My J swapped NSX...sorry for the wait. love or hate

Just quick update. Engine harnesses and dbw system should be ready in a couple weeks thanks to Don at rpm systems. The flywheel is being adressed currently. It will use a custom flywheel but allow a stock or aftermarket nsx clutch.

After some discussion with don at rpm systems we decided for a racecar the j transmission is best to use because of the cheap cost and availability of parts. I will first finish the setup with the c trans to offer anyone that wants to reuse the trans a cheap reliable solution to the c series. After I have maxed out the setup and the trans fails I will switch over to the j trans.
 
Subscribed! Awesome job! I bought my NSX 5 years ago with a blown engine and looked into other engine options, considered the C27 as a temporary to get the car moving, ultimately found a complete used engine and transmission here on prime and got my car on the road. Now, last Tuesday my NSX has a rod knocking, damn! My car has 80,000 miles on it and is well used/enjoyed, it has a branded title due to a previous theft but has not been in an accident however it could use a paint job. So, now the dilemma fix the engine and normal maintenance at least $3,000, sell the car as is $15,000? Or follow this thread and do a J-swap?

OP, I may PM you for details on this swap.

Anxiously watching the progress.
 
Subscribed! Awesome job! I bought my NSX 5 years ago with a blown engine and looked into other engine options, considered the C27 as a temporary to get the car moving, ultimately found a complete used engine and transmission here on prime and got my car on the road. Now, last Tuesday my NSX has a rod knocking, damn! My car has 80,000 miles on it and is well used/enjoyed, it has a branded title due to a previous theft but has not been in an accident however it could use a paint job. So, now the dilemma fix the engine and normal maintenance at least $3,000, sell the car as is $15,000? Or follow this thread and do a J-swap?

OP, I may PM you for details on this swap.

Anxiously watching the progress.

It depends on what you want from the car. If you are after raw power, then a twin turbo J-swap seems the way to go. But, that car will not sound or feel like a NSX. Power will come on early, and the engine will run out of revs at 6500, with a different, lower exhaust note. It will feel more like a Mustang or Corvette. If you are after the "exotic" experience, you should stick with the C, as it will give you that Italian-like high rpm and exotic howl. The more I follow this build, the more it is becoming apparent that this swap would make more sense for the FI crowd, since putting a turbo on a C-series is a risky proposition and any damage is enormously expensive to fix/replace. By contrast, J's are cheap, plentiful and easy to work on with a robust dealer parts stream. Plus, they even have manifolds better suited to bolting up turbos. If I were going to put a twin turbo in my NSX, I would use a J.

I was hoping we would see a 8000 rpm J35 or something similar. Anything less will be a step down from an all motor sense.
 
A year ago at a Japanese auto show a tuner presented a boosted Honda Legend, based on the same J35a8 found in the North American Acura RL.

My daily driver is a 2006 RL and I can attest that it has a great engine.
 
It depends on what you want from the car. If you are after raw power, then a twin turbo J-swap seems the way to go. But, that car will not sound or feel like a NSX. Power will come on early, and the engine will run out of revs at 6500, with a different, lower exhaust note. It will feel more like a Mustang or Corvette. If you are after the "exotic" experience, you should stick with the C, as it will give you that Italian-like high rpm and exotic howl. The more I follow this build, the more it is becoming apparent that this swap would make more sense for the FI crowd, since putting a turbo on a C-series is a risky proposition and any damage is enormously expensive to fix/replace. By contrast, J's are cheap, plentiful and easy to work on with a robust dealer parts stream. Plus, they even have manifolds better suited to bolting up turbos. If I were going to put a twin turbo in my NSX, I would use a J.

I was hoping we would see a 8000 rpm J35 or something similar. Anything less will be a step down from an all motor sense.

I would be willing to bet you could build a high revving J for substantially less than a bored/stroked C.
 
