Jump to content
SAU Community
  • Welcome to SAU Community

    Welcome to SAU Community, like most online communities you must register to view or post in our community, but don't worry this is a simple free process that requires minimal information for you to signup. Be apart of SAU Community by signing in or creating an account.

    • Start new topics and reply to others
    • Subscribe to topics and forums to get email updates
    • Get your own profile page, build reputation and make new friends
    • Send personal messages to other members.
    • See fewer ads!

    Consider joining our newsletter for the latest content updates

    Click here to register

Rb25 Coolant Flow / Hose / Swirl Pot / Header Tank Setup

Recommended Posts

Hey all, this is a bit of a side step from this thread;


What i am trying to work out is the most logical way of properly setting up the cooling system for an RB25, in this case a NEO. This is for race use, so no heater core.

I am trying to get rid of any pipe work that is not necessary for cooling. For example, i no longer have the factory oil warmer, cold start coolant valve, throttle body heating or heater core. I will be fitting a coolant swirl pot and a header tank if deemed necessary. Obviously you do not want any point for coolant to become trapped and not able to flow. I will be avoiding this.

Below are some attached photos.





Here is a legend and the coolant flow path in the factory form of the NEO.

Point 1 - Return from heater core to thermostat housing, (return from oil warmer is also tee'd into this).

Point 2 - Return from front bleed valve to thermostat housing (The return from the throttle body and Point 7 (the front outlet of the runners) connects to this front bleed valve)

Point 3 - Feed to heater core

Point 4 - Feed to oil warmer (This can be completely ignored)

Point 5 - This goes up to the rear bleed valve (Point 8 also connects to this bleed valve and also the feed to the throttle body comes from this bleed valve)

Point 6 - Return From Turbo

Point 7 - Front runner outlet to front bleed point

Point 8 - Rear runner outlet to rear bleed point

NOTE - Both bleed valves will not be re-fitted, Hence the use of a swirl pot for bleeding.

What i am trying to work out, is what to do with the 8 points. How to correctly set them up. I have a few different ideas, but i'm trying to distinguish which one is most suitable, My latest thinking i have come up with:

Point 1 / Point 2 / Point 3 / Point 4 / Point 5 - Delete completely, Remove the fitting completely from the thermostat housing and blank up with a welch plug. Do the same at the rear of the block, remove the fitting in the block completely and blank off with a 3/4" NPT bung. This removes any pockets for coolant to get trapped.

Why delete both fittings completely? Well the original oil warming piping is not required nor do i need the fittings or piping for a heater core. Now most people would say just link the rear of the block to the thermostat housing. But, thinking about it if you retained the heater core as many people do, for 6 months of the year the heater tap is closed completely stopping coolant from flowing through to the heater core and back to the thermostat housing. Therefore i see no reason apart from 1 (Below) to retain those two fittings. If i did have a heater core, it would be shut for 6 months of the year which is the same thing as removing the fittings and blanking.

The only reason i can think to keep a fitting in the rear of the block and in the thermostat housing would be for bleeding purposes if required;

1. Tee the rear of the block to the rear runner; Point 5 to Point 6 - This tee point then runs to the swirl pot.

2. Tee the thermostat housing to the front runner, Point 2 to Point 7 - This also then runs to the swirl pot.

i am unsure if i need to bleed the rear of the block and the thermostat housing to the swirl pot? Or can i simply do as above and just run Point 7 and 8 to the top of the swirl pot? I have different theories on this. My current think is that i do not need to bleed the block.

If you look at an RB26 there is only a fitting in the thermostat housing for a return from the heater core, there is no piping to go to a bleed point. As for the rear i'm not sure what is done. But looking at other peoples swirl pots, i have not seen anyone take a bleed point from the block up to the swirl pot.

Sorry for the long thread, but any help would be appreciated and i hope this can help others.


Link to comment
Share on other sites

7,8,6 to the header tank 1 inch from the top of the tank.

All others blocked off at the block.

Swirl pot in the top of upper radiator hose with another point from its top to the header tank.

Base of the header tank minimum 3/4 hose to the bottom tank of the radiator.

With a swirl pot in top hose block all radiator points including cap.

Without a swirl pot, replace cap with fully sealed cap and use overflow fitting to run to header tank or block cap hole and run a fitting from top of top tank to the header tank.

