subaru FB25 engine interchange

In reply to Skenton :

Hope you feel better soon Skenton. Same here, mine has become my daily and haven't found the time to play with swapping the pins, the 2011 TGVs are still plugged in hanging out in the engine bay without issue and no check engine codes. I did disassemble a spare TGV connector I had to see how the pins come out, pretty easy to do without damaging anything.

I spread out the old and new TGV stuff and got reoriented.

The plastic gears have bent metal tab that act as stops for the butterfly valves open/closed. However, they are rotated about 90 degrees on the new taller gears relative to the old shorter gears. On the new plastic manifold the shorter arc that the tab inserts into is also rotated relative to the old metal TGV body. This means that regardless of the magnet or motor/sensor used the gears have to stay with the manifold they came with and will not fit in the "other" style and engage the motor etc.

On the new plastic manifold with the taller gears the old black magnets need a couple of extra washers to raise them up to the correct level for the Hall effect sensor compared to the new white magnets.

Based on info in the shop manuals and measurements the old style TGV sensors are either on or off and the TGV system is only used at engine start for a few seconds to improve emissions when cold. The newer style TGV system in the plastic manifold seems to have sensor output proportional to the opening angle and is coupled with the throttle system more tightly.

What I was doing when I stopped was measuring voltages with the old, new and hybrid TGV gear/magnet/sensor configurations at different angles. The goal is to make the old Forester computer happy, which might require filing or drilling out the index on the black magnets to allow "reclocking".  Back on it now. I do not want to make irreversible changes until certain.

I just measured voltages on my stock 2011 Forester TGV. Pics below.

Fully open is close to 5Vdc and fully closed is close to 0Vdc. In between, most of the open range is the same as fully open. Only when you get almost closed does it drop to near zero.

I then substituted the motor/sensor from my new 2015 Legacy engine - with the green label on the side. Pics above.

It gives wildly different voltages for open and closed and varies the voltage for the range in between. The new Legacy sensors are not compatible with what we need and can *NOT* be used as substitutes for broken Forester parts.

JDF60 you will have to scroung up a working Forester sensor to replace the one you think is broken.

I also measured the new Legacy TGV after substituting both the Forester sensor/motor and Forester black magnet with appropriate shimming for the height difference. It gave close to 5Vdc when fully open and fully closed and close to 0Vdc at "impossible" angles where the motor/sensor holes did not line up with the manifold holes.

Since the two black magnets are opposite magnetic orientation to each other I tried both and got similar results.

This is not surprising since I have not adjusted the clocking angle yet as that requires a bit of filing on the parts. Which I am delaying until I fully understand what is needed. Probably soon on the agenda though.

But enough for tonight.

OK, I think I got it working. The actual changes were quite simple once I got my mind wrapped around how it all works. Basically drill or file out the key on the black Forester magnets so they can be reclocked relative to the gears and raise them up with some washers since the Legacy gears are taller than those on the Forester and swap them left/right when installing in the Legacy manifold

The first picture is a closeup showing a black Forester magnet with two shimming washers between it and the Legacy gear & spacer.

The second picture is all the components ready for assembly. Take note of the "L" and "R" on the parts. The gears came with the Legacy and L/R are the correct side of the manifold they came from and are returning to. However the black magnets are from the Forester and the L/R refer to where they came from. They are intentionally going into the "other" side of the Legacy manifold. This is because they have opposite magnetic polarities to each other and to deal with the TGV direction change.

In the assembled pictures you can see the before and after locations. I drew a line on the outside edge halfway between the mounting holes to show about where the sensor is located when installed. For the left side the clockwise end of the curved black magnet should be even with the line when closed. For the right side the counter-clockwise end of the curved black magnets should be even with the line when closed. You may have to wiggle it around since the zero range is rather narrow, but you should get a signal near 0Vdc when closed and near 5Vdc in any other position. Essentially start by centering the magnet across from the gear - the Legacy gear is two teeth shorter than the Forester grear so it ends up looking a bit off center when measuring zero volts. Remember to use the sensors/motors from the Forester not those from the Legacy, they are different.