Subscribed! Awesome job! I bought my NSX 5 years ago with a blown engine and looked into other engine options, considered the C27 as a temporary to get the car moving, ultimately found a complete used engine and transmission here on prime and got my car on the road. Now, last Tuesday my NSX has a rod knocking, damn! My car has 80,000 miles on it and is well used/enjoyed, it has a branded title due to a previous theft but has not been in an accident however it could use a paint job. So, now the dilemma fix the engine and normal maintenance at least $3,000, sell the car as is $15,000? Or follow this thread and do a J-swap?

OP, I may PM you for details on this swap.

Anxiously watching the progress.

Best case scenario. $500 full j engine. $ 500-$1000 custom flywheel. $400-$800 custom wire harness ecu setup. Nearly everything else can be used from the factory setup. It will still need little things like coolant lines and custom exhaust. I think once I finish one I may be able to do them fairly cheap.

It depends on what you want from the car. If you are after raw power, then a twin turbo J-swap seems the way to go. But, that car will not sound or feel like a NSX. Power will come on early, and the engine will run out of revs at 6500, with a different, lower exhaust note. It will feel more like a Mustang or Corvette. If you are after the "exotic" experience, you should stick with the C, as it will give you that Italian-like high rpm and exotic howl. The more I follow this build, the more it is becoming apparent that this swap would make more sense for the FI crowd, since putting a turbo on a C-series is a risky proposition and any damage is enormously expensive to fix/replace. By contrast, J's are cheap, plentiful and easy to work on with a robust dealer parts stream. Plus, they even have manifolds better suited to bolting up turbos. If I were going to put a twin turbo in my NSX, I would use a J.

I was hoping we would see a 8000 rpm J35 or something similar. Anything less will be a step down from an all motor sense.

Calling it a mustang or corvette is a huge slap in the face. I was tevving out my old j35 to 7200 rpm. With a basic head cam package it could easily be revved to 8500+.

I thought it was the valvetrain that is the bottleneck for the 6500rpm redline. Maybe a J series with NSX heads for DOHC?

It's the valve springs and cams. Putting upgraded springs and new cams would allow for it to make more power uptop realably.
 
A J35 engine with a 93 mm stroke at 8500 rpm has a piston speed of 26.35 meters per second
A J37 with a 96 mm stroke at 8500 rpm has a piston speed of 27.2 m/s
A C30 with a 78 mm stroke at 8500 rpm has a piston speed of 22.1 m/s

My understanding is that for high speed street gas engines 25 m/s is considered the upper limit for reliable performance.
Competition engines with a higher piston speed, (NASCAR and F1 engines run in excess of 25 m/s and top fuel as high as 30 m/s) need frequent rebuilding and are not seen as suitable for street application.
High revving engines like the C30/32 ,the new Porsche GT3, and Honda sportbike motorcycle engines with 12 k red lines are all big bore short stroke engines to keep piston speed at acceptable levels.

While less than optimal top end breathing can surely hold down an engines top end rpm I understand it's piston speed that will decide if the engine will survive for any length of time.

If I've missed something here and you can build a J37 engine that will rev to 8500 rpm with an OEM life span I would like to be schooled on how to do that.
 
Last edited:
Compare apples to apples. What is the piston speed of a j32 compared to a c32. From any j32 I have seen raising tge rev limiter has never resulted in bearing failure in the bottom end. It usually ends with dropping, or bending valves because the springs are not strong enough.
 
Is the limiting factor the piston to wall interface, or the big and small end bearings not able to handle the reciprocating forces, which far exceed the forces caused by mixture ignition?

If it is the latter, then we can always install titanium connecting rods into a J37.
 
yeah the j series bottom ends tend to take a fair beating but are soooo prone to failure in the head department, even with valves springs retainers etc they are a bit safer but still risky but I say why not, easy and cheap for replacement parts :)
 
Compare apples to apples. What is the piston speed of a j32 compared to a c32. From any j32 I have seen raising tge rev limiter has never resulted in bearing failure in the bottom end. It usually ends with dropping, or bending valves because the springs are not strong enough.

