Race Car Update...

General Gemini Chat
Forum for all gemini based chat & questions
F3ARED
Moderator
Posts: 2705
Joined: Wed Sep 28, 2005 11:55 pm
Location: Melbourne, Victoria

Re: Race Car Update...

Post by F3ARED »

If you have it, you may as well use it. My feeling is it may be a touch small given the restrictor and that its on a 2.3L, but only one way to find out really. When i was looking at running a Piazza manifold on my mrs car, I got frustrated with the drift-tax T28s attract down here. Ended up modifying it for an IHI 3 bolt flange - theres a whole HEAP of relatively cheap/near new IHI VF series turbos ex Subaru/STI out there that nobody seems to want. Something worth thinking about.

Hey Mick, what are the HX35s off in Oz, is it just heavy equiptment? I was looking for one a while back when I had the BMW and was contemplating throwing boost at it. Couldnt find one for love nor money.
Previously GeminiCoupe/GeminiRollingShell/SCAR3D/Jonnoisac#nt

78 TD Gemini Sedan - R.I.P
81 TE Gemini Sedan
DR_GEM
Long Term Member
Posts: 890
Joined: Mon May 30, 2005 7:47 pm
Location: 2196

Re: Race Car Update...

Post by DR_GEM »

The question of whether the t28 will be "ok" on a 2.3 is dependent on a lot of things

But assuming he is going to use it regardless, then the only thing I would ensure is the rear housing is not some tiny 0.48 a:r thing

A turbo will "work" on any engine, but with a restrictive rear housing and a small turbine wheel then it could be a recipe for disaster.

Small rear wheel means more backpressure earlier in the operating range. This is exacerbated by a smaller turbine housing which chokes the path of exhaust gasses even more.

This backpressure causes detonation, reduces your timing range, renders any overlap dangerous (due to reversion), increases pumping losses and transfers additional heat energy from where you want it (acting on the wheel then exiting) to the compressor side and clogs up flow.

There are other downsides but the above gives you an idea of how bad s*#t can get.

53mm wheel of the t28 is ok for the 2.3 displacement but again depends on type of wheel, but if using a bigger cam or other Volumetric efficiency improvements, then this will prove even more of a restriction and exaggerate the above effects

Holsets come on the Cummins engines so are not so rare that you can't find them. Problem is a lot of diy modifiers clued on with the holsets some time back and have been snapping them up when they can.

The holy grail of holsets is the he351ve from memory which are the variable vane turbos that provide insane spool and top end. Like a 6000rpm power band / full boost by 1800rpm to redline
I wish i had something cool and witty in my signature
User avatar
Jonno
Korean Jesus
Posts: 2370
Joined: Tue Jul 25, 2006 5:28 pm
Location: Canberra. Where the Fast Geminis live
Contact:

Re: Race Car Update...

Post by Jonno »

I have an ebay T04E which i had set aside for my 4ze1 turbo for years.

It's new and never been used. If the turbo isn't an IPRA restriction feel free to shoot me an offer on it?

You'd be able to move the T28 pretty easily?
75 TX Coupe
GemiNats > https://www.facebook.com/GeminiNationals
Image
#TurboFamily #Geminats #Bucketship
YT: https://www.youtube.com/channel/UCPKSHb ... akJ04sOYjw
DR_GEM
Long Term Member
Posts: 890
Joined: Mon May 30, 2005 7:47 pm
Location: 2196

Re: Race Car Update...

Post by DR_GEM »

Correction for above holset / the holy grail is the he341ve not the 351
I wish i had something cool and witty in my signature
F3ARED
Moderator
Posts: 2705
Joined: Wed Sep 28, 2005 11:55 pm
Location: Melbourne, Victoria

Re: Race Car Update...

Post by F3ARED »

DR_GEM wrote:The question of whether the t28 will be "ok" on a 2.3 is dependent on a lot of things

But assuming he is going to use it regardless, then the only thing I would ensure is the rear housing is not some tiny 0.48 a:r thing

A turbo will "work" on any engine, but with a restrictive rear housing and a small turbine wheel then it could be a recipe for disaster.

