martinmzfan1

At 67K miles and 5 years in service, I am a bit worried that the "OD Off" light is starting to blink in a noticeable sequence. While driving, I noticed that the ATX was holding gears in a bit longer before shifting (1-2). Then the "OD Off" light popped on and blinked in a sequence. The car still drove normally but I pulled over and turned the ignition off. I restarted the car and drove off. The light did not come on again all day after repeated short errands around town and on freeway. No CEL appeared either.

I toggled the "OD Off" switch a few times and it functioned normally -- steady indicator light and ATX shifted properly from 3-4 (OD). I even checked the fluid while the ATX was hot and found a normal level with no obvious mal smell or discoloration. BTW, the fluid and filter were replaced last summer; my indep. mechanic found nothing unusual.

Arhggg. I'm going to bring her in to my indep. mechanic to check for codes and proper operation this week. Anything I should look forward to? She's been reliable... hate to see a major failure at 67K! Is there an update/reflash for our TCMs? I'm hoping it's a firmware/electrical glitch!

pass the peas

speed sensor?

martinmzfan1

Okay. I'm going to AutoZone and see if any codes are stored in the PCM. Question: Will ATX codes show if the MIL is not lit? This problem is intermittant. I strongly believe that it may be related to the speed sensor. About a couple months ago, the cruise control malfunctioned. It did not engage correctly and/or hold speed. I signed it off as a fluke when I tried it again the next day and it worked fine. So, if the speed control sensor is bad, this may be the problem...

What do you guys think?
Is the speed sensor an easy replacement?

snap-on

You will need to get the code pulled to diagnose any further. Have you had the tranny flushed at 30k intervals?

martinmzfan1

Yes, ATX fluid changed twice up to now (67K). Where is the "speed sensor" for the ATX? Do I need to drop the pan to change it out? Hopefully, I'll get a code pulled out of the PCM. Man... and I just replaced the radiator a couple months ago!

snap-on

It is below the drivers side motor mount. It is a two wire connector and IIRC it is held in place with a single bolt.
I have the inspection readings if you want to check it.
Resistance
250—600 ohms (ATF temperature:–40—160 °C {–40—320 °F})

Since you say you did a radiator is it possible you had coolant contaminating the ATF? Maybe you need a aftermarket cooler?

I really need the code to help any further..

martinmzfan1

snap-on. I did a "simple" radiator change out. The bottom crimp seal (Al to plastic) failed and was leaking out coolant only under pressure. I capped the ATX lines (2) when disassembling the faulty radiator to prevent contamination. No coolant should have entered the ATX lines during the changeout and reconnection. I assume the coolant did not contaminate the ATX fluid in the radiator (they have separate channels)... I'm pretty meticulous. I think only a few ounces of ATX spilled out. The level was still good after checking a week later.

Okay... update: No DTC codes stored in OBD II. The O/D light has not blinked again and the cruise is still okay.

I have noticed that the shift point is about 3000 RPM in 1-2 and 2-3. OD (4) comes in about 3500 or so. Is this okay?

I'll check the "pulse generator" tonight or tomorrow. Do I need to remove it to check resistence? I really appreciate your feedback!!!!

snap-on

The shift points will vary on load request. If you are starting off in a casual manor and the car shifts at 3000 I'd say that is WAY to high. It sounds like the car is in a forced shift based on tranny pump pressure ILO normal speed sensor input.
You do not have to pull the sensor to read the resistance. It is simply a magnet sending pulses as the teeth inside the tranny go by.

martinmzfan1

snap-on: Update... okay, I measured resistence between the two connections on the "pulse generator" [the sensor mounted directly on top of the ATX housing that looks like a rotor/machine steel clip]: 392 Ohms; ATX normal temp at about 100 deg F.

The shifts don't feel forced. When making parking lot manuvers or in bumper to bumper traffic, it shifts at about 2250 RPM -- it does change under moderate load to 3000 RPM when accelerating briskly. What to check next? This is fun... until... we find the culprit!

