I'm guessing it's the speedometer itself, although you might get lucky and discover that it's merely the contact screws being loose.

 

The VSS signal doesn't go straight to the ECU, it goes through the cluster then to the ECU. If your cluster doesn't work, ECU never gets the VSS signal. Your case sounds very extreme, but your first step should be replacing the cluster. Junkyards have cheap clusters.

 

Thank you so much for the quick replies. I'll hit-up the bone yard. Last time I was up there, there were quite a lovely selection of gen 7s baking in the FL sun. I'll read up on the gauge cluster removal section of my Haynes repair manual. Here's to hoping it just the cluster.

 

Be sure you know how to test the raw VSS signal coming from the sensor.
And how to test the massaged signal from speedo to ECM.
You may very well get a cluster that's just as bad as yours.

The problem with swapping parts without testing them 1st is you may end up swapping out a lot more parts than needed until you happen to get lucky and replaced just the right combination that happened to be bad.

 

Thank-you, good point. Testing the VSS itself is pretty straight-forward with a multimeter and some jumper wires running off a 12v battery. I should have taken a moment to check it 1st, but the part was under $20 so replacing it with a new one wasn't cost prohibitive and had I not messed up the original swap, it would have been a fairly quick task.

I'm digging through my Haynes manual now looking for info on testing the instrument cluster. The thought of just randomly grabbing one off a junked gen 7 and hoping it works is not one I relish.

 

Well, you can get some rudimentary testing done with multimeter. But you can't tell what the voltage-peaks are or how many of them are out of spec. Here's what the VSS signal looks like if you measure it in real-time. Top-half of 1st graph shows the raw VSS signal. It needs to be at least 12v peak-to-peak. Then that gets converted into a 0-5v digital signal in the cluster and sent to the ECM.

What happens with a failing VSS is the top-half of 2nd graph. It's not all-or-nothing black-and-white yes/no good or bad. SOME of the waveforms are sub-standard but will still cause the needle on a voltmeter to flicker. But those don't get converted to digital signal. You end up with drop-outs going to the ECM and it'll throw codes and you may or many not have driveability issues.

So... you can break this down into testing segments along the way:

1. test VSS signal at sensor itself, if confirmed bad, then replace it.

2. if signal is good, test signal at cluster-plug. If signal is bad, then wiring in between VSS and cluster is bad.

3. if signal at cluster-plug is good, check digital-output to ECM at cluster, if good, go to step #5

4. If signal digital-output is bad, check for loose connections and screws. If all OK, then cluster is bad

5. Check digital-output at ECM plug, if bad, then wiring between cluster and ECM is bad.

6. If digital-output at ECM plug is good, then ECM may be bad. Tough to diagnosis internally as there's no documentation, so simple test is to swap with known-good ECM to see if problems go away. If so, then ECM is bad.


Without testing individual components and coming up with quantifiable measurements, you're starting on this long chain of swaps:

- swap VSS, darn still problems so..
- swap cluster, nope, still not working
- swap ECM, nope, still not working
- swap transmission solenoids, nope.
- swap transmission, still funky
- swap car with one that works

It very well could be some loose plugs or corroded connectors along the way. But without actually drawing the realtime signals along the way, it's difficult to pinpoint the exact problem area. Sure, you can swap everything between the VSS and ECM, but that's very time-consuming and expensive. And that'll only be effective if you can measure the components from the junkyard cars. They very well could be just as defective as the one in your car. Only way to know for sure is to buy brand-new parts from the dealer. You'll end up with a box of spare parts that you won't need for decades if ever. All the above steps I've outlined, takes all of 15-20 minutes and will highlight the exact component and only the component that's causing the issue. And you won't even have to leave your garage and wander aimlessly around junkyards.

 

As others have mentioned the VSS signal goes through the cluster before hitting the ECM. Without the VSS signal the ECM has a fail safe mode that relies on throttle position and... You guessed it - RPM. So since your ECM doesn't get RPM signal the fail safe mode is in trouble.

I had code 42 for a long time (not 12 though) and recently narrowed it down to loose screws holding the speedometer in place (the screws also conduct signals between the speedometer and the cluster circuit board). Make sure yours is on tight. You need to open the cluster to access the screws.

Edit: you don't need to open the cluster to access the screws, but you do to remove the speedometer (which you could do to visually inspect the solders and capacitors).

 

Wow! DannoXYZ! You rock! Thank you for taking the time to give me a step-by-step troubleshooting guide.

I can get at the instrument cluster and check for loose screws and/or damage with a minimal time investment. And since the thermometer and gas gauge indicators work, perhaps it is just some loose screws or bad solder point.

