FYI You do not want to cut into the head gasket for coolant flow, Subaru did not make the passages smaller for shits and gigs, that is a restriction for thermal convection and turbulent scavenging. Coolant flow does not equal coolant efficiency. You always want the cooling fluid to be touching the hot surface for as long as possible until it has gained thermal equilibrium then rush out.
monumental video, should be a 'must see tutorial' video in every subaru workshop to understand the problem. i removed thermostat just for the sake of experiment of free flow and filled the system with plain distilled water to measure cylinder 4 wall temperature with FLIR temp cam and found out that while system is showing 75celsius water temp the cylinder 4 walls measure up to 175celsius!! other three cylinder walls were around 90celsius! seems like ej20 series have bigger water chambers due to smaller displacement and that is enough to keep it running solid but 2.5 engines with just those few milimeters thinner cylinder and gasket proportions can not manage that design flaw!
Interesting take on piston 4 failures. From what I've come across within the Subaru community, a lean mixture in cylinder 4 stemming from the OEM fuel delivery system was to blame (you did touch on this). Limited cylinder 4 cooling does make sense too though. Thanks for the thorough cooling system break down.
Thanks for the video and explanation. Why does this issue still exists 15 years into production? Yes, Subaru improved other areas of the EJ25, but the cooling "deficiency" remains and is a potential source of continued failure. And thus the aftermarket had to develop an item to rescue the EJ25 from itself. I know the FA series of engines is here, but that doesn't mean we are in the clear (valve crud is next with direct injection).
Installed this mod since I was replacing my clutch. Thanks for the info on the passages. Blowing smoke would have been a nice touch to see everything flow.
Luke, you are a legend! This is an amazing wealth of knowledge on the cooling system! I bought a WRX mechanic special (cylinder 4 misfire) and only found a burnt valve, once I took the heads off. I'm hoping the reverse cooling mod and an AOS will prevent me from having this problem again!
Thank you very much for taking the time to make this video! We all really appreciate it. I really should take one of these engines apart. Maybe later on I guess. Take care!
Thanks for all the info. I'm about to knock out a project on my Forester but right after that I'll be doing this cooling mod on my EJ205 swapped rs and the STI oil pan / Killer B pick up. Can't wait!
Saw a mod on Bucky Lasek that, to me, makes alot of sense. Basically the drivers side fan turns on first, and thst is the cold side of the radiator, also where the thermostat is. He made the passenger side turn on first, that allows for a more uniform and eficient temperature of the radiator. I'll be doing the same mod.
super thorough, thank you for taking your time to bring this video to us.. currently assembling the parts list to build an RA short to replace my ring land failed ej for my 2008 xt
Excellent demonstration. I'm looking to install this cooling mod, the Killer B water pump housing, and maybe a Koyo radiator so I only have to drain, fill, and burp the cooling fluid once.
I have an EJ25 SOHC 4 valve engine in my Gyrocopter. I no longer run the standard water outlet cross over pipe and instead run 2 seperate pipes (one from each outlet) to 2 seperate radiators of the same size and at the same height. The outlets (bottoms) of both radiators are then "Y" pieced together, so that there is no differential feed issues associated with "T' pieces, and are then fed directly back to the water pump. There are no heater circuits, just this cooling circuit. Here is the problem:- All the flow from the waterpump exits the RH side of the engine while the LH side outlet pipe (right at the block) is still cold, indicating zero flow. Restricting the flow from the RH top radiator hose results in the LH side flowing and getting to temp, so no blockages etc. I believe the real issue here is NOT the #4 cylinder cooling issue described here but an overall design flaw seeing the RH side seeing preferential flow through the block/heads (for whatever reason??). I have not yet solved this issue but I would be open to any and all suggestions. I noticed that the original crossover pipe is oddly shaped with a "sort of" restriction (flattened section) built into it which may help address the preferential flow issue from LHS vs RHS??
I wouldn’t shave the ports in the gasket. You need that amount of flow and pressure to have heat transfer from the block properly. And you cannot do it to only one side or the pressure will be too low with the opposite side. If you do both the water pressure will be too low
Thanks for sharing your thoughts on this buddy! I find this topic super interesting. I did not shave the gasket. You’re 100% right about needing to have correct flow and pressure - but it’s really about volumetric flow rate - which is a fine balance to achieve. You need to make sure the coolant has a chance to absorb the heat (i.e., not move too fast) and then be able to carry it away from the heads before things get too hot (i.e., not move too slow, either). And that’s the crux of this heat exchange problem. Does opening up the ports (assuming you did it evenly going in and exiting the heads) make the overall volumetric flow rate through the heads increase or decrease? Constructions usually accurate flow rates, think about a river and the narrow sections. You also want as much volume as possible (when your end goal is heat transfer), assuming it doesn’t flow too fast. So, from a 1st principals engineering standpoint - here’s what we need to achieve. (1) The highest volume of coolant, (2) moving through the heads at a velocity thats slow enough to allow maximum heat transfer. So open up the ports but restrict flow further down the path (if needed). For example, restrict flow at the coolant cross over pipe entry point, for example, which is exactly what some high HP Subaru engine guys are doing! Just some concepts to think about. Does this mean I’m right and you’re not - hell no - We might be on the exact same page here, just needed to dive into the nuances! Cheers buddy and thanks for checking out the videos!! 👊👊
Interesting to see if Porting out the HeadGasket coolant ports will improve block temp/coolant flow. My only Concern is that will it affect the integrity of the Headgasket? Shaving off material from headgasket? Thanks, Luke! Love the content!!!!
