of ATRA. No part of this program should be construed as recommending any procedure which is contrary...

© 2000 ATRA. All Rights Reserved.

��

This manual has been developed by the Automatic TransmissionRebuilders Association (ATRA) Technical Department to be used by quali-fied transmission technicians in conjunction with ATRA’s technical semi-nars. Since the circumstances of its use are beyond ATRA’s control, ATRAassumes no liability for the use of such information or any damages in-curred through its use and application. Nothing contained in this manualis to be considered contractual or providing some form of warranty on thepart of ATRA. No part of this program should be construed as recommendingany procedure which is contrary to any vehicle manufacturer’s recommen-dations. ATRA recommends only qualified transmission technicians per-form the procedures in this manual.

This manual contains copyrighted material belonging to ATRA. No part ofthis manual may be reproduced or used in any form or by any means —graphic, electronic or mechanical, including photocopying, recording,electronic or information storage and retrieval — without express writtenpermission from the ATRA Board of Directors.

Public exhibition or use of this material for group training or as part of aschool curriculum, without express written permission from the ATRABoard of Directors is strictly forbidden.

ATRA and the ATRA logo are registered trademarks of the Automatic Trans-mission Rebuilders Association.

Portions of materials contained herein have been reprinted with permis-sion of General Motors Corporation, Service Technology Group.

© 2000 ATRA, Inc. All Rights Reserved. Printed in USA.

�� 2400 Latigo AvenueOxnard, CA 93030

Phone:(805) 604-2000 Fax:(805) 604-2005http://www.atra-gears.com

��


�� General Motors .................................................... 1

Ford ................................................................... 65

Chrysler ........................................................... 125

Imports ............................................................ 173

Isuzu ........................................................... 173

Mazda.......................................................... 193

Mercedes ..................................................... 222

Mitsubishi ................................................... 233

Nissan ......................................................... 246

Subaru ........................................................ 248

Computer Reprogramming ............................... 253

Reference ......................................................... 260

��


��

��Dennis Madden, Technical Director

Mike BairdWeldon Barnett

Bill BraytonLarry Frash

Steve GarrettEvelyn Marlow

Cliff McCormickRandall Schroeder

David SkoraLance Wiggins

Glenn Troub, Director of Online ServicesIrvin Gers, Online Services

�� Thank you for attending the 2000 ATRA seminar. The peoplebehind the scenes, putting programs like this together don’talways get the recognition they deserve for the effort they putforth. Producing a seminar program of this type requiresmonths of hard work. I would like to thank everyone who had apart in producing this program. I would like to offer a specialthanks to the following persons for spending a lot of eveningsand weekends making sure we produced the best informationpossible:

Larry Frash, who spent hours ferreting-out many of the factsused in this manual, as well as the initial copywriting anddrawing.

Evelyn Marlow, who took great pains to make sure our line artwas as clean as possible, against sometimes overwhelming odds.

Cliff McCormick, whose skill with our digital camera providedus with a crisp and unique collection of images.

Steve Garrett, who was instrumental in collecting the verylatest information for our GM section.

Steve Bodofsky, who designed and laid out our manual, createdthe slide show, and provided much of the editing for this program.

Dennis MaddenTechnical Director

��

��


��

�� Isuzu 4L30E

Solenoid Grounding Kit ....................174

Converter Drainback ........................175

Delays Forward ................................176

Code 41: Ratio Error .........................178

Isuzu NPR

Diagnostic Trouble Codes .................179

Component Locations .......................185

Computer Terminal Identification .....189

Circuit Testing ..................................191

Mazda F4AEL

1–2 Slide-Bump................................193

Mazda G4AEL

2–3 Flare .......................................... 194

Double Engagement into Reverse ...... 202

Misdiagnosed Problems ....................203

Computer-Commanded 2nd GearStarts; Forward Clutch Slip ...........204

3–4 Shift Hunt...............................207

Shifts 4–2–3–1 ...............................208

Shifts 1–3–4 ..................................209

4–2 and 4–3 Flare ..........................211

Codes, Common Causesand Computer Strategies .................. 212

Code 06.........................................212

Code 12.........................................213

Code 55.........................................214

Code 60, 61, 62 and 63 .................215

Rebuild Procedures,Technical Tips and Review ................216

The Transmission ..........................216

The Valve Body.............................. 216

Kill the Hold Mode ............................217

Mazda R4AEL; Nissan RE4R01A, 3A

Replacement Solenoid Connectors ....218

New Design Output Speed Sensor ..... 219

No Turbine Sensor Signal ................. 220

No Converter Charge ........................221

Mercedes 722.3, 4 and 5

2–3 Flare .......................................... 222

Harsh 1–2, 2–3 or 3–4 Shifts.............223

Oversized B2 Servo Seal ...................224

Self-Adjusting B1 Servo ....................225

Slips on 4–3 or 3–2 Kickdown ...........230

Mitsubishi F4A2, 3, W4A3, F4A4 and 5

No 3rd or 4th Gear; Late Shifts ............233

Harsh 1–2 Shift ................................237

Mitsubishi KM Series

No Forward,or Falls out of Forward .....................238

Mitsubishi F4A2, 3, and W4A3

Planet Failure ...................................239

Repeat Pump Bushing Failure ..........244

Nissan RE4F04A

No 3rd or 4th Gear .............................. 246

Reverse Drum Snap Ring Pops Out....247

Subaru 4EAT

Kills the Engine in Gear ....................248

Subaru 4-Speed AWD

Chatters/Binds onSlow, Sharp Turns ............................249

��


��Bad grounds are a real problem, especially with the 4L30E. The standard groundingmethod for the solenoids is a simple spade connector.

Isuzu has a harness kit that includes an eyelet you bolt on the valve body. This newground is far superior to the spade connector, and it’s a good idea to use one on everyrebuild. The Isuzu part number for the kit is 8-96041-331-0.

Early units used a spadeconnector to provide solenoidground. These units oftensuffered from bad grounds.

Isuzu offers a harnessthat provides a morepositive ground for thesolenoids.

��


��

�� Converter drainback can be caused by a leak at the intermediate shaft sealing ring.Isuzu now uses a solid Teflon sealing ring for this location.

Isuzu part number 8-96041-327-0 includes the solid ring and sizing tool. The ring isn’tavailable through Isuzu separately; if you already have the sizing tool, check with yourlocal aftermarket supplier to see if they can supply you with the ring.

��

� ��

��


��!��"��#��A forward delay is usually caused by poor feed to the servo. In many cases this iscaused by a restricted band solenoid. The solenoid screen breaks apart and gets stuckin the solenoid, restricting flow to the servo. It’s a good idea to replace the band solenoidand screen during every rebuild.

The Isuzu part numbers for thesolenoid and screen are:

8-96042-006-0 (Solenoid)8-96041-012-0 (Screen)

��


��

You may also find it helpful to enlarge the servo feed orifice in the separator plate toincrease servo feed. The original orifice is 0.040" in diameter. Enlarging it to 0.050" willhelp reduce the delay.

��!��"��#��$��%

�

��


�� &'�(��A code 41 is typically caused by clutch slip. If you’re working on a 1990 or ‘91 Trooper II,1992 or ‘93 Trooper, or 1991 through ’93 Rodeo, be aware that some units were madewith poorly machined overdrive housing castings. The machining problem was where the4th brake piston outer seal rides, and where the 4th accumulator piston seal rides.