The image of the VSS data in real-time is awesome! I have no idea what instrument is used to capture and display that data. I have very rudimentary diagnostic tools which have been enough up until all the weirdness with this Corolla, but one can never have enough tools! So please, share with me the name of the device that allows one to see that data in real-time and with that accuracy!

 

The image of the VSS data in real-time is awesome! I have no idea what instrument is used to capture and display that data. I have very rudimentary diagnostic tools which have been enough up until all the weirdness with this Corolla, but one can never have enough tools! So please, share with me the name of the device that allows one to see that data in real-time and with that accuracy!

I use a Tektronix 222 digital storage oscilloscope. Pretty old-school, but was a great deal at $30 on fleaby. Came without a manual, had a loose front-bezel, dead-battery and one of the probe-wires was broken. Nothing a download, a little glue and solder couldn't fix! In addition to the screen, I connect it to a laptop and it streams data to a log that can be graphed and analyzed later.

http://www.machine--tools.com/By-Lo...By-Location-/Minnesota-/Tek-tektronix-222PS-portable-10MHZ-oscilloscope-img.jpg

Nowadays, the Pocket Scopes appear to be very handy. Only one channel and doesn't appear to have live-capture/download. But certainly more useful for hopping around engine-compartments and under dashes with. I may just pick one up to play with.

These things have saved me a tonne of time personally and when I had a shop. If I had to pay my guys for all the time it takes to run to junkyards and test-swap parts, no one would get out with a repair-bill less than $2000.

 

I just reviewed the troubleshooting section of the factory manual for DTC 12. And then I re-read your first post.

I agree DannoXYZ has an excellent run-down of how to troubleshoot. I'm adding some commentary here as to what I think might have happened, but by following his steps you could confirm.

Here's what I think: I think you have had code 42 for a while but never noticed (similar to me). DTC 42 doesn't set a check engine light, and it also isn't directly noticeable as long as the ECM gets to operate in fail safe mode. Particularly since your speedo doesn't work. So, let's go forward with the assumption that code 42 is "old". Since the speedo itself doesn't work I bet it's the solders or capacitors on the speedo that are bad unless it's something simple like I mentioned (loose screws, loose connectors). Take the cluster apart and study the speedometer back side. The capacitors can be replaced, and the solders can be re-soldered... But it's up to you to decide if it's worth it versus just getting a "new old one" that works.

So what triggered the issues then, if 42 has been there for a while? It must be that the RPM signal got lost to the ECM, and that all of a sudden fail safe mode becomes impossible. Looking at the troubleshooting for code 12 you need to check the following:

1. Resistance of each pickup coil
2. Check for loose connectors or short circuits in harness and connector between ECM and distributor
3. Check air gap

The manual then states that if all those three check out fine you need to replace the ECM. I have yet to see a bad ECM (I thought I had a bad ECM myself but I didn't). So it's likely one of the other issues.

In terms of priority I would actually seek to fix this Code 12 issue before tackling Code 42. You can live with Code 42 for a while, but Code 12 is what's making the car not drivable.

 

I love this forum!

Thank you 94RollaDad!

I've been so frustrated and I was so close to just towing my Corolla to the shop and throwing money at it. Money I can't afford to throw at it right now, so the car would just sit until mid February. Which is just sad.

I'm pulling the instrument cluster either later today or early tomorrow, depending on the woman + kid's needs. I'll report back. Seriously, I really appreciate all the help.

 

Just another thought, I recently had a friend who owns a 94 with the 1.8. He was having trouble with the transmission acting up, He is an electrician by trade and knew how to read the diagrams. He replaced one of the axle's, no help, thought it was a mount, no help, thought it was the speed sensor, we changed it with a known good one, no help. He told me the circuits go thru the cluster, so we changed it out with a known good one, and that solved his shifting and transmission issues. Just my .02 I have learned may very weird things about these cars.

 

After combing through this board and reading about all the weirdness a failing speedo cluster can cause, I'm hoping that is my issue. I just pulled the cluster out of my 94 Corolla, it was literally 6 screws and 3 plugs to get it out. Less than 5 minutes.

So looking at the back of it, what am I looking for? Burn marks? Do I need to tear it down more? The speedometer is the middle gauge. It looks like those 4 screws come out, 2 screws hold the plastic cover on the front. Pop out the speedometer gauge and inspect the solder joints and any capacitors?

