SaabCentral Forums banner

1 - 1 of 1 Posts

8 Posts
Discussion Starter #1
Hey everyone, first time posting, but I think I have something very interesting to offer.

I bought my first Saab this weekend. A 2003 9-5 2.3t Linear wagon. In the process of assessing its condition and deciding what maintenance and checks should be done, I noticed that CPS failures are quite common.

Due to my job, I have an interesting proposition. I am the lead speed sensor designer at a company that manufactures a variety of sensors, the majority being cam and crank position sensors (speed sensors).

Modern CPS use hall-effect cells and target mounted to the crankshaft (there are dozens of different configurations), but I'm guessing this sensor uses an internal cell and magnet with a metallic 60-2 type target/tone wheel (58 metal teeth with 2 missing teeth to provide a signal reference for TDC location), or at least this is most common now on CPS with internal targets. The reliability of the sensor's internal components have improved over the years due to improvements in the manufacturing processes and I think this is why the original CPS fails--especially with the constant thermocycles it undergoes. The hall-effect switch is a very tiny component embedded in the very end of the sensor and inside this cell are several tiny (think less than the width of a human hair) traces that connect all of the internals of the cell itself--components like this can be manufactured much more accurately now. The way I hear the CPS failures described, it sounds like the soldering of the harness leads to the internal lead frame are failing, or the soldering of the lead frame to the hall cell, or even the internal connections of the hall cell are failing, leading to intermittent signal and symptoms similar to a DIC failure. If you are really curious, you can start here and dig as deep as you want:

I want to attempt to re-engineer this sensor to be more robust. It will be quite challenging since I'll have to thoroughly dissect the original CPS and then attempt to recreate its signal across all engine speeds and temperatures. I do this work often at my job, but we usually have the benefit of starting from scratch and having the OEMs program the ECU to utilize our sensor's signal. But, the ideal end result if I were successful would be the ability to assemble "the last CPS you'll ever need" and ditch the original failure-prone unit at a cost that may be less or slightly more than the cost of an aftermarket replacement.

I have access to an X-ray machine (to analyze the guts of the original sensor), test benches (to determine the necessary signal from a functioning sensor), a machine shop, and the internals from a huge variety of sensors that we manufacture that could potentially "plug and play" into a custom machined housing (the way we manufacture our prototype sensors).

In order to get started on this project, I'm hoping I can get a few people interested enough to send me some old non-functioning CPS that you may have kept around for some reason--functioning would also be great if you are willing to part with one for a good cause, otherwise I'll order a new one when it comes time to determine the sensor's signal. If I can get several non-functioning CPS, I can do "destructive analysis" like sectioning several parts to try to reveal the source of the failures. I will also need the target, but this will be a much rarer component since it is inside the engine and attached to the crankshaft. If you look on this thread (not sure if there is a better way to point to this picture in my post), third picture on the post by Johnnydel, you will see the crank target on the far left of the picture:

It would be nice to have the target, but it might not be necessary, as long as I can determine a few parameters from it, which I should be able to measure while I'm doing the infamous sump drop and sludge check this week.

So, do I have anyone intrigued, or is everyone content buying factory replacement CPS? Since I'm new to Saabs, it's hard to assess just how common CPS failures are and whether a re-engineered sensor is something that could help the community.

Let me know if you are interested in this project and are able to contribute some spare parts for analysis! I will document my findings and make them available in the public domain here for anyone that may be curious about what makes the CPS tick--or not tick in this case!
1 - 1 of 1 Posts