Great video! Here are my thoughts. The 3 ports on the water pump as 5/8" ID, added together total 1-7/8". The upper rad hose is 1.5" therefore more coolant is constantly recirculating in the engine and never seeing the radiator. The cylinder 4 mod just adds to this problem. No wonder why these engines run hot! Thoughts?
From what I’ve seen, adding the coolant mod does even out the coolant temps. Lots of serious racers with decades of experience use them, but you could be right - I haven’t do any test to verify these claims so who knows what the real data would say. Like you’re logic tho. Cheers!
Thanks for the breakdown my man. I too noticed the abnormal exhaust port path on every other cylinder. I wonder why Subaru does that... Just seems like poor flow.
I thought the reason for the dogleg exhaust port was to clear the chassis crossmember on the left side of the car. In order to do a symmetrical port head you would have to redesign that part of the car.
Hi nice video! I already installed the 4 cilynder coolant mod on my 2008 STI and after it, it reaches 101C and before that it did not exceed 98 degrees Celsius. The coolant is complete and there are no leaks. The installation process was ok but I don't know what is happening now. Do you know?
Luke really good video , honestly I thought maybe for a subaru car the Phenix kit would be better because of the install , like the lack of room to work with , the Phenix kit you just use a socket verses crows foot , or cresent ?? so I have an off road application so I probably would go with the larger id hoses etc , also I was hoping you would show the cooling differences between stock and or offroad application also the cooling for the other side.
Interesting... My piston failure was in Cylinder 1... But, thanks for the informative video! I'll have to check into the 'bypass' from cylinder 4, just in case... (ps: not first engine worked on, but first subaru...)
@@SubaruONLY Yes, actually a question on this subject: Is this only for the DOHC head, or can / should this be done on the SOHC head as well? Thank you for your time! :)
ever see what a upgraded rad does for stock cylinder temps u will be surprised biggest thing i noticed was a 30 degree c drop in temps across board once i put a duel core rad that on its own is way better then the cylinder 4 cooling mod also stock wrx/sti rad has same core size as the non turbo cars best bang for money would be a rad upgrade vs messing with cylinder 4 cooling yes i make it cool even but over all coolest temp is a bigger problem at higher power settings also by porting the head gasket u can also control flow direction so u can increase flow speed by porting the rear ports near cylinder 4
Heard people say the twin turbo models use a OEM cooling mod (same design) and tap into the same bung. Why they didn’t continue it for 205s, 255s, and 257s boggles the mind
Yep, those models do!!! Why they didn’t use it for other single turbos seems odd to me too, maybe to save a little cash, which was probably the wrong call. Cheers buddy!!! 🍻
That was a very instructive video. Definitely I'm getting a reverse coolant flow kit. Thanks. Ps - I saw a oil trace at the tmic entrance, what it means?
Is this a problem with engines that just have stage 2 mods? Or does it mainly happen to people trying to push big horsepower? I have a 2006 WRX with 175,000 miles with a downpipe and custom tune. My engine still has great compression numbers. I was curious if I was just a lucky one.
Did I call it a motor or something? I mess that up all the time. Glad to see you checked out the video buddy - love your channel and I always appreciate your advice!!! Thanks and cheers buddy!!! 👊
I wonder how orienting the t for the cooling mod affects it, if any. I set mine up with the coolant line coming from the water pump going straight through the t with the 90 degree part of the t going to the port in the head. I think it's a much cleaner approach, but the way you put the t probably allows for coolant to pass through to the back of the head more easily. Any thoughts?
lol. Yes, I actually ended up orienting it the way you describe for the final install - seemed to line up the hoses better. But I agree with the logic for sure, still think a y-shaped fitting might be better. But just having the mod is the biggest thing! Cheer man! 👊
I did not. But good idea for sure. I actually did go down this rabbit hole a little, but for a slightly different reason. Was curious if I could use some numerical flow models to simulate the flow rates and dynamic heat transfer gradients. Still something I think about pulling together, but ended up too complicated for the time I had available (back then). I’ll still like to pull this off! It would be a major R&D series that I want to see, but not sure if many others would really find it entertaining at all. TBD!
@@SubaruONLY yeah it would be cool to know the cc difference between each exhaust port. It would certainly be a good factor to take into consideration in being able to create an efficient exhaust manifold for the ej engine. Its certainly a volume id want to factor into building a manifold if I knew it haha
This cooling mod might be causing more water to flow around cylinder 4, but why channel that water back to the water pump, when it would have flowed through the radiator? Seems like it could just cause bigger problems.
I can see this mod stealing too much flow for cylinder 2 if the head fitting is to large, then as you pointed out which i didn't know was how the oem knock detection works. So now cylinder 4 is not really setup for early knock detection which maybe a better thing anyway. I think the oem pistons and boost are more of a problem. I got 400whp on petrol 98ron AU and egts and closed deck and the cylinder temps are all within 10 degrees going down the drag strip. On the 2 step im at 900 degrees C so i heat it up.