If you have a wiped out 4th clutch, make sure you check these areas very closely for a roughsurface where the seals ride. If you need to replace the housing you have two choices: Use a1994-or-later housing, or replace it with an update kit. The kit contains a new housing, areplacement plate with gaskets, and two accumulator setups (for various models). TheIsuzu part number for the kit is 8-96041-192-0.

If you use the update kit, checkthe ID tag to make sure you usethe right accumulator setup. TheID tag is located just above themanual linkage shaft on thedriver’s side of the trans.

��

��

Check for these two items: the model code and serial number.

For units with model codes GA, GB, GC, GG, FA, FB and aserial number less than 3887359, use the accumula-tor kit with the longer accumulator spring.

Models FD and FE with serial numbers between3887359 and 4072977 also take the kit with thelonger accumulator spring.

Models FF, FG, GE, GN, GR, GJ, GV, or GO, witha serial number less than4072977, use the accumu-lator kit with theshorter spring.

GA

3747653

��


��

��)*(�� *��The JR403E control system indicates problems in the computer system through theEconomy light on the dash. When you first turn the key on, the light will flash 16 times,to indicate a diagnostic trouble code stored in memory.

Once you know there’s a code in memory, there are two different procedures for retriev-ing and reading the codes, depending on vehicle year. The easiest way to determinewhich procedure to follow is to try the procedure for early systems: If the system dis-plays codes, you chose correctly; if not, look for the diagnostic link connector, requiredfor the second procedure.

On each system, the computer displays the codes by flashing the Economy light on thedash.

(��+�� ,��!��!��To retrieve the codes on early vehicles, set the parking brake. You’ll have to put thevehicle in gear during the procedure, and this will prevent it from rolling.

To prepare to retrieve the codes:

• Turn the ignition key off.• Put the shift selector in D.• Set the Economy/Drive switch to “normal.”

Now you’re ready to begin the procedure to display diagnostic trouble codes:

To indicate a computersystem problem or displaydiagnostic trouble codes,Isuzu trucks flash theEconomy light on the dash.

��


Step 1: Turn the key on, engine off.

Step 2: Move the shift selector to 2.

Step 3: Move the Economy/Drive switch to “economy.”

Step 4: Move the shift selector to 1.

Step 5: Move the Economy/Drive switch back to “normal.”

Step 6: Hold the throttle all the way to the floor.

The computer will begin displaying diagnostic codes by flashing the Economy light onthe dash.

(��+�� ,��!��!��Isuzu trucks display diagnostic trouble codes using an 11-flash sequence. The lightflashes 11 times in a row; the sequence always starts with a long flash — about twoseconds long. It’s followed by 10 shorter flashes.

If there are no problems in the system, all ten flashes will be very short — about 0.1seconds each.

��)*(�� $��%*��$��%

!!

Here’s how early Isuzutrucks display diagnostictrouble code 1.

!!

Here’s how early Isuzutrucks indicate no diagnos-tic trouble codes in memory.

But if the computer identifies a problem in the system, one of those 10 flashes will belonger — nearly a half second long. Count the flashes: The long flash identifies the codein memory.

For example, if the first flash after the two second flash is the long one, you’re looking atcode 1.

��


��

��)*(�� $��%*��$��%

!!

And here’s how early Isuzutrucks display diagnostictrouble code 4.

!!

This is how early Isuzutrucks would display codes1, 4 and 8 at the same time.

If the fourth flash is the longer one, you’re looking at code 4.

If there’s more than one code in memory, the computer displays all of the codes in thesame pass. Here’s how a system would display codes 1, 4 and 8 at the same time:

��-��,��!�.�+��Code Definition

1 Vehicle speed sensor — transmission circuit open or shorted.

2 Vehicle speed sensor — speedometer circuit open or shorted.

3 Throttle position sensor circuit open or shorted.

4 Shift solenoid A circuit open or shorted.

5 Shift solenoid B circuit open or shorted.

6 Overrun clutch solenoid open or shorted.

7 Lockup clutch solenoid open or shorted.

8 ATF temperature sensor circuit is open or the computer power source isinsufficient.

9 Engine RPM signal circuit open or shorted.

10 Line pressure solenoid open or shorted.

��"


��)*(�� $��% ��+�� Once you’ve repaired any problems in the system, you can clear the codes from memoryby disconnecting the #11 fuse for at least 10 seconds. Then replace the fuse.

#��

��


��

(��+�� ,��!��The procedure for retrieving codes on late model Isuzu trucks involves jumping betweenthe two terminals on the diagnostic link connector. Look for the connector just below thebrake fluid reservoir, in the engine compartment.

��)*(�� $��%

Connect a jumper between the two terminals on the connector, and turn the key on, engineoff. The computer will begin flashing the Economy light on and off to display the codes.

(��+�� ,��!��Late model Isuzu trucks display codes using a two digit format: The first series of flashesindicates the tens digit, and the second series indicates the ones digit. Here’s how code32 would appear:

If there is more than one code in memory, the codes will display in order, from lowestnumber to highest. Each code will display three times, then the system goes on to thenext code in memory.

If there are no codes, the computer will flash the Economy light on and off, in a regular,even pattern.

To display the codes, run ajumper wire between the twoterminals, and turn the keyon, engine off.

!!

Here’s how late model Isuzutrucks display diagnostictrouble code 32.

��


��)*(�� $��% ��-��,��.�+��

Code Definition

11 Vehicle speed sensor — transmission circuit open or shorted.

13 Engine RPM signal circuit open or shorted.

15 ATF temperature sensor circuit is open or the computer power source isinsufficient.

17 Inhibitor switch circuit open or shorted

21 Throttle position sensor circuit open or shorted.

24 Vehicle speed sensor — speedometer circuit open or shorted.

31 Shift solenoid A circuit open or shorted.

32 Shift solenoid B circuit open or shorted.

33 Overrun clutch solenoid open or shorted.

34 Lockup clutch solenoid open or shorted.

35 Line pressure solenoid open or shorted.

��+�� Once you’ve repaired any problemsin the system, you can clear thecodes from memory by disconnect-ing the #11 fuse for at least 10seconds. Then replace the fuse.

#��

��


��

��)*( ��

$%��& ��'�&(%

��


��)*( ��$��%

��


��

��)*( ��$��%

$��'�&(%

��


��)*( ��$��%

)�(*� '�'�&(%

�%� &&��+ �

,��&��

�&� *�

��-��

.� ��,��

��


��

��)*( ��-��!��!��

��


��)*( ��-��$��%��!��

��


��

��)*( ��

niP noitcnuF snoitidnoCgnitarepO tnemerusaeM

1 ;hctiwSrotibihnItupnI2launaM

2launaMnI egatloVyrettaB

2launaMtpecxEsegnaRllA stloVoreZ

2 ;hctiwSrotibihnItupnILlaunaM

LlaunaMnI egatloVyrettaB

LlaunaMtpecxEsegnaRllA stloVoreZ

3 hctiwSekarB deilppAladePekarB egatloVyrettaB

desaeleRladePekarB stloVoreZ

4 hctiwSeldI deilppAladeProtareleccA tloV0.1nahtsseL

desaeleRladeProtareleccA stloV51–8

5 A/N—ylraE

kniLcitsongaiD—etaL

6 hctiwSevirdrevO ffOhctiwSevirdrevO egatloVyrettaB

nOhctiwSevirdrevO tloV0.1nahtsseL

7 hctiwSnwodkciK deilppAladeProtareleccA tloV0.1nahtsseL

desaeleRladeProtareleccA stloV8–3

8 A/N

9 hctiwSymonocE lamroNotteShctiwS stloV8–3

ymonocEotteShctiwS tloV0.1nahtsseL

01 ecnerefeRSPT nOyeK stloV5.5–5.4

11 langiSSPT eldI *stloV9.4–0.4

elttorhTnepO-ediW *stloV8.1–1.0

21 rosneSerutarepmeTFTA )Cº01(Fº05tA stloV8.1tuobA

)Cº04(Fº401tA stloV1.1tuobA

31 A/N

41 .feRelttorhTlluF/eldI nOyeK stloV51–8

51 dnuorGrosneS syawlA stloV1.0nahtsseL

61 )noissimsnarTta(1SSV gnitatoRtfahStuptuOhtiW langiSCDevaWerauqS

71 hctiwSelttorhTlluF eldI tloV1nahtsseL

nwoDyawflaHnahteroMrotareleccA stloV51–8

* On some models, TPS voltage is low at idle and increases with throttle opening.

