Subaru Outback Forums banner

1 - 20 of 25 Posts

·
Registered
2009 OB 2.5i 4EAT
Joined
·
298 Posts
Discussion Starter #1
*Automatic transmission*

Hey Folks,

I hate this car :smile2:

Recently the transmission/motor has felt odd. Haven't really been able to say how or why. Today, I got something. I noticed when I'd accelerate from around 20 to 50, the RPMs would increase then hesitate. I got a video of it here: https://photos.app.goo.gl/x4QkkyHt7jEnZrhm2. Throttle was constant during the dip.

I had to stop immediately because of cars in front. But it sort of seems to do this same behavior other times. I'm trying to pinpoint the behavior but it's hard- it just feels off sometimes, like the acceleration is lagging at the top of the gear/near shift.

How should I start to diagnose this?

Transmission was apparently flushed about 20K miles ago with Kendall® VersaTrans LV ATF by previous owner. I have ordered a VAG-COM cable to start doing some logging.
 

·
Registered
2009 OB 2.5i 4EAT
Joined
·
298 Posts
Discussion Starter #2
Welp, I'm an idiot. ATF is low as ****. Below the low lines. Always checked while car was off and I just found out it needs to be checked while car is on. Let's see if I can find something to add....


What can I add? I don't think anything near me will have ATF-HP
 

·
Registered
2006 Outback XT
Joined
·
18 Posts
Depending on how long you ran it low you may have scorched the clutches, keep an eye on it. If the fluid is low you have a leak somewhere, fluid just doesn't disappear and low fluid/leaks are fast way to burn up a trannys.
 

·
Registered
2009 OB 2.5i 4EAT
Joined
·
298 Posts
Discussion Starter #4
Yup I'm aware, I'm already freaking out. Can I add DEXRON III until I can get ATF-HP and swap fluid's?
 

·
Registered
2009 OB 2.5i 4EAT
Joined
·
298 Posts
Discussion Starter #5
Added some Castrol ATF. Found a post online that said Subaru says DEXRON III is OK in a pinch. Level barely touched the bottom of the dipstick. Only added 3/4 qt.. not that much.. so we'll see if that fixes the transmission issues. Really surprised how little was needed. I'll add ATF replacement to my "to do" list.
 

·
I simply cannot abide useless people.
2006 2.5i and 2002 3.0 wagons.
Joined
·
12,306 Posts
Yup I'm aware, I'm already freaking out. Can I add DEXRON III until I can get ATF-HP and swap fluid's?
You can. Expect it to be not quite as smooth as per norm. You can find ATF HP on Ebay for not too painful.
 

·
Registered
2009 OB 2.5i 4EAT
Joined
·
298 Posts
Discussion Starter #7
You can. Expect it to be not quite as smooth as per norm. You can find ATF HP on Ebay for not too painful.
Yeah I found the Idemitsu stuff, I'll do a few drain/fills when I do my oil changes. Full flush looks a little gnarly.
 

·
I simply cannot abide useless people.
2006 2.5i and 2002 3.0 wagons.
Joined
·
12,306 Posts
If you buy a case, as I'm getting ready to do for the nameless one, you should get either 3 or 4 drain and fills out of the 12 quart case.

You can do them rather quickly after a short drive to get the fluid up to temp and after manually putting it in each gear for a few seconds, but I personally prefer to do them over a few weekends to let the fluid circulate and clean some.
 
  • Like
Reactions: 1 Lucky Texan

·
Master Caster
2005 XT, Mildly Modified...2006 XT Limited, Highly Modifed
Joined
·
16,144 Posts
Yeah I found the Idemitsu stuff, I'll do a few drain/fills when I do my oil changes. Full flush looks a little gnarly.
Just get some new fluid in and through it. I would take the suggestion on running a 100 miles or so between drain and fills for good circulation. It may be a panic over nothing. It's good you caught it when you did. It's not the 1st trans to ever run low.

Let this be a reminder for all to check the tranny while running.
 

·
Registered
03 H6 OBW & 06 WRX Sportwagon
Joined
·
17,704 Posts
wouldn't hurt to double check level and appearance of the coolant since the trans coil is in the radiator.


but I bet everything's OK.
 

