DSMTalk Forums: Mitsubishi Eclipse, Plymouth Laser, and Eagle Talon Forum banner

4G64 swap

255K views 470 replies 125 participants last post by  rself 
#1 ·
Has anybody done it? Seen it? Heard of it?

I'm about to do it.
 
#2 ·
With the availability of 4G63 SOHC's and the advantage of running almost half a liter more displacement, why not? 1.5mm more bore and 12mm more stroke on every piston.

The swap is actually very easy. The only aftermarket parts that will be needed are pistons, and maybe rings...I've even figured out what you need to use to use your 4G63 DOHC head on the 4G64 2.4L block and, of course, 1G big rods.

So, in reality, what's not to like about the possibility?

More displacement, More Torque, No Crankwalk.



[Edited by awd92gsx on 12-06-2001 at 08:46 AM]
 
#3 · (Edited by Moderator)
The swap is actually very easy. The only aftermarket parts that will be needed are pistons, and maybe rings...I've even figured out what you need to use to use your 4G63 DOHC head on the 4G64 2.4L block and, of course, 1G big rods.

There's your answer. Yeah me and my buddy are going to put one in his 92 Laser RS. The only thing we need to work out is the compression with the DOHC head on it. We don't want to end up with a 13:0-1 ratio or a 5:0-1 ratio. Have to do some messuring and math in the near future.

Clinton
1990 Talon Tsi Awd @20 psi
 
#4 ·
Putting one in a 1G is going to be a little different...

There are no engine mounts on the block on a 2G 4G64, which means you have to use one from a truck or an older Galant, or maybe even a Dodge Caravan...

However, the older 4G64's don't have the right kind of timing belt tensioner, so something else will have to be done.

I actually have a few ideas about that one, too...but need to check some things first...
 
#6 ·
Which gasket are you referring to? The head gasket? You can use the head gasket from a '95 Galant GS DOHC 2.4L. I have compiled a list of the parts that a N/T Spyder or Galant needs to do the DOHC conversion.

http://www.geocities.com/awd92gsx/4g64swap.html

I've just got to do the list for doing the 4G64 into the 2G turbo conversion now.

The 4G64 block is, to me, a better proposition than a stroked 4G63. If nothing else, the 4G64 has 1.5mm more bore on every cylinder and I've never seen a 4G64 crankwalk...
 
#7 ·
awd92gsx said:
The 4G64 block is, to me, a better proposition than a stroked 4G63. If nothing else, the 4G64 has 1.5mm more bore on every cylinder and I've never seen a 4G64 crankwalk...
I've never seen a 4G64 with oil squirters, or with a knock sensor. Also I've never seen a 4G64 run 10s with the stock block. And last time I checked it wasn't the 4G64 that was rated as the 2nd strongest stock 4-cylinder engine.
 
#8 ·
That's because you didn't realize the 4G64 and the 4G63 are based on the same casting and having oil squirters put in is a simple task for any competant machine shop, besides, are oil squirters really needed? How many fast cars are out there that don't even have oil squirters?

You also didn't realize the 4G64 has all of the holes that are needed to run a knock sensor and halfshafts...

Try as hard as you might, there is no replacement for displacement.

A 2.4L with all the same mods as a 2.0L will be just as strong, make more power, and be more reliable in the long run.
 
#9 · (Edited by Moderator)
Oil squirters help a lot on a street driven turbocharged car. You should double check about the hole for the knock sensor, because even the NT 4G63 block doesn't have that hole. I don't see why they would make one for the 4G64. Also, the equal length half-shafts really don't help as much as some people say it does. But if you got some laying around you might as well put them in.

Now as for your 2.4L being more reliable in the long run. That is very doubtful. But putting that aside, that engine has some nasty stroke, and as such has some nasty rod angles (even worse than Hondas). Also because of the long stroke the piston velocity is tremendous at high RPMs. All of this, simply put, means there's no revving this "badboy" past redline. Or else your project will be cut short VERY quickly.

Learn a little bit about engines and you will see that the 4G63 setup is very nicely matched for a turbocharged engine. Hence the reason why you can put so much power through them.
 
#10 · (Edited by Moderator)
There are many street driven turbocharged cars out there without oil squirters, but in any case, putting squirters in the 4G64 block is very easy (remember, they are the same block, one is just milled 6mm more than the other and has a few more holes drilled through the deck, but, these are only oil drain holes from the head to the pan and are not pressurized.)