��"


niP noitcnuF snoitidnoCgnitarepO tnemerusaeM

81 A/N

91 ;hctiwSrotibihnItupnIlartueN

lartueNnI egatloVyrettaB

lartueNtpecxEsegnaRllA stloVoreZ

02 ;hctiwSrotibihnItupnIevirD

evirDnI egatloVyrettaB

evirDtpecxEsegnaRllA stloVoreZ

12 dioneloShctulCnurrevO nOdioneloS egatloVyrettaB

ffOdioneloS stloVoreZ

22 dioneloSpukcoL deilppAhctulCretrevnoC stloV51–8

desaeleRhctulCretrevnoC tloV0.1nahtsseL

32 hctiwSymonocE nOyeK;noitisoPlamroNnihctiwS egatloVyrettaB

nOyeK;noitisoPymonocEnihctiwS tloV0.1nahtsseL

42 )retemodeepSni(2SSV gnitatoRelbaCretemodeepShtiW langiSCDevaWerauqS

52 rosneSMPRenignE ffOenignE CAstloVoreZ

gninnuRenignE CAstloVgniyraV

62 ;hctiwSrotibihnItupnIesreveR

esreveRnI egatloVyrettaB

esreveRtpecxEsegnaRllA stloVoreZ

72 A/N

82 rewoPevilA-peeK syawlA egatloVyrettaB

92 noitingI nOyeK egatloVyrettaB

03 noitingI nOyeK egatloVyrettaB

13 dnuorG syawlA tloV1.0nahtsseL

23 dnuorG syawlA tloV1.0nahtsseL

33 lortnoCerusserPeniLdioneloS

eldI stloV5.2–5.1

elttorhTlluF stloV5.0nahtsseL

43 .loSlortnoC.sserPeniLrotsiseRgnipporD/w

eldI stloV41–5

elttorhTlluF stloV5.0nahtsseL

53 AdioneloStfihS nOdioneloS egatloVyrettaB


63 BdioneloStfihS nOdioneloS egatloVyrettaB


��)*( ��$��%

��


��

/��"�0��&12��34��Mazda has a replacement 1–2 accumulator spring for resolving a 1–2 slide-bump com-plaint. The replacement accumulator spring replaces both the large and small originalsprings.

The Mazda part number for the spring is FU68-21-227C.

�

��


/��0��21��"��A 2-3 flare in a G4AEL can be very difficult to diagnose. To add to the frustration, youhave the customer, the center manager, and maybe even the dog pressuring you to fixit… now! Because of the additional pressure, you’re often forced to do things that won’treally fix the problem; rather, they simply overcome it or cover up the problem. One oftwo things is almost certain:

• It’ll probably come back fried within 6 months, and…

• Since you never really found the problem, it’s very likely that the next five G4AELtransmissions you rebuild will do the same thing.

There are a number of things that cause problems in the 3–4 clutch drum. Many ofthem can be prevented during the original rebuild.

��&'��+��Leak check the air bleeds with solvent. If theyleak, try reseating them by tapping them care-fully with a small punch. If that fails to correctthe leak, throw the drum away or machine it fora new checkball capsule.

Try to get 3rd clutch clearance between 0.030"–0.045".The shorter the distance the piston has to travel, theless volume necessary to apply the clutch.

� �

�

��


��

��2'��+��--��Take the extra 15 minutes to disassemble the differential completely. The output gearhas two Teflon rings that seal the 3rd clutch drum.

The best working ring available for this application is machined Teflon. You can identifythese rings by their absence of molding marks; all you should see are machining marks.Because these rings are machined, they’re held to a much tighter tolerance.

/��0��21��"��$��%

�

�

��


��'��+��0��When you reassemble the differential, pay

close attention to the 2–3 accumulator. Agood quality Teflon or rubber ring will

work well. The main point is goodquality.

� � � � � � � � � Always Scotchbright thebores when using rubber rings.

/��0��21��"��$��%

One of the most critical parts of the 3rd clutch circuit is the Teflon ring on the inputshaft. It can be difficult to find a ring to fit properly. Sometimes the only alternative is tocustom fit your own. Butt cut a larger ring to fit.

�

��


��

��'�0��Air checking the 3rd clutch is critical to avoid a 2-3 flare. If you don’t air check the 3rd

clutch properly, you can easily create a problem.

The transmission must be completely assembled with the pump bolted to the case be-fore you apply any air to the 3rd clutch.

If you try to air check the 3rd clutch before bolting the pump in place, air pressure willforce the input shaft upward. This allows the Teflon ring to come out of its bore andseparate. When the drum comes back down the ring is cut. In some cases you won’tnotice the cut ring.

/��0��21��"��$��%

�/��&(%

��


��5'��+��.��4��!While you’re rebuilding the valve body, there are some things that should become part ofyour normal rebuild procedure.

• Flat sand all surfaces of the valve body with 180–220 grit sandpaper. Use a ma-chined surface or a thick piece of glass; never use a file. Many of the circuits in-side the small channel cover and directly below it are for 3rd gear pressure, makingthis a very critical area for flatness.

• Enlarge the 3rd clutch apply orifice to 0.041".

/��0��21��"��$��%

�/��&(%�.��

��


��

The 1–2 accumulator is commonly overlooked when dealing with a 2–3 flare. 3rd clutchpressure routes between the two O-rings of the 1–2 accumulator.

During the 2–3 upshift, the computer turns the 1–2 shift solenoid off, which exhausts2nd apply pressure. This computer strategy causes some hard-to-find problems.

One of those hard-to-find problems is a 2–3 flare. This occurs when the 1–2 shift sole-noid hydraulic circuit reacts faster than 2–3 shift solenoid hydraulic circuit. This re-leases the band before the 3rd clutch has even begun to apply. This is a difficult problemto identify when it’s in the vehicle, and impossible to find if someone yanked out thetransmission before it was diagnosed properly.

/��0��21��"��$��%

��*��&�&%�+�+��(�(��+��"/��!��

��"��


��&(%

��6'��+��(��There is no accurate test for a 2–3 flare, because the only test that works dependsgreatly on your judgement.

With a 0–100 PSI gauge on the 3rd gear pressure tap, monitor how the pressure risesduring the very beginning of the flare. If you see about 30 PSI, the problem is usuallyinternal. But if there’s 0 PSI at the beginning of the flare, it’s because the 1–2 shift valvedownshifted quicker than the 2–3 shift valve.

There’s another way you can identify whether a 2–3 flare is due to a shift valve timingproblem or an internal leak: If the 2–3 flare is caused by an internal leak, it will usuallyget worse as the transmission gets hot. A 2–3 flare caused by shift valve timing problemstend to get less noticeable when hot, because the shift solenoids receive more volume asthe oil thins out.

