Transmission "grinds" in neutral& drive, Engages hard into Reverse

1986nate · #1 03-22-2010, 03:01 AM

Alright, this is the problem I'm having in the FWD laguna. It has done the exact same thing in the 93K original transmission that was in it as well as the roughly 150K known working transmission that had (what turned out to be) a sheered reduction gear and thats why it was removed from the previous car. I took both the ring gear and the pinion shaft from the 93k transmission as well as the reduction gear, bearings and shaft as well as the pinion shaft gear and swapped them onto the 150k transmission. That's the history on what was done. Other parts switched over from the original transmission include the electronic part (sorry, i don't know how else to put this) of the gear selector that is on the side of the transmission case as well as the hard lines from the trans to the rubber lines.

Now, thats the history, here's the symptoms that Both transmissions had... Both shift fine into reverse but they engage pretty harsh compared to all other SVXi I've ever been in. This to me seems like there is high pressure when shifting into reverse as it makes the entire car shake when the gears engage. Other than that, no issues in reverse. Drives just fine and does what it should.
When I shifted both transmissions into Neutral or Drive (or any forward drive position on the gear selector) the speedometer would immediately go to about 20 mph. When in drive after about 5 seconds, the speedometer will go up to about 35 mph and I hear a "grinding" noise coming from the transmission as if the gears are trying to engage but can't. Also, the transmission will grind if I move the gear selector back to R or P too quickly even if it wasn't grinding yet when in D. Meaning, I have to move from D to N, wait about 2 seconds, shift from N to R and let it fully engage, and then move to P or else I hear the same grinding noise-no change in sound at all, literally the same noises.
Things such as solenoids can be ruled out as 2 separate complete transmissions are unlikely to both have problems like that with solenoids. The diff was in great shape when removed from the car and that same 93k was used on the new transmission-also, reverse works great when driven about 50 yards so that would also rule out the diff.

Here are some thoughts I have. Just what seems "logical" at least right now granted I'll be honest and don't really know "jack" about how it all works but am slowly learning

1. TCU problem of some sort causing signals to be haywire causing not enough pressure to actually engage the forward gears properly.
2. transmission resistor-seemingly high pressure making it engage quickly and hard into Reverse and possibly no pressure when in drive?-similar to above
3. Possibly the electrical part in the gear selector lever on the side of the transmission. This was a part I had to use from the old transmission as the replacement did not have this. Just something random, not sure it actually has anything to do with it.
4. Another just random thought off the top of my head is the speed sensor somehow. The speedometer goes up in speed even though it doesn't move in Drive or Neutral. It does work properly in Reverse unless the speedo isn't supposed to move in reverse?

So, I do have extra Trans resistors that I could test out. I do have spare TCU's but only from AWD cars. Is it possible to use these to test? obviously they will show a trans code for no Sol C but other than that, can it be used to try and see if the stock is at fault in some way?

Really could use anyone's suggestions or inputs. I don't care if you're an expert on Subaru's or not. I'm just looking for some ideas to try and test out as well as confirmation on the TCU question for testing.
Many thanks to anyone in advance as I just spent about 20 hours on Saturday and Sunday doing a trans R&R that ended up doing nothing

-Nate

Edit: I suppose I should have also put that there is no blinking power light signifying any transmission codes.
I also tried unplugging each of the transmission wiring looms to see if I could have it default to "limp home"/safe mode and might work. But when doing this, the car would not start. Don't know if that really means anything.

92 SVX · #2 03-22-2010, 07:44 AM

The jumping speedo coupled with the grinding in both N and D signifies to me that the shaft inside is moving but not making good connection to the diff and the wheels.
Put it another way if you had all wheel off the ground and put the car in drive then tried, without using the brakes, to put in park it would make a similar sound.

NiftySVX · #3 03-22-2010, 12:56 PM

I've got a spare FWD 94 tcu I could lend you, which is no problem, but I believe you are correct in the assumption that the AWD conrol box would run it. The FWD won't have the wires in the harness for the AT OIL TEMP FWD lights or the FWD circuit but this shouldn't matter. Also it will get upset it doesn't see speed sensor 1.