The piston speed of an 86 mm stroke J32 at 8500 rpm is 24.36 m/s (about 11 % higher than a C30.)

I think you stated " I was revving out my old J35 to 7200 rpm. With a basic head cam package it could easily be revved to 8500+."
I have no idea what head and valve train modifications would be necessary to rev your J35 to 8500 rpm.
You are the expert there.
My only point is that for a J35 to rev to 8500 rpm piston speeds will be very high - almost 20 % higher than a C30/32.
That will put severe strain on pistons rods crank and may compromise reliability.

I don't think this is about comparing apples to apples, perhaps it's more about physics and the exponential increases of component mass at high speed.
I believe one of the reason we have titanium rods is to reduce the reciprocating mass which reduces the stress at the high rpm of a C30/32 engine.
Perhaps one of the reasons the J series engines are robust is that they are not high revving engines.
 
The piston speed of an 86 mm stroke J32 at 8500 rpm is 24.36 m/s (about 11 % higher than a C30.)

I think you stated " I was revving out my old J35 to 7200 rpm. With a basic head cam package it could easily be revved to 8500+."
I have no idea what head and valve train modifications would be necessary to rev your J35 to 8500 rpm.
You are the expert there.
My only point is that for a J35 to rev to 8500 rpm piston speeds will be very high - almost 20 % higher than a C30/32.
That will put severe strain on pistons rods crank and may compromise reliability.

I don't think this is about comparing apples to apples, perhaps it's more about physics and the exponential increases of component mass at high speed.
I believe one of the reason we have titanium rods is to reduce the reciprocating mass which reduces the stress at the high rpm of a C30/32 engine.
Perhaps one of the reasons the J series engines are robust is that they are not high revving engines.

I'm not claiming to be an ex expert by any means. Just stateing that upgraded valve, springs and a cam that would make power that high up in rpm is definatly do able. It was a race car and probably had less than 5k miles put on it the 2 seasons I raced it. What needs to be done to a c30/32 to be able to rev and make power to 8500 rpm? I have seen more bottem end failures out of the c series than I have out of the j series. I'm sure there is more to it than just rev amount, but does the 90 degree vs 60 degree have any benefits when it comes to the strain on the pistons/rods/crank?
 
The piston speed of an 86 mm stroke J32 at 8500 rpm is 24.36 m/s (about 11 % higher than a C30.)

I think you stated " I was revving out my old J35 to 7200 rpm. With a basic head cam package it could easily be revved to 8500+."
I have no idea what head and valve train modifications would be necessary to rev your J35 to 8500 rpm.
You are the expert there.
My only point is that for a J35 to rev to 8500 rpm piston speeds will be very high - almost 20 % higher than a C30/32.
That will put severe strain on pistons rods crank and may compromise reliability.

I don't think this is about comparing apples to apples, perhaps it's more about physics and the exponential increases of component mass at high speed.
I believe one of the reason we have titanium rods is to reduce the reciprocating mass which reduces the stress at the high rpm of a C30/32 engine.
Perhaps one of the reasons the J series engines are robust is that they are not high revving engines.

I'm not claiming to be an ex expert by any means. Just stateing that upgraded valve, springs and a cam that would make power that high up in rpm is definatly do able. It was a race car and probably had less than 5k miles put on it the 2 seasons I raced it. What needs to be done to a c30/32 to be able to rev and make power to 8500 rpm? I have seen more bottem end failures out of the c series than I have out of the j series. I'm sure there is more to it than just rev amount, but does the 90 degree vs 60 degree have any benefits when it comes to the strain on the pistons/rods/crank?

I think the trick with the J35 would be, as suggested, lighten the rotating mass as much as possible by using light forged rods and SOS-style lightweight pistons. Also, a precision balance and blueprint would help with supporting 8000+ rpm. I wonder if you could use the OEM engine management (from say, a TL-S) with FIC to manage the fuel/timing requirements of the revised cams and pistons.
 
