R34 Gtt Front Mount Intercooler Installation

[V28VX37]

Hi all, I just spent the past couple of days installing a front mount intercooler kit on my R34 GTT and I thought I'd share some of my experiences.

Picking the kit

I spent ages looking for a decent second hand kit. I specifically wanted a return flow type to avoid cutting the battery tray and running silly piping over the engine - it's a skyline not a wrx. I almost gave up and bought a Cooling Pro Stealth kit from Just Jap when another forum member put their Trust turnflow kit up for sale. These are pretty rare units and it was hard to find decent information but SAU came to the rescue once again (see here and here.)

I did some numbers before buying the kit, here's a list of core volumes:

• Stock SMIC: 3.6 litres (~3,600 cm3)
• Cooling Pro: 8.6 l
• Trust turnflow. 9.2 l
• Blitz CS: 12.5 l (about 30% larger than Cooling Pro)

So even though the Trust looks a lot smaller than the Cooling Pro it's in fact larger thanks to that extra 1 cm of depth! It's about 2.5 times the size of the stock SMIC so should be a good upgrade from that perspective, too.

Another interesting thing about the Trust is that the end tanks are on the top and bottom rather than at the ends - hence the "turnflow" or "vertical flow" name. It looks funny compared to the standard horizontal coolers but one of the benefits is greatly reduced distance the air has to travel through the cooler and the piping. It's probably around half of something like the Blitz kit. Being narrow it also appears to obstruct radiator airflow less - the guy I bought it from reported increased engine temps under comparable conditions after upgrading to an ARC cooler.

Installation

I decided to do the installation myself and found these two DIY's invaluable:

- Taking Off Gtt R34 Front Bumper With Pictures

- Installing Return Flow Fmic on R34 GTT

I was a bit worried about the kit as it's designed for an R33 but it fitted the 34 without a hitch. The installation was straightforward enough but very time-consuming. I haven't done anything like this before and it took me the best part of three days.

I used some very basic tools for tidying up the cooler: tweezers for straightening the fins and a cable tie to clean between them. Have a look at the before and after photos below. There's a good summary on the process here. I also did a final sweep with an air compressor.

The Trust cooler has three mounting brackets: top, bottom and right hand side (from driver's seat) of the cooler. Top and bottom lined up perfectly but bolting up the bottom was an issue. There's a hole in the chassis down there but no nut, and since it's a boxed in section getting a bolt in was a major PITA. There is a larger hole on the front side of the chassis bar and I finally managed to get a bolt through after some serious fiddling with line and cable ties. I also had to grind the bolt beforehand, as the base wouldn't fit through otherwise. Since the bottom tanks sits fairly low I cut a hole for it in the plastic under tray (dremel FTW).

The intercooler piping lined up perfectly with the stock SMIC joiners under the airbox. I also didn't have to fiddle with the power steering pump nipple as it's part of the stock piping next to the airbox. I did need to relocate the horns and also adjust the little (temperature?) sensor in front of the small cooler on the bottom right.

Cutting the reo bar wasn't as bad as I expected either. It seems that with the horizontal flow kits you need to shave off a bit all along the reo bar, however with this kit I only had to make some space for the top corner of the cooler & piping on the driver's side (see pics below.) Most of this can be reverted too just by bending the flaps back up and spot welding.

Trimming the front bar was hard, I probably spent half a day just on that. I removed the whole mid section of the bar and had to trim back the surround half a dozen times before it fit on properly. I also had to make some space for the piping. I first used a dremel but the cutting discs seem to heat up and break very quickly when working on plastic. Once out of those I pulled out the angle grinder and instantly wished I'd done that sooner - so much quicker and easier to do longer runs, too. A flap disc (for metal) worked a treat.

The last bit was the mesh, I used the basic stuff you can buy from Supercheap. I just folded suitable sizes around the inside edges of the front bar and used cable ties to attach them. On the bottom side I ran the ties through holes I drilled in the underside of the front bar so they're not visible. It was all good until I ran out of light at sunset..

