98 aw4 tc lockup switch

[kastein]

My thoughts exactly. Or because the CEL only comes on if there is more than one error condition. I notice their unit doesn't allow control of the TC, either. It would be interesting to know if using their device sets a code internally, but doesn't light the CEL.



That's above my pay grade, but would almost certainly work if it could be done.

Taking a different approach, I wonder if it wouldn't be possible to simply take over the TCM's inputs rather than it's outputs. There are only three three of them. The two shaft speed inputs appear to be little more than 5V pulses whose frequencies directly correlate to shaft speeds. The TPS is a basic voltage divider with one resistor being variable, thus providing a voltage proportional to throttle position. So (theoretically) one could measure the appropriate pulse frequencies and TPS voltage for each forward gear. Programming a micro-controller to produce those signals on demand shouldn't be all that difficult. Smoothly switching between the real and the false inputs without upsetting the proverbial apple cart might be tricky though...

As long as the TCM is not comparing notes with the ECU regarding the values of those 3 inputs (which are also visible to the ECU), we should be able to lie to the TCM and let it do the shifting for us, no?

Admittedly, that's idle speculation on my part. I don't know that I'd go to that much trouble at this point. What I have now works more than well enough to suit my needs -- besides it's already installed and ready to go. Now that I've finally broken down and purchased a code scanner, I don't worry so much about an occasional CEL. And once the novelty wears off, I suspect I'll only use the manual shifter when I really want it for holding 1st or 2nd gear on the trail -- in which case I'll probably never see another CEL. Time will tell...

You could do it this way, but it would be fairly complex. Hacking / reverse engineering is one of my hobbies, as soon as I find a way to read the code out of the microcontroller, it should not take too long. Did an entire PC BIOS in a weekend once while disassembling by hand, and this microcontroller (Fujitsu MB89665) only has 16kB of code storage and a much simpler instruction set (136 instructions, 1-byte opcodes plus 1-2 bytes of operand) than the x86. But... so far I have been unable to find a company selling the adapter that will allow me to plug the microcontroller into my EPROM burner/reader.

By the way kastein, what value resistors are you using in your testing now?

332 ohm, 2 watt right now. I'm going to try some 100 ohm and 1000 ohm sometime soon.

EDIT: got a form generated email from the company I asked about the MCU programming adapter, they're in Japan and the characters/font did not translate well, almost deleted it thinking it was spam:

Subject: $@$*Ld9g$;$r<u$1IU$1$^$7$?(B : InquiryInquiry form to Sunhayato$B$K$D$$$F$N$*Ld9g$;!"$"$j$,$H$&$4$6$$$^$7$?!#(B

$BC4Ev<T$K%a!<%k$rAw?.$$$?$7$^$7$?!#(B

Thank you for contacting us.

$B%5%s%O%d%H3t<02q<R(B
http://www.sunhayato.co.jp/

aperwork:looney:

 

[raneil]

332 ohm, 2 watt right now. I'm going to try some 100 ohm and 1000 ohm sometime soon.

I may try the 100 ohmers this weekend, if I can get to it. Will advise.

EDIT: got a form generated email from the company I asked about the MCU programming adapter, they're in Japan and the characters/font did not translate well, almost deleted it thinking it was spam:

aperwork:looney:

That looks familiar. I don't think it's from Japan. I bet it's from your TCM. It's mocking you like mine is mocking me. :scared:

You've been warned.

 

[R.DesJardin]

How about just for test puroposes connect another set of trans solenoids to the system? :dunno:

 

[kastein]

Could do that, but I'm really starting to think that the dummy load resistors can be any value (from my reading of the spec sheet) but that they will only keep codes P0750/0753 (and analogous solenoid B / C codes) at bay.

If I can find my bin of real high power resistors somewhere or other (a 14 ohm resistor, within the 11-15 ohm spec'd range for the solenoids, will dissipate around 14 watts at normal auto DC power) I will try those, but I don't think the outcome will be any different than the 330 ohm resistors I'm trying or the 1k resistors raneil is using.

 

[raneil]

Hopefully it's not inappropriate to dredge this thread up again. But I thought a follow-up post would be in order now that I've spent some quality time on the trail with the new setup.

Overall, I was quite pleased with the results. The shifter performed perfectly, though the CEL came on twice -- both times while descending relatively steep grades in 1st or 2nd gear with the tach approaching 3,000 RPM or so. It seems the computer is indeed comparing input and output shaft speeds, but it allows a fair bit of variation from the expected values before throwing a code and lighting the CEL. Keeping the RPMs down (below say 2,500 RPM?) kept the light dark. Very easy (for me) to live with.