Thanks for another great video it was really informative. You didn't say where the last bit of the t pipe goes can you help please. have you got a video on pcv as I went to delete that and getting confused
I would not open up the gasket ports to match. The diameter reduction could be intended to work like orifice plates ( reduce diameter to increase flow velocity) opening the ports will slow down flow increasing resistance time coolant sit inside the head, not good. The kits are theoretically a good idea but there need to be some kind of thermodynamics Modelling to determine the correct hose side diameter, again you mentioned a larger Diameter will increase flow, which I believe is not a better design without Empirical data. Larger flow with the mod will stagnate heat transfer from other areas of the engine causing heat soak in other areas of the engine. This “can “ be a solving one problem but creating another elsewhere.
Just experience my 21 sti being heatsoaked today. Car felt super sluggish, slow throttle response, dropping to 600-700rpm. After finding out its normal for hot weather, the 4 cylinder mod sounds about right... a
One question. I use Subaru platform EJ251 with reverse manifold option to swap Volkswagen T3 vanagon . The problem related to cylinder 4 was same not matter ? What do you recommend …. Thanks in advance
would this mean that the cooling mods are recycling hot uncooled coolant because they go through the heatercore and straight back into the chambers? looks like they are never are cooled by the radiator? why route the coolant to the heater core instead of to the crossover or outlet towards the radiator to be cooled? thanks
No, it still goes to the radiator… when the coolant goes to the heater core, the next place is back to the crossover pipe, then over to the radiator inlet at the top. Cheers!!!
I made my own for 20 bucks. XD. Order this, you will have to grind down the 2 barbs to make the oem coolant hose barely fit over. 3/4" in (20mm) T Aluminum Heater Hose Barb Water Coolant Fitting Adapter Metal. And this Adapter Coolant Barb Kit M20x1.5mm to 20mm Barb. Then go to napa get heater hose, harbor freight, get hose clamps call it a day.
Where doesn’t does that line connect to? Is it going back to the lower inlet on the heater core.. then outlet from heater core line going back to the other line on the right facing back of engine?
It seems like it would prioritize cooling the head and slow down flow around the cylinders. I have always been on the fence on this mod. I wonder if drilling that air bleed hole in the headgasket on the top right side of cylinder 4 would be better to allow more flow in that stagnant pocket. My motor melted piston 3 "without the piston 4 coolant mod" from my poor tuning, too much compression and too much boost. Piston 3 has the turbo coolant passage out of the head.
Great video! very helpful, by the way, I have the FB20 engine 2.0 in a 2012 Subaru,6sp auto.dual trans ,..got just over a hundred k.runs great, is this mod 4 kit good for my car,..?? p.s. Enjoy a little spirited running on highways..
Hey Luke, on my ej251 the passenger side cylinder head has no exit of water other than out the top of the block. In other words... both cylinder heads are the same. Couldn’t one also do a reverse cooling mod for the passenger side head as well? Since there is also a plug on that cylinder head. Not that those cylinders run hot like #4 but still it should help? Unless coolant has a completely different flow pattern on the passenger side cylinder head?
I didn’t for several reasons (however, still seriously considering this for a future build) - flow rate is also variable and although opening up those ports lets more volume in, some engine builders have shared their R&D with me and making sure the coolant doesn’t pass through the heads too quickly is a major issue (basically, you want the coolant to pull as much heat out of the system as possible and if the coolant passes through too quickly, it won’t have a chance to absorb the heat and carry it away), so there are more factors at play than I originally mentioned in the video. More to come on this as I dig deeper. Thanks and cheers! 👍
Got it. I've still been thinking to do it. As I plan to run Evan's waterless coolant. Yet I'm still waiting to beable to afford my v8 ej207. I have quite a few things I'm considering for my build. I'm not building the block but considering everything else.
I just finished watching this and really feel I gained a lot of understanding of what is going on with cooling in my engine. I do have one question/idea. From my understanding, the hottest coolant is pulled out of the block at cylinder number 4 and then routed directly back into the water pump and straight back into the block. Is this correct? Is there a way to plumb this from out of the block at cylinder four and into the intake for the radiator to cool the fluid before returning it to the pump and the block?
Great video and i too had the same thinking as yours, even if it's there to create an extra low pressure exit on #4 could it not be rerouted back to the rad instead of WP or maybe like he said the connection at the WP is a low pressure suction so maybe thats why. Pity no reply, this guy knows his subaru.
For some reason my OEM thermostat it's not open, there it's not leaking, no air pockets, no head gasket issue (I guess) but some times the thermostat won't open, I have to put in parking, then press the gas pedal until reach 3-3.5k rpm and for some reason the coolant starts flowing through the cooling system... Any idea? This is the first well explained cooling system flow process I seen, Thank you for your time for it..
I’d start with a new thermostat for sure. Make sure you orient it so the little bleed hole is on the upper side, that way air pockets can pass though. Should be good after that! Good luck! 👍
Have a look at the toyota 4age 20v coolant flow it will make u sick going bk to the Ej , the bring the colder Copland in from the rad across all the intake then down into the pump back up through the block and across the exhaust valves bringing the heat out to the rad , helping cool the intake all the time , The ej is not a reverse coolant mod it’s more an assist
Good video mostly. And I took you pretty serious up until the point you said you were going to cut your brand new head gaskets 🤦♂️ ... can you say NEVER!!! and no I wouldn't cut them I would just buy the upgrade version of the gaskets with those holes already made bigger 🤷
Can you tell me what the thread pitch of the M20 coolant plug (at 36:15) on the back of the driver head is? I'm making my own reverse cooling kit, and my M20x1.5 to AN adapter doesn't seem to fit right.