Line pressure calculation does take into account vehicle speed, to a pretty high degree. I am not certain but I can assume fairly certainly that base line at 20 mph would be significantly higher than base at idle in N( N is always the same as P) which is around 70. The dropping resistor will have a major effect on base line at below about 15% throttle opening, rasing it to levels that could damage if it is not used IMHO, (mine would jump to more than 140) and it is certainly noticable. This is not dependent on gear position, it makes this change to it across the board moving ratio or neutral, so I do not see how the resistor could be at fault for these symptoms, though it may be bad in any case. The best way to figure this out is to measure line from the main port on the side of the pump housing. (These numbers come from diagniosis of my own unit, which is heavily modified hydraulically, yours should be similar assuming the pump is in good condition and you are not leaking main line somewhere internally)

The switch on the side of the trans is just that, a switch. It tells the TCU what the position of the manual valve is in the main body. It is adjustable, and you can confirm this by counting the clicks on the shaft as it comes out the case and watching the indicator. In removing and installing these trans units I have had to adjust the switch and the cable more often than not. If there were a problem with the manual valve itself, it could cause some odd issues. I have not seen this but it is possible. Did you switch valve bodies in the respective transmissions? This could be your issue. The AWD valve body is different i believe because of the addition of the oil line to the transfer clutch, but they may be interchangeable.

You can only unplug the trans connector on the drivers side, as the other one goes to the aforementioned switch, actually that's all it goes to, and if the car doesn't know it's in P or N it will not turn the starter. once it is running you can unplug it but it will not allow you to shift out of park unless you bypass the shift lock.

I am really thinking there is something odd going on with the information the TCU is getting, as we say in the business "garbage in, garbage out". If it thinks the car is already rolling along at 20 mph it will raise line. BUT I do not believe the TCU corrects for throttle or speed in neutral anyway, I think it is always just a value near 70. I think this is maintained using only fluid temperature correction. This is assuming all things are working obviously.

The line pressure correction for a moving gear should not go into effect until the trans is already in that gear.

The speedo does work in reverse so that is normal.

I wonder, is there a problem with the speedo drive gear on the front diff? Or the driven gear on the sensor itself?

I'm also wondering about this grinding noise. The park pawl on these is on the reduction gearset, isn't it? If they are rotating and the pawl stays engaged it makes a nasty noise, and that may have been what took down that other set of gears.

Anyway, back to work. sorry the post is disorganized. My thoughts are not really coherent.

NiftySVX · #4 03-22-2010, 12:58 PM

I've got a spare FWD 94 tcu I could lend you, which is no problem, but I believe you are correct in the assumption that the AWD conrol box would run it. The FWD won't have the wires in the harness for the AT OIL TEMP FWD lights or the FWD circuit but this shouldn't matter. Also it will get upset it doesn't see speed sensor 1.

Line pressure calculation does take into account vehicle speed, to a pretty high degree. I am not certain but I can assume fairly certainly that base line at 20 mph would be significantly higher than base at idle which is around 70 for N. The dropping resistor will have a major effect on base line at below about 15% throttle opening, rasing it to levels that could damage if it is not used IMHO, (mine would jump to more than 140) and it is certainly noticable. This is not dependent on gear position, it makes this change to it across the board moving ratio or neutral, so I do not see how the resistor could be at fault for these symptoms if it is low pressure in D though it may be bad in any case. The best way to figure this out is to measure line from the main port on the side of the pump housing. (These numbers come from diagniosis of my own unit, which is heavily modified hydraulically, yours should be similar assuming the pump is in good condition and you are not leaking main line somewhere internally)

The switch on the side of the trans is just that, a switch. It tells the TCU what the position of the manual valve is in the main body. It is adjustable, and you can confirm this by counting the clicks on the shaft as it comes out the case and watching the indicator. In removing and installing these trans units I have had to adjust the switch and the cable more often than not. If there were a problem with the manual valve itself, it could cause some odd issues. I have not seen this but it is possible. Did you switch valve bodies in the respective transmissions? This could be your issue. The AWD valve body is different i believe because of the addition of the oil line to the transfer clutch, but they may be interchangeable.