·
Registered
2009 OB 2.5i 4EAT
Joined
·
298 Posts
Discussion Starter #11
wouldn't hurt to double check level and appearance of the coolant since the trans coil is in the radiator.


but I bet everything's OK.
Well there's definitely seepage from the driver side front axle, but I think that's just the differential.

I think the low fluid is from a leak in the oil cooler connection that I clamped up about 3-4 months ago. So, she might have been going on a while with low fluid. I had been checking because of that, but never while the car was on. But definitely, good idea. I have Friday, I'll be under the car inspecting.
 

·
Registered
2009 OB 2.5i 4EAT
Joined
·
298 Posts
Discussion Starter #12
So experienced the same issue twice today, at lower speed (20-30). Under constant throttle the car would accelerate then suddenly drop 200-300 rpms, adding throttle would immediately jump it back up. Engine seems normal otherwise- maybe sluggish? Or perhaps I've ust gotten used to the power jump from moving into OB from my old Civic.

So perhaps this is an engine issue? I'm getting a VAG-COM cable soon here, is there some general logs I could take? Should I try to reset the ECU or TCU 'learning'(I've seen this mentioned a few times, not familiar with it)? Only engine related thing done lately was I put in a new air filter.
 

·
Master Caster
2005 XT, Mildly Modified...2006 XT Limited, Highly Modifed
Joined
·
16,144 Posts
Only engine related thing done lately was I put in a new air filter.
Ok, check your worm clamp off the air box then.

Logging data will definitely help to see if you have an engine issue or not. What about fuel pressure. I would agree it could possibly NOT actually transmission related.
 

·
Registered
Joined
·
18,289 Posts
So experienced the same issue twice today, at lower speed (20-30). Under constant throttle the car would accelerate then suddenly drop 200-300 rpms, adding throttle would immediately jump it back up. . . . I'm getting a VAG-COM cable soon here, is there some general logs I could take? . . .
Torque converter lock-up (and release)?

Under low/moderate throttle acceleration (relatively low engine torque), the TC will partially lock up when the difference between engine rpm and torque converter turbine rpm is in the 400 rpm range.

Here's a graph showing various 4AT parameters on a fairly steady acceleration. The difference between engine rpm and turbine rpm (torque converter "slippage") is scaled on the right vertical axis. All other parameters are on the left side vertical axis. Overall time is about 36 seconds.



In the graph, the car is initially stopped. At around line 61 the accelerator pedal is advanced to about 20%. The rpm-turbine difference jumps from about 600 (idle, in gear) to over 1000, and the car begins to accelerate (Front kph).

As the car accelerates, the rpm-turbine difference drops to almost 200 rpm, but during that time the accelerator position was increased as well from slightly less than 20% to 30%.

At around line 166 the rpm-turbine difference jumps back up to 600 rpm. This reflects the 1-2 gear change. The turbine speed drops down due to the lower ratio of 2nd gear and drops more than the engine at that time.

Then, around line 211, the rpm-turbine again drops to around 400 rpm. At that time, Lockup d/c increases from 5% to 35%, and rpm-turbine speed drops to zero. (Even under partial duty cycle application to the lockup solenoid, there's enough pressure to lock the TC clutch under moderate throttle.)

Only after getting into 4th gear, and having little or no rpm-throttle difference, is the Lockup d/c increased to 95% to provide full lockup.

The "partial lockup" is more sensitive to the throttle. Because at 35% full AT line pressure isn't being applied, the torque handling capability of the clutch is reduced. Consequently, any addition to the throttle will cause the Lockup d/c to drop back to 5%, thereby releasing the clutch, and preventing slippage at the clutch friction surfaces. That release will result in the rpm-turbine difference increasing from zero and the engine rpm (tachometer) going up accordingly.

Logging the transmission data under the same conditions as you described above should reveal if it's the TC lock up system operating.
 

Attachments

·
Registered
2009 OB 2.5i 4EAT
Joined
·
298 Posts
Discussion Starter #15
... How did you just have that handy? haha.

Thank you for the extremely informative write up. I really wish this had engine RPM on it, that would confirm my observations prior to data-logging. If this is the case, I wonder if something has changed to make it more apparent, or if I have just become hyper-sensitive to my car (which is true for sure).

Was this made with btSSM or FreeSSM?

EDIT: My RPM goes down- would that be something different? Steady increase with a very sudden drop.

I think I need to do some logging and first gain an accurate understanding of what's occurring.
 