Each time the CEL came on, I quickly reset it. Considering the amount of time the manual shifter was in use during the week, two resets did not seem at all excessive. I can't say it's perfect since my goal was to keep the CEL completely dark at all times. However, it works well enough that I doubt I'll spend much more time doinking with it -- at least for now.

Here we are atop American Flag Mountain (12,713 feet). Good stuff! atriot:

 

[tbburg]

Interesting work. I have to admit, after the first page I was thinking, "Damn! I paid 150.00 for a pair of resistors?" Now I feel a little better, but also feel bad about that, because I don't like to think of myself gloating over other peoples failure. Sounds like you guys(separately, and/or between the 2 of you)have got it about 1/2 figured out.

Couple questions:
Are you both using a TC lockup control in conjunction with the gear control? If so, could it be the TC lockup at some pre-set RPM that is throwing the code?

When the ECU throws a trans. code and sets off the cell, does the transmission still work normally in the auto mode, or does it go into limp home mode?

 

[raneil]

Interesting work. I have to admit, after the first page I was thinking, "Damn! I paid 150.00 for a pair of resistors?"

Well, there's a bit more to it than that...

Now I feel a little better, but also feel bad about that, because I don't like to think of myself gloating over other peoples failure. Sounds like you guys(separately, and/or between the 2 of you)have got it about 1/2 figured out.

Couple questions:
Are you both using a TC lockup control in conjunction with the gear control?

I have the TC lockup wired to a separate switch. It can be used in either manual or auto mode with no ill effects. Frankly, I never used it during the week on the trail. I have found that leaving the TC under the control of the TCU even while shifting the tranny manually works great.

If so, could it be the TC lockup at some pre-set RPM that is throwing the code?

During my testing (with the TC always under computer control), the CEL came on only under compression braking that reved the engine significantly. I can't say for sure where the threshold is, but it seems to be around 2,800 RPM or so -- and that's a rough estimate, at best. There may well be other circumstances that light the CEL, but I haven't encountered them yet. Once I figured this out, and made an effort to keep the RPMs to a more reasonable level during steep descents, I was 100% CEL-free.

When the ECU throws a trans. code and sets off the cell, does the transmission still work normally in the auto mode, or does it go into limp home mode?

It still works fine, both in manual and auto mode. The CEL being on has no impact on the Jeep's behavior/performance.

 

[tbburg]

thanks for the reply. I've read the trans acts pretty crappy when it's in limp home mode(full hydraulic control)and picked up a rumor somewhere that if the trans computer has solenoid failure codes it shifts into limp home. Glad to hear that's wrong.

 

[kastein]

Depends on the vehicle. My parents old chrycrap minivans went into limp mode. My 96 did also, but only when it hiit the rev limiter doing 80 in the left lane due to a severe lack of clutch function. I don't think the test rig has done so but I'm not sure, at some point I'll remember to wire in the indicator lamps I got so I can tell what the TCU is asking for while manual shifting.

 

[kastein]

Update -

I still have not gotten it to stop setting off the CEL while tooling around town at 20-30mph due to the 1/2 to 3/OD shift failure detection (at least, that's what it seems like to me.) I have some ideas as to how to do this though, not sure if they'll work.

Also, I was hanging out with a couple of the other local guys last night and one of them had a 97 XJ, so we unplugged my testing harness from my 98 and plugged it into his 97 and took it for a spin. I'm really glad I built that harness so it could be unplugged and moved easily between 98-01s and 97s instead of hacking up my engine harness!

It seems to have worked quite nicely in his rig, we didn't get a single CEL and he was romping on it pretty hard in 1st and 2nd, it would have been in 3rd several times if not for the override. I think we tested it enough to prove that a 97 can be fooled with just the dummy load resistors, but I'm not 100% sure yet. So if anyone with a 97 wants to build the circuit raneil posted (which is damn close to what I came up with, I just didn't bother with the relays since I was in a hurry building the testing rig and didn't feel like adding extra complexity, it's just a proof of concept) and test it out daily driving for a few days in a 97, that should settle that question.

Next things I need to try - I'm going to put a 97 TCU in my 98 and see if the 98-up ECU is happy talking to a 97 TCU. If so, all you gotta do to avoid a CEL in a 98-01 is hit the junkyard, buy a 97 TCU, stick it in your 98-01, then build the circuit raneil posted. If not, I'll need to do more research.