The Equal length halfshafts are an absolute must if you have an AWD. If you are to put it in an AWD it must have the holes for the driver's side axle bracket (which it does)

With .4 more litre of displacement, you shouldn't have to hit as high of an rpm to make the same power as you would in a 2.0L. If you look at a 2.6L you'll find it has the same stroke issue that the 2.4L has, but it has more than proven itself to be able to handle extremely high horsepower figures (Sakura). Even chevy guys understand the benefits of added stroke, hence the 350 to 383 conversion.

There are no proven fixes for 2G 4G63 crankwalk. After your crank walks a couple of times I'm sure you'll see the benefits of running a '64. Do a little more homework on the 4G64 and you'll see it has no reliability issues that the 4G63 has.
 
#11 ·
beerglass, you quoted the post exactly as i read it.. and still you didnt point anything out that points to how he beefed up the block? i am not worried about the rods or pistons, i was thinking about the actual crankshaft bearing caps getting busted out from 350+ hp when they were made to handle alot less.. that was my point thats all..
 
#12 ·
The caps are the same caps that are used on the 2G 4G63.


The ONLY differences between a 4G64 (2.4) block and a 4G63 (2.0) block are:

1) The 2.0 block was shaved 6mm more during manufacturing
2) The 2.4 had the holes for the oil drain tubes drilled through, whereas the 2.0L does not, although the tubes are in place
3) The 2.0L has the holes drilled and tapped for the oil squirters, although the 2.4L has the spots there to do it, so it's an easy mod, although, if I were doing that, I'd just do 1G squirters.
4) The cylinders on the 2.0L are 85mm in diameter, the cylinders on the 2.4L are 86.5mm in diameter.

As far as the block goes, that's it...the other differences are:

1) Different rods, pistons, and crank but the rods from a 1G will fit on the 4G64 crank

2) SOHC 16 valve cylinder head (a tiny few 95 Galants received a DOHC

Structurally, there are no differences in the 4G64 block than in the 4G63 block, but, for some reason, no 4G64 has ever crankwalked. I've heard somewhere that 4G63's tend to do that ;)
 
#13 ·
Compression Ratio's?

As Clinton said, we're thinking of giving this a go. My main concern is the compression ratio.

I'm not 100% sure that simply stating that one needs 1G pistons that are 1.5mm overbore is going to cut it.

Now, I'm no engine designer, but I'm pretty sure that compression ratio is made up of 2 (maybe 3) things.

1: Combustion Chamber Volume (cc's, usually)
2: Piston Dish Volume (cc's as well)
(Possibly) 3: Bore and Stroke

I'd think the bore and stroke wold have something to do with it as well. Think about it... you have x amount of air/fuel being compressed with 1997cc's (4G63), while the 2350cc's (most 4G64's) would have substantially more volume per cylinder. The volume of a 2350cc engine may be trying to be compressed in soemthing sized for 1997cc's... Then again, I'm fat and ugly, so I could be talking out of my ass. I'm simply not sure.

Anyone have some formulas for deriving compression ratios? I'm quite interested in this.

On another note, older Hyundai Sonata's came with a 4G64 (~92 or so). I know a couple guys have swapped 4G63's into Sonata's, so I'd feel that a 4G64 out of one of those would drop into a DSM. I'd figure they'd have the same mount locations, etc. Also, the availability of Hyundai's in boneyards is quite frequent.

Imagine... a DSM with "Powered By Hyundai" on the back of it. :D
 
#15 ·
Talk to someone at Doug's Dynopower, they've been turboing the 2.4 for a while, they'd prolly know they specs for 8.5:1 compression for the SOHC and DOHC.
 
#17 ·
This is very long and probably a bit confusing. Read it a couple times before you reply.

You guys (with one exception ;)) are forgetting one very important thing, GEARING.
Remember gear ratios x the final drive ratio affect how much force (torque) the drive wheels get. Higher gears equal less force (torque) at the wheels. The longer you can stay in lower gears the better, as this will allow you to accelerate faster.

HP and torque cross at 5252rpm after that HP increases with RPM, UNTIL the torque drops faster than the rpms are rising. Once torque and HP drop your rate of acceleration in that gear will decrease, so you'll shift. As we know already once you grab that higher gear the amount of force put down at the drive wheels will decrease.

So if you can figure out how to keep the torque curve on its "plateau" up to a higher rpm than you can rev higher while making more hp. That will let you stay in a lower gear for a longer period of time. Which will allow you to get more WORK done.

