Bouncing speedometer

[p-Bar]

Fred,

When I owned a 99 gasser the speedo was bouncing and the ABS light came on so I installed a new sender in the rear diff and used my cheapy code reader to reset it.

 

[FM]

Do you have to reset or clear codes before the new sensor takes effect?



Can anyone tell me what the difference, if any, is between the "vehicle speed sensor" shown on 8w-30-71 and the "wheel speed sensor-rear(abs)" shown on 8w-34-3.



Also, if the sensor on 71 is in the rear axle, why does it show 3 circuits on a 2 wire connector?



Electricly challanged,

Fred

 

[rbattelle]

Wow, those are confusing!

Notice the connector diagrams on page 8W-80-107 indicate a unique 3-wire vehicle speed sensor for the diesels.

I don't see why two twisted pair wires are shown for the vehicle speed sensor when there are only supposed to be 3 wires. Unless it's 1 twisted-shielded pair and the shield is connected to the APPS ground line.

So then the question is where on the vehicle is the vehicle speed sensor? The way I read the diagrams, a VSS is present only on automatic-equipped trucks. If so, then it receives its data from the PCM (manual trucks don't have a PCM).

Then, of course, you ask where does the ECM get it's speed signal from on manual trucks?

Check out page 8W-30-82! It shows clearly that on M/T trucks the vehicle speed signal is routed to ECM connector C2, cavity 11. It's fed by the ABS controller (which we know gets its signal from ABS sensors).

This implies that ECM connector C1, cavity 22 is unused on manual trucks (27 and 52 may also be unused on manual trucks - I didn't look into it).

If I'm right, then the ECM is getting its speed signal on manual trucks through the ABS module. Perhaps the ABS module is the problem, after all (if all the wiring is good).

Ryan

 

[FM]

Dr. Ryan, That was great!



Why do you see 2 twisted pair and not 3? Does that diagram imply those 3 curcits are twisted together



What is a twisted pair? A pair as in phone line "pair" or a pair instead of a single wire and that pair go to a single pin, each wire in the pair carring part of the same load?



Where do you see the vss is used only on autos?



I see no guages in section 8w-40 nor do I see a curcit from the ecm to the speedometer. Am I visually impaired as a bat? It's pasta under the hood and pasta in the book only much straighter and at right angles.



I'll see if I can look up the "APPS ground line".



Many Thanks,

Fred

 

[PatrickHansen]

I had similar symptoms until my crank position sensor was replaced.

 

[rbattelle]

Dr. Ryan, That was great!

Why do you see 2 twisted pair and not 3? Does that diagram imply those 3 curcits are twisted together

What is a twisted pair? A pair as in phone line "pair" or a pair instead of a single wire and that pair go to a single pin, each wire in the pair carring part of the same load?

Where do you see the vss is used only on autos?

I see no guages in section 8w-40 nor do I see a curcit from the ecm to the speedometer. Am I visually impaired as a bat? It's pasta under the hood and pasta in the book only much straighter and at right angles.

I'll see if I can look up the "APPS ground line".

Many Thanks,
Fred

Twisted pair is a cable made up of 2 wires literally twisted along their length. Back when the telephone system was being developed, someone (Bell?) discovered that if he twisted the wires together it reduced signal noise. So twisted pair wire is normally used in sensitive instrumentation when it's important to get a low noise signal.

Twisted-shielded pair wire is the same as twisted pair, except there's a metallic shield that surrounds the pair of wires. That shield must be connected at one end to a ground, which provides a path for noise to follow. It's based on the idea of a Gaussian surface, which is way more than you wanted to know.

So in the diagram there are 3 wires shown, but all 3 are labeled "twisted pair". Which doesn't make any sense, since 3 twisted-pair cables would be 6 wires total. So I was thinking maybe they have 1 twisted-shielded pair with 1 wire carrying 5v, one wire carrying the speed signal, and the shield carrying the ground.

It's not explicitly stated that the VSS is only in automatic trucks, but it certainly seems to imply that. If you look at the service manual volume 1, page 8E-12, the VSS is discussed in the section about how the PCM works. Since M/T trucks don't have a PCM, I deduced that the VSS is probably an automatic-only item.