It depends on what you want from the car. If you are after raw power, then a twin turbo J-swap seems the way to go. But, that car will not sound or feel like a NSX. Power will come on early, and the engine will run out of revs at 6500, with a different, lower exhaust note. It will feel more like a Mustang or Corvette. If you are after the "exotic" experience, you should stick with the C, as it will give you that Italian-like high rpm and exotic howl. The more I follow this build, the more it is becoming apparent that this swap would make more sense for the FI crowd, since putting a turbo on a C-series is a risky proposition and any damage is enormously expensive to fix/replace. By contrast, J's are cheap, plentiful and easy to work on with a robust dealer parts stream. Plus, they even have manifolds better suited to bolting up turbos. If I were going to put a twin turbo in my NSX, I would use a J.

I was hoping we would see a 8000 rpm J35 or something similar. Anything less will be a step down from an all motor sense.

I am mostly interested in just getting my car back on the road. I like the swap idea, I have driven many of the Honda/Acuras with the J-engines and they are great smooth engines. I think Honda even saw the limitations of the 90 degree C engine and they stopped after the C35 which needed a balance shaft. I believe the new NSX is going to be a J35 variant with the SH-AWD, basically the new RL SH-AWD drivetrain although maybe they will surprise us with a DOHC J35. Mazda, Toyota, Nissan all have DOHC 60 degree V-6's why hasn't Honda done one yet. I think the new NSX deserves one, and the previous NSX buyers will expect one.

I am not sure I "need" an engine to run up to 8000 rpm, however I am sure it is possible with one of the 86mm stroke J engines, the longer stoke engines might be doable with enough money, the F20C revs to what 9000, so it can be done.
 
Last edited:
I think the trick with the J35 would be, as suggested, lighten the rotating mass as much as possible by using light forged rods and SOS-style lightweight pistons. Also, a precision balance and blueprint would help with supporting 8000+ rpm. I wonder if you could use the OEM engine management (from say, a TL-S) with FIC to manage the fuel/timing requirements of the revised cams and pistons.

It is very difficult to mess with timing on the fic and you can't raise the factory rev limit. I will be using the tl type s ecu on mine with hondata flashpro. Without upgraded cams in the motor it is pointless to raise the rev limit because the power starts dropping off considerably at 6800 rpm. I will post some dynographs later.
 
Best case scenario. $500 full j engine. $ 500-$1000 custom flywheel. $400-$800 custom wire harness ecu setup. Nearly everything else can be used from the factory setup. It will still need little things like coolant lines and custom exhaust. I think once I finish one I may be able to do them fairly cheap.



Calling it a mustang or corvette is a huge slap in the face. I was tevving out my old j35 to 7200 rpm. With a basic head cam package it could easily be revved to 8500+.



It's the valve springs and cams. Putting upgraded springs and new cams would allow for it to make more power uptop realably.

I have already found several J35A3 engines in my area for about $400. Will the NSX flywheel fit on the J engine, I will re-read the entire thread to see if you tried that already I imagine you did. I have a spare NSX flywheel so I may just run it up to my engine supplier to see if it fits. Custom wire harness/ECU? I would like to see this as plug and play as much as possible. I know I have a OBD1 NSX and all the J engines are OBD2 so I will have to look into solving that issue, I know people put OBD2 engines in OBD1 cars, I just didn't pay attention as to what ECU they were using.

I think the torque of the J35 will be welcome in the NSX.
 
I'm not claiming to be an ex expert by any means. Just stateing that upgraded valve, springs and a cam that would make power that high up in rpm is definatly do able. It was a race car and probably had less than 5k miles put on it the 2 seasons I raced it. What needs to be done to a c30/32 to be able to rev and make power to 8500 rpm? I have seen more bottem end failures out of the c series than I have out of the j series. I'm sure there is more to it than just rev amount, but does the 90 degree vs 60 degree have any benefits when it comes to the strain on the pistons/rods/crank?