Just an idea, when you are talking about opening up the head gasket for coolant flow, you could open up only the cylinder 4 ports to allow more flow through that side, but leave cylinder 2 holes a normal size. Could help number 4
What about the coolant flow, of the cylinder #1 & #3? So time much was spent on the #2 & #4, is there the same issue with the other side? Why not do the mod with the other cylinder head?
Hello Luke, first of all, thanks so much for your videos, very informative and a great help! Got a question about coolant flow when removing the OEM oil normalizer/cooler. Do you know if deleting this will cause a problem with coolant flow? I have an aftermarket oil cooler and don't have an OEM oil normalizer. Thanks Luke!
@@SubaruONLY Thanks Luke, maybe if you have the time, you could make a video that shows how the coolant expansion tank works on the side of the intake manifold? I read it cools the turbo by the process of convection when the car is turned off and cooling down.
Need your help wizard.. my 05 sti is sputtering under boost only.. fuel pressure is good.. coils are new and tested.. new ngk sparks.. new oem fuel injectors.. compression is fine.. leak down is fine .. took engine apart rebuilt it same.. tested all pig tails on harness and fine.. vaccum is fine.. no boost leak.. idles 100% fine please help
Good question and I’ve thought the same. It helps bring down the coolant that is being recirculated in the system - BUT if there is a stagnant pocket of coolant that gets trapped in the back of the rear left head, that coolant isn’t getting cooled by the radiator, so no help for that issue (which could be the big problem). So I recommend both for best results!
![Shareปลาวาฬ ส่งท้าย SS1 พาเปียโน ขึ้น Ferrari ก่อนไปขาย แลก Nissan l [Nickynachat]](http://i.ytimg.com/vi/KAknGkTNIvg/mqdefault.jpg)
FYI You do not want to cut into the head gasket for coolant flow, Subaru did not make the passages smaller for shits and gigs, that is a restriction for thermal convection and turbulent scavenging. Coolant flow does not equal coolant efficiency. You always want the cooling fluid to be touching the hot surface for as long as possible until it has gained thermal equilibrium then rush out.
This is the answer I was looking for, engineer?
Nice. Didn't think about that.
Are you sure?
Increased flow always means increased heat transfer. Sorry
@@martinburnett3852Ya but if it's not even touching and/or continuously touching the hot surface how r u supposed to get maximum cooling?
monumental video, should be a 'must see tutorial' video in every subaru workshop to understand the problem.
i removed thermostat just for the sake of experiment of free flow and filled the system with plain distilled water to measure cylinder 4 wall temperature with FLIR temp cam and found out that while system is showing 75celsius water temp the cylinder 4 walls measure up to 175celsius!! other three cylinder walls were around 90celsius! seems like ej20 series have bigger water chambers due to smaller displacement and that is enough to keep it running solid but 2.5 engines with just those few milimeters thinner cylinder and gasket proportions can not manage that design flaw!
Whoa - this is awesome little experiment you did man! Crazy how much higher #4 was - thanks for sharing! Very cool. 👊
Interesting take on piston 4 failures. From what I've come across within the Subaru community, a lean mixture in cylinder 4 stemming from the OEM fuel delivery system was to blame (you did touch on this). Limited cylinder 4 cooling does make sense too though. Thanks for the thorough cooling system break down.
This is probably the ultimate Subaru EJ cooling video!
Thanks for the video and explanation. Why does this issue still exists 15 years into production? Yes, Subaru improved other areas of the EJ25, but the cooling "deficiency" remains and is a potential source of continued failure. And thus the aftermarket had to develop an item to rescue the EJ25 from itself. I know the FA series of engines is here, but that doesn't mean we are in the clear (valve crud is next with direct injection).
Installed this mod since I was replacing my clutch. Thanks for the info on the passages. Blowing smoke would have been a nice touch to see everything flow.
Luke, you are a legend! This is an amazing wealth of knowledge on the cooling system!
I bought a WRX mechanic special (cylinder 4 misfire) and only found a burnt valve, once I took the heads off. I'm hoping the reverse cooling mod and an AOS will prevent me from having this problem again!
Thanks buddy. Glad these videos help!!! 👍
Thank you very much for taking the time to make this video! We all really appreciate it. I really should take one of these engines apart. Maybe later on I guess. Take care!
Love your channel you have one of the most detailed explanations on our engines.
Keep up the great 👍🏻 work.
Thanks for all the info. I'm about to knock out a project on my Forester but right after that I'll be doing this cooling mod on my EJ205 swapped rs and the STI oil pan / Killer B pick up. Can't wait!
Saw a mod on Bucky Lasek that, to me, makes alot of sense.
Basically the drivers side fan turns on first, and thst is the cold side of the radiator, also where the thermostat is. He made the passenger side turn on first, that allows for a more uniform and eficient temperature of the radiator.
I'll be doing the same mod.
Just saw that too!! Really interesting. Gonna look into that too for sure!
Dude you're an absolute gem. Your content is so good!
Ahhh, thanks man! 🙏🙏🙏
A 45 minute video worth watching! Great info!
