Getting to know our 911's: Performing some basic tests before there is a problem

[ischmitz]

Ok, the first test tells you that the pin of the fuse block where the red lead is attached to is getting +12V once you crank or the engine runs. Think of this test as testing the pressure of a garden hose. The pressure (+12V) registers on the MM.

The second test is a pressure difference on a small restriction (fuse) in the garden hose. You measure before and after the restriction. Once water (current) flows through the restriction (fuse) there is a little pressure difference. This pressure difference (voltage drop) proves there is current going through the fuse and into the fuel pump.

The second test is helpful to diagnose a potential bad fuel pump. Imagine the wire from the fuel pump to chassis GND is broken. When you crank you will still see +12V at the fuse (1st test). But no current will flow. The second test will read zero at all times.

Ingo

[sugarwood]

Thanks Ingo.

I am having trouble understanding test #2.
Test #1 makes sense to me: There is 12V of "pressure" in the circuit, and the MM splices into the circuit and sees the same 12V.

Why doesn't test #2 also see 12V? It's part of the circuit, isn't it?
Can you explain why I am not getting the identical reading for both tests?
What is fundamentally different about them?
Why does one see the full 12V while the other does not?

[ischmitz]

Both sides of the fuse have almost the same voltage with reference to GND. A voltage is a potential (pressure) difference. Convention is that voltages are reported referenced to GND and GND is at zero. For this you put the negative lead of the DVM to the chassis. If you repeat test 1 on the other side of the fuse you will see a slightly different value.

In the second test you measure +13.20V on one side and +13.04V on the other side of the fuse (both referenced to GND). If you measure on either side you measure 13.20V - 13.04V = 0.16V

Does that explain it better?

[sugarwood]

Ok, so I need to think of a D-- DC voltage test as the difference between the pressure (voltage) read by both leads ?

In test #1, the red lead reads 13.2V, and the black lead reads 0V. Hence the difference (voltage) is 13.2V

In test #2, the red lead reads 13.2V, and the black lead reads 13.04V. Hence the difference (voltage) is .16V = 160mV. And the fuse has introduced a little bit of resistance (heat dissipates or faint light, like a filament.)



I will now re-read your fuel pump example below and make sense of it.

So, if the fuel pump GND wire is broken, the circuit is broken.
However, doing test #1 at the fuse and chassis will show a pressure difference of 12V b/c power is still being supplied, even though it's not being used.
So, when a broken circuit is being powered, you will still see a 12V difference when you compare pressure in the wire vs. the chassis.
Doing test #2 will show the same voltage before and after the fuse, b/c there is no current flowing, so the fuse doesn't act as a resistor.
Since there is no voltage difference, test #2 would read 0V. Is that about right?

What kind of test is test #1 called?
What kind of test is test #2 called?

Originally Posted by ischmitz

The second test is helpful to diagnose a potential bad fuel pump. Imagine the wire from the fuel pump to chassis GND is broken. When you crank you will still see +12V at the fuse (1st test). But no current will flow. The second test will read zero at all times.

[T77911S]

your way of looking at it to me is screwy but it works.

your statement here is very good.

"So, if the fuel pump GND wire is broken, the circuit is broken.
However, doing test #1 at the fuse and chassis will show a pressure difference of 12V b/c power is still being supplied, even though it's not being used. "

this can be true. and even though this test does not mean there is voltage AT the pump, it is really to verify that the DME relay is working and providing power to the fuse.
troublshooting comes down to checking easy things and the most likely things that go bad first. its also about starting at a place that will cut things in half. like checking for spark first.

both tests are VOLTAGE tests.
the other test i told you to do is a CURRENT test.
to do this test the fuse has to be REMOVED. the red lead on the MM needs to be moved to the 10A connection and the meter set to A-DC10. (the pump current is right at the limit of the meter. if it is too high it will just blow a fuse in the meter). connect the leads to each clip of the fuse holder like you did. run the pump. the current actually goes THROUGH the meter now.

just remember this. measure voltage reference to ground. IE, connect one lead to gnd.
current has to go thru the meter. so something has to be removed. a fuse, one end of a wire.
like remove the neg cable on the battery and connect the meter to the cable and the post. but the meter has to be set up to read current.