You can only unplug the trans connector on the drivers side, as the other one goes to the aforementioned switch, actually that's all it goes to, and if the car doesn't know it's in P or N it will not turn the starter. once it is running you can unplug it but it will not allow you to shift out of park unless you bypass the shift lock.

I am really thinking there is something odd going on with the information the TCU is getting, as we say in the business "garbage in, garbage out". If it thinks the car is already rolling along at 20 mph it will raise line. BUT I do not believe the TCU corrects for throttle or speed in neutral anyway, I think it is always just a value near 70. I think this is maintained using only fluid temperature correction. This is assuming all things are working obviously.

The line pressure correction for a moving gear should not go into effect until the trans is already in that gear.

The speedo does work in reverse so that is normal.

I wonder, is there a problem with the speedo drive gear on the front diff? Or the driven gear on the sensor itself?

I'm also wondering about this grinding noise. The park pawl on these is on the reduction gearset, isn't it? If they are rotating and the pawl stays engaged it makes a nasty noise, and that may have been what took down that other set of gears.

I do not see how the speedo could move unless the diff carrier is rotating. The noise in N is real odd, because no clutches or the band should be engaged, and the reduction and diff gears are stationary, so there are literally no gears to grind...

Now, If your selector lever says N and the TCU thinks N but the trans itself is in D...then that would account for low line in D but normal in R, assuming the TCU sees R when the trans is actually in R
Anyway, back to work. sorry the post is disorganized. My thoughts are not really coherent.

NiftySVX · #5 03-22-2010, 01:07 PM

I've got a spare FWD 94 tcu I could lend you, which is no problem, but I believe you are correct in the assumption that the AWD conrol box would run it. The FWD won't have the wires in the harness for the AT OIL TEMP FWD lights or the FWD circuit but this shouldn't matter. Also it will get upset it doesn't see speed sensor 1.

Line pressure calculation does take into account vehicle speed, to a pretty high degree. I am not certain but I can assume fairly certainly that base line at 20 mph would be significantly higher than base at idle in N( N is always the same as P) which is around 70. The dropping resistor will have a major effect on base line at below about 15% throttle opening, rasing it to levels that could damage if it is not used IMHO, (mine would jump to more than 140) and it is certainly noticable. This is not dependent on gear position, it makes this change to it across the board moving ratio or neutral, so I do not see how the resistor could be at fault for these symptoms, though it may be bad in any case. The best way to figure this out is to measure line from the main port on the side of the pump housing. (These numbers come from diagniosis of my own unit, which is heavily modified hydraulically, yours should be similar assuming the pump is in good condition and you are not leaking main line somewhere internally)

The switch on the side of the trans is just that, a switch. It tells the TCU what the position of the manual valve is in the main body. It is adjustable, and you can confirm this by counting the clicks on the shaft as it comes out the case and watching the indicator. In removing and installing these trans units I have had to adjust the switch and the cable more often than not. If there were a problem with the manual valve itself, it could cause some odd issues. I have not seen this but it is possible. Did you switch valve bodies in the respective transmissions? This could be your issue. The AWD valve body is different i believe because of the addition of the oil line to the transfer clutch, but they may be interchangeable.

You can only unplug the trans connector on the drivers side, as the other one goes to the aforementioned switch, actually that's all it goes to, and if the car doesn't know it's in P or N it will not turn the starter. once it is running you can unplug it but it will not allow you to shift out of park unless you bypass the shift lock.

I am really thinking there is something odd going on with the information the TCU is getting, as we say in the business "garbage in, garbage out". If it thinks the car is already rolling along at 20 mph it will raise line. BUT I do not believe the TCU corrects for throttle or speed in neutral anyway, I think it is always just a value near 70. I think this is maintained using only fluid temperature correction. This is assuming all things are working obviously.

The line pressure correction for a moving gear should not go into effect until the trans is already in that gear.

The speedo does work in reverse so that is normal.

I wonder, is there a problem with the speedo drive gear on the front diff? Or the driven gear on the sensor itself?

I'm also wondering about this grinding noise. The park pawl on these is on the reduction gearset, isn't it? If they are rotating and the pawl stays engaged it makes a nasty noise, and that may have been what took down that other set of gears.

Anyway, back to work. sorry the post is disorganized. My thoughts are not really coherent.