I think the C30/32 will do 8500 rpm now if the fuel shut off would allow it.
It's an engine configuration that is designed to rev high with acceptable stresses.
In Jon Martin's 3.8 liter engine the stroke was only 88 mm, so piston speed at 8000 was much less that a J35 or J37 with their longer strokes.
Jon said he felt the biggest challenge to overcome making power with the C30/32 was improving the ability to get more air/fuel in but rpm was not an issue.


No question improving an engine's ability to breath will increase power and much work has been done to make power in this area alone.
If we look back at all the work done to small block Chevy's over the years, the first things builders did was install more cam, large valve better flowing heads, and stronger valve springs, all designed to increase the engine's ability to pump more air/fuel mixture and rev higher without valve float. After those gains were made then work might be done to reduce reciprocating mass and beef up the bottom end. But if a built 5 liter (4"bore x 3"stroke) Chevy small block could be revved to 8000, piston speed was only 20 m/s- less than our C30/32.

Perhaps the reason more C30/32 engines fail is because builders are adding stress through forced induction or longer stroke causing the engine to run closer to its design limits.
I'm not aware of a large number of stock or near stock C30/32 engines failing unless they have had significant mods. but I'm not an authority on this for sure.

My understanding is a V6 engine can't be balanced because of the odd number of cylinders on each bank. I've read however that 60 degrees offers a better balance than 90 degrees. Our C30 90 degree has split crank journals to mimic a 60 degree firing order.

Your J35 at 7150 rpm has the same piston speed as the C30/32 at 8500 rpm.
I suspect the C30 has a lower reciprocating mass as well
If you could get your J35 to 8000 rpm through breathing and valve spring improvements you will have racing engine piston speeds. At the 8500 you say is possible with breathing you are at NASCAR piston speeds.
At 8000 rpm or more it would seem you would need to do what Honcho suggests and reduce reciprocating mass as much as possible because of those very high piston speeds and stresses inside the engine.
Probably better to go with a lower revving boosted J35/7 ,than trying to make a low rev engine design into a high revving engine
 
Last edited:
You want the J37A4 from the 2009 Acura TL SH-AWD model.

Quoting http://en.wikipedia.org/wiki/VTEC

As popularity and marketing value of the VTEC system grew, Honda applied the system to SOHC (Single Over Head Cam) engines, which share a common camshaft for both intake and exhaust valves. The trade-off was that Honda's SOHC engines benefitted from the VTEC mechanism only on the intake valves. This is because VTEC requires a third center rocker arm and cam lobe (for each intake and exhaust side), and, in the SOHC engine, the spark plugs are situated between the two exhaust rocker arms, leaving no room for the VTEC rocker arm. Additionally, the center lobe on the camshaft cannot be utilized by both the intake and the exhaust, limiting the VTEC feature to one side.

However, beginning with the J37A4 3.7L SOHC V6 engine introduced on all 2009 Acura TL SH-AWD models, SOHC VTEC was incorporated for use with intake and exhaust valves. The intake and exhaust rocker shafts contain primary and secondary intake and exhaust rocker arms, respectively. The primary rocker arm contains the VTEC switching piston, while the secondary rocker arm contains the return spring. The term "primary" does not refer to which rocker arm forces the valve down during low-RPM engine operation. Rather, it refers to the rocker arm which contains the VTEC switching piston and receives oil from the rocker shaft.

The primary exhaust rocker arm contacts a low-profile camshaft lobe during low-RPM engine operation. Once VTEC engagement occurs, the oil pressure flowing from the exhaust rocker shaft into the primary exhaust rocker arm forces the VTEC switching piston into the secondary exhaust rocker arm, thereby locking both exhaust rocker arms together. The high-profile camshaft lobe which normally contacts the secondary exhaust rocker arm alone during low-RPM engine operation is able to move both exhaust rocker arms together which are locked as a unit. The same occurs for the intake rocker shaft, except that the high-profile camshaft lobe operates the primary rocker arm.