I took some numbers from a dyno chart of a 99GSX (dre). It's major mods are a stock 1G head, crower cams, stock bottom end, FP green, large spearco FMIC, usual fuel mods, PROefi stand alone. This pull was at 20psi on 92 octane pumpgas:
http://www.twingles.com/ideck/datalog/dyno092101b.jpg
Granted it's not the best dyno chart in the world due to something causing the choppy torque/hp curves, but it's enough to get our job here done.

It makes 345 foot lbs of torque at around 5000rpm and it starts to drop rapidly at around 6,800rpm. This netted 401hp at 6,800rpm.
Now let's say that dre did something to keep the torque curve from dropping until 9,000rpm (hard I know and the required mods would probably raise the torque peak, change it, and this is all going to be impossible on pump gas. Bear with me, this is all theoretical).

Horsepower * 5252

Torque = -----------------

RPM

If we plug in the numbers, 345 foot lbs of torque times 9000 all divided by 5252 we get ~591hp.
So we gained 190hp and allowed ourselves to rev another 2,200 rpm. That allowed us to stay in gear a lot longer, so it increased the total amount of WORK done in a given time frame (remember if you shift earlier you decrease the amount of force at the wheels, which will decrease the amount of work done than if you were in a lower gear).

Now let's make up some mythical numbers for our 2.4l engine. Let's say that with it our torque peaked around the same rpm, but a much higher torque value, 425 foot lbs. BUT we can't rev over 7,500 rpm due to the rod angles and piston speed of the 4G64 at high rpms, even though our torque curve might not be dropping.
When we plug in the numbers at our 7,500 rpm redline we get:
571hp. So in the end we are making LESS peak hp and shifting earlier, decreasing the amount of work we could get done in a given time period.

Less work in a given time period will mean it takes a longer period of time to move a mass a certain distance. If you do less work it is going to take longer to move 3000lbs a 1/4 mile.

So in other words our super torquey 2.4l will be slower than our hyper strung 2.0l.

Given the right valvetrain modifications the 4G63 can rev to 9k, although I do not have dyno charts to show if torque falls off well before that rpm or not.

A short word on torque vs hp though. Torque is all you feel when you drive, hp doesn't affect how hard your car "pulls" at all. The 4G64 powered car would feel a hell of a lot faster than the 4G63 powered car on the street, BUT it wouldn't actually be any faster. If you just want your car to slam your head into the seat the 4G64 would be worth it for sure, but if you actually want to be faster it might not be in the end.

Another thing to remember is that torque is multiplied by your gear ratio. 400 foot lbs of torque is going to end up being a shitload more going through your tranny in 1st and second than 345 foot lbs is. You're going to kill second gear syncros like mad.

I can't see this ever making the car faster in the 1/4 mile. Yes if our mythical 4G64 setup made 413 foot lbs of torque it's peak power would match that of the 4G63s. BUT it would still require you to shift 2,200 rpm lower and the amount of work done would still be less for a given time period.

You need to rev higher, not make more torque.


The replacement for displacement is HIGH RPM TORQUE AND GEARING! :)
 
#18 ·
Theories are often overshadowed by the real world.

In theory, a bumblebee can't fly.

Having more horsepower and torque doesn't kill synchros, especially if you actually lift while you shift.

We don't really know what the rpm limit of the 2.4L is, a bit more research can be done on this, but, remember there are quite a few engines out there with the same rod angles (Sakura for example, which runs the same rod angles as the 2.4L) that are making as much, if not more, horsepower and torque than the 2.0L and are runner much faster in the 1/4 mile than any 2.0L at the moment...

We have seen one 2.4L crank crack after hitting 9500 rpm, but, at that rpm, you can't really shift anyway and totally defeats the purpose of running more stroke.

If you look towards the domestic crowd and compare a 383 to a 350, you'll find that a built 383 will (most oftentimes) out run a 350. And the same usually applies to a 454. With everything else being equal, the larger displacement engine almost always wins, even with a somewhat lower rpm limit.

But, with all of that put aside, there are still advantages to running a 2.4L. If nothing else, crankwalk prevention and the low cost and high availability of the 4G64 block would make me want to run one in a 2G.

It might not be the "end all" solution, but, it never hurts to try. If you recall, quite a few years ago, people were claiming you couldn't run a 20G on the street because of the "increased lag" and that anything much bigger than a 20G would actually run slower on the track.

There's a couple of local cars that are in the process of gathering the necessary parts for the swap. One car ran a 12.6 on a completely stock bottom end with a 20G, but usually pulled consistent 13.1's. His engine blew (tried to use NOS, once), so lemme get the swap done on his car and then we'll pull some more numbers and see if there are any performance losses or gains.
 
