subaru FB25 engine interchange

In reply to mildensteve :

Ok so  it was a newer forester engine. I guess at least you won't have the tgv codes to worry about. 
 

curious about the motor mounts and harmonic balancer. I used a 2015 legacy engine and it had the same mounts. I also thought either harmonic balancer would work. I think I used the legacy version. But now I can't remember for sure. Now I'm wondering if using the incorrect version could have anything to do with my misfire code. 

In reply to jdf60 :

Harmonic balancers are tuned to the rotating mass of the crankshaft/pistons/connecting rods/flywheel/flex plate/torque converter/trigger wheel/etc with the firing order and thrust angle of the connecting rods.  Long way of saying your old harmonic balancer is tuned to your installation.  That is why I reused my old one instead of the newer one that was from a CVT car.  I don't think you will see any misfire from the harmonic balancer being mismatched since it is not an ignition trigger like some engines do.

Did you use the fuel injectors from the new or old engine?  There are many reasons for a misfire ranging from coils, plugs, injectors, fuel, computer, crank/cam position sensors, even valves.  Without a real factory/dealership level diagnostics you will be doing a lot of troubleshooting to figure out the cause.  I would start with the basics, which it appears you are doing.

And yes, not dealing with the TGV codes is the only reason I went this way.  But I really want to see the solution to this.  Hopefully Skenton can do it.  It appears he certainly has the patience.

BTW, did you ever change out the cam sensors?

In reply to mildensteve :

Agreed about balancer likely not causing misfire, but I'm reaching. Out of ideas. I think the misfire codes are generated by the cps and if something were vibrating due to me using the wrong balancer I thought maybe it would cause inaccurate cps readings.  The balancer itself  appeared identical. Other than the oil pump shaft was built into the legacy balancer and the forester was a two piece design. But still interacted and worked the same. But could have different harmonics I'm sure. I'll verify which one i used. I agree we should use our vehicles version since the only differences in engine rotating assembly is the flywheel,  flex plate and pickup wheel which stays with our vehicle. 
 

I have ruled out everything I can think of except ecu.  Now I'm reading through a misfire diagnostic doc someone shared on another site.  The only reason I really care about the code is because I think it's deactivating the traction control and cruise. The car is running fine otherwise. 
 

i had to use my forester injectors  as they are not interchangeable. But I swapped them around and no change in cylinder misfire. I haven't had time to change the cam or crank sensor yet. I held off some because another guy said if those were bad I'd see other symptoms amd codes. 
 

Thanks for your inputs! on with the search! 

In reply to jdf60 :

When we got to Tulsa to pick up the engine it turned out to still be in their Kansas City facility. Oops! So, they are paying to ship it to Oklahoma City along with some other orders and I will meet the truck at whatever shop the other stuff is going to. Probably today, waiting for details. My daughter and I had a fun road trip though. FYI it is car-part.com which in internet terms is a very old site which explains the sort of odd name. My favorite mechanic told me about it years ago.

Edit

I sure hope we get my engine delivery sorted out today. I really want a one piece manifold to experiment with this weekend so we can finally get to the bottom of the TGV changes.

In reply to Skenton :

Any luck?  I did look on that car-parts.com. They seem to be a referral site where sellers advertise parts for sale. Lkq has a lot of stuff on there.  I did see some local 2015 legacy engines advertised in Tulsa. Better deals that what I got from lkq   

jdf60 said:

In reply to Skenton :

I did look on that car-parts.com.

It seems they have a lot of listings for the same parts.  Next time I will remember that website for a comparison.  Seems like a good place to search.  Not sure about shipping though. 

I had a little heart to heart discussion with Robert the midway Auto Tulsa manager Friday. He did confirm that my engine made it to Tulsa from Kansas City. He was not happy to find out what people had told me they would do and then didn't follow through with or the lack of communication about the brakes on their delivery truck failing which is delaying things. He was going to "talk" to some people about updating dispatch info and promised to take care of me. Probably Tuesday since they will be playing catch up on Monday. Things always take longer than expected . . .

Yes, car-part.com (note the singular) is just listings from yards all over the country. I suspect is originated on Compuserve in the dialup pre-internet days. I have had good luck finding things there over the years. If you click on price to sort in descending order the asterisk tells you which page has the lowest actual price before the listings switch to "call for price". One time I got some grade B alloy rims from South Dakota way cheaper than I could find locally even with shipping.