This compilation is excellent. Thanks man
Thank you for breaking this down so clearly and thoroughly. This is a great video!
Thanks!
I got this mod, I bought my VA at 58,000mi. I'm at 10% in a leakdown test. Hopefully this mod helps when I start making big power.
That was absolutely brilliant thank you very much your knowledge of fluid dynamics is second to none well done.
super thorough, thank you for taking your time to bring this video to us.. currently assembling the parts list to build an RA short to replace my ring land failed ej for my 2008 xt
Nice… GL man!!!
OUTSTANDING video Luke! Great job! 👏
Excellent demonstration. I'm looking to install this cooling mod, the Killer B water pump housing, and maybe a Koyo radiator so I only have to drain, fill, and burp the cooling fluid once.
Nice! 👌
This video is amazing, thanks for putting all the effort into making this, I now understand the cyl 4 design flaw.
Glad it helped! Cheers!!! 🍻
I have an EJ25 SOHC 4 valve engine in my Gyrocopter. I no longer run the standard water outlet cross over pipe and instead run 2 seperate pipes (one from each outlet) to 2 seperate radiators of the same size and at the same height. The outlets (bottoms) of both radiators are then "Y" pieced together, so that there is no differential feed issues associated with "T' pieces, and are then fed directly back to the water pump. There are no heater circuits, just this cooling circuit.
Here is the problem:-
All the flow from the waterpump exits the RH side of the engine while the LH side outlet pipe (right at the block) is still cold, indicating zero flow. Restricting the flow from the RH top radiator hose results in the LH side flowing and getting to temp, so no blockages etc. I believe the real issue here is NOT the #4 cylinder cooling issue described here but an overall design flaw seeing the RH side seeing preferential flow through the block/heads (for whatever reason??). I have not yet solved this issue but I would be open to any and all suggestions. I noticed that the original crossover pipe is oddly shaped with a "sort of" restriction (flattened section) built into it which may help address the preferential flow issue from LHS vs RHS??
I wouldn’t shave the ports in the gasket. You need that amount of flow and pressure to have heat transfer from the block properly. And you cannot do it to only one side or the pressure will be too low with the opposite side. If you do both the water pressure will be too low
Thanks for sharing your thoughts on this buddy! I find this topic super interesting. I did not shave the gasket. You’re 100% right about needing to have correct flow and pressure - but it’s really about volumetric flow rate - which is a fine balance to achieve. You need to make sure the coolant has a chance to absorb the heat (i.e., not move too fast) and then be able to carry it away from the heads before things get too hot (i.e., not move too slow, either). And that’s the crux of this heat exchange problem. Does opening up the ports (assuming you did it evenly going in and exiting the heads) make the overall volumetric flow rate through the heads increase or decrease? Constructions usually accurate flow rates, think about a river and the narrow sections. You also want as much volume as possible (when your end goal is heat transfer), assuming it doesn’t flow too fast.
So, from a 1st principals engineering standpoint - here’s what we need to achieve. (1) The highest volume of coolant, (2) moving through the heads at a velocity thats slow enough to allow maximum heat transfer. So open up the ports but restrict flow further down the path (if needed). For example, restrict flow at the coolant cross over pipe entry point, for example, which is exactly what some high HP Subaru engine guys are doing!
Just some concepts to think about. Does this mean I’m right and you’re not - hell no - We might be on the exact same page here, just needed to dive into the nuances! Cheers buddy and thanks for checking out the videos!! 👊👊
SubaruONLY I couldn’t have said it better myself!
Interesting to see if Porting out the HeadGasket coolant ports will improve block temp/coolant flow. My only Concern is that will it affect the integrity of the Headgasket? Shaving off material from headgasket? Thanks, Luke! Love the content!!!!
Great video! Here are my thoughts. The 3 ports on the water pump as 5/8" ID, added together total 1-7/8". The upper rad hose is 1.5" therefore more coolant is constantly recirculating in the engine and never seeing the radiator. The cylinder 4 mod just adds to this problem. No wonder why these engines run hot! Thoughts?
From what I’ve seen, adding the coolant mod does even out the coolant temps. Lots of serious racers with decades of experience use them, but you could be right - I haven’t do any test to verify these claims so who knows what the real data would say. Like you’re logic tho. Cheers!
Correction. 5/8 is only the heater core bypass; 1/2" for the other two.
Thanks for the breakdown my man.
I too noticed the abnormal exhaust port path on every other cylinder. I wonder why Subaru does that... Just seems like poor flow.
thanks, same. Cheers!!!
I thought the reason for the dogleg exhaust port was to clear the chassis crossmember on the left side of the car. In order to do a symmetrical port head you would have to redesign that part of the car.
Ahh. Think you have a good point - THANKS! I’ll take a closer look at that for sure. Maybe a TSS fab front cross member...
Hi nice video! I already installed the 4 cilynder coolant mod on my 2008 STI and after it, it reaches 101C and before that it did not exceed 98 degrees Celsius. The coolant is complete and there are no leaks. The installation process was ok but I don't know what is happening now. Do you know?
Mate of all the special forces subie lovers out there your defiantly the best ..
Man, thanks!!! Hope to grow the channel to reach more people, we’ll see where this whole thing goes!!! 👊
Luke really good video , honestly I thought maybe for a subaru car the Phenix kit would be better because of the install , like the lack of room to work with , the Phenix kit you just use a socket verses crows foot , or cresent ?? so I have an off road application so I probably would go with the larger id hoses etc , also I was hoping you would show the cooling differences between stock and or offroad application also the cooling for the other side.