Link to comment
Share on other sites

I was hoping you would see this thread Brad. Your very helpful with this sort of stuff.

1. Point 6 is the return from the turbo, you've said to run this to the header tank. Where do i run the turbo return line to now or have you mistakenly said to run point 6 to the header tank.

2. Ok so if i run a header tank and coolant swirl pot; The top of the swirl pot to a point on the header tank using -4 or -6?, Points 7, 8 and 6? To the header tank in -4 or -6, 1" from the top. Very bottom of swirl pot to the bottom radiator tank or can it be tee'd into hose? with a 3/4" minimum hose. Use a blanking cap to block of the radiator fill neck and also block off original overflow point - does the neck need to welded shut or can i do what i stated? New overflow point is from the header tank.

If i dont use a swirl pot i think its pretty self explanatory. it was what to do with the block fittings which was doing my head in. I assume you do this regularly with RB's brad, do you block the thermostat housing with a welch plug and just put a bung in the rear fitting like i suggested?

I saw in the other thread that you manufacture swirl pots to suit RB's. Are they made in such a way that you simply cut an OEM hose to suit, insert the swirl pot and your done? Do they clear clutch fans with no issues?

Thanks mate.

Link to comment
Share on other sites

Use the turbo water line as a turbo waterline and just run the two from the inlet manifold to the header tank.

I normally use -6an for the header tank lines from the engine and -4an from the radiator or swirl pot.

You can block off the radiator filler either way. I usually cut the filler off and weld on a round 3mm plate.

You can T the header into the lower radiator hose but its easier to run a designated fitting in the radiator.

I've also done the bottom of the header to the thermostats fitting.

It's not ideal as its on the wrong side of the thermostats but does work.

Usually welsh plug the block at the front and use an old diff filler plug to block the rear fitting off

Yeah they fit with a cut factory hose or two lengths of silicon hose

Link to comment
Share on other sites

OK no problems.

In your experience, is it much better to use a swirl pot?

I havn't got a radiator yet, so when i buy one maybe i'll look into something custom - with no filler cap and a -12 or -16 fitting in the bottom tank for the header tank return. I was looking at a racepace 42mm unit, maybe they can help me here.

The two bleed points from the runners to the top of the header tank are 1" from the top (The very top of the header tank, or the overflow nipple?) Where does the bleed point from the swirl pot feed into on the header tank? Same height or lower?

Is your swirl pot generic for all RB's, or do you change them around a bit between R32 / 33 / 34 / 20 / 25 / 26?

The only issue i can see with a swirl pot is trying to get the header tank a substantial amount higher than the top of the swirl pot.

Link to comment
Share on other sites

I don't always use a swirl pot.

I normally make my own radiators and leave the filler neck off.

Then run a -4an fitting off the corner of the top tank which runs to the header tank.

-12an is okay for the bottom of the header.

Rb20,25 &26 swirl pots are all different.

The header tanks are car specific too.

All entry's to the header tank should be an inch from the tanks top. Then the new radiator car above that

Link to comment
Share on other sites

This is an off the shelf PWR core which I've just put a modified cap on the filler next and welded a -4an fitting to the tank.

Also welded up the hole where the original overflow would have come out the filler neck.

Works fine but the swirl pot would give 100% confidence there is no air lock in the radiator as its bleed point is higher


Link to comment
Share on other sites

Ahh yeah no problem Nice work!

Yeah well i think i'll run a swirl pot. Why not. So long as it clears the clutch fan (Still yet work out if i can use a clutch fan).

I've heard a few people saying the PWR rad's arn't worth the money you pay for them.

The off the shelf 42mm rad to suit an R31 from them is well over $800 and is only single pass.

I'll have to make up a custom header tank as i've got things everywhere in the engine bay - Having tubs doesn't make this any easier either. I'll probably place it behind the drivers strut tower. That way it will be higher than the engine and the swirl pot.

Oh, do you remember what size welch plug you use for the thermostat housing off the top of your head? I've just pulled it out and measured the diameter - 18.4mm approximately.

Link to comment
Share on other sites

Just keep in mind that Risking has posted a cross flow radiator. If you are running a traditional radiator its probably a good idea to run a non pressure cap and bleed to the top tank to your header tank. The wide top tank vs narrow cross flow tank means the raised section for the radiator cap is a pretty opportunistic place to bleed to your header tank

Link to comment
Share on other sites

Troy is correct. Even get the top tank tapered if its a custom radiator.