here is how a test light can be used. connect it to the body. then just probe what ever you want to see votlage on. the actual voltage is not really impotant. charging problems you need the voltage. the other way a test light can be used os to connect it to 12v, then probe and check for ground.
here is the problem with a MM. a MM is a hi resistance device. IE, it does not need really any current to measure voltage. this can send you down the worng road. a test light (TL) is a low resistance and needs current to light the bulb.
i have seen severla people here working on a brake light (or some light) problem. they measure 12v at the light and the ground is good but the brake light will not work. thats becuase there is a hii resistance due to a bad connection that reduces the current so the light will not light but the MM reads 12v. if they had used a TL, the TL would not light (after all it is light bulb too) and they would know power is bad at the bulb and they would not be pulling their hair out trying to figure out why it does not work.
checking for shorts.
if a fuse is blowing you cant put power to the circuit, obviouly. buy you can remove the fuse and connect the TL in place of the fuse. now the TL is a load that reduces the current so wires dont melt and you can look for the short. a MM will do the same thing here, but what is easier to read, a light bulb or a number. OR, you can connect the TL to 12v and look for what is shorted to ground.

i hope this is not too much info but you seem like you are getting a graps of things.

[ischmitz]

Originally Posted by T77911S

both tests are VOLTAGE tests...

Even inserting an Amp meter instead of the fuse is a voltage test as the Amp meter has an internal shunt and measures the voltage across that shunt. Test #2 uses the fuse as shunt.

Doing test #1 is great for checking the DME relay. This cuts the fuel pump branch in half. If #1 passes #2 is the quickest and most telling as it shows a bad fuse (measure +12V), a working circuit (measure ~ 0.1V - 0.2V) or bad pump/open wiring downstream (measure 0 volts).

The problem with a test light is that it itself has an appreciable internal resistance and alters the circuit to much once serious currents flow. Image using in inline with the battery and then crank.

[sugarwood]

Ingo, I am just going to paste some personal notes here, as a way of archiving these thoughts.
Anyone is welcome to comment or correct. To review,


Test 1: So, if I "splice" my red lead into any active wire and ground my black lead, I will be "observing" how much pressure is in the live circuit. This is a voltage test, and simply tells me if the wire is being supplied with power.

Test 2: So, if I "splice" both my red and black leads into any active wire, I will still be "observing" the pressure difference (or voltage drop) in that section of live wire. In this case, I was measuring around the fuse. What is the point of this test again? Was it to verify that, since there IS a voltage drop across the fuse, that means some amount of current has to be flowing through it ?


Let's say we are troubleshooting a no start and think there might be a frayed wire.
When you do a test #1 voltage/pressure test, it will read 12V, b/c there is continuity, and the circuit is closed. This is deceptive.
You would need to do a voltage drop test, like this test #2


To test for a frayed (+) cable on the fuel pump or starter, this is test #2, except you are measuring voltage drop across the start and end of the (+) starter cable, and not a fuse.
With no power applied, if the cable was perfectly intact, you would read 0V since there is no voltage drop.
With no power applied, if there was a frayed wire, you would still read 0V, since there is connectivity, and MM's top lead and bottom lead would both observe 12V volts of pressure?

But, what happens at full load with a frayed wire?
Now, the full pressure can't make it across the frayed part?
So, that's when you'd see a difference, and get a reading?
So, does a voltage drop simply occur when there is resistance (fuse, fray, etc)
This creates 2 different pressure (voltage) readings on either side of the "resistor" ?


So, the question about test #2 (voltage drop) is how MUCH is acceptable? Like in the fuse test, some amount of voltage drop is normal (160mV in the fuse case) This proved current was flowing. But, if there is a frayed wire, there will be a much higher voltage drop reading, since the voltage difference will be greater on either side of the break. So, you need to know X voltage drop means there is a problem in the circuit. Is that on the right track?