Would have liked to see some measurements of the ID on both the head fitting and the t fittings for both kits but overall great video.
Thanks mate. Bloody awesome info. I'm going to install the reverse coolant mod.
Regards Noel.
Australia
Thanks for the support - cheers! 🍻
Interesting... My piston failure was in Cylinder 1... But, thanks for the informative video! I'll have to check into the 'bypass' from cylinder 4, just in case... (ps: not first engine worked on, but first subaru...)
Sorry to hear about the engine issue but thanks for the support here. Good luck! 👍 Let me know if you have any questions!!!
@@SubaruONLY Yes, actually a question on this subject: Is this only for the DOHC head, or can / should this be done on the SOHC head as well? Thank you for your time! :)
ever see what a upgraded rad does for stock cylinder temps u will be surprised biggest thing i noticed was a 30 degree c drop in temps across board once i put a duel core rad that on its own is way better then the cylinder 4 cooling mod also stock wrx/sti rad has same core size as the non turbo cars best bang for money would be a rad upgrade vs messing with cylinder 4 cooling yes i make it cool even but over all coolest temp is a bigger problem at higher power settings
also by porting the head gasket u can also control flow direction so u can increase flow speed by porting the rear ports near cylinder 4
Thanks!! Great info, explained really well.
Thanks!!!
Heard people say the twin turbo models use a OEM cooling mod (same design) and tap into the same bung. Why they didn’t continue it for 205s, 255s, and 257s boggles the mind
Yep, those models do!!! Why they didn’t use it for other single turbos seems odd to me too, maybe to save a little cash, which was probably the wrong call. Cheers buddy!!! 🍻
The bottleneck of each fitting is the smallest part of the barb. It will only flow that much it cannot flow more.
Nailed it!!! 👌👌👌
That was a very instructive video. Definitely I'm getting a reverse coolant flow kit. Thanks. Ps - I saw a oil trace at the tmic entrance, what it means?
Prob oil from the PCV system - I’d add a good Air-Oil Separator - that should solve it!!! 👌
@@SubaruONLY ok, I thought my turbo is finish. ( 05 OBXT)
Turbo could be bad too, but the PCV system is most often the issue. Good luck 👍
Great resource! Thanks, Luke. :)
Is this a problem with engines that just have stage 2 mods? Or does it mainly happen to people trying to push big horsepower? I have a 2006 WRX with 175,000 miles with a downpipe and custom tune. My engine still has great compression numbers. I was curious if I was just a lucky one.
If you have good compression, wouldn’t worry about it. It’s more pronounced with increased boost over factory range.
Succinct and thorough explanation and demonstration.
Good teaching teacher
Great explanation… thanks!
Engine.
Did I call it a motor or something? I mess that up all the time. Glad to see you checked out the video buddy - love your channel and I always appreciate your advice!!! Thanks and cheers buddy!!! 👊
@@SubaruONLY yeah, the whole intro had me a bit triggered with the EJ motor, EJ motor, etc. Lol
I wonder how orienting the t for the cooling mod affects it, if any. I set mine up with the coolant line coming from the water pump going straight through the t with the 90 degree part of the t going to the port in the head. I think it's a much cleaner approach, but the way you put the t probably allows for coolant to pass through to the back of the head more easily. Any thoughts?
lol. Yes, I actually ended up orienting it the way you describe for the final install - seemed to line up the hoses better. But I agree with the logic for sure, still think a y-shaped fitting might be better. But just having the mod is the biggest thing! Cheer man! 👊
Strange question, did you ever measure the exhaust port volumes? Thinking for a true, equal length manifold.
I did not. But good idea for sure. I actually did go down this rabbit hole a little, but for a slightly different reason. Was curious if I could use some numerical flow models to simulate the flow rates and dynamic heat transfer gradients. Still something I think about pulling together, but ended up too complicated for the time I had available (back then). I’ll still like to pull this off! It would be a major R&D series that I want to see, but not sure if many others would really find it entertaining at all. TBD!
@@SubaruONLY yeah it would be cool to know the cc difference between each exhaust port. It would certainly be a good factor to take into consideration in being able to create an efficient exhaust manifold for the ej engine. Its certainly a volume id want to factor into building a manifold if I knew it haha
Heck yea it would. I hear ya. Someone out there gotta do this, maybe some already on it. Cheers buddy!
This cooling mod might be causing more water to flow around cylinder 4, but why channel that water back to the water pump, when it would have flowed through the radiator? Seems like it could just cause bigger problems.
Good point. I will think this through. Also we wouldnt want to draw all the cooling pressure from cyl.2.
I can see this mod stealing too much flow for cylinder 2 if the head fitting is to large, then as you pointed out which i didn't know was how the oem knock detection works. So now cylinder 4 is not really setup for early knock detection which maybe a better thing anyway. I think the oem pistons and boost are more of a problem.
I got 400whp on petrol 98ron AU and egts and closed deck and the cylinder temps are all within 10 degrees going down the drag strip. On the 2 step im at 900 degrees C so i heat it up.
interesting. good points to consider.
