95 Windstar AX4S transmission ATF fluid flow

[fix4all]

Recently had a tranny filter/oil change since shifts were becoming erratic. Prior, had a rebuild about a year ago and tranny was satisfactory for the most part. After the recent service the tranny went bi-polar. Always take my vehicles to this one shop for many years and they are top-notch with oil/filter service. Never issues. Suspecting that the shop that did the rebuild botched things since they dropped the tranny at least 6 times. Long story, 6 months worth, that I wont get into and now out of business. I'm guessing they may have used some additive to get things going rather than fixing the underlying issue. Original problem was the common 1st/reverse gear cracked piston. They corrected that but shifts were very erratic after, so the saga began...

Suspect the addition of new fluid may have changed viscosity and frictional properties, thus why the slip, shudder and shifting issues after the fluid change. Added various additives myself with some degree of help. Not willing to waste any more time decided to do a complete flush with Mercon V. So I did with Super Tech brand. Shifting improved with less shudder and was almost perfect so long as the ATF was under 150 deg. Once the fluid heated beyond that (up to 230 deg) shifting again went to hell.

The service shop said the ATF smelled burnt although the color was not too far off from normal. What I noticed when doing the complete flush is that virtually no fluid was flowing out the return line when in park or neutral when idling, even after the ATF had warmed. Thought that was odd. Ended up doing the flush with the engine in park with the throttle open some. That was the only way I could get a fair steady flow. The stock cooler was bypassed after that with a aftermarket cooler.

Barrowing a high-end scan tool was able to monitor tranny status in real-time. It appeared that the EPC voltage would drift about 1/2 a volt with the ignition on (engine-off). Went through the entire circuit and connectors/wiring back to PCM and could not find any resistance or voltage variations. Did not swap the PCM. The resistance at the EPC solenoid measured a steady 4 ohms (circuit disconnected). Other transmission related sensors appeared OK monitoring. Most had been changed prior to the rebuild. VSS, Range switch & TPS. MAFF cleaned. Engine runs well. Misfires (on scantool) minimal. TSS and engine RPM are very close at idle/neutral. Of course, varies some due to TC slip during driving.

Was clued-in that this appeared to be a pressure related issue since occasionally when the OD light would flash (code always consistent, non-specific generic incorrect 2-3 gear ratio). Of course, the PCM would ramp-up pressure (limp mode) and at that point the car would shift nearly perfect in every instance even at elevated temps.

Speeding forward (no pun intended) I placed a 3.3 ohm 25 Watt resistor in parallel with the EPC solenoid wires near the bulkhead connector. Observed that the lower overall circuit resistance raised the EPC voltage on the scantool. The car shifted perfectly thereafter, except for an occasional semi-hard 2-3 shift! After about 20 minutes of driving the ATF is well above 200 degrees. No codes or OD flashing any longer. Plenty of torque. TCC locks up as it should. Still little to no change in external ATF flow at low RPM's.

This is baffling. There's apparently adequate pressure with the mod, still why very little to no flow to the cooler when at low RPM's? Could the tranny internal thermal bypass valve be sticking (the cooler has none). Don't want to dump a load of $$$ on this issue. Hoping to get a couple more years if possible with the Winny.

Was even thinking of connecting the cooler supply line to the EPC test/service port. Would that bypass the internal valve?

Any thoughts or ideas from the community?

[tomj76]

I had long post written that was lost when I tried to post it (session timed out). It must have been too long.

I don't know why the EPC voltage would increase by adding a resistor in parallel with the EPC circuit. It seems like it should be the opposite (more current, more voltage drop). Can you check it with an oscilloscope (EPC is pulsed)?

You might have a hydraulic leak inside the transmission (e.g. at a pushing passing hydraulic fluid into a clutch pack. When I rebuild my 1996 AX4S, I found several bushings with higher than expected (I couldn't find clearance specs on the bushings). In the same fashion you might have a seal that is leaking.

Have you checked pressure levels directly?

[fix4all]

Don't you hate that!

You are absolutely correct. The voltage measured is from the PCM side connection to GROUND. That is what the scantool also shows. The lower the resistance value of the solenoid (PCM to V+), the greater the pressure the PCM commands.

I blocked off the supply line from the trans and connected the inlet of the cooler to the EPC test port. Immediately could detect the cooler warming significantly at idle and beyond. The vehicle ran cooler, of course. The shifts and slipping, however, were not acceptable in this configuration, perhaps due to altered fluid distribution. I did throw a pressure gauge on the port prior and idle pressure was about 15 PSI and at 2K RPM around 30 PSI.

Could be what you mention about internal leakage. The AT is not cooling adequately and already the fluid smells burnt after a couple hundred miles of freeway driving since the recent flush with SuperTech ATF. Will flush/replace fluid again, this time with Schaeffer full synthetic, and limit driving to short hops in town. Don't have the courage nor money to do another rebuild at this point.

Unfortunately, do not have an oscilloscope any longer.

[tomj76]

Well, your description of how the EPC works makes sense with what happen with the resistor. The EPC regulates pressure by pulse width modulation of the current to the solenoid. The voltage is switched rapidly between GND and B+, but the PCM controls the amount of time at either condition. By controlling the time the pressure varies. If the GND state provides maximum pressure and the B+ state provides minimum, then adding the resistor would result in higher pressure.

Also, I just read a piece on flushing that was dead set against it, mainly because (1) fluid from the transmission to the cooler is not 100% of the flow from the pump, but instead only a fraction of the total resulting in a dilution of the old ATF rather than replacement, and (2) because service stations that flush ususally don't change the filter. The site (agcoauto.com) suggested a traditional service of removing the fluid and filter through the pan. Their argument seemed logical to me.

My approach is as follows

(1) Siphon out most of the fluid through the dipstick tube.

(2) Remove the outlet tube (to the radiator) and direct it to a suitable container. Start the engine while watching for fluid to exit from this tube and shut down as soon as the flow stops .

(3) Drop the pan, replace the filter and clean the pan. It is worthwhile to cut open the filter to see if any debris has accumulated inside.

(4) When reattaching the pan, watch for striping the threaded holes. I stripped several because the bolts I removed had aluminium embedded in the threads. I ended up putting in about 12 helicoils to repair all the holes that I stripped before I understood the problem. Chlorine bleach will dissolve aluminium without damaging the steel, so you can soak them to clean them.