Some specific tips

• All the tools are pretty basic but you'll need plenty of them. Try to gather them beforehand so you don't have to run off to get missing stuff in the heat of the battle.
• Loosen or remove the plastic inner guards around the front tyres before you start, this makes it heaps easier to get the two front bar side bolts out.
• Tape up all corners of the front bar as soon as you pull it out. You'll be shuffling it around a fair bit and scuffs are highly likely otherwise.
• Make sure the (hard) intercooler piping doesn't scuff where it goes through the metal under the airbox. Add rubber/padding/duct tape if necessary.
• Painting or rust proofing any bare metal is good idea.
• Test the basics before you put everything back together (ic piping leaks, lights, indicators, foggies etc.)
• Take photos as you go so you remember what goes where when you start reassembling the front.
• Don't do this in a rush. There's nothing worse than having to work until midnight because you need to be somewhere at 8 am the following morning...

Photos

I've included some photos below. The car is absolutely filthy here thanks to some country driving, and the headlights need some serious polishing. Scuffs on the front bar are courtesy of the previous owner

1. Stock front bar with old plates

2. Trust vertical flow intercooler kit

3. Intercooler before tidying up

4. Intercooler after tidying up

5. Tools for tidying intercooler

6. Box-o-nuts

7. Trial mount, compare FMIC size to stock SMIC on the right

8. Trial mount of piping

9. Need clearance

10. Tools of the trade

11. Trimming the reo bar

12. Got clearance!

13. Trimmed reo bar

14. Hard piping, straight bolt-on

15. Reinstalled reo bar

16. Front bar before cutting

17. Front bar inside, the whole boxed section in the middle was removed

18. Almost done

19. The finished product. Needs a good clean though!

Cheers!

[V28VX37]

Interesting, I never realised the new GT-R uses a vertical flow intercooler design too, must be something in it.

The following photos are shamelessly borrowed from 09GTR's 09gtr Lives Again thread.

Stock intercoolers:

HKS intercoolers:

I wonder why vertical flow units are not more popular for the R32/33/34's?

[Addicted2BoostR34]

Gday mate could you up load some pics of the fog lights or tell me where you got them from cheers

[discopotato03]

Yes I wonder too , I'm still looking for one of these by the way . Anyone open to handsome bribes ?

Cheers A .

Edited August 24, 2011 by discopotato03

[fastjetjockey]

Nice write up. Timely too as it's something I'm going to be undertaking pretty soon, but with a Blitz CS Return Flow.

[V28VX37]

Gday mate could you up load some pics of the fog lights or tell me where you got them from cheers

Yes I wonder too , I'm still looking for one of these by the way . Anyone open to handsome bribes ?

Hey guys, sorry for the late followup. The front bar and fog lights are standard R34 series 1 parts as far as I know - are yours different?

Nice write up. Timely too as it's something I'm going to be undertaking pretty soon, but with a Blitz CS Return Flow.

Thanks mate, how did you go with the install?

[discopotato03]

Interesting, I never realised the new GT-R uses a vertical flow intercooler design too, must be something in it.

Yes there is "a lot" in it .

More tubes has always been better than longer ones because more flow paths equals less resistance to flow . The cooling area ia about the same flow restriction is not .

Also note the horizontal tanks mean you don't have the return pipe under/over your typical front mount return flow system to rob core area or add volume to the IC system .

take a closer look at the OPs second picture and note that these vertical flow intercoolers are made from two cores fitted side by side . If we could come up with a couple of OE from something same sized cores the fab work wouldn't be real difficult to make the same sort of thing .

If anyone has access to light truck wreckers can you please look at whats being used on some of the turbo diesel top mount units because good second hand cores "in parallel" may just be the ticket .

I think a lot of the work in making custom aftermarket intercolers is punching the end plates and welding around each tube . More welds more time greater cost .