This time I sorted by distance from my zipcode, which said 109 miles to Midway Auto in Tulsa OK. Unfortunately whoever made that listing had the location wrong. Lesson learned, confirm *everything* when calling before placing the order.

Edit

I received the engine today (Monday). Robert was good to his word about straightening it out and I called and thanked him. The engine is still strapped into my daughters truck under a tarp. Freezing rain and snow started while I was meeting the Midway truck in Oklahoma City.

Skenton said:

I received the engine today (Monday).

Let the fun begin! 

In reply to mildensteve :

Engine unloaded and Pblaster sprayed on TGV actuator bolts. Tally Ho!

In reply to Skenton :

Good deal!

I believe this design used hex head instead of Phillips on the tgv so they were easier to get off. 
one thing I found odd on the subaru engines was that none of the bolts seemed to be very tight. Except the few that were corroded on like exhaust manifold and some others. Maybe it's just because I'm used to working on old cars and tractors :) 

jdf60 said:

one thing I found odd on the subaru engines was that none of the bolts seemed to be very tight.

Agree.  But if you use the FSM torque values the requirements are pretty low as well.  Makes it easier to assemble and disassemble.  Just be careful not to over torque and strip the bolts.  Ask me how I know... 

I pulled out the left white TGV magnets from my 2016 Legacy engine while getting one of my tools for another job. Yes the 2016 actuator uses hex heads instead of phillips. The clocking of the shaft mounting slot is different from the 2014 and older black magnets as expected. The sloped top surface surprised me. It does not seem to be wear since the plastic sticks up above the magnet all along the slot. I was in a hurry and only grabbed one so i could not mix them up . . . Looking at it now, I think the plastic is assembled as two pieces with the magnet going into a hollow space from below before the bottom is attached. Dunno if it can be dis/re-assembled yet. Also have not checked magnetic poles yet. More later.

Edit

The left white 2016 TGV magnet attracts my probe on the clockwise end just like the 2011 black magnet did. I am about to go put it back and check the right one. Here is a better picture of the sloped top. I think 2016 is a little different from 2015 from what I can see in pictures.

Edit2

Yes indeed, the right white 2016 TGV magnet also attracts my probe on the clockwise end, unlike the 2011 black magnet where the magnetic poles are reversed from the left to right sides. Unless I am missing something this means the Forest computer will be getting the wrong TGV signal voltage on one side due to this change. I will dig out the USB tester and check actual signal voltages at the actuator connector sometime soon.for open/closed on both sides.

In practical terms the idea of drilling/filing the clocking slots on the old Forester magnets and using them on the newer Legacy Donor engine with the one piece manifold still looks good. If the white magnets can be dis/re-assembled with the one of the magnets flipped over that might do to make it compatible with the old Forester computer too.

In reply to Skenton :

Nice work! That would be nice if it were that simple. I will have to wait for a long weekend when we don't need the car and get back into the manifold to try to fix it. I'd really like to get the codes removed eventually. I think it would fix my traction control issue. Someone said Subaru is very picky about allowing cruise and traction control to work when the cel is on. 

thanks for sharing all your efforts here!

jdf60 said:

In reply to Skenton :

Someone said Subaru is very picky about allowing cruise and traction control to work when the cel is on.

You will not get your cruise or traction control back until you get the CEL fixed.  So a worthy effort. 

Should have time to pull the new manifold today. Then I can play with the test gear with a warm desk and recliner instead of a cold garage floor. :-) Will update with what I find.

Edit - Got the manifold off. Also got a Covid booster today, my lungs are bad enough as it is :-(  I expect to be tired and nap(s) tomorrow, probably a good day to break out the USB test set up and multimeter again.

In reply to mildensteve :

Good to know. 
one  thing I figured out with my new Bluetooth code reader. The misfire code is not active. It is archived I think it said. But I can't clear it out. So that explains why it doesn't act like it's misfiring. Question is, will fixing the tgvs turn off the CEL with the archived misfire code that won't erase? 

jdf60 said:

In reply to mildensteve :

So that explains why it doesn't act like it's misfiring. Question is, will fixing the tgvs turn off the CEL with the archived misfire code that won't erase? 

My understanding of the TGV's is that they work like an old time choke although computer controlled and extremely complicated.  As long as they stay open I wouldn't expect a misfire, just some prolonged heating time and potentially higher emissions when cold.  As far as the misfire codes... who knows why they are still there if you don't have any misfires.