/��0��21��"��$��%

��


"��

� � � � � Never modify this plate unless you’re faced specifically with a 2–3 flare. If youperform this modification on a unit that’s working properly, it could cause a2–3 bind.

While the 2–3 shift solenoid can cause a 2–3 flare, it’s highly unlikely. Don’t waste yourtime and money on a set of new shift solenoids; it’s very rare to find a bad solenoid. Tocorrect a shift valve timing problem, you can enlarge the 2–3 solenoid feed hole to0.041". This will lessen the time the 2–3 shift valve takes to move after the solenoidcloses.

�"/�� .�� !�(�

/��0��21��"��$��%

��"�"


/��0��(��A double engagement into reverse on a G4AEL occurs when the reverse clutches applybefore the low/reverse clutches.

Because the reverse clutch assembly uses a belliville spring to release the apply piston,clutch clearance should be 0.015"–0.020" for a two-friction clutch assembly, and0.020"–0.025" for three-friction clutch assembly. The tight clearances are to make surethe belliville spring doesn’t overtravel.

Because reverse clutch clearances are so tight, the low/reverse clutch pack must also main-tain a tight clearance. Total clearance on the low/reverse clutch should be 0.030"–0.040".

��


"��

/��0��/��*��The G4AEL computer never reads actual vehicle speed during normal operation (nocodes in memory). That’s because the computer only uses the signal from the outputspeed sensor if the input speed sensor fails. Instead, the computer calculates vehiclespeed from the input RPM, based on the gear command.

This works just fine as long as the transmission is working properly. But when internalproblems occur, such as forward clutch slip or sticking shift valves, to name a few, thecomputer can no longer calculate vehicle speed accurately. What’s more, the computerdoesn’t look for ratio errors. Because of this, many symptoms that may seem to becaused by a computer-commanded electrical problem are actually misdiagnosed oroverlooked internal problems.

Problems such as 2nd gear starts, 3–4 shift hunting, shifts 4–2–3–1, or shifts 1–3–4 canbe very difficult to diagnose without looking at the signals the computer receives. Thischart shows the vehicle speed, based on input RPM and gear range. This will allow youto see what the computer sees. The difference is that you, unlike the computer, aresmart enough to recognize a problem when you see it. We’ll refer to this chart whendiscussing these types of symptoms.

MPRtfahStupnIrotareneGesluP)zH(ycneuqerF

)HPM(deepSelciheV

1 ts raeG 2 dn raeG 3 dr raeG 4 ht raeG

05 01 3.0 6.0 9.0 3.1

001 02 7.0 2.1 9.1 6.2

002 04 3.1 4.2 7.3 3.5

003 06 0.2 6.3 5.5 9.7

004 08 6.2 8.4 4.7 0.11

005 001 3.3 0.6 2.9 0.31

0001 002 6.6 0.21 0.91 0.62

0002 004 0.31 0.42 0.73 0.35

0003 006 0.02 0.63 0.55 0.97

0004 008 0.62 0.84 0.47 0.601

0005 0001 0.33 0.06 0.29 0.231

0006 0021 0.93 0.27 0.111 0.951

��"��


��3 ��2��,"��#�� +�� A computer-commanded 2nd gear start is usually caused by forward clutch slip. If theforward clutch slips while the vehicle is stopped, the computer will assume the vehicle ismoving. Remember, the computer calculates vehicle speed from input RPM and gearrange: If the forward clutch slips enough, the computer will command second gear.

7�� .��-��Disconnect the input speed sensor (pulse generator) located on top of the case. This willcreate a code 55, forcing the computer to use the output speed sensor. If the computernow commands 1st gear starts, the problem is a slipping forward clutch.

/��0��/��*��$��%

,��+��0��& �

��


"��

/��0��/��*��$��%

�+��"�81) Use good quality, machined teflon rings

2) Check for cracks in the forward piston

��

��3 ��2��,"��#�� +��$��%

��"��


3) Set clutch clearance to 0.030"–0.040"

4) Install a rubber sealing ring from an AXOD over-drive servo cover in place of the lower teflonsealing ring

/��0��/��*��$��% ��3 ��2��,"��#�� +��$��%

�

��


"��

�1��+�-��9��

�� Even though this condition appears to be commanded by the computer, the problem isusually caused by a very small piece of dirt stuck in the 3–4 shift solenoid feed circuit.

7�� .��-��Monitor the 1–2 shift solenoid signal. If the 1–2 shift solenoid is the only one being com-manded to cycle on and off, the problem is in the 3–4 shift solenoid feed circuit.

�+��"�8Because the problem is usually caused by a very small piece of dirt stuck in the 3–4shift solenoid feed circuit, disassemble the valve body and inspect the 3–4 shift solenoidfeed orifice. If the hole isn’t plugged, try enlarging it to 0.041". This should overcome anysmall leaks caused by normal wear.

/��0��/��*��$��%

:�+��*�� Other possible but unlikely causes could be a bad 3–4 shift solenoid or a cracked endplug for the 3–4 shift valve.

�/��%�!&�� .�� !�(�

��"��


�+�-��121�1&

�� A 4–2–3–1 upshift is usually caused by the 1–2 and the 2–3 shift solenoids wiring beingcrossed. The picture shows the proper placement for the two shift solenoids.

/��0��/��*��$��%

7�� .��-��Use the shift solenoid pattern command chart with the ratio chart to see which patternsgive you the proper gear ratios. If you use your imagination while looking over yourresults you’ll be able to verify whether you have the wires crossed to the 1–2 and 2–3shift solenoids.

�+��"�8If the valve body cover is off, switch the wiring for the 1–2 and 2–3 shift solenoids totheir proper positions (green wire on top, blue wire in the center).

If the side cover is still bolted to the case and you’re in a hurry, simply cut the 1–2 andthe 2–3 solenoid wires and cross them.

"/�� 12��3��4

��


"��

�+�-��&1�1�

�� If the computer is commanding 1–2–3–4 but the transmission is shifting 1–3–4, it canonly be caused by a leak at the large diameter seal on the 1–2 accumulator. Since 3rd

clutch apply pressure feeds between the two sealing rings on the 1–2 accumulator, servoapply pressure can only leak past the large diameter ring if 3rd gear apply pressure isn’tthere to stop the leak. Once in 3rd gear, servo apply pressure doesn’t leak.

7�� .��-��Verify the computer is commanding 2nd gear and that 4th gear is good. If so, the problemis a cut 1–2 accumulator seal.

�+��"�81) Inspect the bore for wear

2) Scotchbrite the bore

/��0��/��*��$��%

��"��


3) Install a good quality rubber O-ring

/��0��/��*��$��%

�+�-��&1�1��$��%

� � � � � When using rubber O-rings, make sure the O-ringdoesn’t get cut by the square feed passage duringassembly.

��


"��

�12��1��"��

�� Although a 4–2 flare is much more common, 4–2, 4–3 and 3–2 flares can all be causedby a worn forward sprag.

7�� .��-��To verify whether the sprag is at fault, disconnect the 3–4 shift solenoid electrical con-nector. When the 3–4 solenoid is off, the coast clutch stays applied in 1st, 2nd and 3rd

gears, which helps support the forward sprag. Although this won’t allow 4th gear, drivethe vehicle making several 3–2 kickdowns. If the flare has improved, the sprag is bad.

� � � � � While inspecting the forward sprag assembly, keep in mind a little wear goesa long way. Be extremely picky.