To the OP can you please measure the vertical heights of the cores between the tanks and the overall width which we can half to determine the width of each core Trust used .

Anyway there is a lot to like about this system and I think its as good as it gets whilst having standardish plumbing in the engine bay . It doesn't cause legal hassles going back in the inlet manifold side and if you can still get 250-300 Kw potential thats good enough for a few stealth bombers . Actually if hand made there may even be the opportunity to have a front mount that didn't need any bodywork cut which is the other 50% of the legal issue .

A .

[GTSBoy]

There is a downside to vertical flow intercoolers.  The longer tanks, particularly the inlet side tank, make it a bit harder to ensure an even distribution of flow across all the core tubes.  It's a similar problem as plenum design really.  If the inlet tank is approx the same cross section along its whole length, then the flow will tend to crowd to the far end, leaving the tubes near the inlet pipe under-utilised.  You can try to taper the tank, and you can try to include flow conditioning devices (turning vanes etc) but it's hard to squeeze stuff into the space available.

I have seen this happen.  The old Supra intercoolers with tubular top and bottom tanks were particularly bad.  But you also see it on various fabricated ones too.  If you measure the core temperature at the left and right hand ends, you often see quite a lot of difference.  The hot end is opposite the inlet/outlet end.  That's where all the heat goes because that's where all the flow is.

The other problem, which is not always a problem, is that with more tubes and hence lower velocity (which is good for reduced pressure drop), you also get a reduction in the convective heat transfer coefficient between the air and the inside of the tube walls.  Lower velocity will do that.  So there is a slight (thermal) efficiency penalty to take (all other things being equal, such as intercooler mass, core area, etc).  In a side tanked version, the smaller number of tubes leads to higher velocity and better thermal efficiency.  The increased velocity is offset by the longer tubes, and so the flow time in the tubes is not compromised.  With the increase in heat transfer coefficient, you should see a lower exit temperature than in a vertical flow core.

If you couple the efficiency penalty with the quite likely reduction in core usage from poor inlet distribution, vertical coolers can suck quite hard compared to side tanked coolers.  I'm sure the new GTR is set up with carefully designed and flow tested inlet tanks and the overall design would be smack in the middle of the good heat transfer coefficient region.  But there's no guarantee that any aftermarket kit or thrown together cooler will be like that.

FWIW, I don't have a problem with vertical flow coolers.  If they allow you to fit in a big core with the most efficient piping layout, then they're a fine choice.  Just may not be as thermally as good as the alternative option though.

Edited December 18, 2011 by GTSBoy

[zebra]

To add to above, the R35 uses two small separate intercoolers, not one large one, so the flow distribution is not a big issue.

But yes JZA70 intercoolers arent that great, but work fine for a streeter

[V28VX37]

Interesting discussion guys, thanks. So to recap on vertical flow intercoolers:

Pros:

• Shorter air travel distance so less pressure drop and potentially better throttle response
• Efficient piping layout for return flow configuration (doesn't need long return pipe)
• No battery tray cutting as with non-return-flow coolers

Cons:

• Potentially uneven flow & heat distribution
• End tank design is probably even more important than in horizontal flow coolers
• Rare

For up to 300rwkw or so it probably doesn't matter what aftermarket cooler you pick though

One thing that no one's mentioned is the shape of the core vs. the cutout in the front bar - i.e. the available cooler surface area vs. actual air flow. In other words, if only 70% of the cooler is exposed its efficiency is respectively reduced, correct?

To address this I would have thought a thicker core with smaller dimensions is better than a larger but thinner one, no? And this obviously doesn't depend on whether it's horizontal or vertical flow.

To the OP can you please measure the vertical heights of the cores between the tanks and the overall width which we can half to determine the width of each core Trust used .

From my notes the overall core dimensions for the Trust are L=520mm, H=252mm, D=70mm so the individual cores would be 260 x 252 x 70mm (or 4.6l each).

