My career as a diagnostic technician has spanned over 40 years and I can say with certainty that I am still learning. What keeps any good diagnostic tech moving forward is the fact that there is always something new to learn and the quest for knowledge is what separates the average technician from the industry leaders. The best learning experiences I have been involved in is when I venture outside the typical comfort zones of familiar engine problems and into jobs that are not in my so-called wheelhouse.
The featured vehicle in this month’s article is one such example.
Applying drivability tactics to transmission concerns
I am by all admissions not a transmission expert, but I have been involved in repairing many transmission-related problems. As a matter of fact, one of my better customers is a local transmission shop that sends me many vehicles that have transmission problems that turn out to be electrical in nature. This 2002 Acura RSX is one such example.
The transmission shop wants to know if I can test the computer to determine if it is working properly. They have rebuilt the trans, but the original problem remains. This 4-cylinder, 5-speed automatic Acura has a problem of going into neutral at the 4th gear shift. The shifting is normal from 1st to 2nd, and 2nd to 3rd, but when attempting to go into 4th the engine revs way up and no engagement into 4th gear occurs. Sometimes the transmission will shift into 5th gear if the driver keeps on the gas.
When I first took a look at the vehicle, I connected the Honda HDS scan tool and pulled codes which are displayed in figure 1. The factory scan tool has a split screen look with the module data appearing in the left-hand screen and scan tool help or service information displayed on the right side if you have a current Honda/Acura website subscription active. The DTC help displayed seems to point to a mechanical problem inside the transmission. Nothing about the P0780 code seems to mention anything concerning an electrical fault as a possible cause. While I would not call the three possible causes listed as a comprehensive list, it does seem to suggest that the code set criteria is most likely caused by a mechanical issue in the transmission.
Keep in mind that any solenoid is an electro-mechanical device, meaning it can fail either electrically, (open, shorted or high resistance coil), or mechanically, (stuck or frozen armature preventing mechanical movement). The first two items mentioned state mechanical problem with shift solenoid A or Linear solenoid B so the circuits inside the PCM that monitor electrical function are not reporting a problem. Codes P0753, P0758, P0763, P0768 and P0773 are circuit codes for shift solenoids A through E and none of these codes are present. Code P0780 is described in service information as a “mechanical problem in hydraulic control system”, so I think I’m simply going to have to verify electrical circuit integrity and send the car back to the trans shop. They always think there must be a “bad” wire somewhere. Only testing will tell.
Let the game begin!
The Acura is road tested with the scan tool connected and the following recording is captured. The problem is seen in figure 2, when the 4th gear shift occurs, the engine RPM and Main Shaft speed flares way up once 4th gear is commanded, 4th gear has not engaged. While the scan tool can show solenoid commands, it cannot verify proper circuit operation such as voltage levels and solenoid movement that a scope can uncover. Testing will continue with a lab scope.
Because I am not familiar with this transmission’s operation it is necessary to do some research into the operation and electrical control of the transmission, so I can check the operation of the PCM. Service information states the transmission shifts into 4th gear by turning off Shift solenoid C and turning on Shift solenoid A as shown in the shift chart in figure 3. This seems to point to shift solenoid A as the major player in completing a shift into 4th gear.
With this information I decide to connect my Pico 8-channel scope to all the shift solenoids, Linear solenoids A and B, and the last channel to shift solenoid A with a current probe. With the current probe connected I will be able to confirm that shift solenoid A does indeed move by observing if a “pintle hump” is present when current flows though the solenoid.
I will point out here that the shift solenoids for this transmission are feed side switched, not ground side controlled like many domestic applications. The PCM supplies 12 volts through a High Side Driver to turn a shift solenoid On. My scope connections can be seen in figure 4. The linear solenoids control the hydraulic pressure that is supplied to the actual clutch packs through a high rate duty cycle control which can control the apply rate of the clutch pack and hence shift feel. The shift solenoids control hydraulic pressure to the shift valves inside the valve body which in turn control shift timing.
