This new Gears column, “Thinking it Through,” came from my discussion with ATRA President Dave Wilkes while visiting his shop in Ventura, California. We discussed training methods and article/ seminar topics when he said, “I’m going to use something you said to me years ago.” He referred to an incident about 20 years earlier with an AX4S shift problem. We went for a road test, and somewhere along the discussion, I said, “Okay, Dave, let’s think this through.” I had no idea what the cause of the problem was just yet, but I knew what it wasn’t. We continued thinking it through until we arrived at the most probable cause of the problem.
This is the spirit of this column. We’ll cover diagnostic techniques primarily centered on discovering what isn’t the cause of a problem, thereby limiting the possible causes. This approach is the essence of diagnosis. I wrote about this in a book, “The Art of Diagnosis,” thirty years ago. While the book’s subjects are considered “oldies” by today’s standards, the approach is timeless.
Let’s get started. The customer complaint for this article is a harsh 1-2 shift. It’s a 2007 Chevy Silverado with a 5.3L engine and 4L60E transmission. It’s a good subject for this exercise because everyone is familiar with it, including tech students just entering the industry. The customer says that sometimes the shift is violent. She’d noticed it for about a month, but it hadn’t been such a problem as to rush to get it looked at.
Okay, she dropped off the vehicle; now what? I prefer sitting in the passenger seat while the customer demonstrates the complaint. That’s not always possible; in this case, it wasn’t. That being the case, I’d drive the vehicle as is; I wouldn’t check the fluid or codes (if the check engine light is on). Why? Any changes you make could affect the complaint. Say you clear the codes or top off the fluid if it’s low. You could change the transmission’s behavior and not experience the complaint.
You go on a test drive, and it shifts fine at light throttle. You try it again from a complete stop, but now at half throttle. This time, you feel something happening, and then there is an aggressive shift to 2nd gear. The 2-3 and 3-4 shifts feel fine, just as they did at light throttle. You try it again at full throttle, and the 1-2 shift is even more aggressive. The 2-3 and 3-4 shifts feel as they should for the heavier throttle.
What do we know at this point? Unless there’s some odd pressure spike during the 1-2 shift, we’re pretty sure line rise is working correctly. If you’re unfamiliar with the pressure system, it’d be worth checking it out. Figure 1 shows the 4L60E pressure-control system. No features affect line pressure differently in 2nd than other gears, so there’s nothing to work with. We also know the likelihood that the Orificing check ball for second gear (Figure 2) didn’t wear out and blow through the separator plate. We know this because the shift felt great at light throttle. Had the ball worn through the plate, it’d also have a harsh shift at light throttle.
What else have we learned? We felt “something” happen just before the harsh shift, and that something got worse at heavy throttle. It felt like it was trying to shift but couldn’t; then came the harsh shift.
The shop manager who took the customer’s information asked a few questions and learned that the vehicle hadn’t been worked on other than an engine oil change three months ago. This means that the customer’s complaint happened out of the blue. That’s good to know.
Now, it’s time to check the fluid level and condition. This vehicle uses a dipstick, so it’s straightforward. Have the engine running with the transmission in Park. Remove the dipstick, wipe it clean, reinsert it into the tube, and then remove it to check the level. The fluid was full, and though it was darker than normal, it didn’t smell burnt. For our purpose, we’ll consider it normal.
Let’s stop for a minute and consider other transmissions. While the 4L60E is pretty forgiving regarding fluid level, others are particular. Pay close attention to the factory procedure. Some manufacturer procedures require the transmission to be at a specific temperature. You’ll want to use a scan tool to verify the fluid temperature and then check the level as specified.
In addition, we’ll do a pressure test. We’ll do this for two reasons:
- It’s a good habit to have. To know what bad looks like, you need to know what’s normal.
- We may be facing an odd computer malfunction that spikes the pressure on the 1-2 shift, and we wouldn’t want to assume that the pressure rise is okay.
Of the two reasons, number one is the better one. This vehicle’s computer system is simple and is unlikely to have a glitch during the 1-2 shift, so it’s good to practice checking the pressure behavior of vehicles that work. This way, it’s easy to spot pressure problems when you see them.