#19 ·
awd92gsx said:
If nothing else, crankwalk prevention and the low cost and high availability of the 4G64 block would make me want to run one in a 2G.

It might not be the "end all" solution, but, it never hurts to try. If you recall, quite a few years ago, people were claiming you couldn't run a 20G on the street because of the "increased lag" and that anything much bigger than a 20G would actually run slower on the track.
Yes I can see how you might want to do if you got crankwalk on a 2g. 6 bolts are getting harder to find now. It would probably be fun on the street.
I just don't think that someone with a perfectly good 4G63 should be building up a 4G64 because they think they are going to run some amazing time.

I remember those days, pretty funny.
 
#20 ·
Ok, I have some questions that go along with this post. Lets say you can take a 2.0 block and stroke it, could you possibly stroke the 2.4 to say like 2.7. I have been comtemplating this combination, and noticed this post. I have been researching the idea of a 25G turbo, and if you could get a 2.7 out of a 2.4 and actually be able to put the 2.0 head on, you would have a rocking idea for some major horsepower. The thing is, how reliable would that motor be under that amount of force? Any ideas?
 
#21 · (Edited by Moderator)
Why run the 2.4L? More useable torque across the entire power band will result in a better driving car. No Crankwalk means you actually get to drive your car.

The entire buildup can be done for about $1300, maybe $1400.

There is NO replacement for displacement. With everything else being equal, the bigger engine will always produce more power.
 
#22 ·
Nick 92 TSi AWD said:
The replacement for displacement is HIGH RPM TORQUE AND GEARING! :)
I agree that everything you said is factually correct. In fact, this is the Honda formula. Just look at the S2000 with the 8900rpm rev limit. If a Stang could rev to 8900rpm (and had the cam and intake to support airflow at that speed) it would make 400hp.

However, the question still remains, would a DSM driveline survive at those engine speeds? Would you even be able to shift at 9000rpms?
 
#23 ·
Yes if you build the tranny with the right components. Shepard makes around 30 passes shifting at 9k before rebuilding his tranny.
But everything has his limits, if you're trying to get a mid 11 when your traps are only 116mph of course you're going to kill your tranny.

Do remember though that the whole speel I wrote up is only talking about things in a drag context. The 4G64 will make for less lag and better transient response. It would be a good idea for autoX and roadracing.
 
#24 ·
after what nick said I can not say a word :D well done. Aboiut the NT block ya it will hold but no oil jets will lead to top ring wear. not a HUGE problem but will shorten the motors life but it sounds like a slap happy motor any way. Can't beat it for the money and that is why we love it. I am on the same ground that size is not the answer and that RPM's and more boost is. finding the ballance is the thing. As most honds are going to the 1.8 and stroking down to 1.6 cranks for RPM's ( some said to be in the 10K range ) and a TON of boost ( 35+PSI ) are getting it done. The block in the race cars and the blocks in streey cars are two different things, the block in Sakura is all we know. w do not know what crank is in it and many lie :D but it's all good. it's weight and gearsXpower that we all crave and some would like to do it in other ways. Jim Z was the fist that I know of that did the 2.4 crank in the 4G63 and reported a "better down low" but I never saw numbers, but remember that they went "bad" But plaged with trany probs is the last I have heard.

Best of luck Bill and let us know how it goes.

BM-
 
#25 · (Edited by Moderator)
A 2.4 is basically a stroked 2.0. The rod angle are already getting towards a point where you don't want to go nd this engine shouldn't be further stroked out. Why a 25g? a 20g goes 10's. A hybrid(RED/GREEN) can go faster, whats the point?
 
#26 ·
I just don't think that someone with a perfectly good 4G63 should be building up a 4G64 because they think they are going to run some amazing time.

Sorry I dont think this swap is for running some amazing times. Someone is out to help the 2g guys with this Crankwalk thing and he gets ridiculed. I wanna see someone else try come up with a fix.


I would rather build a 4g64 while I still have a prefectly good 4g63 so when the time comes that I do get the walk. All I have to do is pull it out and put in the 4g64 that DONT get the walk. It would seem to be a easier swap than a JDM 6 Bolt.
Considering you have to do a lot to put a 1g motor in a 2g. The 4g64 block and 63 head are already set up for the 2g's wiring harness. I'd rather do that than build a seperate 4g63 that will likely get the walk also.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top