92 SVX · #6 03-22-2010, 01:48 PM

Quote:

Originally Posted by NiftySVX

I've got a spare FWD 94 tcu I could lend you, which is no problem, but I believe you are correct in the assumption that the AWD conrol box would run it. The FWD won't have the wires in the harness for the AT OIL TEMP FWD lights or the FWD circuit but this shouldn't matter. Also it will get upset it doesn't see speed sensor 1.

Line pressure calculation does take into account vehicle speed, to a pretty high degree. I am not certain but I can assume fairly certainly that base line at 20 mph would be significantly higher than base at idle which is around 70 for N. The dropping resistor will have a major effect on base line at below about 15% throttle opening, rasing it to levels that could damage if it is not used IMHO, (mine would jump to more than 140) and it is certainly noticable. This is not dependent on gear position, it makes this change to it across the board moving ratio or neutral, so I do not see how the resistor could be at fault for these symptoms if it is low pressure in D though it may be bad in any case. The best way to figure this out is to measure line from the main port on the side of the pump housing. (These numbers come from diagniosis of my own unit, which is heavily modified hydraulically, yours should be similar assuming the pump is in good condition and you are not leaking main line somewhere internally)

The switch on the side of the trans is just that, a switch. It tells the TCU what the position of the manual valve is in the main body. It is adjustable, and you can confirm this by counting the clicks on the shaft as it comes out the case and watching the indicator. In removing and installing these trans units I have had to adjust the switch and the cable more often than not. If there were a problem with the manual valve itself, it could cause some odd issues. I have not seen this but it is possible. Did you switch valve bodies in the respective transmissions? This could be your issue. The AWD valve body is different i believe because of the addition of the oil line to the transfer clutch, but they may be interchangeable.

You can only unplug the trans connector on the drivers side, as the other one goes to the aforementioned switch, actually that's all it goes to, and if the car doesn't know it's in P or N it will not turn the starter. once it is running you can unplug it but it will not allow you to shift out of park unless you bypass the shift lock.

I am really thinking there is something odd going on with the information the TCU is getting, as we say in the business "garbage in, garbage out". If it thinks the car is already rolling along at 20 mph it will raise line. BUT I do not believe the TCU corrects for throttle or speed in neutral anyway, I think it is always just a value near 70. I think this is maintained using only fluid temperature correction. This is assuming all things are working obviously.

The line pressure correction for a moving gear should not go into effect until the trans is already in that gear.

The speedo does work in reverse so that is normal.

I wonder, is there a problem with the speedo drive gear on the front diff? Or the driven gear on the sensor itself?

I'm also wondering about this grinding noise. The park pawl on these is on the reduction gearset, isn't it? If they are rotating and the pawl stays engaged it makes a nasty noise, and that may have been what took down that other set of gears.

I do not see how the speedo could move unless the diff carrier is rotating. The noise in N is real odd, because no clutches or the band should be engaged, and the reduction and diff gears are stationary, so there are literally no gears to grind...

Now, If your selector lever says N and the TCU thinks N but the trans itself is in D...then that would account for low line in D but normal in R, assuming the TCU sees R when the trans is actually in R
Anyway, back to work. sorry the post is disorganized. My thoughts are not really coherent.

This brings up a very good point. Could the axles not be meshed with the diff for forward? the diff spinning would explain a lot.

NiftySVX · #7 03-22-2010, 03:16 PM

Also I am not sure why this posted 3 times.

1986nate · #8 03-22-2010, 04:25 PM

Quote:

Originally Posted by NiftySVX

I've got a spare FWD 94 tcu I could lend you, which is no problem, but I believe you are correct in the assumption that the AWD conrol box would run it. The FWD won't have the wires in the harness for the AT OIL TEMP FWD lights or the FWD circuit but this shouldn't matter. Also it will get upset it doesn't see speed sensor 1.