CURRENT TEST:
The 911 fuse #6 for fuel pump is 25A, so it sounds like a 10A MM can't handle a current test for fuel pump. But, for learning's sake, for a true current test, you put the MM into current test mode (10A), the MM is now designed to create a draw. This is the difference b/w a voltage test (observe the circuit) vs a current test (create a wattage draw to consume power) and this is what dictates when current will actually flow INTO the MM. Simply being connected to the circuit doesn't mean you will draw current, the device must be designed to "suck in" the current, like fan or bulb or something. What makes the MM "suck in" and draw current when it is in 10A mode? Where does the current go? Heat? Light? Motion?

[sugarwood]

Originally Posted by ischmitz

Test #2 is the quickest and most telling as it shows
1) a bad fuse (measure +12V),
2) a working circuit (measure ~ 0.1V - 0.2V) or
3) bad pump/open wiring downstream (measure 0 volts).

I am not sure I understand the difference between 2 and 3 above.
When the circuit is broken/open, I won't get ANY voltage drop (resistance) b/c it's not carrying any current at all.
So, on a powered closed circuit, a little resistance means there is current, while zero resistance means the circuit is broken somewhere?



BAD FUSE TEST: Circuit is powered. Switch is OPEN. I connect the MM to points B & C.
Point B is seeing 12V of pressure, and point C is seeing 0V of pressure.
I should read 12V, since 12V - 0V = 12V, and there is a huge pressure difference (voltage drop).

WORKING CIRCUIT TEST: Circuit is powered. Switch is CLOSED. I connect the MM to points B & C.
Point B is seeing 12V of pressure, and point C is seeing 12V of pressure.
I should read 0V, since 12V - 12V = 0V, since there is no pressure difference (voltage drop) at points B and C.
In reality, I will read some nominal amount of voltage drop, due to resistance like heat loss from wire, etc. (All circuits have some built in resistance?)

OPEN WIRING DOWNSTREAM TEST: Circuit is powered. Switch is OPEN. I connect the MM to points A & B .
Both point A & B are seeing 12V of pressure. I should read 0V, since there is no pressure difference (voltage drop) at A & B.
I don't get ANY resistance b/c the broken circuit is not carrying any current at all.
So, on a powered closed circuit, a small voltage drop (some resistance) means there is current, while zero voltage drop (INFINITE resistance) means the circuit is broken somewhere?

To summarize, a voltage drop test:
12V = Break in wire being tested. Power is supplied. Current not flowing.
.1V = Resistance in wire being tested. Power is supplied. Current is flowing.
0V = Break in downstream wire (but not wire being tested). Current not flowing.

[ischmitz]

We are getting into the weeds here. My test #2 was to measure voltage drop across the fuse. The fuse can be modeled as a resistor with about 0.05 Ohm value. If a current flows through that fuse there will be a voltage drop proportional to the current according to Ohms law:

U[Volt]=I[Ampere]*R[Ohm]

In your case lets assume there are 4 Ampere flowing. 0.05 Ohm * 4 Ampere = 0.160V

If the circuit is not flowing current because the pump is broken or not connected to chassis GND the current is zero and thus the voltage drop is zero. The resistance of the fuse isn't changing but since there is no current going through it your voltmeter will measure zero voltage across it. Try this: Connect any resistor to your volt meter in setting VOLT. You will always measure zero. Once you change the setting on your voltmeter to OHM the voltmeter will pass a small current through the resistor and measure the voltage drop.

Cheers,
Ingo

[mysocal911]

Originally Posted by ischmitz

We are getting into the weeds here. My test #2 was to measure voltage drop across the fuse. The fuse can be modeled as a resistor with about 0.05 Ohm value. If a current flows through that fuse there will be a voltage drop proportional to the current according to Ohms law:

U[Volt]=I[Ampere]*R[Ohm]

In your case lets assume there are 4 Ampere flowing. 0.05 Ohm * 4 Ampere = 0.160V

If the circuit is not flowing current because the pump is broken or not connected to chassis GND the current is zero and thus the voltage drop is zero. The resistance of the fuse isn't changing but since there is no current going through it your voltmeter will measure zero voltage across it. Try this: Connect any resistor to your volt meter in setting VOLT. You will always measure zero. Once you change the setting on your voltmeter to OHM the voltmeter will pass a small current through the resistor and measure the voltage drop.