You’ll also want to scan for any codes. You may not have any current codes that’ll turn on the check engine light, but you may find historical data that offer a clue to the problem. Another thing is that, just like using the pressure gauge, it’s good practice to check the vehicle with a scanner. The more experience you get with both, the easier it is to spot maladies that otherwise might get you off track.
Let’s get back to the diagnosis. Attach the pressure gauge so you can watch it while driving. I prefer routing the pressure gauge so I can hold it with my left hand and that it’s nowhere near an exhaust pipe, which might burn the hose and cause it to burst.
We fire up the engine and check the pressure at an idle in Park. It’s around 55 psi. In reverse, the pressure rises to about 80 psi. There’s no reason to stall test it in reverse, but if you did, it’d rise to about 300 psi. In Drive, we’re back to around 55 psi. Doing a stall test gets us to about 165. The real test is what it does on the road. At a medium-throttle takeoff, the gauge gets to about 110 psi. At around 25 mph, it dips to about 90 psi. There’s no pressure spike when the harsh shift occurs during the 1-2 shift. Its behavior is like other 4L60Es we’ve tested before. The heavy-throttle test results are also as expected. It’s around 140 psi at takeoff, dips a bit as the car gains speed, and no pressure spike during the harsh 1-2 shift. The pressure test passes as normal.
Referring back to the “Shift- Feel” article, everything either checks out okay or doesn’t apply. That is, everything except item three, the accumulator. In fairness, an experienced diagnostician would recognize this problem on the initial road test. The telltale was the fact that something happened before the harsh shift. That something was the accumulator moving too quickly. The action causes a low-pressure condition at the servo. The transmission tries to shift, but the low-pressure condition doesn’t allow the band to apply enough to stop the drum; it just slides. Then, when the accumulator bottoms out, it delivers full pressure to the servo, causing a harsh shift. This is called a slide bump and is often confused with a harsh shift. On earlier units, like the 4L60 with a TV cable, I’d commonly see techs adjust the cable to lower throttle pressure. This would make it worse because the pressure at the servo is even lower when it tries to shift. That’s another telltale. The problem with this unit was that the 1-2 accumulator spring broke. This allowed the accumulator to move more quickly than otherwise, resulting in pressure being too low at the servo to shift in time.
Okay, now we have a suspect to verify. We’d drop the pan and remove the 1-2 accumulator housing only to discover we were right; the accumulator spring was broken. At this stage, you may find evidence in the pan that suggests further damage, or you might replace the accumulator spring, service the transmission, and be the hero for this customer.
Now, let’s examine this complaint further under different circumstances. What if this unit had just been rebuilt and had the same problem? We’d go through the same series of tests and “Think it Through.” We know the accumulator spring isn’t broken, but the transmission exhibits the same problem. Here, it could be that the rebuilder sanded the reverse drum, which could cause a slide bump. Or maybe they reversed the accumulator assembly order and installed it with the springs in the housing first, then the piston. Here again, the same complaint has a different cause.
What about a different transmission? Perhaps it’s a Honda where we know they’re particular with friction material. We’d go through the same process. We’d verify that pressure and pressure rise were normal (we’ve checked the pressure on units that worked right, so we know what to expect). Honda generally has pressure taps for each gear, so we verify the pressure behavior during the 1-2 shift. We might question a few things at this point:
- Did the rebuilder install the cushion in the 2nd clutch drum?
- Were the friction plates compatible (correct) for this unit?
- Did the installer use Honda oil during the fill process?
We could go through the list of shift-feel factors and either question them or rule them out. While the systems vary from unit to unit, the factors for shift feel do not. Armed with this knowledge it doesn’t matter whether we’re working on a 6L80 with a harsh (or soft) 3-4 shift, a 4R100 with a soft 2-3 shift, or any other unit with a shift-feel problem. The answer to getting it fixed is knowing the systems at play and then… Thinking it Through.
Some systems, like the 41TE and 62TE, keep line pressure high all the time and vary the pressure to each shifting clutch based on throttle and load. The 41TE doesn’t have a mainline pressure port. Instead, it has pressure ports for each clutch. To know the mainline pressure, attach a gauge to a clutch tap, say the 2-4 clutch, and watch its behavior. Once it completes the shift into second, it’ll spike to line pressure. The point is that you must know and understand how a system works in order to test it properly. And sometimes you have to be creative to get what you’re looking for. Testing more on units that work will quickly make you an expert in pressure systems diagnostics.