But yes JZA70 intercoolers arent that great, but work fine for a streeter

What size are those coolers, out of curiosity?

[zebra]

550x70x280ish

2 1/2" in/out

[V28VX37]

550x70x280ish

2 1/2" in/out

That's a pretty decent size core for a stocker, almost 11 litres... Must have a suboptimal end tank design or something if it's considered no good.

[zebra]

just a CHEAP mass produced thing, the tubes are rather small etc

[discopotato03]

GTSBoy everything I've been taught about radiators of any description is that the slower the air or fluid velocity through the tubes the better chance it has of rejecting heat because it spends more time exposed to the cooling air flowing over the outsides of the tubes and fins . I can't see how higher velocity meaning less time to cool can work better if everything else is the same core design wise .

If you look at the tanks on that Trust IC they reduce in volume at the far ends so I'd say the idea was to attempt to even out the air speed along the length of the tanks .

Anyway given a choice I think its the better compromise .

Also , BTW , I think the old Supra ones are supposed to be poor flowing things because of the tube and tank design rather than their layout . Had they used that tube/tank design with lesser longer tubes they would have been totally useless IMO .

A .

[GTSBoy]

Nup. Fundamentals of convective heat transfer theory in conduits. Higher velocity = higher convective heat transfer coefficient.

Side story: That old wives tale about slowing the flow down has come about as a way to describe what happens when you don't run a thermostat. The mechanic-speak for what happens is that the coolant runs around too fast and doesn't spend any time in the engine picking up heat and doesn't spend any time in the radiator to drop the heat. I have heard it used to explain why engines without a thermostat overheat and also to explain why they run cold. Can't have it both ways. The simple fact on that one is that by design the radiator should drop the water to a fairly low temperature (compared to the peak temperature), and it should enter the engine at the coolest location and flow towards the hottest locations before leaving. The idea being that we want the hottest locations to be the hottest locations because it improves thermal efficiency - if you have cold coolant at the head you lose too much heat to the cooling system and waste fuel. But that's a side issue. If you double the flow rate of coolant in a system, then the coolant will not heat up to the same temperature as if it were running at normal flow, and then it doesn't have to cool down as far in the radiator either. Let's say that double flow equals half the temperature rise. On that basis, you can consider that doubling the flow rate doubles the cooling capacity of the system, because you can now use the extra capacity to pump in extra heat and take the coolant back up to its old maximum temperature. This is of course limited by the capacity to put heat in (in the engine) and to take it out (on the ambient side of the radiator). In reality, because making the coolant flow faster increases the convective heat transfer coefficient, you will more than double the cooling power of the system (at least as seen from the wet side - the same limitations exist on the engine and ambient sides of the system).

In the case of these long or short tubes in coolers, you have to keep in mind that we need to keep all other things equal. Us assuming that we're keeping the core area the same is a rough way of balancing out the internal flow area vs the change in residence time in the tubes themselves. More shorter tubes with lower velocity (compared to fewer longer tubes) means

• possibly the same total time spent in each tube (compared to the sideways flow cooler)
• lower heat transfer coefficient meaning that less heat can be moved from internal air to the tube wall in that same time duration
• equals lower thermal efficiency.

The Supra cores would have to be the nastiest bit of tank design ever contemplated. But they were priced to suit the task and worked after their own fashion. A long long way from a good design, but also a lot cheaper and easier for Toyota's engineers to source and package at the time.

[discopotato03]

Well we will have to agree to differ because as I said everything I've been taught about radiators intercoolers and oil coolers says the opposite . Next time you see an oil cooler catalog note they generally get taller as in add more tubes to increase the core area and flow area (tube number) .

Cooling water radiators are a bit different because they have to be fitted in the available area and that includes plumbing as well .

Anyone that took a thermostat out to increase cooling didn't understand that how water flows around a cylinder head is very important . The correct way to go about that is to remove the flat valve from the thermostats surrounding plate and refit it .