/��0��/��*��$��%

.��&��5 &+��6&��((�5&��

��"�"


/��0�� ;� �� Almost every trouble code in every vehicle causes the system to switch to an alternateoperating strategy of one form or another. Many of these can be used to your advantage,as long as you can recognize them. Here’s a list of common strategies and fixes for themost common transmission-related codes.

��6Definition: VSS (output speed sensor)

Sensor Location: Either on the differential, driven directly from the speedometer gears,or inside the speedometer in the dash.

How the Code Sets: The computer must see all of these conditions to set a code 06.

1) Receive a Manual D, S, or L signal from the inhibitor switch.

2) Receive a signal from the input speed sensor (pulse generator).

3) Receive no signal from the output speed sensor.

Common Causes: Melted speedometer gears, bad wiring, blown dash fuse (in someapplications).

Causes of False Codes:

1) An inhibitor switch that signals Manual D, S or L while the transmission is actuallyin neutral or park. Forward clutch slip can easily set this code.

2) When retrieving codes, 6 short flashes represent a code 06 while 6 slightly longerflashes represent a code 60. Most code 06 problems are actually a misread code 60.(This is when knowing computer code strategies can really help you).

Computer Strategy: When a code 06 sets, the only action the computer takes is to flashthe Hold or Manual light.

��


"��

��&2Definition: TPS (Throttle Position Sensor)

Sensor Location: Mounted on the side of the throttle body.

How the Code Sets: The computer must see one of these conditions to set a code 12.

1) Less than about 0.02 volts

2) More than about 4.8 volts

Common Causes: Bad TPS, no 5.0V reference signal from the computer to the TPS, badground wire, and bad connections.

Causes of False Codes: None

Computer Strategy: Very late upshifts.

/��0�� <�$��%

��"��


��55Definition: Pulse generator (input speed sensor)

Sensor Location: On top of the case, directly in line with the forward drum.

How the Code Sets: The computer must see both of these conditions to set a code 55.

1) Signal from the output sensor.

2) No signal from the pulse generator; the minimum AC threshold voltage for the inputspeed sensor is about 0.5 volts AC.

Common Causes: Bad pulse generator, poor connections.

Causes of False Codes: Other than coasting downhill with key on, engine off, there areno false codes possible.

/��0�� <�$��%

��


"��

��6�;�6&;�62��6�Definition:

Code 60: 1–2 shift solenoid open or shorted circuit.Code 61: 2–3 shift solenoid open or shorted circuit.Code 62: 3–4 shift solenoid open or shorted circuit.Code 63: Lockup solenoid open or shorted circuit.

Solenoid Location: On the valve body.

How the Code Sets: The computer must see one of these conditions to set a code 60,61, 62 or 63:

1) Less than 0.20 amps through the solenoid circuit.

2) More than 1.5 amps through the solenoid circuit.

Common Causes: Pinched wires inside the transmission, bad section in harness wiring(especially in the area around the battery box), bad connection at the solenoids or atthe computer.

Causes of False Codes: None

Computer Strategy: The computer simply stops using the solenoid that set the code.Solenoids without codes are used normally. This presents some very odd shift com-plaints as shown below.

Code 60: 1st and 3rd gear only.

Code 61: Shifts 3–4–3–4.

Code 62: No 4th gear, 2–3 flare in some cases.

Code 63: No lockup.

The Fix: Only a few things cause shift solenoid codes.

1) Bad solenoid… very uncommon.

2) Bad computer… very uncommon.

3) Bad wiring or connections… almost every time.

The easy fix: run four brand new wires from the computer connector all the way tothe shift solenoid connector.

/��0�� <�$��%

��"��


/��0��(��*��;��+��(��#

�+��1) Stay as far away from internal parts interchange as you can.

2) Always use good quality machined Teflon rings.

3) If you custom cut rings to reduce leaks, use a butt cut instead of a scarf cut.

�+��.��4��!1) Remove all valves.

2) Flat sand each casting.

3) Thoroughly clean all components before reassembly.

4) Enlarge the TV balance to 0.055".

5) Enlarge the TV feed orifice to 0.075".

6) Enlarge the 3–4 clutch apply orifice to 0.041".

7) Scotchbrite the accumulator bores and install rubber rings.

8) Be careful assembling the accumulators; make sure the square feed passages don’tcut the rubber rings.

9) Leave out the checkballs under the thin cover on all applications.

10) Check the TV cam for wear. If it’s worn, don’t worry; a new one is available fromMazda P/N FU01-21-369A.

11) With the exception of the accumulator retaining plates, leave all of the valve bodygaskets out. Tests have shown that leaving these gaskets out allows the valve bodyto maintain its torque, and the valve body won’t warp as easily at high tempera-tures.

12) Torque the valve body bolts to 70–80 inch-pounds.

��


"��

/��0��+��9��/��When selected, the Hold Mode tells the computer to provide 2nd gear starts and shift into3rd at a very low speed. In this mode you will also have no kickdown and, if driven longenough, the transmission will likely die. The only usable function is to cancel overdrive.Unfortunately many customers like to use this feature, not realizing what they’re doingto the transmission.

During some extensive research, we wanted to see what strategies the fluid temperatureswitch would cause. It took 2 hours of driving to realize what this switch would do whenshorted. It turned the wicked Hold Mode into a sweet OD cancel switch. Whether or notit’s a secret input created by Mazda engineers (because they hate the Hold Mode too)…who cares? We’ve tested it on enough models to feel confident that it’s not a fluke.

If you want to turn the Hold Mode into an OD cancel switch, disconnect the fluid tem-perature switch and short the two wires from the harness connector together.

��.��

��5��&��'�&(%

��"��


/��(�0��,�)��(��(�&0(�� It’s not uncommon for the solenoid connectors on these units to get brittle and break.Mazda offers replacement connectors to address this. The Mazda part numbers for theseconnectors are:

BV36-19-010 ................Brown ConnectorBV36-19-01X ...............Gray Connector

� �(& ��2� '

� �(& ��0��7

� �(& ��22� '

��


"��

/��(�0��)�#��:��The original design output speed sensor had a molded resin body that would crack overtime. This problem usually sets a code 6. The new design sensor has a steel body, and ismuch more durable. The Mazda part number for the new sensor is:

BV72-21-5H2 (2WD)

BV74-21-5H2 (4WD)

��'1�&��++��&��

� 8��4

��1��+�� 8��4

��""�


/��(�0��,�)��(��(�&0;��0)��Some Mazda R4AELs and Nissan RE4R01As and 3As use a turbine shaft sensor. Thissensor accesses the turbine shaft through a hole in the stator support.

The turbine shaft used on these models have splines machined into the shaft, whichallows the sensor to measure its RPM. If you need to replace the turbine shaft makesure the replacement shaft has the machined splines. The splines are not very large,and are easy to miss.

Using a non-machined turbine shaft in an application that uses the sensor will set a “noturbine shaft signal” code.

��


""�

/��(�0��,�)��(��(�&0;��0)�� +��Some Mazda R4AELs and Nissan RE4R01As and 3As use a turbine shaft sensor. Thissensor accesses the turbine shaft through a hole in the stator support. Using a statorsupport with an access hole on models that don’t use the sensor will prevent the con-verter from developing the necessary converter charge.

��

��"""


/��=22>��21��"��

During the 2–3 shift, the B1 band releases while theK1 clutch engages. If working (mainline) and modu-

lator pressures are normal, and the band adjustmentis correct (0.3 mm) you can fix a 2–3 flare by removing

one of the B1 release springs. Both springs are similar intension, so it doesn’t matter which one you remove.