That would be great, just to try and rule things out one by one. I'll send you a PM with info and let me know how much shipping is. as soon as I figure out if its a culprit in any way, i can send you more money for the unit and if there is no change after everything, I'll ship it back to ya

Line pressure calculation does take into account vehicle speed, to a pretty high degree. I am not certain but I can assume fairly certainly that base line at 20 mph would be significantly higher than base at idle in N( N is always the same as P) which is around 70. The dropping resistor will have a major effect on base line at below about 15% throttle opening, rasing it to levels that could damage if it is not used IMHO, (mine would jump to more than 140) and it is certainly noticable. This is not dependent on gear position, it makes this change to it across the board moving ratio or neutral, so I do not see how the resistor could be at fault for these symptoms, though it may be bad in any case. The best way to figure this out is to measure line from the main port on the side of the pump housing. (These numbers come from diagniosis of my own unit, which is heavily modified hydraulically, yours should be similar assuming the pump is in good condition and you are not leaking main line somewhere internally)

The switch on the side of the trans is just that, a switch. It tells the TCU what the position of the manual valve is in the main body. It is adjustable, and you can confirm this by counting the clicks on the shaft as it comes out the case and watching the indicator. In removing and installing these trans units I have had to adjust the switch and the cable more often than not. If there were a problem with the manual valve itself, it could cause some odd issues. I have not seen this but it is possible. Did you switch valve bodies in the respective transmissions? This could be your issue. The AWD valve body is different i believe because of the addition of the oil line to the transfer clutch, but they may be interchangeable.

The valve body was not switched over so it is the valve body from the "new" transmission. meaning, both transmissions did it with 2 different valve bodies.

You can only unplug the trans connector on the drivers side, as the other one goes to the aforementioned switch, actually that's all it goes to, and if the car doesn't know it's in P or N it will not turn the starter. once it is running you can unplug it but it will not allow you to shift out of park unless you bypass the shift lock.

I was probably in a rush/pissed off and didn't have the one for the selector all the way pushed in. I will have to try to double check this. If I can get it to work in "safe mode" or whatever you want to call it, that could rule out a couple things.

I am really thinking there is something odd going on with the information the TCU is getting, as we say in the business "garbage in, garbage out". If it thinks the car is already rolling along at 20 mph it will raise line. BUT I do not believe the TCU corrects for throttle or speed in neutral anyway, I think it is always just a value near 70. I think this is maintained using only fluid temperature correction. This is assuming all things are working obviously.

The line pressure correction for a moving gear should not go into effect until the trans is already in that gear.

Could this account for the speedometer climbing to about 35 mph and the actual grinding starting after about 5 seconds of being put into "D" or could the grinding and raise in speedo reading be the fact that the transmission was trying to upshift or something?

The speedo does work in reverse so that is normal.

I wonder, is there a problem with the speedo drive gear on the front diff? Or the driven gear on the sensor itself?

I'm also wondering about this grinding noise. The park pawl on these is on the reduction gearset, isn't it? If they are rotating and the pawl stays engaged it makes a nasty noise, and that may have been what took down that other set of gears.

The parking prawl is in fact in the reduction gearset, and I double checked multiple times to see that it was working properly-grabbing the park "cog" when put in park and releasing when moving the gear selector to all other positions.

Anyway, back to work. sorry the post is disorganized. My thoughts are not really coherent.

I will have to apologize as the car is about 15 miles away so I cannot just simply go out and test things as I figure them out. I will probably not have a chance to go out and check things out until this weekend but possibly get a chance earlier. I will however be printing any suggestions out so I don't forget any one thought/option.

I thought of another random idea. Like i said, i'm just throwing out ideas. Is it possibly for the "spider gears" inside the front diff to somehow be damaged that it would allow it to move one way but not the other? I guess I probably should have used the other differential off the "new" transmission rather than reinstalling the "old" one.

BTW, for more info on the "new" 150k trans. I got it from wikkedjuggalo as he had (thought) he blew the front differential on it. Upon opening everything up to swap the good and bad bits, the differential was in good shape but the reduction gear was sheered off.

1986nate · #9 03-22-2010, 04:55 PM

Also, I would like to add as I wonder if this has anything to do with the problems. According to the PO, the car was bought from the dealership in October for $4k. Upon disassembly of everything, it was quite obvious that the transmission had been removed before. Could this be that a dealership did a "quick fix" to get a car sold and it failed again soon after? Like they got a cheap trade in for a low mileage SVX, didn't feel like doing a proper fix and chose to just rig it to work long enough?