I strongly suggest making a test jig based on the posted factory diagnostic circuit like mine to measure voltages between signal and ground. Since we reuse the Forester wiring harness I just cut up the Legacy harness to get a TGV connector and cut up an old USB cable to provide +5Vdc.  The 1K Ohm resistor between 5Vdc and signal is buried inside the electrical tape.

I have not looked into swapping the direction pins 4/5 yet but Sodapopdave said it was pretty simple.

Ask if you have questions

I took a look at the connector for pin swapping. You pry out the center and the little plastic tab holding the pin is visible on the side. Release that and pull the wires out the back. I will locate my jeweler screw drivers and practice on this one tomorrow before tackling the Forester harness. I think just be gentle and it is simple indeed.

Pins 4 and 5 control the direction, just swap their positions and reassemble to reverse the TGV rotation. They are the row of two pins in the connector.

Just for grins I reinstalled one of the white Legacy magnets in the Legacy manifold while using the Forester sensor/motor. It measured close to five Volts in all positions. The only way i could get a measured signal near zero volts was to manually hold the thick end of the white magnet low over the sensor. When it was raised up a bit the measured signal changed back to near five volts. Maybe if the magnet was shimmed up a bit closer to the sensorit could be made to register.

It is a moot point since the white magnets are too long to reclock significantly without being cut down and they both have the same magnetic poles orientation so only one side could possibly work in any case.

Bottom line is the stock white Legacy magnets and Denso sensors/motors are incompatible with the Forester computer. Sell them on Ebay or something because they are of no use for this engine swap! Modding the black Forester magnets and reusing the Forester TGV sensors/motors on the Legacy donor engine seems to need golden however.

Howdy y'all.  My buddy and I did this swap two years ago, been working great.  2013 Legacy engine in a 2011 Forester.  Was wondering if anyone replaced the starter.  I'm inclined to use a 2013 Legacy starter, but RockAuto specifies that its for CVT transmission.  We can't remember which engine contributed the starter during the swap. Does anyone have any input?

In reply to jbranham :

I used my factory 2011 starter with my 2017 Forester engine swap if it helps, I think you would use the 2011 starter as well. 

I do not remember it being discussed as a problem in any of the swapping threads I read. I will take a look at the two starters I have from 2011 Forester and 2015 Legacy engines and see how they compare/differ.

Edit

Brain short ZZZZZZZZ. I went over and pulled the wiring harness off the Legacy donor engine this morning and realized that the starter mounts from the tranny side so I do not have it since I did not pull that engine myself. So, I will also be reusing  my Forester starter. The same would probably be true for all the folks who used LKQ engines.

I would suggest getting a starter meant for the original Forester engine and transmission not the Legacy donor.

Jdf60 and Sodapopdave, I know you both said your Forester had turned into a daily driver but you were still interested in getting the TGV issue fixed "properly". Is my description of the magnet modification clear? Is it still something you plan to try. Feedback welcome.

Now that I am feeling better and the TGV analysis/mod is complete my  next goal is pulling the front cover on the Legacy donor engine to get to the cam ends.

In reply to Skenton :

Thank you for all your work with this! I definitely plan to do the proper fix, I do have another vehicle so I can afford to take mine off the road for a few days to do the swap just need to find a few free evenings. The only thing I'm a little unclear on is the clocking of the magnets but I'm sure once I get my hands on it and re read the posts it'll be clear. Would you say your findings confirm the information in ChayNZ posts or anything would do differently (besides swapping the pins on the harness). I love the fact that it looks like I don't have to hack up the harness and extend wires, that just invites a ton of extra failure points.

In reply to Sodapopdave :

Yes, ChayNZ had it right. Just a bit short on details of why. Without pictures I had trouble visualizing but pretty much what I ended up doing. Now I understand about the magnetic poles and why too. I do not remember who first suggested swapping the pins but I liked it.

As for clocking, I was surprised how narrow the ranges of angles that generate zero volts turned out to be. Essentially five volts everywhere and then BANG one point where you get zero. I had the multimeter hooked up and turned the magnet slowly back and forth to nail it down. After each magnet move I just held the sensor/motor in place to get a voltage reading. I had the manifold off the engine using the diagnostic harness jig.

If you think of the curved  gear and maget as two arcs their centers should be roughly across the middle from each other like these parentheses (0) where "0" is the nut on the shaft. That gets you close as a starting point.