EDIT: oh, and my scangauge II has mysteriously stopped properly clearing my CEL. It reads the codes but they're mostly P0000 and it refuses to clear them properly. I think I pinched the cable under something, going to bring the 98 to autozone and have them read+clear the codes today and see if that fixes it.

 

[raneil]

Thanks for the update. Glad to hear you're still making progress.

Next things I need to try - I'm going to put a 97 TCU in my 98 and see if the 98-up ECU is happy talking to a 97 TCU. If so, all you gotta do to avoid a CEL in a 98-01 is hit the junkyard, buy a 97 TCU, stick it in your 98-01, then build the circuit raneil posted. If not, I'll need to do more research.

I'd be very interested to know if this works and if it improves the CEL situation at all.

So the 97 TCU will plug directly to the harness in a 98+?

 

[kastein]

It should - same connector. I believe the pinout is compatible as well, the only thing added externally was a second speed sensor, the ISS. The sensors changed type as well but that doesn't change the way it's wired, it only changes the way the signals are interpreted within the TCU.

I will try and pick up a 97 TCU at the junkyard this weekend and give it a spin. Not sure if there will be a 97 at the yard, or if I'll have time though.

 

[raneil]

I will try and pick up a 97 TCU at the junkyard this weekend and give it a spin. Not sure if there will be a 97 at the yard, or if I'll have time though.

I hear ya. It's always tough when you're looking for one particular year model. They never seem to have the one I want...

Good luck. I look forward to hearing your results.

 

[slimpartywagon]

Whats the update on this. I know it is several months old, but looking at making a shifter set up for my Cherokee... I know mine is pre-obdII but this electrical stuff fascinates me.

 

[O-Gauge Steamer]

Now I have yet another project...

 

[kastein]

I don't have a resistance decade box, never needed one before. I don't believe they are needed as both raneil and I had good luck getting rid of the solenoid codes with 100 or 1000 ohm resistors of the 1/2 to 3 watt variety (1000 ohm you can get away with 1/2 watt, 100 ohm you'd need a 3 watt.) It looks like 15mA or so (15 volts max on the auto power system, at 1000 ohms) fools the SI-5154 load detection circuitry fine, so I left it at that.

I agree with your understanding on PCB layers, but it doesn't appear to be true here. I haven't finished shaving the one I have yet. It could also be more than 4 layers, I'm honestly not sure, I just know that it's at least 4 by how blurry the inner layer silhouettes were when I used a bright light to look through the PCB around the pins on the more important/interesting parts (one of my standard techniques I apply before ripping a board apart.) It's also a fairly complex board, they did a very good job of utilizing real estate efficiently, so I would not be surprised if it has at least one internal signal layer, or at least one internal layer that serves as a signal layer on the area of board with the signal processing and logic circuits and as a power or ground plane over by the solenoid drivers.

I've been pondering the ISS/OSS signal conditioning circuitry recently, nothing significant to report yet though.

 

[O-Gauge Steamer]

As an aside, I was the Drill/Rout/Lamination Process Engineer for Advance Circuits up in Aurora. I left that industry and went into my "other" industry, microelectronics (read Intel chip manufacturing) until I retired in 2008.

 

[raneil]

It's been many months since I've toyed with any of this. But as I recall, there was no difference between using 100-ohm and 1k-ohm resistors. As long as there is a load present, I don't think the resistance value is particularly important. Using the higher values allows for lower-wattage (read: cooler running) resistors.

So far, I've been quite pleased with the results using 1k-ohm, 1/2-watt resistors.

Still, I think the TCU is doing more than just checking for the presence of a load in each solenoid circuit. I'm convinced it's also comparing the relative speeds of the input and output shafts to verify that the gear it has selected has in fact been engaged. It's pretty tolerant and allows for a fairly wide variation before it gets angry and throws a code, though. So it's easy to live with -- at least for me.

 

[kastein]

I'm fairly certain you are correct. I tested this out on a friend's 97 (no input speed sensor, so the TCU can't tell if it's in the right gear) and he thrashed it pretty hard and got no codes, but it throws codes on my 98 that indicate the TCU knows it is not shifting the transmission. It doesn't throw codes relative to failed solenoids anymore though.

I'm trying to figure an easy cheap way to fake the TCU into thinking it is in the gear it asked for, but everything I come up with seems to involve a microcontroller or a few oscillators that have to be set to the appropriate frequencies.

 

[raneil]

Were you ever able to test a '97 TCU in your '98? That sounds like an easy fix if it works (and you can find a '97 TCU!).