Many moons ago some special tuning/competitions departments used to sell what they (in BMCs case) called a blanking plate which was a tubular device to replace the thermostat in old Cooper S race/rally engines . It ensured that the coolant flowed properly around and out of the thermostats end of the head casting .

If maximum effort cooling is needed in competition apps and another I'll mention in a sec "split cooling" is used . This is where coolant flows in the top of the engine and through the head and down through the block passages and out the bottom to be returned to the radiator/s .

The Locomotives we run around in at work have GE 7FDL V16s in them and the way to boost power reliably in them from in the 3s to 4400 gross Hp was to use the split cooling technique .

Anyway getting back to radiator type cores . The tube design is very important because the idea is for the tube to absorb as much heat as possible from the fluid or gas running through them and the conduct it out to the fins or gills if you like .

These days they virtually always use "turbulators" inside the tubes to break up whats called the boundary layer which wants to insulate inside of the tube from whatevers going through it . Aside from adding more internal surface area these turbulators encourage turbulent flow because this is what it takes to get the medium to expose most of its volume to the tube material so it can conduct the heat away .

Now we know that turbulance doesn't do a whole lot for flow velocity but the whole point of the cooling core is to cool whilst posing the least flow restriction . So therefore more tubes or air paths means you have a better chance to make the air turbulent enough to give up the heat but overall not slow it down over less flow paths creating a pressure rise .

I'm pretty sure Corky Bell mentions the correct way to increase an intercoolers capacity is to increase the number of tubes rather than increase the tubes lengths .

Realy the GTRs intercooler tends to be different in the tanks to many aftermarket ones especially those that don't have a lot of tank depth . Also the intention was to have the turbos on one side of the engine and the throttles the other so its not hard to see why they are layed out the way they are .

Anyhow to each their own and I'll continue to look for the elusive Trust turnflow kit or if I can find two suitable cores get one made .

A .

[GTSBoy]

I think you're misunderstanding what I mean by "thermal efficiency". I literally mean how much heat it can move per unit area of core. I'll get back to that in a sec. Your oil cooler example and Corky's recommendation on how to increase the capacity of a core are both about one thing - increasing capacity. In the oil cooler case the reason you need to increase capacity is because you have more oil to cool. The hot and cold oil temperatures are always about the same for any given application type (ie engine, transmission), and the amount of pressure available to drive oil through the cooler remains the same (ie, how much pressure drop you can afford) and so to increase capacity you have to increase tube area. For the Corky intercooler recommendation, it's essentially the same thing. But for any fixed core area, so long as he balance between number of tubes and length of tubes remains valid (which it should for any reasonable sized core) then the longer tubes will offer more pressure drop and less heat transfer coefficient. It will be less efficient in terms of boost loss and more efficient in terms of heat transfer. It's simply not possible to argue against it. The convective heat transfer coeeficient is a function of the Nusselt number, which in tun is a function of the Reynold's and Prandtl numbers. And Re goes up with velocity. The thing abotu turbulators in cooler tubes is that for the ones that are really just internal finning (ie they go all the way across the tube) for the most part they are as much about increasing surface area as they are about increasing "turbulence". More surface area is worth a lot more to the heat transfer than modifying the boundary layer, given that the flow inside cooler tubes is highly turbulent anyway (the Reynold's number is really high inside any inlet tract when the engine is running.) cheers

[DarkRaven]

Great write up! Is there anyway to get the pics back? I know this is an old post but I am just about to do this to my Gtt and the pics would help so much

[owen1r]

Great write up! Is there anyway to get the pics back? I know this is an old post but I am just about to do this to my Gtt and the pics would help so much

I recently completed it with the help of the second link, more than enough.

[Kinkstaah]

If you want to cut the reo bar, that is... given the kit specifically mentions that you don't need to cut it.

(you do)