A question for folks who have done this swap. What did you do when swapping the rocker covers and timing cover?

Did you use Fel-pro gaskets, use Permatex sealant or do something else? Any tricky spots need extra care? How about O-rings, did you use new O-rings as recommended or just reuse the old ones? Any leaks/problems?

In reply to Skenton :

I used new OEM gaskets with some ultra grey permatex on the rocker covers, where the manual recommends. Used ultra grey for the front cover as well and for the oil pan. No leaks here. I've done a lot of timing chains and always use the grey permatex without issue. New o rings under the cover as well, they are cheap and not worth the risk of losing oil pressure if they break down.

In reply to Sodapopdave :

Thanks, sounds like a plan. I think I will replace the crankshaft front and rear main seal since they will both be accessible during the parts swap. As you said "they are cheap" :-)

Sodapopdave said:

In reply to Skenton :

I used new OEM gaskets with some ultra grey permatex on the rocker covers, where the manual recommends. Used ultra grey for the front cover as well and for the oil pan. No leaks here. I've done a lot of timing chains and always use the grey permatex without issue. New o rings under the cover as well, they are cheap and not worth the risk of losing oil pressure if they break down.

Ditto.  Same same here.

Skenton said:

In reply to Sodapopdave :

Thanks, sounds like a plan. I think I will replace the crankshaft front and rear main seal since they will both be accessible during the parts swap. As you said "they are cheap" :-)

A lot cheaper than redoing the seals after engine installation.

I was just looking for valve/rocker covers gaskets and noticed they seem to change between 2016 and 2017. That might matter to people with a 2017+ donor engine wanting to move the Forester covers for the sake of the coils direct fit. If using the short spacer fix with the Legacy covers it might not matter.

Sodapopdave, I believe you used a 2017 Legacy donor,  what did you do with the valve covers/coils? Now I am wondering how the gaskets changed.

In reply to Skenton :

Mine is actually a 2017 Forester engine - not sure if it's the same as a 2017 Legacy but I have the VIN and know it came out of a wrecked 2017 Forester. I used the 2011 rocker covers and 2011 coil packs and boots, perfect fit with no washers. The gaskets looked the same to me, the only difference in the valve covers I saw was that one set put the coil packs further out then the other. 

In reply to Sodapopdave :

Thanks for the information, good to know.

Going back through Rock Auto the valve covers for Forester and Legacy looks to use the same Fel-pro gasket kit part number up until the switch from FB25B to FB25D in 2019/2020.

I guess I just fumble fingers it or something the first time. Sorry for the noise.

In reply to Skenton :

I also did the valve cover swap from the 2011 to the 2016.  No spacers, works perfect with the 2011 coils.

In reply to obsequious_knight :

That's my video, bro! I've been following this thread to see if there is a clean way to make the TGV system work without my extension harness hack-job. It makes my day that someone found my video helpful. I just completed my 4th swap of this kind with a 2016 Legacy engine into a 2011 Forester. Everything went without a problem, but this time I removed the TGV flaps inside the manifold and left the Legacy solenoids attached and hooked up. I'm considering muting the engine codes for TGV related flags using a Tactrix 2.0 port and ECUflash software. I'm going to try that tonight and see what happens.

In reply to stvnswld :

Good to hear from you and thanks for the video.

I am pretty sure that moving the old Forester solenoids and modding the white Forester magnets plus swapping the direction pins 4 & 5 as described above does the trick at least as measured with a volt meter and the manifold on the bench. My actual engine swap has been stalled over winter due to my lung issues but I am chipping away at it again. Today spent dealing with winter flat tires so I can push it out of the garage to use the engine hoist again.

If anyone has tried it with a running engine please report results.

I have swapped pins 4&5 on the TGV harness and swapped my OEM 2011 TGV manifolds side to side and they are operating properly with no TGV related check engine codes.

To clarify, my OEM TGVs are just hanging out in the engine bay but plugged in to avoid a check engine light. I have not swapped the actuators/magnets to the newer manifold just yet but this adds more confidence that a simple swap of the pins is the only modification needed to the wiring harness to get the newer style TGVs to work with the older TGV actuators.

I'm hoping to swap the magnets and actuators to the new manifold very soon for a final and permanent fix.