An effective measure against 2–3 flareproblems is to use the self-adjustingservo. The part number for the self-adjusting B1 servo is 124 270 12 32.

If you upgrade to the self-adjustingservo you must also use the later-model servo guide. The part numberfor the later-style guide is 140 277 08 40.

9��7��&7��8 ��0��

��!:��;�+&��8 ��0��

��


""�

/��=22>��&�9��+�&12;�21��1��+�-��Mercedes has released new accumulator assemblies for the 722.403 (190D 2.2L) and414 (190D 2.5L) units to combat harsh shift complaints. You can resolve these com-plaints by using these parts:

Harsh 1–2 Shift ........................... 124 270 00 35Harsh 2–3 Shift ........................... 126 270 03 35Harsh 3–4 Shift (722.403) ............ 126 270 04 35Harsh 3–4 Shift (722.414) ............ 126 270 05 35

)"��((��& �1�/��%�!&4

)��((��& �1"/��%�!&4

2��((��& �1�/"��%�!&4

��""�


/��=22>��:��42��Transmissions remanufactured by Mercedes may have an oversized B2 servo bore.Mercedes addresses worn B2 servo bores by machining them larger. If you get one ofthese units, you’ll need to use the oversized sealing ring. Machined cases will have IDnumbers (89 for 722.3, and 76 for 722.4) stamped in the bottom of the servo bore.

The Mercedes part numbers for the oversized rings are:

140 277 01 55 (722.3)

202 277 00 55 (722.4)

��


""�

/��=22>�;��5��-30�?��4&��Starting in 1991, Mercedes used a self-adjusting B1 servo. You may use this servo inany 722.3 – 5 transmission. But if the unit you’re working on already has a self-adjust-ing servo, you may run into problems reusing the servo. Resetting the self-adjustingservo isn’t so easy.

��!• Fully extend the servo: The best way is to put the servo in the case — without the

servo guide and spring — and blast the servo with air. This will move the servo tothe fully-applied position, and allow you to release the cover. You don’t need tosecure the servo with the snap ring to extend the servo; the servo extends just finewithout the snap ring.

��

��""�


• Once the servo extends, remove the assembly from the case. Now you should beable to separate the cover and pin from the servo.

/��=22>�;��5��-30�?��4&��$��%��!�$��%

• Use narrow snap ring pliers to open the pin-lock mechanism in the servo cover.

��


""�

• Remove the snap ring that secures the diaphragm spring.

/��=22>�;��5��-30�?��4&��$��%��!�$��%

��(��!• Drop the servo pin into the servo, with the rounded end facing up.

�

��""�


• Place the diaphragm spring over the pin.

/��=22>�;��5��-30�?��4&��$��%��(��!�$��%

• Use a ½" deep well socket to push the diaphragm spring onto the pin, and downto the seat.

�

��


""�

• Install the snap ring into the servo, securing the diaphragm spring.

• Snap the cover onto the pin.

• Assemble the servo and cover assembly into the case.

/��=22>�;��5��-30�?��4&��$��%��(��!�$��%

�

��"��


/��=22>��1��12�� #Mercedes uses a complex system for reapplying the B2 band for a 4–3 kickdown, andthe B1 band for a 3–2 kickdown. An often-overlooked aspect of this system is the reac-tion valves in the case.

The reaction valves control servo apply. If the reaction valves leak, the bands won’t applyfully.

Testing the reaction valves is easy:

• Remove the valves from the case.

2��(&� <��8�

2"��(&� <��8�

��


"��

• Blow compressed air into the feed holes; air should blow out the exhaust port.

• Compress each reaction valve in a vise. Never overtighten the vise; is doesn’t re-quire a lot of force to close the valves.

• Remove the pin retainer.

/��=22>��1��12�� #�$��%

��"�"


• With the reaction valve compressed, blow compressed air into the feed hole.

Air shouldn’t leak out of the exhaust port. If the reaction valve leaks, replace it.

/��=22>��1��12�� #�$��%

��


"��

/��+�"�02;��;�@�0�;�"�0��5)��+��,��+�-��Starting in 1991, Mitsubishi incorporated several temperature-related modified shiftstrategies. Depending on the model and ATF temperature, the modified strategies rangefrom late shifts, to no 3rd and 4th gear, or simply no 4th gear.

If you experience any of these conditions, check the temperature sensor reading with a scantool and see if it’s accurate, and if it falls within the specifications detailed in this text.

� � � � � � � � � � � � The temperature sensor reading should be at room temperature first thing inthe morning, after the car has sat for several hours.

0--��.�+��Models equipped with F4A automatic transaxle:

1991-98 3000GT1992-98 Diamante1994-98 Galant1995-98 Eclipse1992-96 Expo and Expo LRV1993-98 Mirage

The following chart shows which models have which strategy. Also notice that somemodels have what’s called “fuzzy logic” and “Invecs II.” These are self-tailoring strategiesthat vary shift timing based on the driver’s driving habits.

elciheV enignE

:taedoMtfihS

erutarepmeTlamroN erutarepmeThgiH erutarepmeTwoL

TG000389–1991 L0.3 ONOCE/rewoP IedoMpmeThgiH II&IsedoMpmeTwoL

etnamaiD69–2991 L0.3 ONOCE/rewoP IedoMpmeThgiH II&IsedoMpmeTwoL

etnamaiD89–7991 L5.3 II-SCEVNI IIedoMpmeThgiH IIedoMpmeTwoL

espilcE89–5991 obruTL0.2 cigoLyzzuF IedoMpmeThgiH II&IsedoMpmeTwoL

espilcE8991 CHOSL4.2 cigoLyzzuF IedoMpmeThgiH II&IsedoMpmeTwoL

redypSespilcE89–6991 obruTL0.2 cigoLyzzuF IedoMpmeThgiH II&IsedoMpmeTwoL

redypSespilcE79–6991 CHOSL4.2 cigoLyzzuF IedoMpmeThgiH IedoMpmeTwoL

VRLopxE/opxE69–2991 L4.2,L8.1 dradnatS IedoMpmeThgiH II&IsedoMpmeTwoL

tnalaG89–4991 L4.2 cigoLyzzuF IedoMpmeThgiH II&IsedoMpmeTwoL

egariM69–3991 L8.1 dradnatS IIedoMpmeThgiH II&IsedoMpmeTwoL

egariM89–7991 L5.1,L8.1 II-SCEVNI elbacilppAtoN elbacilppAtoN

��"��


*:@�(A� :):�/��The POWER or ECONO light on the instrument panel will come on when using theseshift modes. ECONO (meaning ECONOmy) is used for most driving situations and pro-vides the best fuel mileage. Pressing the A/T MODE button on the shift indicator putsthe transaxle in POWER mode. When in the POWER mode, the transaxle generally staysin lower gears longer before upshifting, and also downshifts more quickly during decel-eration.

"��!��/��On F4A transaxles that use fuzzy logic, the transmission control module (TCM) modifiesshift timing for smooth shifting when traveling up and down hills. Artificial intelligenceis built into the TCM to automatically simulate the natural shift patterns that a driverwould use with a manual transmission. The TCM analyzes inputs from the brake switchsensor, throttle position sensor, and vehicle speed sensor. In addition to the normal A/Tshift pattern, “fuzzy” shift controls have multiple shift patterns to suit five differentdriving conditions:

1. level roads (same as standard A/T mode)

2. uphill at low speeds

3. uphill at high speeds

4. downhill in 3rd gear

5. downhill in 2nd gear

When driving uphill in fuzzy logic shift mode, the transaxle will stay in a lower gearrange to maintain power instead of hunting between gears. The TCM takes frequentsamples of vehicle speed and throttle opening. It compares each sample with a previoussample to know when to prevent upshifts that would detract from climbing performance.

