.

Steps:
1) Set multimeter to the 20K (20,000) Ohm scale.
2) Measure the secondary coil resistance.
3) Done

------------------------------------------------------------

A bad ICM can also sometimes cause weak spark. AutoZone tests ICMs for free. Also, here's an ICM testing DIY:

http://www.team-integra.net/forum/bl...eshooting.html




C) All plugs lack spark

Do these steps in order:

a) Check whether the timing belt snapped or stripped. This is most easily done by removing the oil filler cap and, while a buddy cranks the engine, by peering into the hole with a flashlight to look for evidence that the camshaft is turning (e.g., movement of the rocker arms). If the belt snapped (=no rocker arm movement), install a new timing belt, tension it properly, and then compression test the cylinders. The latter step is essential as major engine damage may have occurred when the timing belt snapped/stripped.

b) Check whether the important G101 ground wire on the thermostat housing (see FAQs sticky) is loose or corroded. If so, clean it thoroughly with steel wool and tighten. By the way, if G101 is disconnected or loose, the engine will lack both spark and fuel.

c) Check whether the CEL remains ON when the key is turned to ON(II). If so, pull the code(s) (see Codes sticky). If a code for one of the distributor sensors (codes 4, 8, or 9) or the igniter/ICM (code 15) is retrieved, then the distributor subassembly or igniter/ICM, respectively, is possibly bad. Further troubleshooting here would be:

--->(for sensor codes) measure the resistance of the sensors (spec = 350-700 Ohms) and their corresponding wires for a short or open.

--->( for code 15) (i) have AutoZone test the igniter/ICM, (ii) test the Blk/Yel wire attached to the coil and igniter/ICM for battery voltage with the key in ON(II), and (iii) test the Yel/Grn wire running between the igniter/ICM and the ECU for a short or open.

d) Check whether battery voltage is reaching the coil and igniter on the Blk/Yel wire in the distributor when the key is turned to ON(II). If not, there may be a bad wire from the ignition switch (or for 98-00 Civics, a blown fuse 9 in the under-hood fuse/relay box). A bad ignition switch is another possibility but, in this case, the engine will lack both spark and fuel.

e) Check whether the coil or igniter/ICM in the distributor is bad. Again, there are resistance tests for the coil (see FAQs sticky and above), but take the igniter/ICM to AutoZone for testing.

f) In some cases, lack of spark is due to a bad ECU. This is usually only considered if all tests described above yield negative results.

[casper_hb]

My swap has no spark??? i check all those are good.
98 integra LS motor swap into civic ex chasis
using LS harness

HELP????

[nasbc]

I can add to this. If your injectors are hard to get at or just another method just unplug ur fuel pump relay yes some fuel will till be injected the 1st few revolutions this wont hurt anything after pressure has dropped it wont inject more anymore fuel because the fuel pump wont be supplying any.

[serversurfer]

Since the "test the coil" section of the FAQ just links to the appropriate manual page, I guess I'll ask here.

I recently replaced my plugs — I'd gotten the car used and the old plug were thrashed — and I was checking the spark as I installed them by grounding the tip against the valve cover, as I'd seen a friend do in the past. I did wipe down the valve cover before doing this, but I don't know if the cover has a legit coating or is just old, but I don't know if I'd describe it as "bare metal."

In any case, I'd say the spark I got from all four plugs was more of the dim blue/orange variety than the bright white, but I don't have a lot of experience here. I did get a bright white spark to arc from the tip to the valve cover by holding the tip a few millimeters away from the cover, but the only sparking between the tip and the electrode seemed decidedly less impressive.

So I decided to test my coil. I just received a slick DMM from Amazon, and I was getting ready to test the coil. The manual notes that when testing, the ambient temperature should be 68ºF, but it's currently 97ºF here, with a heat index of 105ºF. How will that affect the resistances I read with the DMM? Should I just pull the coil, and bring it in to the house for a few hours to normalize before I test it, or will a 30º varient in ambient be "no big deal"?

Also, I don't need to be concerned with messing up my timing, do I? Everything just magically goes back together correctly?

[RonJ@HT]

Originally Posted by serversurfer

Since the "test the coil" section of the FAQ just links to the appropriate manual page, I guess I'll ask here.

I recently replaced my plugs — I'd gotten the car used and the old plug were thrashed — and I was checking the spark as I installed them by grounding the tip against the valve cover, as I'd seen a friend do in the past. I did wipe down the valve cover before doing this, but I don't know if the cover has a legit coating or is just old, but I don't know if I'd describe it as "bare metal."

In any case, I'd say the spark I got from all four plugs was more of the dim blue/orange variety than the bright white, but I don't have a lot of experience here. I did get a bright white spark to arc from the tip to the valve cover by holding the tip a few millimeters away from the cover, but the only sparking between the tip and the electrode seemed decidedly less impressive.

So I decided to test my coil. I just received a slick DMM from Amazon, and I was getting ready to test the coil. The manual notes that when testing, the ambient temperature should be 68ºF, but it's currently 97ºF here, with a heat index of 105ºF. How will that affect the resistances I read with the DMM? Should I just pull the coil, and bring it in to the house for a few hours to normalize before I test it, or will a 30º varient in ambient be "no big deal"?