Here's a pic:

 

You have to remove the front glass (see tabs on front) and then lift out the black bezel (if I recall correctly). That might also have tabs - can't remember. Once everything is off in the front, you can unscrew the four screws for the speedometer and gently lift it out. You're looking for broken solders and leaky capacitors. If everything looks nice, assemble tightly again, clean the connectors with some rubbing alcohol or some electronic cleaner, and put back in. See if the VSS code still persists, the fact that you've now tightened everything might have helped.

 

Further disassembly was pretty straight forward. A few clips to get the housing apart and those 4 screws to free-up the speedometer gauge.

Upon close scrutiny, nothing jumps out as the cause for the gauge not functioning. I was looking for a bulging capacitor or a burnt solder joint. Nothing. The thing appears to be pristine. Well I'm still gonna replace it. I want a tach (don't we all, heh) but I'll take what I can get. I need to make sure the car is drivable then I can mess with tweaking parts.

Pay day on Friday and I can get a replacement part. I guess I'll clean this one up and re-install it though.

Thanks for every one who has contributed to this thread. I am really appreciative of all your time and energy.

 

Probably nothing, but those 4 speedo screws look black or corroded?

 

They are black. I had a moment of excitement when I 1st glanced at the back of the instrument cluster and saw those 4 screws were not all shiny like the ones holding in the other gauges. But when I un-screwed them, they're just black screws.

Got the panel re-assembled and cleaned everything that I could with some isopropyl alcohol. As soon as the wife wakes up and I can hand her the baby, I'm gonna re-install the instrument cluster and see if it clears up my codes. Wouldn't that just be nice if all it took was some cleaning! We'll see in a bit I guess.

 

Got the panel re-assembled and cleaned everything that I could with some isopropyl alcohol. As soon as the wife wakes up and I can hand her the baby, I'm gonna re-install the instrument cluster and see if it clears up my codes. Wouldn't that just be nice if all it took was some cleaning! We'll see in a bit I guess.

It might clear DTC 42 but not DTC 12 as far as I can understand. You'll need to inspect the items I previously listed for that. RPM does indeed run to the cluster for tach setup, but judging by the wiring diagrams it doesn't run through the cluster to get to the ECM. Maybe start off by inspecting the connectors on the distributor.

Regardless - good luck!

 

OK, just re-installed the gauge, hooked the battery back up and started her up. The check engine light came back on pretty quickly, so obviously something is still amiss, not that I really thought pulling the instrument cluster and cleaning the contacts would have made everything perfect, but still, lol.

Anyway, took her for a spin, the speedo is still dead...BUT...I noticed my trip odometer was functional! WTH. I drove the car a mile and the odometer counted it! The trip odometer was aligned weird, so I hit the trip reset to zero it back out and it stopped working again. But I drove the car another mile and the odometer totally registered it.

The car is shifting fine again and it's not idling weird at stops. But the car drove fine on it's initial test drive, then was weird the next day. It seems that after it sits for a bit, it drives great for a few miles before it stops shifting right.

I'll keep yall updated as I continue troubleshooting. :/

 

Pull the codes again. As mentioned DTC 42 doesn't set the check engine light. You didn't really do anything to address DTC 12 with what you did there, so it was expected that the CEL would come back on. But pull the codes to see if you got rid of DTC 42.

 

If it would help in this step, and it appears a lot of people have VSS-to-ECM signal issues, I have a circuit that massages the 12v analogue signal into a 0-5v digital signal for the ECM. You can them bypass the cluster completely and send the signal directly from VSS to ECM.

If shifting then functions smoothly, we can know for sure the cluster is bad. Although replacing the cluster with known good one will do the same thing. Trick is the time to find a working cluster, yank it, pay for it and swap it.

Hmmm, perhaps we can start a tool-sharing club? Have a box of tools and spares on-hand that would be too costly for any one person to assemble. Then lend it out as needed, perhaps with a deposit.

 

A few members have offered spare parts for testing, just paying shipping. Speedy offered a known good dizzy to a few people for testing.

 

OK, updates on the weirdness with the shifting problems. So initially the speedo, the odometer and the trip odometer weren't working at all. Cruise control was dead too. OBD was showing a code 12 and a code 42.

I swapped the VSS and the odometers and cruise control started working again. I took the car out yesterday and it drove fine for about 5 miles then it stopped shifting correctly again. It wouldn't go faster than about 25 mph. I turned off Over-Drive and I limped back home on back roads. After a mile or so, it was suddenly shifting properly again. I turned O/D back on, put on the radio and the a/c and the car immediately dropped into high rpms and wouldn't go above 25mph. I turned off the a/c and the radio and it drove fine the last mile home.