When driving downhill, the TCM downshifts to provide engine braking without the driverrepeatedly pressing the brake pedal. This automatic downshifting is sometimes experi-enced even when driving on short, gradual slopes or flat roads.

The transaxle won’t enter fuzzy logic shift mode if the selector lever is not in “D” or if theATF reaches temperatures specified for High Temperature Mode or Low TemperatureMode. Also, the fuzzy logic mode isn’t used if the TCM judges the throttle position sensorto be operating improperly, or if the transaxle enters the failsafe mode.

/��+��"�02;��;�@�0�;�"�0��5)��+��,��+�-��$��%

��


"��

�).� �3��$��.�+�� !��%�/��Used only on 1997 and later Mirage and Diamante models, INVECS-II shift control oper-ates similar to fuzzy logic. It has an added feature of learning the driver’s shift patternpreferences.

Shift control patterns are modified to favor the driver’s style of driving. For drivers whoprefer a vigorous style of driving, the transaxle stays in a lower gear until the enginespeed becomes relatively high, to provide a sporty feel. For drivers who prefer a moreleisurely driving style, upshifting occurs at lower engine speeds. For drivers who fre-quently press the brake pedal when driving downhill, the transaxle downshifts earlierthan normal.

If the battery is disconnected, INVECS-II must relearn the driver’s shift pattern preference.

For diagnosis, INVECS-II can be disabled with a scan tool, provided the scan tool hasthe proper software.

This upshift pattern chart demonstrates shifting differences with fuzzy logic andINVECS-II. The heavy lines indicate normal A/T up shift patterns. The shaded areasindicate the range of speeds at which “fuzzy logic” and INVECS-II controls shifting.

/��+��"�02;��;�@�0�;�"�0��5)��+��,��+�-��$��%

� ��

� � & % � � � / � � 0 �� 1 � 2 3 � � 4( � 5 � 6 � ' 2

� � � � � � � � � � � � � � � � � � � ! � � �

� � � 7

� 7

� 7

��&��

��

� �'

��'

��

� � � � � � � ) � � 0 � � ) � � � � � * � � 8 .

��"��


9��+��/��To avoid ATF overheating, the transaxle enters this shift mode when the ATF tempera-ture reaches above 125° C (257° F) at vehicle speeds below 70 KPH (44 MPH). Underthese conditions, the transaxle will shift between 1st and 2nd gears only, not into 3rd gear.When the ATF temperature decreases to 110° C (230° F) or below, the transaxle returnsto normal shift mode.

� � � � � On Eclipse models, the ATF temperature for entering High Temperature ModeI is 115° C (239° F). The transaxle returns to normal shift mode when theATF temperature is 105° C (221° F) or below.

9��+��/��To avoid ATF overheating, the transaxle enters this shift mode when the ATF tempera-ture reaches above 125° C (257° F) at vehicle speeds below 65 KPH (40 MPH). Underthese conditions, the transaxle will only shift between 2nd and 3rd; not into 4th gear.When the ATF temperature decreases to 110° C (230° F) or below, the transmissionreturns to normal shift mode.

��#��/��When the ATF temperature reaches 20° C (68° F) and below, upshifts are delayed tocompensate for the reduced power that normally accompanies lower operating tempera-tures. The transmission returns to normal shift mode when the ATF temperaturereaches above 20° C (68° F).

��#��/��When the ATF or engine coolant is cold, upshifts between 2nd and 3rd and between 3rd

and 4th are delayed to help the catalytic converter warm up more quickly. This conditionlasts only one or two minutes after engine startup.

/��+��"�02;��;�@�0�;�"�0��5)��+��,��+�-��$��%

��


"��

/��+��"�02;��;��@�0�9��+�&12��+�-�A harsh 1–2 shift is often caused by a shorted servo switch. One of the more commonreasons for this is having the wrong E-clip on the servo pin. The proper clip is very nar-row. Using a standard clip for a shaft the size of the servo pin will short the switch. Thecomputer uses the servo switch to initiate the shift-feel strategy of the pressure controlsolenoid.

The E-clip isn’t available separately; if yours is missing, you’ll have to purchase a newservo.

��

��"��


/��+��/��)��"��#��;��"��-�"��#��A KM-series unit with no forward ranges, or falls out of forward, can be caused by astuck N-D control valve, or an N-D control valve sleeve that is installed backward. Thereare other problems that can cause this condition, such as a bad sprag or rear clutchleaks. Here’s a simple test to help find the problem:

1. See if the trans goes in gear in manual low. If it still doesn’t move forward, suspect arear clutch leak, such as sealing rings or a missing checkball in the pump body.

2. If the vehicle moves, shift to manual second.

• If the trans shifts to second, suspect a bad sprag.

• If the trans binds up, the sprag is in backward.

• If the trans shifts to second and then quickly falls out of gear, the N-D controlvalve is either stuck or the sleeve is installed backward.

�:=�� &� ��8�

��


"��

/��+��"�02;��;��@�0�*��"��There are three main factors involved in planet failures on these units:

1. A stuck damper clutch control valve: Overtightening the valve body or filter can causethe damper clutch control valve to stick. Make sure you don’t overtighten the valvebody and filter bolts. The torque specs for these bolts are:

Valve Body Bolts .......... 36 – 48 in-lbsFilter Bolts ................... 48 – 60 in-lbs

=��5��&(%� &� ��<��8�

��"��


2. A clogged turbine shaft: Make sure you blow out the turbine shaft if you suspect atorque converter failure or have an existing planet failure problem. You can removethe plug at the end clutch end of the shaft to make it easier to blow out the shaft.Replacement plugs are available at the dealer.

MF665531 — 10 mm (0.400") 1989–1992MD712132 — 8 mm (0.320") 1992–1998

/��+��"�02;��;��@�0�*��"��$��%

�

��


"��

/��+��"�02;��;��@�0�*��"��$��%3. A clogged radiator: If the planet fails on one of these units, consider replacing the

radiator. These radiators have a habit of clogging with converter clutch material.Unfortunately, a cooler flow test won’t always reveal a clogged filter. All too often, thecooler will flow fine for awhile, and then later restrict cooler flow.

If you must replace the planets, make sure they are the exact same version as those thatcame out of the unit. You can have all sorts of slipping and ratio-error codes as a resultof using the wrong planets.

These charts and illustrations show the details of the different planetaries.

soitaRraeGyratenalP

raeG oitaRediW oitaResolC

1 ts 648.2 155.2

2 dn 185.1 884.1

3 dr 000.1 000.1

4 ht 586.0 586.0

esreveR 671.2 671.2

The information on these pages was provided courtesy of Bob Nuttall of Nuttman, Inc.