Also, I don't need to be concerned with messing up my timing, do I? Everything just magically goes back together correctly?

Remove the coil from the distributor and bring in the house to do the test.

You won't affect the ignition timing if you don't remove the entire distributor.

Originally Posted by RonJ@HT

Resistance testing the coil
Remove coil from the distributor for tests. Make measurements inside a room at about 70F. Note that Ohm scale recommendations below are irrelevant if you are using an digital multimeter that auto-ranges.

Primary coil resistance
The spec reading for the primary coil reading is very low resistance, so you need to use the lowest Ohm scale (e.g., 20 Ohms) for testing.

Steps:
1) Set multimeter to the lowest Ohm scale.
2) Touch the two meter probes together and note the reading. This reading represents the internal resistance of your multimeter.
3) Measure the primary coil resistance and note it.
4) Subtract the meter internal resistance reading from the primary coil resistance reading to obtain the actual primary coil resistance.

Secondary coil resistance
This is much easier.

Steps:
1) Set multimeter to the 20K (20,000) Ohm scale.
2) Measure the secondary coil resistance.
3) Done

[serversurfer]

Okay, I pulled the coil and brought it in to the house for a few hours, where it was 72º. At first I tried touching the probes to the screw connectors, and I was getting readings that varied wildly, but seemed to be centered around 1.5-2.5 Ω. When I touched the probes to the bottoms of the holes the screws sink in to, I got a much more stable reading of 0.8 Ω, though sometimes it would jump up to 0.9 Ω or even 1.0 Ω. Once or twice, it dropped to 0.7 Ω, but it went up to 0.9 Ω fairly regularly. I tried measuring the internal resistance of the device as you advised, but when I touched the probes together, the reading just bounced all over the place. Perhaps because the DMM is auto-ranging? In any case, the secondary coil seemed to be in spec, showing just shy of 16 KΩ.

So I guess my question now is, is my coil "going bad"? Should I go ahead and order a new one? This is my first time testing anything, so I'm not sure how to interpret my results, and I don't know precisely how coils fail, or the symptoms they show when they do. I don't know if a coil is something that wears out over time, or just fails catastrophically one day. I suspect it's the former, simply because we're testing it when we suspect reduced performance. Mine seems to be barely within spec, if not just out, but I don't know if that's a sign of impending doom, or if it means everything is fine.

[RonJ@HT]

Try to borrow a higher quality digital multimeter. I only paid $15 for my Craftsman multimeter and it gives very steady low Ohm readings.

[serversurfer]

I have this one. It seemed to review fairly well, but like I said, I'm new to this.

So what about coils and how they fail? Do they come from the factory at 0.6 Ω and slowly increase in resistance as they age until they're no longer considered useful, or do they suddenly go from ~0.7 Ω to ~0.9 Ω, and that's how you know it's "broken," or what?

[RonJ@HT]

Originally Posted by serversurfer

I have this one. It seemed to review fairly well, but like I said, I'm new to this.

I suspect that you may not be making good contact between the two probes or between the probe and coil.

Set the meter to measure Ohms. Then firmly press the two metal probes together in an X configuration. Hold very steady to see if the Ohm reading stabilized.

So what about coils and how they fail? Do they come from the factory at 0.6 Ω and slowly increase in resistance as they age until they're no longer considered useful, or do they suddenly go from ~0.7 Ω to ~0.9 Ω, and that's how you know it's "broken," or what?

I think both can happen - gradual and sudden failure.

Gradual by age and usage.

Sudden catastrophic failure by not doing regular ignition system tune ups or by removing a plug wire while the engine is running or cranking.

[Int3graz]

good info

[serversurfer]

Well, after picking up a new set of test leads which included alligator clips at Radio Shack, a couple of sub-Ohm resistors at Radio Shack, and ultimately, a $15 coil on Amazon, I've learned a few things.

1. I'm apparently not a born surgeon.
2. While the Equus 3320 seems to be a fairly nice meter, the included probes, maybe not so much.
3. Alligator clips are my friends.
4. I may have been a little OCD with my definition of "stable reading."
5. Coils do indeed fail in a way that causes "wandering readings" across the primary terminals.
6. This may not be the only way coils fail.
7. Good coils make your engine strong.
8. It's difficult to develop proper testing skills when testing parts which fail in a way that duplicates poor testing procedure.

I'm thinking of replacing my ignition wires as well, just to see if there's a difference, basically. Is there any point in getting the snazzy blue NGK wires, or is that like getting Monster cables? Should I just go with Monoprice, err, Duralast instead? In a review, a user said the NGK wires had about half the resistance of the wires he took off, which sounds great, but I don't know if that makes a difference in this particular application. Regardless though, the caps on my current wires don't fit very well, which makes me sad.

[pecas]

hello my first time here need help no spark on my 2000 si obdo b16 swap tested ecu dizzy on friends car and it worked gd but when i put it on mine it dnt work

[HondaPartsHero]

What distributor is it?

[pecas]

all i know its a vtec dizzy but i think is a gsr