Im doing a tube front end on an RX7 currently that's getting a laid down radiator and fairly elaborate header/swirl pot system to cope with the air and cavitation problems.

Find the pro fabrication race engineering face book page if you want to see a lot of the cooling system going together. Everything I do gets put up on there

Link to comment
Share on other sites

heres mine... some feedback would be good. it seems to work well, Car still gets hot after multiple hard laps but does not boil.


cap on rad and swirl pot are not pressure caps. the canister at the back of the engine is the highest part of the cooling system and has pressure cap on it.

3x fittings welded onto the rb26 coolant rail to remove any localised boiling points.

bottom of swirl pot goes to top of the radiator tank.. so the idea is to remove any air bubbles.

Link to comment
Share on other sites

Just keep in mind that Risking has posted a cross flow radiator. If you are running a traditional radiator its probably a good idea to run a non pressure cap and bleed to the top tank to your header tank. The wide top tank vs narrow cross flow tank means the raised section for the radiator cap is a pretty opportunistic place to bleed to your header tank

Sorry mate, not fully sure what you mean? Do you mean if i have a down flow radiator that i must bleed from the top tank to the header tank? This of course would be done. I would just do it from the filler neck. But if i have a swirl pot this shouldn't be needed as the fitting at the top of the swirl pot back to the header tank should be higher than the radiator top tank.

Just had a geez at your facebook page Brad. I'm guessing you havn't uploaded any photos of cooling setups yet - can't seem to find any.

Link to comment
Share on other sites

Nice work Brad! The swirl pots do look good!

I'll grab one of your swirl pots when i'm ready!

Simon, i could only suggest that maybe you bleed from that buldge section as well as the radiator neck on the rad.

Link to comment
Share on other sites

since we have a thread up may as well chime in.

i'm running a header set up on a rb20 with no swirl pot and i only fill up from the header tank thats mounted next to the brake booster , rad cap isn't pressurized but header cap is of course.

i have a standard 8/10mm hose from the highest point on the header tank going to the rad cap nipple and then the hoses from the intake manifold join up to the header tank.

my main problem is i always have to do a water flush if i've drained the rad to get the thermostat to open other wise car will just over heat and bottom of rad will be ambient temp. my thinking is theres a pocket of air or something near the thermostat and somehow a water flush helps?

other then that works fine

so open to opinions

Edited by Dan_J
Link to comment
Share on other sites

More details of how you've plumbed up the header tank?

Is the base of the header tank linked to the bottom rad hose?

So the radiator filler neck is connected to the top area of the header tank?

How exactly have you connected the top of the engine to the header tank? You should simply run a line from the original bleed point to the header tank.

Link to comment
Share on other sites

Re the Hypertune swirpot Risking posted. I am 99% sure it works fine and there is absolutely F. all difference but technically you shoud try (space permitting) to have the inlet tangent to the circumference of the swirl pot about half way up. You should then run the outlet from the centreline of the swirl pot from a lower point then the inlet. Ideally the bottom but not going to happen on an RB so lower is generally acceptable.

Link to comment
Share on other sites

More details of how you've plumbed up the header tank?

Is the base of the header tank linked to the bottom rad hose?

So the radiator filler neck is connected to the top area of the header tank?

How exactly have you connected the top of the engine to the header tank? You should simply run a line from the original bleed point to the header tank.

no the bottom of the header tank is connected to the heater joiner near the block

( might be the one that goes to the thermostat or the one right off the block, need a closer look)

and yes the radiator filler neck nipple is connected to the highest point (below the cap) on the H/tank.

i'm not running any coolant on the top half of the intake manifold so no bleed point

pic below for a rough idea of my set up


Edited by Dan_J
Link to comment
Share on other sites

You are correct Troy. For cavitation purposes that's the best way to do them.

Ill post up some photos of the Supercar swirl pots which are done the same way you mentioned.

Couple of issues with the setup above.

Bottom of header needs to go to the radiator side of the thermostat housing preferably as low was possible

You must have a bleed from the inlet manifold or cylinder head. It's a cavity for air locks, even it's a drilled and taped 1/8th npt thread with a 5/16ths nipple and hose.