Cheers,
Ingo

Voltage Drop = .05 ohms X 4 amps = .200 volts (200 mv)

[T77911S]

Originally Posted by ischmitz

Even inserting an Amp meter instead of the fuse is a voltage test as the Amp meter has an internal shunt and measures the voltage across that shunt. Test #2 uses the fuse as shunt.

Doing test #1 is great for checking the DME relay. This cuts the fuel pump branch in half. If #1 passes #2 is the quickest and most telling as it shows a bad fuse (measure +12V), a working circuit (measure ~ 0.1V - 0.2V) or bad pump/open wiring downstream (measure 0 volts).

The problem with a test light is that it itself has an appreciable internal resistance and alters the circuit to much once serious currents flow. Image using in inline with the battery and then crank.

using a MM to measure voltage that way can be confusing to those that dont really have a grasp. plus its really not a GOOD way to do it. if the circuit is open at the load, you could still have power at the fuse but the meter would read 0v and send you down the wrong road so i would never suggest to measure voltage that way. this could lead to replacing the DME relay when it is not bad.

like the MM, to measure voltage with the TL its best to reference ground for the reasons above.

i suggested to use the TL in line or in place of the fuse for checking for current drain or when something is blowing the fuse, not for checking for voltage.


always measure voltage to ground and you can verify grounds by putting 12v on them with a test light.

[T77911S]

Originally Posted by sugarwood

I am not sure I understand the difference between 2 and 3 above.
When the circuit is broken/open, I won't get voltage drop ANY resistance b/c is not carrying any current at all.
So, a little resistance means there is a closed circuit, while zero resistance means the circuit is broken somewhere?




(YES. VERY GOOD)BAD FUSE TEST: Circuit is powered. Switch is OPEN. I connect the MM to points B & C.
Point B is seeing 12V of pressure, and point C is seeing 0V of pressure.
I should read 12V, since 12V - 0V = 12V, and there is a huge pressure difference (voltage drop).


(YES VERY GOOD)WORKING CIRCUIT TEST: Circuit is powered. Switch is CLOSED. I connect the MM to points B & C.
Point B is seeing 12V of pressure, and point C is seeing 12V of pressure.
I should read 0V, since 12V - 12V = 0V, since there is no pressure difference (voltage drop) at points B and C.
In reality, I will read some nominal amount of voltage drop, due to resistance like heat loss from wire, etc. (All circuits have some built in resistance?)

(PARTLY CORRECT)OPEN WIRING DOWNSTREAM TEST: Circuit is powered. Switch is OPEN. I connect the MM to points A & B .
Both point A & B are seeing 12V of pressure. I should read 0V, since there is no pressure difference (voltage drop) at A & B.
I don't get ANY resistance b/c the broken circuit is not carrying any current at all.
So, a little resistance means there is a closed circuit, while zero(NO. INFINITE RESISTANCE) resistance means the circuit is broken somewhere?

here is why the last one is partly riight.
you are correct in the A & B are seeing 12v thus you should read 0v.
you are wrong in thinking that the open circuit has anything to do with resistance at A & B.

here is another way to look at it.
THEORETICALLY any point along a wire no matter how long is electrically the same point. thats because in theory that wire has no resistance.
to measure voltage you HAVE to have reistance.
there are 2 extremes. removing a fuse= infinty and putting the fuse back in = zero.
in between these 2 you will measure some voltage drop across the resistance (R).

to measure voltage you have to have resistance. to play with the formula ingo gave you. a fuse has "no" resistance. a wire has "no" resistance.


i am reluctant to say this without a LONG explanation but i am saying it because i am trying to keep electrical trouble shooting simple for now.
for the most part you are only concerned about opens and shorts. here is why i am syaing this. an OPEN (blown fuse, broken wire) will keep something from working. a SHORT will blow fuses. anything in between can do strange things and these problems can be harder to find. here is what hi resistance can do.
slow starting. the clicking of the solenoid with no start. bad charging. dim lights.
and again, with a MM you can measure 12v but not have enough current to operate the load.
load= a light bulb, starter, radio.


some people look at this like water flowing thru pipes.
your pressure gauge is a MM. it is placed before a shut off valve. with the valve open, all the water flows, there is no R (short or fuse) so the gauge reads 0v. turn the valve half way and now you have R. the pressure gauge starts to go up. close the valve. now the gauge reads the full power (12v) that the pump (battery) is supplying.