The difficulty of incorporating VTEC for both the intake and exhaust valves in a SOHC engine has been removed on the J37A4 by a novel design of the intake rocker arm. Each exhaust valve on the J37A4 corresponds to one primary and one secondary exhaust rocker arm. Therefore, there are a total of twelve primary exhaust rocker arms and twelve secondary exhaust rocker arms. However, each secondary intake rocker arm is shaped similar to a "Y" which allows it to contact two intake valves at once. One primary intake rocker arm corresponds to each secondary intake rocker arm. As a result of this design, there are only six primary intake rocker arms and six secondary intake rocker arms.

- - - Updated - - -

Attached pdf shows all the J37A4 engines available throughout North America right now, as reported by car-part.com, including a nice one right near me in Baldwin Park, CA. They are all around $2000 and up.
 

Attachments

That engine is produced in Ohio, in ...

the Anna Engine Plant, Honda’s largest engine plant in the world, produces fuel-efficient V6
and four-cylinder automobile engines for a wide variety of Honda and Acura passenger cars and
light trucks. The plant’s annual production capacity of 1.18 million engines means it produces enough
engines to stretch, end-to-end, from Cincinnati to Cleveland and back again. The Anna plant also
manufactures driveshafts, crankshafts, camshafts and cylinder sleeves.​

http://ohio.honda.com/pressroom/files/2012_Driving_Force_in_Ohio_Brochure.pdf

Maybe Honda will decide, one day, to offer an engine retrofit kit for all versions of the NSX.
 
I have parts coming in monday. I will post more progress pics next week as more testing goes. Im very close to haveing it figured out. I will update when they come.

new goal. make nsxpo with the car.
 
That engine is produced in Ohio, in ...

the Anna Engine Plant, Honda’s largest engine plant in the world, produces fuel-efficient V6
and four-cylinder automobile engines for a wide variety of Honda and Acura passenger cars and
light trucks. The plant’s annual production capacity of 1.18 million engines means it produces enough
engines to stretch, end-to-end, from Cincinnati to Cleveland and back again. The Anna plant also
manufactures driveshafts, crankshafts, camshafts and cylinder sleeves.​

http://ohio.honda.com/pressroom/files/2012_Driving_Force_in_Ohio_Brochure.pdf

Maybe Honda will decide, one day, to offer an engine retrofit kit for all versions of the NSX.

Retrofit/Swap kit….hummm..

Parts list..

Well the J-series will bolt to the NSX trans, so pick up a J-series of your liking. I like the idea of the J35 with the Type-S cams. Interestingly, Honda uses the J35 in their Outboard V6 motors and they use the K-series in their 4 cylinder outboard motors.

Based on part # 90429-PC6-900 it looks like Honda used the same bolt pattern on the business end of the crank for all B, C, D, F, G, H and J series engines, so finding a flywheel/clutch combo that will bolt up to the J and fit inside the NSX trans should be possible, I am highly optimistic that the NSX flywheel/clutch will bolt up.

Front motor mount, Rouge has this pretty much figured out.

Passenger side motor mount, I think an adapter could be made to go from the J block mount to the stock NSX mount.

OBD2 to OBD1 adapter for older NSXs

Exhaust adapter, I could see an adapter pipe going from the J35 head to the stock NSX cats, so you could use any of the NSX exhausts.

I know this seems overly simplified, but this just may be one of the most viable swaps/retrofits around.

- - - Updated - - -

I have parts coming in monday. I will post more progress pics next week as more testing goes. Im very close to haveing it figured out. I will update when they come.

new goal. make nsxpo with the car.

Ahh, the agony of waiting.
 
Always use the ECU that came with the donor engine, and in this case it is the TL's ECU and that is a OBD-II.

Then, you can get it serviced at any Acura dealership, since the Honda Diagnostic System will recognize the ECU as coming from a 2009 TL.
 
A huge upgrade for auto nsx owners would be to get a modern 6 speed auto tranny. Is that possible?
 
Back
Top