I'm texting the PO and getting a little info from in on how it "blew". He said it happened while driving that It just wouldn't move all of a sudden when driving it. I tried asking him if he was taking off from a stop and accelerating hard just before at all but he hasn't responded.

Another thing to add is that when draining the front diff oil, there weren't any large pieces or chunks but the magnetic plug was loaded with shavings but I assumed that normal as gears like that have a "break in" period I believe where the gears have to "wear in" to each other when they're new.

The spider gears in the diff seem like something that could be likely. This would explain the speedo moving, correct? Or where does the speed sensor get its main signal from? How often do those really fail or just go bad? And is it possible for them to work properly one way and not the other? Also, when removing and reinstalling, the axles spun and grabbed very erratically. EX. sometimes would spin with normal resistance for a full half turn while other times I could only turn one about an 1/8th turn, have to spin the other one a little, and continue going back and forth until I got them where I needed to get the axle pins in position to where I could remove, and then later, reinstall the pins. This was with the trans in Park as I forgot to put it into neutral. And the axles shouldn't spin at all when in park, correct?

Just trying to think of ideas as things seem odd in my head and remembering things. Sorry for the long and confusing posts.

oab_au · #10 03-22-2010, 05:25 PM

It won't be electrical. More than likely the planet and side gears in the differential. The speed sensor is driven off the parking cog in the box. So it rotates and the sensor reads, till the slack is taken up. Should have used the other differential center.

Harvey.

NiftySVX · #11 03-22-2010, 06:12 PM

The only way I can see the speedo registering speed is because the diff carrier is rotating. It has to be in the final drive. I really don't see any other possibility.

The earlier post was me going thru my usual thinking aloud, but I think we got ahead of ourselves trying to diagnose the problem from the inside out.

1986nate · #12 03-22-2010, 09:48 PM

Quote:

Originally Posted by oab_au

It won't be electrical. More than likely the planet and side gears in the differential. The speed sensor is driven off the parking cog in the box. So it rotates and the sensor reads, till the slack is taken up. Should have used the other differential center.

Harvey.

Yes Harvey, I am kicking myself for the differential

The problem is I didn't think anything would be wrong with the differential since it drove in reverse perfectly normal.
Now, the other problem is the replacement transmission that I got (the 150k one) had a sheered reduction geer in the tailhousing. I am worried that with as much stress as that took, that the front differential may also be at a point of distress that it could fail at any time.

also to clear up what i was saying, when the car was in park and I was doing the work, the axles were able to move independantly of each other but then at random times would catch. This was the same both times. I was too tired to have it fully click in my head. Something seemed odd but didn't think much of it. So this again would point to a problem in the differential. I am thinking that this is the only thing that could cause all of this. Its either a spider gear issue or an issue with bad bearings in the diff carrier that allow things to line up when in reverse and in other gears, things are able to move/shift so that things are off.

Also, here are pics of the sheered reduction gear

(again, not caused by me, was on the replacement trans I got that was thought to have had a blown differential...

92 SVX · #13 03-23-2010, 07:39 AM

You could have a perfectly good trans then, can you rebuild the diff?

1986nate · #14 03-23-2010, 11:02 AM

Quote:

Originally Posted by 92 SVX

You could have a perfectly good trans then, can you rebuild the diff?

Yes, it would appear as that lol. So i have 2 good transmissions, both originally from 94 fwd models. 1 (more than likely) bad diff, and the other differential could be stressed since there was enough torque for the reduction gear to be sheered.
It is easier just to swap an entire carrier with the matching pinion shaft than getting new as well as much cheaper.

1986nate · #15 03-27-2010, 10:16 PM

Well, as I had pretty much come to conclusion, it was in fact the spider gears in the front diff that were obliterated. The only reason Reverse worked was because everything was so messed up that it was able to somehow "jam" things together that allowed reverse to work. It is now up and running, and I have a great running 93,000 mile SVX

Edit: Also, many thanks to everyone who helped me out with this issue. Especially svxfiles and Niftysvx who helped me out with how to remove the reduction gears, and also just making sure that I was doing things properly.

In all honesty, it should've been something I noticed right away when the axles turned freely and independently of each other but being overtired just screws you up sometimes.