[rusnak]

On a completely different subject, you should take the time to check your distributor bushings for wear.

Remove the cap, and check the rotor for side-to-side movement. Almost any movement will mean that the rotor will strike the inside of the cap, causing the rotor to be destroyed and preventing spark. If this happens while the engine is running, then fuel will continue to flow into the cat - destroying it and stranding your car with a red hot glowing cat.

Preventive maintenance should be done on the fuel lines on top of the engine.

Carry a spare DME relay and fuses.

Useful checks are resistance checks of the speed sensor, reference sensor, CHT sensor, and if you have a wideband O2 gauge, the "lean stop" and "rich stop" tests (remove the oil tank cap and replace. remove the fuel pressure regulator vacuum hose and replace).

You can also peform the voltage output tests on the AFM that are not mentioned in Bentley, but are covered in this forum. There is also a DME test plan that has been posted for "no start" conditions when there is no spark.

[sugarwood]

I fixed my sentence above:
"So, on a powered closed circuit, a small voltage drop (some resistance) means there is current, while zero voltage drop (INFINITE resistance) means the circuit is broken somewhere?"

Originally Posted by T77911S

some people look at this like water flowing thru pipes.
your pressure gauge is a MM. it is placed before a shut off valve. with the valve open, all the water flows, there is no R (short or fuse) so the gauge reads 0v. turn the valve half way and now you have R. the pressure gauge starts to go up. close the valve. now the gauge reads the full power (12v) that the pump (battery) is supplying.

Ok, in your pipes example, you're describing test #2 (opposite sides of the fuse/resistor/valve)

With valve open, no resistance, no voltage difference.
So, in a voltage drop test, 0V means full current, provided there is power being supplied.
(Since 0V can also mean no power, no current.)

With valve half closed, you have resistance, pressure difference is X.
(There is full pressure/voltage before the valve, and less (?) pressure after the valve)
So, in a voltage drop test, I get some non-zero value.
This means there is a resistor in the circuit, power is supplied, and current is flowing.

With valve totally closed, you have infinite resistance (simulates a break in the current), pressure difference is entire voltage.
(There is full pressure/voltage before the valve, and zero pressure after the valve)
So, in a voltage drop test, full voltage value (12V - 0V = 12V) means break in circuit. Power being supplied, no current flowing.

Right?

To summarize, a voltage drop test:

12V = Break in wire being tested. Power is supplied.
If voltage drop is 12V, that's a huge drop in voltage between points A & B.
Current not flowing.

.1V = Resistance in wire being tested. Power is supplied.
Since a wire has SOME resistance, you will always have a voltage drop in an active current.
Current is flowing.

0V =
Case 1: Power is not flowing at all. No voltage drop, since no voltage.
Case 2: If there is power, and current flowing, then voltage drop of 0V is impossible. There is always some resistance.
If break was within length of wire being tested, you'd see 12V voltage drop.
So, with no voltage drop whatsoever, there must be a break in downstream wire (but not wire being tested).
Current not flowing.

[T77911S]

Originally Posted by sugarwood

I fixed my sentence above:
i am reluctant to say yes because this really is not a good way to check a circuit. but yes.
"So, on a powered closed circuit, a small voltage drop (some resistance) means there is current, while zero voltage drop (INFINITE resistance) means the circuit is broken somewhere?"




Ok, in your pipes example, you're describing test #2 (opposite sides of the fuse/resistor/valve)

YES
With valve open, no resistance, no voltage difference.
So, in a voltage drop test, 0V means full current, provided there is power being supplied.
(Since 0V can also mean no power, no current.)

YES
With valve half closed, you have resistance, pressure difference is X.
(There is full pressure/voltage before the valve, and less (?) pressure after the valve)
So, in a voltage drop test, I get some non-zero value.
This means there is a resistor in the circuit, power is supplied, and current is flowing.