·
Registered
Joined
·
18,289 Posts
... How did you just have that handy? haha.
Logs that I've made over the years and saved. The graph is from a June 2017 log.

Was this made with btSSM or FreeSSM?
That log was made using BtSsm. FreeSSM does not record, it only displays live data, so no way to save monitored data for review. (Romraider is another program that can both display live data and record it.)

Thank you for the extremely informative write up. I really wish this had engine RPM on it, that would confirm my observations prior to data-logging. . . . EDIT: My RPM goes down- would that be something different? Steady increase with a very sudden drop.
Here's a graph from the same log with engine rpm and turbine rpm separate.



The time scale is the same as the previous graph. In that graph, the Lockup d/c rises to 35% at around line 206. In this graph the engine rpm reaches about 2600 at line 216, and then drops to around 2400 where it meets the turbine rpm around line 230. The actual time between line 206 and 216 is about 1/2-second and between 206 and 230, it's barely over a second, from lock up command to sync.

The "sawtooth" pattern in the engine and turbine rpm reflects the effect of the transmission upshifting as the car accelerates.

If this is the case, I wonder if something has changed to make it more apparent, or if I have just become hyper-sensitive to my car (which is true for sure).
I tend to be more sensitive to operation and noises after I've worked on the car, so I wouldn't be surprised. However, perhaps with the new fluid it's actually working better.

I think I need to do some logging and first gain an accurate understanding of what's occurring.
Agreed.
 

Attachments

·
Registered
2009 OB 2.5i 4EAT
Joined
·
298 Posts
Discussion Starter #17
That is very interesting. Thank you for the data.

That 200-230 region is very interesting and looks like it might be something similar to what's going on.

I'm going to check the levels again, and do some logging tomorrow or saturday when my cable comes in. Good to know BtSsm is so powerful, I'll be making that purchase.
 

·
Master Caster
2005 XT, Mildly Modified...2006 XT Limited, Highly Modifed
Joined
·
16,144 Posts
That is very interesting. Thank you for the data.

That 200-230 region is very interesting and looks like it might be something similar to what's going on.

I'm going to check the levels again, and do some logging tomorrow or saturday when my cable comes in. Good to know BtSsm is so powerful, I'll be making that purchase.
It's logging data that's powerful. Folks like @plain OM who help interpret the data are even more...powerful.

Getting to know you car, vs treating it like an appliance.
 

·
Registered
Joined
·
18,289 Posts
That is very interesting. Thank you for the data.

That 200-230 region is very interesting and looks like it might be something similar to what's going on.

I'm going to check the levels again, and do some logging tomorrow or saturday when my cable comes in. Good to know BtSsm is so powerful, I'll be making that purchase.
Note that BtSSMsm is somewhat different than Romraider. With Romraider (RR), once the ECU and Logger definition files are downloaded and set up so that RR can use them, and RR connects to the ECU, it will then automatically display the list of parameters and switches that are available from the ECU (and TCU) and can be viewed just by checking them off in the list. The two definition files are fairly comprehensive, covering a wide range of Subaru years and models.

BtSSM is somewhat different. It comes with only a relatively small selection of preset parameters that it can display/log. Others can be added, but this has to be done manually. The on-line manual on the BtSSM website has instructions for this. Also, BtSSM seems to have a far smaller range of year/model coverage. In some cases, if a particular ECU isn't included, a "proxy" can be set up using the definitions for a different ECU, but that's another step.

Also, RR works with Windows or Linux, BtSSM is Android.

My recommendation would be to go with RR initially, see what parameters are available from your ECU and TCU, and get some experience using the real-time display and the recorded logs.

I started with FreeSSM, switched to RR for it's display and recording advantages (which I still use on occasion), and then added BtSSM on a dashboard tablet that provides a range of data whenever the car is being used, in other words, a second "dashboard", without having to deal with the laptop.

For more info on BtSSM, http://www.btssm.com/, and this thread.
 

·
Registered
2009 OB 2.5i 4EAT
Joined
·
298 Posts
Discussion Starter #20
In case anyone is interested, I took a minute to do the math...

For a 10 qt capacity and 4 qt's removed per draining, running through 12 qts (so 3 changes) replaces 78.4% of the original ATF.

Formula is OLD ATF=10 * 0.6^(# of changes)
 
1 - 20 of 25 Posts
Top