The reason you don't see gauges in section 8w-40 is that the gauges are not fed in a "traditional" way. Rather, they're fed on the PCI bus. Think of the PCI bus like the USB cables on your computer - they carry lots of data with only a couple wires. A multiplexer mixes the signals together from the various sensors in the engine, passes that along the PCI bus as a binary signal, and then a demultiplexer in the instrument cluster breaks the signal back into its component parts and distributes it to each gauge as appropriate.

You can see where the PCI bus connection occurs on page 8W-18-4 and 8W-40-5.

So, there is no speedometer wire going from the ECM to the speedometer. It's transmitted in computer code via the PCI bus.

At least, that's how I interpret the drawings. I'm certainly no electrical engineer. I keep waiting for someone to post "Ryan, you dunderhead, that's not the way it works at all!".

Ryan

 

[FM]

How did you know the crank sensor was bad? Were there symptoms other than the wild speedo and cruise trouble?





Well you're certainly more of an EE than I.



I see what you're saying about the shielded pair.



I looked at the reference on 8e-12. Odd way of wording it. If a vss is dependant on a pcm then if there's no pcm then stands to reason there's no vss. The same reference says the pcm supplies the 5v power but 8w-30-71 shows the 5v supplied by the ecm. Yea I'm confused.



So I'm thinking there's a common problem that's messing with both the speedo and the cruise. If the bus comunication with the speedo is one way, if, then the problem has to be from the ecm back. The only "boxes" involved are the ecm and the abs controller. The only curcits are g7, b113, and b114. Assuming the boxes are good, because they're expensive and I don't want them to be bad, hows that for sticking my head in the sand, if I pull the connectors on both ends of the curcits and check continuity between the pins of those curcits and adjacent pins I should locate the bad curcit except that its intermitant and may or may not complete the short while I'm testing.

Then if I can't find a bad curcit it has to be a box unless it's a bad ground somewhere or another bad sensor that affects the way the ecm processes the data. That should be simple enough, right?



Arrrrg,

Fred

 

[rbattelle]

I looked at the reference on 8e-12. Odd way of wording it. If a vss is dependant on a pcm then if there's no pcm then stands to reason there's no vss. The same reference says the pcm supplies the 5v power but 8w-30-71 shows the 5v supplied by the ecm. Yea I'm confused.

Well, these manuals are far from perfect. They're very good, but not perfect. Sometimes I think the people writing the manual weren't quite sure how the different truck configurations work. This is a good example.



I find it highly unlikely you'll find a direct short using the test method you suggest (although you can't hurt anything this way, so there's no harm in trying it). Like you say, you'd have to be very lucky because the problem is probably intermittent.



I wonder if a DRB-III scanner would find anything if it probed the ABS controller. I don't know whether it's possible for the ABS controller to be partially faulty and not turn the light on the dash.



Try your simple test procedure and see what happens. Who knows, maybe there's crud in one of the connectors and by removing and installing it you'll get it fixed up.



Otherwise, if the wires are all good it might actually be worth it to (gasp!) take it to a dealer and ask them to hook the DRB-III up to it.



Ryan

 

[FM]

Well, I never found the cause and quit looking. After all, how critical is knowing your speed. So the other week I'm parked sitting in the truck in a suit with the ac on waiting for an interview. I'm watching the trip meter and all of a sudden I see it advance while the speedo is bouncing wildly. Tripped me out! I keep watching, it keeps advancing. I had no idea up to this point that all these years it was putting false miles on. In addition it had recently been surging the engine although I'd not realised there was a causal relationship with the speedo. So I go to the dealer and after a day of searching they find that the wire harness going down over the bell housing to the 6 speed is "stretched tight through a metal tie down and is causing signal interference". No breaks or shorts. They rerouted the harness such that it was no longer tight and no more problem. It's been 2 weeks and has not recurred. Cost me a pant load but I'd never have found it with no physical damage to the harness.

 

[Prairie Dog]

Wow, never heard that one. How many miles did it add while you were watching?

BTW, thanks for the update.

 

[SDrake]

While tracking down a speedo problem similar to yours I recently found the connector to my ABS controller sitting atop the ABS pump was corroded so bad two pins were entirely gone. Wiggling the plug made the speedo needle wiggle. Sent the controller off to have it rebuilt. don't know yet whether that will entirely solve the speedo issue but I got my speedo back for a while before sending the controller off. wouln't hurt to have a look. It is the 14 pin connector on my 03.

 

[FM]

It added about 3/10ths in 30 minutes.



The ignition switch had problems and they replaced that first but they came and talked with me after replacing it and said it had no effect on the speedo.