It's gotta be a sensor failure... something is causing the "O/D Off" light blinking sequence and intermittant cruise control malfunction. The shifts feel smooth at times and rough at other times. The unit shifts through all four gears including reverse and O/D hold functions fine.

martinmzfan1

Okay... Update: I definately feel shifts are intentionally delayed. This is the problem. What's causing it? I contacted a couple Mazda dealers' and got some more info. One likely cause of delayed shifting in my ATX is a bad Shift Solenoid "E". According to him, techs found Proteges entering the shop for "O/D off" blinking and similar symptoms replaced this faulty part and achieved success (at least for 12 months or so).

Another Mazda dealer said they don't even bother rebuilding or replacing ATX internal parts like solenoids. They are only authorized to replace whole ATXs when an internal fault is diagnosed.

I didn't remember the blink sequence... and the OBD II showed no codes, although it was a generic scanner -- Mazda may have codes past 1000?

Moral dilemna: 1) have the solenoid replaced, 2) wait until it gets worse, 3) replace ATX or 4) sell the car. FYI: I'm leaning on 1 or 2.

snap-on

Sadly the art of tranny repair has gone by the wayside at most dealerships. That goes for most manufacturers. I wonder if you can find a shop with a NGS? They can use this to electrically command the solenoid.

martinmzfan1

Okay... I went to check the Throttle Position Sensor (TPS) on throttle body; It's registering 1150 to 4590 Ohms between two connectors (not the signal one) from closed throttle to WOT. So it's working; didn't check voltage output (0-5v).

I'm trying to get TPS operating specs.

UPDATE... just returned from a recommended independent transmission mechanic. verdict: He used a sophisticated Snap-On (red box) code reader with a software card for Mazda. One code returned: P0745 -- Pressure Control Solenoid Malfunction. He stated that the fluid looked a little dirty but okay (only 10K miles).
Code cleared and took a test-drive. No blinking O/D off light. Re-scanned car. No codes were returned. I get a free re-test anytime.

So my moral dilemna is solved. Car stays. Problem is pinned at the pressure control solenoid or its wiring harness. I'll probably wait until the O/D light flashes again and have it re-scanned. If it's the same code, I'll change out the solenoid. Besides having some slightly delayed shifting (mechanic didn't see anything wrong), the car is okay. I am going with the wait and see approach.

UPDATE again...

How difficult is it to drop the pan and replace the one solenoid myself?
Can someone provide instructions? I'll have the fluid & filter replaced again since the pan is down. I've never dropped the ATX pan & changed fluid before. At least it's mounted directly on the bottom and forward. I know it's a messy job! What is very BEST fluid I should use... something about redline?

martinmzfan1

Okay... well the OD light has started to blink again. I'm going back to ATX shop today for a rescan of the TCM and PCM. If it's the same DTC as before (P0745 -- pressure control solenoid malfunction), I will have the solenoid replaced and ATX reserviced (new fluid & filter).

Questions:

Do I have to replace the filter screen or can I just clean it out?
How easy is it to drop the pan and replace the valve-body solenoids? Can I do it under jack stands "relatively" easily or is a lift really the only way?
What is very BEST fluid I should use... something about redline?

Thanks

snap-on

INSPECT RESISTANCE

Inspect resistance between ATX connector (transaxle case side) terminals D and I.

Is resistance within 2.4—7.2 ohms?

I'm not sure I would pull the pan just yet. Check for power to the connector and a good ground.

martinmzfan1

Okay, I checked resistance between terminals D and I on the ATX case housing side: I got an OPEN connection (dead open) -- no reading. Which terminal on the control harness lead has power for the solenoid? D or I?

I believe at this point it's a bad solenoid, burnt/damaged wire inside the valve body (very doubtful) or bad connection at solenoid wiring harness inside valve body (doubtful again).

My money is on the dead pressure solenoid. Sound right? The 'E' Shift solenoid is also prone for failure in our Proteges too as I heard.

The ATX is shifting roughly (like a Mack truck) and O/D light is now flashing continually on EVERY start now. Arhgggg.

The pressure control solenoid (21-1L1A) has been ordered along with a new ATX filter. Do you have the replacement procedure for the solenoid?? Please attach. Do I have to drop the valve body too?