YES
With valve totally closed, you have infinite resistance (simulates a break in the current), pressure difference is entire voltage.
(There is full pressure/voltage before the valve, and zero pressure after the valve)
So, in a voltage drop test, full voltage value (12V - 0V = 12V) means break in circuit. Power being supplied, no current flowing.

Right?

To summarize, a voltage drop test:

12V = Break in wire being tested. Power is supplied.
If voltage drop is 12V, that's a huge drop in voltage between points A & B.
Current not flowing.

this is a very low voltage and i would not trust it to tell me a circuit is good
.1V = Resistance in wire being tested. Power is supplied.

maybe, maybe not. always consider one end of a wire = to the other end. don't rely on this. the voltage drop is too low. the only thing that will tell you something here is if you have a big voltage drop indicating a bad wire or bad connection making resistance hi.
Since a wire has SOME resistance, you will always have a voltage drop in an active current.
Current is flowing.

0V =
MAYBE Case 1: Power is not flowing at all. No voltage drop, since no voltage.
NO Case 2: If there is power, and current flowing, then voltage drop of 0V is impossible. There is always some resistance.
YES If break was within length of wire being tested, you'd see 12V voltage drop.
MAYBE So, with no voltage drop whatsoever, there must be a break in downstream wire (but not wire being tested).
Current not flowing.

you HAVE to stop trying to test a circuit like this. there are too many variables and interpreting your readings can be hard if you don't have a lot of electrical background.
there are a lot of situations that could come up that i just can't cover or predict.

test for voltage with one lead connected to ground. that will ALWAYS tell you power is good from the battery to where you are testing. (there is the exception that you could still have resistance in the circuit between the battery and the point you are testing) then you move down the circuit and test again. the key to being a good technician is keeping it simple and knowing how to cut the system in half so you don't have to start the battery and test EVERYTHING one part at a time.
i think you are getting a GREAT understanding of voltage drop across a component but this is really only good for testing a single part, like one resister, or a coil like on a relay.
this is the reason a test light is made the way it is with one end having a clip to clip it to ground (or power) and a sharp point to easily probe

[T77911S]

i use to do auto electric work for a living and i have worked for the FAA as an electronics tech for over 20yrs.

when i did auto electric work "we" almost never used a MM.
what ever you use connect one lead to ground (the body) and probe for voltage with the other.
test lights are great for checking fuses. just touch the top and bottom of each clip on each fuse. much faster and easier than a MM. this also checks the fuse connection.

here is how i fixed my brothers no start after he put in a new switch and starter.
with a test light, i probed the yellow wire a the starter as he tried to start. the light was kind of dim. i could easily compare it to the battery volt at the start. went to the front trunk and found a relay for the car alarm that had a bad connection. i removed the relay and it has never not started since.
the yellow wire was getting voltage but not enough current to pull in the solenoid. i did not check the voltage but i bet it had 12v there.
the point is all checks were made to ground.

[sugarwood]

CURRENT TOPIC: Check Fuel Pressure (at fuel rail)

Ok, I decided to move onto this step.
First, determine your fuel rail fitting size.

First remove the air duct.


Next place a rag, and get your 19mm wrench


I verified my fitting in M12x1.5



So, next step is to get a fuel gauge that fits.
1) The $100 Harbor Freight kit supposedly works.
2) Try to make my own female M12x1.5 fitting? Has anyone built their own?
3) Or this option for $35.

[winders]

Option 3 works fine.......

[sugarwood]

In the meantime, I would now like to test for spark.
I will need a 7th spark plug. Just get NGK BPR6ES?

Since my car currently runs fine, should I disable the motor somehow before cranking? How?
I just don't know if it's bad to have the engine accidentally start up with the 6th spark plug not connected.

[ischmitz]

I assume you want to test for spark with a functioning engine by disconnecting one lead from a plug and run a spare plug. For that pretty much any spark plug works. Most use an old discarded one.

I would prevent the engine from getting fuel so it doesn't start while you do the test. Either pull the fuse for the fuel pump or disconnect all six injector connectors.

While you are at it maybe get an LED test light and see how the coil "sees" spark pulses on its primary. And then there are inline spark test lights - they hook into the high-voltage line and have a small discharge lamp that lights up when there is a spark.

Ingo