1994, 78K miles, 5.8L, Elect. Shift transfer case, E4OD transmission w/Mercon V ATF looks/smells OK.

Have developed a very intermittent trans problem lately. Will be accelerating slowly, or on and off the gas in traffic just after a slow accell, and suddenly the trans breaks free and engine revs head up. I ease up on the gas quick to avoid clutch burn, and after a flub or two, it lock in again and everything is fine. Once when it did it, I quickly locked out OD via the stalk-end switch, that had no effect (I had thought that maybe OD was going).

Later on, I carefully counted shifts versus light acceleration speed, and I figured if the intermittent problem did occur at a shift, it was probably the 2 -> 3 shift.

I thought I'd get out the code reader today, as it can read the 600-level trans codes.

For KOEO: Hard Codes stored were 111, test pass. For Continuous Memory Codes, Code 452 VSS Signal Too Small was stored.

It appears the only trans output speed/road speed info the PCM gets is from the squatty VSS sensor on the rear axle. From the VSS sensor to 4WABS Control Module, then from 4WABS Control to PSOM, from there to PCM. Have not noticed anything funny with Odometer or Speedometer, but then with an intermittent, might not.

Questions:
- Does an intermittent VSS sensor sound like a possible transmission control problem? (like maybe the momentary loss of speed info causes trans control pressures to drop lower, which allows clutch to slip and breakaway)??

- It is so hard to see the VSS up on the top of the diff, between two casting ears, and with the rear sway bar not that far over it.
I have looked up two VSS sensors online, the BWD S8319 which I can get locally at O'Reilly, and the SMP SC84 on RockAuto. Those two look exactly alike to me, but I am wondering if they look taller than the factory VSS I have, which looks like it has a very flat thin mounting plate on it so overall outside height sticking up looks shorter. Anything important on Bronco VSS sensors that I should know about?

 

take that rear diff sensor out when you go to the auto parts store to match it up. I would start there.

when was the last time you had the trans serviced and did you do the work or a shop?

 

I do everything. 22K miles ago at 56K miles I drained torque converter, blew out coolers and lines, changed filter, gasket, and switched to Mercon V from the original Mercon. Everything looked fine.

The problem has occurred a total of 3 times over a period of about 6 weeks minimum. There is no way I can just go out and duplicate it! I figured something might/hopefully be stored in continuous memory, and there was!

 

I would not suspect a bad VSS if the speedometer is not fluctuating. The VSS rarely go bad on these trucks. Aftermarket versions are a crap shoot. Measure the VSS signal at the PSOM. It should increase linearly to ~3.5 volts as you increase speed to 30 MPH. I bet you have a PSOM issue sending the signal to the computer ( PCM ).

 

Yes, a problem with the VSS will cause trans control problems.

Specifically what is happening is that the signal is probably going to zero, so the software thinks you've stopped and shifts to first gear. There is no engine braking in first gear, so it feels like neutral.

VSS failures are not at all uncommon.

 

I'm going to stick with my original statement. We see VSS issues in the 1987 to 1996 F-series forums come up regularly. Rarely has replacing the VSS fixed any of them. Sometimes the problem has become worse. Just my two cents worth.

I respect Mark's wealth of knowledge but too many times I have seen someone replace a VSS because of that code and did not resolve the issue. Proper diagnosis is the key.

 

I don't disagree with this at all. You're correct. Proper diagnosis is the key.

In your previous post I read it as you thinking that VSS failures are rare. They are not.. But just replacing the VSS because of a code is often the wrong thing to do. It could be the VSS, the wiring, the PSOM, or even sometimes a problem inside the differential.

 

I agree, testing would be the best. However, the problem happens so rarely, only happens for maybe 5 seconds or so, and then is fine, and would require quite a bit of disassembly at the proper time... on the road... when it's happening...

I was able to use a friend's wheel-type lift, so changeout of the VSS sensor was easy. No filings or pieces on the old one, physically looked real good. The few ABS ring teeth I could see with a mirror and light looked fine. Can't rotate wheels on the wheel lift, so all I could see was what was up right then.

Drives fine with the new VSS, but so did it before with the original one, until the rare problem happens. So I guess time (a LOT of time), will tell.

I later connected my DVM on AC Volts across the old VSS, and brought the edge of a piece of steel rapidly back and forth past it, and could get 350 mV or so. But that would be way slow as compared to operation, plus its hard to keep the gap small while wavering a piece of metal without hitting it.