Link to comment
Share on other sites

could i get some advice on if mine is correct?


as per pic... swirl pot one has a blank cap on it and has 3 bleeder lines from coolant rail. the red hose goes to the top of the radiator outlet and the cap is blank off, the blue line goes to a 2nd swirl pot canister from the top of the swirl pot. Shown in purple is a 10mm hose half way up the canister which is joined to the bottom radiator hose. the 2nd canister has a 1.1 bar pressure cap with a pipe going to over flow.

feedback appreciated

  • Like 1
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Latest Posts

    • If the timing check shows timing bouncing all over the place despite commanding a fixed value in the ECU then the most likely explanation is either the CAS drive pin on the cam has worn, the CAS drive gear has worn, or both. I suspect a number of the more severe cases of timing scatter are actually due to this issue.
    • I might try out the Direct Clutch twin plate, they run a mixed setup. 1x plate is organic 1x plate is ceramic  Which in theory makes it driveable, from what I hear it seems to be quite nice to drive from a handful of people. I've yet to experience this myself.
    • I was running the direct clutch services 1000hp rated twin plate clutch, I thought it was really nice, good pedal feel, not too heavy, a little ‘bitey’ on initial grab, but nothing excessive, got used to it pretty quickly, it held just over 500kw for over a year, when the car was pulled apart had it checked, was told plate wear indicated it had done what they considered’ bed in stage. Lol it will be going back in the car when it all goes back together. 
    • Right, so wheeeere were we. We were commuting around without a swaybar at the front for awhile. Until I heard... a...noise. A bad noise, a loud.. ticking...in-time-with RPM noise. I also noticed my power steering leak just wasn't going away. Wisely, I went to consult with an actual mechanic this time for a second opinion/further guidance. Upon hearing the sound, he immediately diagnosed it as Lifters (which is what it sounded like) but given I heard it from the bellhousing I thought it could be something in the bellhousing. Turns out the lifters are in the block, and that all immediately made sense. Phew, no need to worry about clutch/gearbox things, I have to worry about lifter things instead which as just mentioned, are in the block. Also: 1)      “Did you know your belts are misaligned and are starting to get chewed through?” Uh no, no I didn’t know that. Turns out my new underdriven alternator pulley that was a direct bolt on, was not a direct bolt on, and needed to be machined down as below it is compared to the one on my stock alternator. 2)      “Your aircon dye is leaking out the bottom and generally everywhere” Turns out the seals on my new AC system were 20 years old, and an old gasket had blown, all very surprising to me. These were replaced and system holding pressure now and no longer leaking.. 3)      Power steering leak was traced to the solenoid that was no longer being used and half disassembled from nearly fouling on headers. An O-ring was sourced, only to find that the housing of the solenoid body itself had a crack, and when removed found that the other side of it was flat. So a plate was machined up to just blank off the solenoid anyway, as the O-rings were in the rack itself. So no new rack needed. Hooray!   4)      I then learned my Power Steering had been broken since the conversion. What I thought was “Heavy but clearly working power steering” all this time was actually completely and utterly f**ked. The car now drives with slightly lighter than-OEM Nissan steering which was massive, massive, massive MASSIVE improvement which I didn’t expect! Still, the lifters thing. Then I had an idea, as my dipstick was still not connected to the car since the gearbox in (I had to get the bracket back from the other people). It was one of those last gasp ideas you have when you desperately want the problem to be something else. I had always historically overfilled my oil, and wondered if I was getting oil leaks (turns out I do not actually have ANY oil leaks.. but didn’t know it at the time) because something was just overfilled and it seemed to track. So I turned on my Accusump, filled up, then closed the valve. Left the car on overnight, checked the oil. ¾ of the way to H. Perfect. The way I always checked the oil. Still, mysteriously low oil pressure. I’m talking like 0.5 bar on the OEM gauge (which works in a R34). I then realized my accusump and oil cooler (a 25 row big boi) and lines are above the sump. I also remembered how I always check the oil overnight, after the car has sat. So as soon as I was starting the car, that oil cooler and lines would get re-filled again with oil from the sump, not exactly leaving enough in the f**king car. I went back to pick up car from mechanic and dumped a few liters into it and mysterious sound is now gone… and my oil pressure is a much healthier 3.5 bar at idle and moves according to RPM. That was a relief, to say the least. Still haven’t given  it a lot of RPM, as need to do my 1000km to get that gearbox all run in.
  • Create New...