With the scope connected and a slow time-base of 5 seconds per division, the Acura is test driven through the 4 shift points and all 4 shift events are captured on one screen. Once back at the shop, the pattern is analyzed. It is clear the PCM turns off shift solenoid C and then turns on shift solenoid A when 4th gear is commanded and the current probe confirms that there was pintle movement from shift solenoid A. The linear solenoid waveforms have been removed for clearer analysis of the shift solenoids. Figure 5 shows the whole test drive, figure 6 adds labels to each waveform and highlights the shift points, and figure 7 is a close-up view of shift solenoid A voltage and current when commanded on. The current probe clearly shows there is a pintle bump that indicated the shift solenoid did indeed stroke and eliminates the possibility that the shift solenoid is mechanically stuck or jammed.
PCM OK — So what else?
After reviewing the waveform, I concluded the PCM was doing its job and the problem must lie inside the transmission since there are no clutches or bands held applied by any electrical component. The shift solenoids are being operated as designed and I did not see why a different computer would change anything.
The transmission shop picked up the car and went through the trans again but found nothing wrong. They returned the car to me for another look. This time I connected the scope like before but added a Pico pressure transducer to the 4th clutch pressure port on the front of the transmission. I road tested the car on my lift and watched to see if pressure was applied to the 4th clutch circuit. The 4th clutch is applied in both Reverse and 4th gear. To my surprise, the pressure transducer confirmed the 4th clutch received pressure in reverse but not during 4th gear apply as seen in figure 8.
Wondering if something could be blocked in the valve body, I told the transmission shop what I found and if they knew if a valve body restriction was possible. They did not believe this could occur and did not want to take the transmission out again. It was decided to try a different PCM as a last hope. I did not think this was going to cure this issue but I was dead wrong. After installing a used PCM the Acura shifted flawlessly! I could not explain the reason so I connected my scope for a third time to capture a known good waveform. This is when the problem was finally determined. Once the waveforms were compared a subtle difference stood out.
The “aha!” moment
While I was mostly looking at shift solenoid A operation, I did not pay much attention to shift solenoid C or the downward spikes that were present on most of the solenoid turn off points. Once I zoomed in on the solenoid turn off commands an interesting problem was noticed. Just as a ground-controlled solenoid creates an upward spike when commanded off, these High Side Driver solenoids produce a downward spike when turned off due to magnetic induction in the coil. Figure 9 shows a close up of shift solenoid A as it is commanded off. The spike as well as another pintle closing bump can clearly be seen. These spikes could be seen on each solenoid turn-off event except solenoid C.
When I looked at solenoid C the waveform voltage did not return to zero volts but instead hung around 2.1 volts and no downward spike was present. This meant there is leakage across the transistor driver and some current is still flowing!
With 2.1 volts applied to a 14-ohm solenoid there is about 150 milliamps of current flow, (ohm’s law). While I can see this electrical problem and I know that the car is fixed, I still don’t know why this could keep the 4th gear shift from occurring. I will need to go back to service information and determine how the hydraulic system works. Fortunately, my Mitchell 1 repair information system had detailed descriptions of how the transmission hydraulic circuit operated. The current flow through the HSD for shift solenoid C prevents the solenoid from returning to its Off position which keeps fluid pressure applied to shift valve C. The hydraulic circuit can be seen in figure 11.
If hydraulic pressure is not removed from shift valve C it cannot return to its home position and uncover the pressure apply port 5G. Linear solenoid B supplies 56 pressure to shift valve C and is then connected to the 5G port and routed to shift valve B and becomes 4th clutch pressure 40 which is sent to the 4th clutch. Similar to a restriction, when the shift valve C does not stroke back to the right, the 5G port cannot connect to circuit 56 pressure from linear solenoid B. Figure 12 is a closer look at shift valve C and the 4th clutch hydraulic circuit ports, the hydraulic circuit is highlighted in red.
This Acura was a great learning experience for me and has made me pay greater attention to circuits controlled by high side drivers when I scope test any system using them. Keep in mind that many Chrysler engine control systems use HSDs so you may encounter a similar condition somewhere down the road. For now, I feel better knowing I have restored my lack of drive.