I put the old VSS into a baggie, connected it to my DVM on ohms, and into a pot of water with a cooking thermometer (yeah, a real Mercury one!) and slowly heated. EVTM lists the VSS as 1400 ohms nominal.
1378 ohms @ ambient F
1400 before 100 F
1500 @ 122 F
1550 @ 145 F
1600 @ 160 F (that was at the 20 minute mark, as I was heating slowly)
1650 @ 177 F
1700 @ 195 F
The water temperatures listed are leading the actual inside the coil temps due to thermal lag, as the resistance continued to rise for a while after I took it off the burner.

All the data showed was just the positive temperature coefficient of many turns of copper wire. I was doing it to see if it went open at all for an instant... it did not, but then being so intermittent of a problem, chances of seeing something were about zero, if it IS the problem.

Time will tell...

 

If the Speedometer never wavers when the problem happens my money is on a flaky connection on the PSOM. It's a worth a look IMHO.

 

And what would I be looking for?
I've read a few of the PSOM threads here on FTE, I certainly could have missed a thread that would tell me what to look for. If you could steer me towards a particular one. I can handle PCB electronics work, it's this other part that bothers me --->

The thought of doing anything with the instrument panel does not thrill me... here in the land of sun, the heat and high UV light turn dash plastic crispy. And I've had to improvise when switches fell through their panels, their mounting bosses crumbled, had to flatten off what was left of the bosses, and drill a hole from the front through the boss, to mount the switch back on with screw, metal spacer, washer and nut for each end of the switch.

 

Look for leaking capacitors and/or loose components.

courtesy of subford

 

Thanks.

Well, today it glitched just once in a bunch of drive cycles. I think it was during or right after a 2-3 upshift, was light on the gas, it's hard to play back what happened previously when you don't expect it. Engine revs suddenly went up, I automatically eased up, it locked in without any jolt. The speedometer did not change, it held constant.

After that occurred, I had two more drive cycles after that till I got home. At home, I checked for KOEO codes, both the hard codes and the continuous memory codes were 111, pass. It's possible that EEC-IV might integrate events such that just ONE occurance in a drive cycle may not set a code in memory to avoid nuisance, don't know, don't think I've ever seen EEC-IV software algorithms really addressed anywhere with enough detail.

If I had NOT gotten the VSS signal too small code before as described above, I probably would have changed out the MLPS, as it seems most 1994 owners have done it already.

I guess I could be fighting two different sources, but it would seem odd that the problem symptoms are exactly alike!

 

The EEC-IV does need to see a fault multiple times before setting a code.

A bad MLPS does have the same symptoms. A bad MLPS will cause the trans to shift to first gear for a short time, and first gear will overrun, so at higher speeds, such as during a 2-3 shift, the effect is the same as going to neutral. A bad vehicle speed sensor can make the EEC-IV think the truck has stopped, so it shifts to first gear. Same thing happens.

 

Thanks Mark. I'll keep watching out to see if the VSS code eventually gets set. If I never see a code, then on to MLPS replacement.

 

UPDATE
Knock on wood... and I may regret posting this... but since I changed the original 1994 MLPS out using a SMP (Standard Motor Products) update kit (new different MLPS and new connector housing and separator plate, connector seal, and repinning instructions), I have had no more trans issues.

A few days later, I did have to take the new re-pinned connector back apart, as I had lost the switch closure in Neutral required for the 4WD Control Module to allow shifting into 4 LO. I found that the individual connector socket for that circuit had not securely latched into the new plastic connector housing, and evidently had backed out of place when I plugged it into the new MLPS.
Trying out 4 LOW was the one test I forgot to do before I had put it all back together!

With it apart again, I could not get the connector socket for that circuit to latch to the harness housing. I temporarily swapped socket positions with another circuit in the housing to see where the problem was. The original socket would latch in a different position, the 2nd circuit socket would not latch in the suspect position, so it was that cavity in the plastic housing that was the problem. I tried pushing it in further, too far, and it latched! I just had to force the socket back under the front lip then, and all was good, very secure. Maybe the plastic molding of that housing cavity was not perfect.

The hardest part of MLPS replacement and connector repinning is unplugging the trans solenoid connector over on the passenger side, to get enough harness slack to be able to work on the MLPS harness connector. The exhaust pipe positioning, and the frame rail, make it very hard to get to the solenoid connector. The arm, wrist and fingers, at least mine, were not made to contort that way, and deliver the effort needed. I had pins and needles in my thumb and wrist for a week afterwards.

It's done... I hope... for ever. Or maybe I'm whistling in the dark, as the VSS code I got originally doesn't track with MLPS issues! I'll keep whistling...