��"�"


/��+��"�02;��;��@�0�*��"��$��%

reirraCyratenalP

epyTsnarT oitaR

A B C D E F

llarevOthgieH

lanruoJDO

enilpSthgieH etalPniT stloB

enilpSretemaiD

1-571MK ediW "992.3 "081.1 "019.0 talF desiaR "398.4

5-671MK ediW "053.3 "592.1 "058.0 deppetS desiaR "581.5

0-771MK esolC "053.3 "592.1 *"051.1 talF desseceR "581.5

12A4F ediW "063.3 "592.1 "058.0 deppetS desiaR "581.5

,22A4F,5-571MK23A4W,23A4F

ediW "334.3 "592.1 "259.0 deppetS desiaR "581.5

,32A4F,8-771MK23A4W

esolC "334.3 "592.1 *"531.1 deppetS desseceR "581.5

33A4W&33A4F esolC "926.3 "592.1 **"051.1 talF detiviR "654.5

.potehtdnuoratucrefmahcdetagnolE**.refmahcafodaetsnipotehtdnuoranitucpetS*

��


"��

/��+��"�02;��;��@�0�*��"��$��%

raeGnuSdrawroF

epyTsnarT oitaR

A B C D

llarevOthgieH

lanruoJDO

htooTtnuoC

DIevoorG

1-571MK ediW "291.3 "020.1 62 enoN

5-671MK ediW "510.3 "690.1 62 enoN

0-771MK esolC "002.3 "690.1 92 1

12A4F ediW "510.3 "690.1 62 enoN

22A4F&5-571MK ediW "002.3 "690.1 62 enoN

23A4W,32A4F,8-771MK esolC "591.3 "690.1 92 1

23A4W&23A4F ediW "591.3 "690.1 62 1

33A4W&33A4F esolC "044.3 "690.1 92 enoN

raeGnuSesreveR

1-571MK ediW "782.2 "075.1 43 enoN

5-671MK ediW "072.2 "545.1 43 enoN

0-771MK esolC "762.2 "545.1 43 1

12A4F ediW "072.2 "545.1 43 enoN

22A4F&5-571MK ediW "762.2 "545.1 43 2

23A4W,32A4F,8-771MK 43esolC "762.2 "545.1 43 1

. �'��0��

��8��+��0��

��"��


/��+��"�02;��;��@�0�(��*��4��+��"��There are several possible causes for repeat pump bushing failure. Some of the thingsyou should check are:

1. Proper case ground. The best way to check for a proper ground is to measure thecase for voltage.

• Start the engine and turn a bunch of accessories on.

• Connect the negative lead of your voltmeter to the negative battery post.

• Connect the positive lead of your voltmeter to the transmission case.

If you have more than 0.01 volts at the case, add a ground strap.

You may also have a poor connection at the negative post of the battery. Make surethe connection is clean and tight.

2. A missing torque converter pilot bushing: The crankshaft has a bushing that worksas a pilot for the torque converter. It’s not uncommon for this bushing to seize to theconverter pilot; when you remove the trans, the pilot comes out with the torque con-verter. Then, when you install the replacement converter, the crankshaft has no pilotbushing, so the converter mounts off-center. Always make sure the converter yousend out for rebuild doesn’t have a pilot bushing stuck to it. And more importantly,make sure the bushing is in the crankshaft when you install the transmission.

��&+��+%��%�+�� !��!!��&5�� &+��8��

��


"��

3. If all else fails, you can remove the O-ring from the inner pump gear. This will allowfor more pump bushing lubrication.

/��+��"�02;��;��@�0�(��*��4��+��"��$��%

�

��"��


)��(��"��0)��+��A complaint of no 3rd or 4th gear is often caused by a worn outer piston seal in the highclutch drum. This can occur just weeks after rebuilding the unit. The problem could becaused by the type of seal you use, or the inside surface of the drum may be too roughfor the seal. Nissan has a replacement drum with a smoother seal surface. The Nissanpart number for the drum is 31412-80X07.

If you still have this problem after replacing the drum, consider using factory seals. Thepart number for the seals are:

Inner Seal ...........31527-80X01Outer Seal ...........31527-80X02

�%�(*�&%�+��+! �� %�++

��


"��

)��(��"��0(��(��*��:��The reverse drum snap ring can pop out of its groove, causing the unit to lose reverse.Nissan has a replacement drum with a design change to address this problem. Whenyou replace the drum, replace the snap ring, too. The replacement drum uses twodished cushion plates.

The Nissan part numbers are:

Drum .................................... 31511-80X03Snap Ring ............................. 31506-80X12Dished Cushion Plate ............ 31535-80X07

��"��


��0��+��A misassembled lockup control valve will apply the converter clutch in every gear. Thiswill kill the engine when you put the transmission in gear.

Some service manuals show the lockup valve arrangement incorrectly. Make sure youassemble the lockup valve with the spring installed into the valve body first, then thevalves.

> (*�5�� &� ��<��8��

��


"��

��+!��<��8��2 �7

� ��

��3��0@� +��A4��#;��+��A problem in the transfer clutch on these units can cause thevehicle to chatter or buck while turning a corner slowly whilebraking.

The computer inputs that can affecttransfer clutch operation are:

• Vehicle Speed Sensor (VSS)• Throttle Position Sensor

(TPS)• Inhibitor Switch• Manual Switch

The transfer clutch is controlled byduty solenoid C. When the solenoidreceives a 95% duty cycle signal, thetransfer clutch is fully released; at25% the transfer clutch is fully ap-plied. During failsafe or power loss tothe system, the solenoid will put theunit into full time all-wheel drive.

Here’s how to check the transfer clutch systemon these vehicles:

Step 1: Check for diagnostic trouble codes (DTCs). The procedure for checking troublecodes varies from year to year, and model to model. Always check your shopmanual for the proper procedure for the vehicle you’re working on.

Step 2: Disable the transfer clutch.Insert a 15-amp, blade-typefuse in the FWD switch, asshown in the graphic.

.3=�� (& �

=� 55��+�+& �

��"��


Now check the vehicle: If the chatter is gone, you know there’s a problem in the AWDcontrol system (electrical or hydraulic). If the problem is still there with the fuse inplace, the problem is probably internal; the transfer clutch is probably seized.

Here’s how to check the system operation:

• Connect a pressure gauge to the transfer clutch tap.

��3��0@� +��A4��#;��+��$��%

• Connect the positive lead of your digital voltmeter to transmis-sion harness terminal 11.

��+!��&(%

� � � �

� � � �

� � � � �

� � � � � � � �

��


"��

• Connect the negative lead of your digital meter to a good ground.

• Set your meter to DC Volts.

��3��0@� +��A4��#;��+��$��%

egnaRrotceleS

edoMDWA edoMDWF

)MPR008–006(eldI llatS llatS

esreveR ISP11–7 ISP411–401 ISP0

evirD ISP11–7 ISP411–401 ISP0

rotareleccA desaeleR TOW TOW

Check 1: Engine idling and the fuse inserted in the FWD switch.

You should see 8–14 volts at duty solenoid C, and have zero PSI to the trans-fer clutch circuit.

Check 2: Engine idling and the fuse removed from the FWD switch.

Transfer clutch voltage and pressure should vary with throttle opening: Pres-sure should be low at idle, and increase as you open the throttle. Duty sole-noid C voltage should be high at idle, and drop as the throttle opens. Com-pare your readings to the chart.

• If the voltage doesn’t respond properly to throttle opening, check the circuit be-tween the solenoid and the computer.

• If the voltage responds properly to throttle opening but pressure doesn’t, look for abad solenoid or a sticking transfer clutch valve.

• If the voltage and pressure respond properly to throttle opening, look for a dam-aged transfer clutch assembly; the transfer clutches are probably seized.

� � � � � � � � � � � � A condition that feels somewhat similar can be caused by a variation in tiresize, such as one new tire with three old ones. Always check the tire sizesand condition before beginning a major repair to correct a chatter in turns.

=�&7�� 1��+��++� 4

.3=�'�&(%

0� ��

�

0�

��?

(�

("0> 0> 0>

.�+�2

��

of ATRA. No part of this program should be construed as recommending any procedure which is contrary...

Documents

Transcript of of ATRA. No part of this program should be construed as recommending any procedure which is contrary...