The misfire is probably a "pending" code. Things like misfire need to happen several times before they are considered "real". I believe they are auto cleared after a certain number of starts without recurring. Ignore it for a while until we get the TGV sorted out. The auto clear feature may be inhibited while CEL is illuminated.

TGV is an emissions only feature on our old Foresters. When partially closed it increases the air velocity to improve fuel/air mixing during startup of a "cold" engine. According to the FSM on newer engines it was integrated with the throttle control somehow. I did not study that very carefully since we are still using the old Forester computer with our donor long block.

I ended up sleeping yesterday and did not touch the manifold but I just located my test setup, so maybe this evening.

Skenton said:

TGV is an emissions only feature on our old Foresters. When partially closed it increases the air velocity to improve fuel/air mixing during startup of a "cold" engine.

And boy is it noisy air flow.  You can hear them going closed by the noise increase, kinda like the VW's with their aux air pump.  After about 30 seconds or so the idle drops down and the TGVs open back up.  I did a test on mine by disconnecting the TGV when cold just to see what would happen, other than a CEL and cruise flashing.  Real quiet even cold.

In reply to mildensteve :

Well I had t really driven the car much since it's my sons. . But I did take it for a drive the other day.  To me it seems like it's not running quite right. It seems to be slightly underpowered and I think I can feel a very faint surging. But it's not noticeable on the rpm or speed.  And it's running great at idle. I do notice cold starts are a little rougher but nothing I'm concerned about.  I had my Bluetooth scanner app open and cleared codes after it was warm and two of the tgv codes came right back. It's like it knows they are "stuck" during warm operation. Makes me think it's using them for something else. 
 

if you get an inkling, unplug them and drive it and see if you can tell anything while driving. 
 

does your traction control light also come on?  
 

I did try the trick of using the old tgvs on the car while it was running. But one of the old ones actually is stuck open and won't move so I couldn't clear that code. The stepper motor is bad I think. That might explain the CEL before the engine swap. 

It is rainy here and I sat down with the manifold and my multimeter. At first glance the signal voltages are strange, running about a half volt higher on one side than the other when both open and closed. Maybe there is enough variation in magnets and sensors that they need calibration at startup or something. It did not occur to me to trade either the actuators or magnet side to side and see if the issue tracked the sensor or magnet until after I put everything away out in the shop shed for the night. The magnet parts seem visually identical in terms of clocking angle of the shaft mounting holes and also the location of which magnetic pole. I cannot measure the magnet strength with the equipment I have.

I had enough trouble keeping contacts on the actuator pins that I am thinking about cutting off one of the wiring harness connectors that came with the Legacy engine and wiring up something more reliable then remeasuring.

You can probably temporarily swap a Legacy actuator/motor for the stuck one on the old Forester TGV and repeat the trick to test if that avoids some TGV codes and "fixes" the cruise and traction.

Edit

One thing I did notice was that the new white magnets are longer the old black magnets and when the TGV is rotated full range the white magnets go from one end over the sensor to the middle over the sensor instead of one end over the sensor to the other end over the sensor like the black magnets. This may imply a reduced signal voltage swing and probably different biasing in the sensor circuit to the new computer. Since we are keeping the old Forester computer I begin to suspect that the white magnets truly are incompatible with it and we will indeed need to reclock and reuse the black magnets to get the necessary TGV signal voltages.

Edit2

Here is a picture of my 2011 black magnet on top of the 2016 white magnet

Has anyone had issues with the torque converter fitment with a 2015 legacy into a 2013 forester?

I just did all the swaps as mentioned in this thread, put it back in and the TC doesn't fit. The spindle is too long to fit in against the crankshaft. 
im wondering if my TC has been replaced with a aftermarket knock off. 

In reply to Roobasu :

https://grassrootsmotorsports.com/forum/grm/the-dreaded-subaru-torque-converter-situation/92399/page1/

Read the thread link above carefully, there are apparently 3 separate sets of turn, wiggle and push until it drops in some more to fully seat. Folks mangled their TC trying to force it.

In reply to Skenton : mine is on the crank side. The spindle is a tad too big  

Jdf60 put a 2015 Legacy engine into his Forester, which I believe is 2013 like yours. Maybe he can shed some light.

Pics of the problem parts/region might help the discussion too.