Luke ...... well done mate. 👍👍👍👍👍👍👍
Thanks buddy!!! 🤙🤙
Thanks for another great video it was really informative. You didn't say where the last bit of the t pipe goes can you help please. have you got a video on pcv as I went to delete that and getting confused
Seems like the kit hose is mean to be on the tail of the “T” which then taps into the heater core, not on one of the arms…
You could be right…. But I think it probably doesn’t matter. Cheers!! 🍻
I would not open up the gasket ports to match. The diameter reduction could be intended to work like orifice plates ( reduce diameter to increase flow velocity) opening the ports will slow down flow increasing resistance time coolant sit inside the head, not good.
The kits are theoretically a good idea but there need to be some kind of thermodynamics Modelling to determine the correct hose side diameter, again you mentioned a larger Diameter will increase flow, which I believe is not a better design without Empirical data. Larger flow with the mod will stagnate heat transfer from other areas of the engine causing heat soak in other areas of the engine. This “can “ be a solving one problem but creating another elsewhere.
I agree 100%, need to set up controlled experiments and run some modeling. Let’s do this! We just need a funding partner 🤔
Just experience my 21 sti being heatsoaked today. Car felt super sluggish, slow throttle response, dropping to 600-700rpm. After finding out its normal for hot weather, the 4 cylinder mod sounds about right... a
Excellent 👍
One question. I use Subaru platform EJ251 with reverse manifold option to swap Volkswagen T3 vanagon . The problem related to cylinder 4 was same not matter ? What do you recommend …. Thanks in advance
would this mean that the cooling mods are recycling hot uncooled coolant because they go through the heatercore and straight back into the chambers? looks like they are never are cooled by the radiator?
why route the coolant to the heater core instead of to the crossover or outlet towards the radiator to be cooled?
thanks
No, it still goes to the radiator… when the coolant goes to the heater core, the next place is back to the crossover pipe, then over to the radiator inlet at the top. Cheers!!!
@@SubaruONLY awesome! thank you :)
I made my own for 20 bucks. XD. Order this, you will have to grind down the 2 barbs to make the oem coolant hose barely fit over. 3/4" in (20mm) T Aluminum Heater Hose Barb Water Coolant Fitting Adapter Metal. And this Adapter Coolant Barb Kit M20x1.5mm to 20mm Barb. Then go to napa get heater hose, harbor freight, get hose clamps call it a day.
Good call man. Love to hear the ingenuity - cheers!!! 👊
Amazing video sir. Well well done
Thank you! Really appreciate you taking a sec to share that. Cheers buddy, glad it was interesting.
My heater core is damaged, so I had to loop/bypass it. Can I still run the Cylinder 4 cooling mod off the inlet/outlet heater core lines bypassed?
Yep. Totally.
@@SubaruONLY Thanks Luke! 🙏🏻
@@Mike_44 cheers and good luck! 👊
@@SubaruONLY 👊🏻 Thanks!
Where doesn’t does that line connect to? Is it going back to the lower inlet on the heater core.. then outlet from heater core line going back to the other line on the right facing back of engine?
I am wondering if this also applies to the FB engines, and if this type of kit would work on an FB20 I plan to drop into my 2013 Forester.
eye-opening video, great thought. a+++
Thank you. Cheers!
What about flowing from the heater core line to the head?
It seems like it would prioritize cooling the head and slow down flow around the cylinders. I have always been on the fence on this mod. I wonder if drilling that air bleed hole in the headgasket on the top right side of cylinder 4 would be better to allow more flow in that stagnant pocket. My motor melted piston 3 "without the piston 4 coolant mod" from my poor tuning, too much compression and too much boost. Piston 3 has the turbo coolant passage out of the head.
Can you also do the same mod for cylinder 3? It also has the same plug you can tap into or is that unnecessary?
Great video! very helpful, by the way, I have the FB20 engine 2.0 in a 2012 Subaru,6sp auto.dual trans ,..got just over a hundred k.runs great, is this mod 4 kit good for my car,..?? p.s. Enjoy a little spirited running on highways..
can this "reverse cooling mod" be done on non turbo models?
Hey Luke, on my ej251 the passenger side cylinder head has no exit of water other than out the top of the block. In other words... both cylinder heads are the same. Couldn’t one also do a reverse cooling mod for the passenger side head as well? Since there is also a plug on that cylinder head. Not that those cylinders run hot like #4 but still it should help? Unless coolant has a completely different flow pattern on the passenger side cylinder head?
Hey Luke did you end up making the head gasket kidneys larger to increase the flow? If so how did it work out for you?
I didn’t for several reasons (however, still seriously considering this for a future build) - flow rate is also variable and although opening up those ports lets more volume in, some engine builders have shared their R&D with me and making sure the coolant doesn’t pass through the heads too quickly is a major issue (basically, you want the coolant to pull as much heat out of the system as possible and if the coolant passes through too quickly, it won’t have a chance to absorb the heat and carry it away), so there are more factors at play than I originally mentioned in the video. More to come on this as I dig deeper. Thanks and cheers! 👍
Got it. I've still been thinking to do it. As I plan to run Evan's waterless coolant. Yet I'm still waiting to beable to afford my v8 ej207. I have quite a few things I'm considering for my build. I'm not building the block but considering everything else.
Those are about the most legit starting point you can use - short of going full closed-deck! Legit man, love to hear about it once it gets rolling.