I checked the battery and it's a cheap-o 425 CCA economy battery. I don't know if the battery is contributing to these weird problems but it is a very small battery and the weather has been considerably colder outside than it normally is. Like high 60s whereas it's usually in the mid 80s around here, if not hotter. I know cars can do all kinds of odd things when the voltage drops too low. I hate these people who cheap out on the battery! The difference between this battery and the correct one is only like $20! I get it, maybe they didn't have the money at the time they needed the battery, but ugh, c'mon!

I ordered a new instrument cluster off ebay and it arrived about 30 mins ago. I tossed it in and the speedometer is working! I pulled the negative battery cable off to clear the CEL (hopefully). After I eat dinner I am going to re-connect the battery and see if the CEL goes away. The back-light/night illumination bulbs in the new cluster are bad and it's dark out at friggin 5:30 lately so I only drove to the end of my block but the car seems to be driving fine. Lights were on and it wasn't doing anything weird. I wish I had more time to drive-test it but my work week starts again tomorrow. I will update again real fast tho once I reconnect the battery after dinner. Here's to hoping it will run fine now that the VSS is changed and the speedo is functional!

**UPDATE** OK so the Code 42 is gone but the Code 12 CEL is still present. I'll drive it around tomorrow and see what she does.

*UPDATED AGAIN* - I know the code 12 can be a distributor issue. The car is sometimes making an odd, high-pitched whining noise when it's first started up or after it's driven for a few minutes. One of the plug wires is odd too. I need to do a tune-up.

 

...
I swapped the VSS and the odometers and cruise control started working again. I took the car out yesterday and it drove fine for about 5 miles then it stopped shifting correctly again. It wouldn't go faster than about 25 mph. I turned off Over-Drive and I limped back home on back roads. After a mile or so, it was suddenly shifting properly again. I turned O/D back on, put on the radio and the a/c and the car immediately dropped into high rpms and wouldn't go above 25mph. I turned off the a/c and the radio and it drove fine the last mile home.

I think this is a sign of a wiring-problem such as a short. Radio-installers often hack up your wiring and leave a trail of damage.

I checked the battery and it's a cheap-o 425 CCA economy battery. I don't know if the battery is contributing to these weird problems but it is a very small battery and the weather has been considerably colder outside than it normally is. Like high 60s whereas it's usually in the mid 80s around here, if not hotter. I know cars can do all kinds of odd things when the voltage drops too low. I hate these people who cheap out on the battery! The difference between this battery and the correct one is only like $20! I get it, maybe they didn't have the money at the time they needed the battery, but ugh, c'mon!

Not likely the battery is contributing to problems. The only function of the battery is to start the car and if it can start the car, it's fine. Once the car's running, there's absolutely zero output needed from the battery. All of the power-consumption needs are supplied by the alternator and an additional amount is used to re-charge the battery. Once the car starts, you can actually disconnect the battery and throw it into the trunk for the remainder of your trip. Although not a good idea as the disconnection process can damage the VRM/rectifiers in the alternator.

 

And now the car won't start. Wifey was driving it when it just died. Battery is fine. Engine cranks but it won't fire. WTF?! Ugh.

 

That's part of why I recommended going after DTC 12 first, since that lack of RPM signal could mean there's something more fundamental at issue in the distributor. And lack of start could be the same. Was this a cold start or attempting to start it when it's still hot?

You need to start digging into troubleshooting the distributor since that error code points to something already wrong there.

 

Thank you. Multiple signs were pointing to the distributor. And of course this happened when I was at work today. The woman was trying to get the baby to mellow out by taking him for a drive. She got 3 blocks away from home and she says the car just died. No niose, no weirdness, just lost all power and rolled to a stop.

So we all know fuel, spark and compression to start an engine. I'm leaning towards a fuel issue as the car was running. She was at a red light, light turned, she accelerated car died and she rolled to a side street.

It sat for an hour or more. It cranks. Battery is fine. Has gas, added 2 gallons just in-case the gauge was wonky. Just wont fire up.

 

At a minimum check for spark before focusing on the fuel system. Also check all fuses, especially the EFI.

 

Fuses are all good. Turned the key and listened for the fuel pump and didn't hear a thing. I am not gonna mess with it anymore tonight. I have to go to bed shortly. I guess she sits til my next day off. Boooooo

 

Perhaps the relay is bad.

 

And you can listen to the injectors with a stethoscope or some vacuum hose, and see if you hear them snapping/firing.
But I suspect a spark issue too.
I believe that on a Ford EEC4 system, if the distributor is going wonky, it can cause the system to not deliver fuel. They definitely can do the fuel pump trick above.

 

Not a good idea on any computer-controlled car, I've always been led to believe.