Thanks! Very good video
Do you have videos on the 2001 3.0L H-6 engines. Of course, I've got a clean LL Bean Outback but needs the head gaskets.
Hi. None yet, but lots of things overlap for sure!
Is this a problem with 1990 ej22 engines?
Thank you regards awesome explanation
You are welcome!
I just finished watching this and really feel I gained a lot of understanding of what is going on with cooling in my engine. I do have one question/idea. From my understanding, the hottest coolant is pulled out of the block at cylinder number 4 and then routed directly back into the water pump and straight back into the block. Is this correct? Is there a way to plumb this from out of the block at cylinder four and into the intake for the radiator to cool the fluid before returning it to the pump and the block?
Great video and i too had the same thinking as yours, even if it's there to create an extra low pressure exit on #4 could it not be rerouted back to the rad instead of WP or maybe like he said the connection at the WP is a low pressure suction so maybe thats why.
Pity no reply, this guy knows his subaru.
The heater not affected whit this mod?
No. All good.
Hey u said u from northern California.. wondering where in norther cali u from in Santa Rosa cali
I’m in Concord, but moving the Sac / Roseville area soon. Cheers man! We’re pretty close!!!
For some reason my OEM thermostat it's not open, there it's not leaking, no air pockets, no head gasket issue (I guess) but some times the thermostat won't open, I have to put in parking, then press the gas pedal until reach 3-3.5k rpm and for some reason the coolant starts flowing through the cooling system... Any idea? This is the first well explained cooling system flow process I seen, Thank you for your time for it..
I’d start with a new thermostat for sure. Make sure you orient it so the little bleed hole is on the upper side, that way air pockets can pass though. Should be good after that! Good luck! 👍
Have a look at the toyota 4age 20v coolant flow it will make u sick going bk to the Ej , the bring the colder Copland in from the rad across all the intake then down into the pump back up through the block and across the exhaust valves bringing the heat out to the rad , helping cool the intake all the time ,
The ej is not a reverse coolant mod it’s more an assist
Do you know the size of the hex socket for the plug for cylinder 4 on the head to do the cooling mod. Thanks!
Great video thanks man
Good video mostly. And I took you pretty serious up until the point you said you were going to cut your brand new head gaskets 🤦♂️ ... can you say NEVER!!! and no I wouldn't cut them I would just buy the upgrade version of the gaskets with those holes already made bigger 🤷
Can you ad the mod on the other side also?
Thanks you very much!
Can you tell me what the thread pitch of the M20 coolant plug (at 36:15) on the back of the driver head is? I'm making my own reverse cooling kit, and my M20x1.5 to AN adapter doesn't seem to fit right.
Just ordered a M20x1.5 NPT thread adapter but maybe Subaru uses BPT thread pitch.
Just an idea, when you are talking about opening up the head gasket for coolant flow, you could open up only the cylinder 4 ports to allow more flow through that side, but leave cylinder 2 holes a normal size. Could help number 4
NO
What about the coolant flow, of the cylinder #1 & #3? So time much was spent on the #2 & #4, is there the same issue with the other side? Why not do the mod with the other cylinder head?
Tim Salzmann The right side head already has something like this, to feed the turbo. This mod evens it out
Jesus took you long enough, been waiting here
Hello Luke, first of all, thanks so much for your videos, very informative and a great help!
Got a question about coolant flow when removing the OEM oil normalizer/cooler. Do you know if deleting this will cause a problem with coolant flow? I have an aftermarket oil cooler and don't have an OEM oil normalizer.
Thanks Luke!
You’ll be fine, they can be deleted. 👌
@@SubaruONLY Thanks Luke, maybe if you have the time, you could make a video that shows how the coolant expansion tank works on the side of the intake manifold? I read it cools the turbo by the process of convection when the car is turned off and cooling down.
great👍 thank you
Btw, the kits are great, but overpriced.
That has a cost of production of less than 5$!
30-50$ would be a more realistic price.
we've already done this ....no they are not (see other video on this )
I have the coolant mod, still blew an engine
34:00 after installation, STi feels lighter and stable.
Huh???
Is the cylinder 4 an issue on na 2.5 ej in 8' legacy?
Not much of an issue for these NA motors.
how many times ya gonna say actually
Where does the other side of the T goes too.. new to the subie world?
All good buddy! It connects to one of the lines (return line) that go to you heater core.
‘Sko Orediggers! 1:14
Has anyone used the coolant gel through the cooling systems rather than the half water version readily available
Never heard of that.
Epic
👍
Need your help wizard.. my 05 sti is sputtering under boost only.. fuel pressure is good.. coils are new and tested.. new ngk sparks.. new oem fuel injectors.. compression is fine.. leak down is fine .. took engine apart rebuilt it same.. tested all pig tails on harness and fine.. vaccum is fine.. no boost leak.. idles 100% fine please help
Tony Guzman Check crank sensor spacing
Tony Guzman Also clean your MAF and then reset the ECU
Hello...Does a radiator upgrade solve overheating problem?
Good question and I’ve thought the same. It helps bring down the coolant that is being recirculated in the system - BUT if there is a stagnant pocket of coolant that gets trapped in the back of the rear left head, that coolant isn’t getting cooled by the radiator, so no help for that issue (which could be the big problem). So I recommend both for best results!
I installed a koyorad radiater and the cooling did improve on my car 06 sti.