Small rear wheel means more backpressure earlier in the operating range. This is exacerbated by a smaller turbine housing which chokes the path of exhaust gasses even more.

This backpressure causes detonation, reduces your timing range, renders any overlap dangerous (due to reversion), increases pumping losses and transfers additional heat energy from where you want it (acting on the wheel then exiting) to the compressor side and clogs up flow.

There are other downsides but the above gives you an idea of how bad s*#t can get.

53mm wheel of the t28 is ok for the 2.3 displacement but again depends on type of wheel, but if using a bigger cam or other Volumetric efficiency improvements, then this will prove even more of a restriction and exaggerate the above effects

Holsets come on the Cummins engines so are not so rare that you can't find them. Problem is a lot of diy modifiers clued on with the holsets some time back and have been snapping them up when they can.

The holy grail of holsets is the he351ve from memory which are the variable vane turbos that provide insane spool and top end. Like a 6000rpm power band / full boost by 1800rpm to redline
Never actually considered reversion on a turbo engine but now that you say it, I know ive heard it before...just cant remember whos car it was on/what the setup was. Makes a really distinctive noise, almost like cavitation but with air rather than water as the fluid. Easiest way to fix or reduce reversion is to form a step in the exhaust path; if the port diameter is 36mm, then making the runner diameter on the manifold 40-42mm will help stop gas being pulled back in while the exh valve is opened. Wont help with backpressure though.

They love the Holsets in the US, havent really seen them used much here though and wouldnt know where to start to look for one. People are getting huge numbers out of them, almost perfectly suited to the M50B25 i wanted to run it on!
Previously GeminiCoupe/GeminiRollingShell/SCAR3D/Jonnoisac#nt

78 TD Gemini Sedan - R.I.P
81 TE Gemini Sedan
DR_GEM
Long Term Member
Posts: 890
Joined: Mon May 30, 2005 7:47 pm
Location: 2196

Re: Race Car Update...

Post by DR_GEM »

I will respond properly tonight , but reversion is an absolutely huge consideration for turbo setup - much much more so than an n/a engine
I wish i had something cool and witty in my signature
DR_GEM
Long Term Member
Posts: 890
Joined: Mon May 30, 2005 7:47 pm
Location: 2196

Re: Race Car Update...

Post by DR_GEM »

So yeah reversion / turbo sizing / cam timing / ignition timing.

One thing that isnt generally discussed is the importance of intake : exhaust pressure ratio

We must remember than an engine is essentially an atmospheric pump. The more atmospheric pressure you can ingest and expel, and the faster/ more efficient you can do this, the more power you make (well technically torque but anyway)

With smaller turbine wheels and turbine housings, intake to exhaust pressure ratio will be higher than a larger, more efficient turbo hot side

To understand how this would impact a combustion engine and its volumetric efficiency, we need to understand fluid dynamics. Take a naturally aspirated motor for example. When a piston moves down the bore with the inlet valves open, the general concensus is that air is "sucked" in by the moving piston (think a syringe pulling liquid into the body of the syringe when you pull back).

This is technically not correct, as what is really happening is the higher ambient pressure in the atmosphere is flowing to the lower pressure that is in the cylinder to "fill" the vacuum that is created. This vacuum is filled with air/ fuel.

With a forced induction engine, again its a misconception that air is "forced" through to the combustion chamber by the turbo or supercharger. Infact the volume of air able to be ingested itself is unchanged for any given motor. Turbo / super charging is merely packing the air more densely and thus allowing more oxygen per particle of air to be ingested, thus creating a bigger bang.

Boost is a measurement of inlet restriction, so the more boost you run, yes the more pressurized the intake will be, but the less efficient your engine is in ingesting this air. Remember boost is pressure increase above atmospheric pressure. Pressure can only exist with a restriction.

Theoretically speaking, if you could create an unrestricted, continuously flowing and efficient engine, then you would never be able to build boost because the engine will use all the available particles of air /oxygen to make power without creating an inlet restriction. However this is an impossibility, as anything with work has inefficiencies and losses. So boost is here to stay.

Why do I mention all this? Because its critical in understanding the role of valve timing in assisting an engine's efficiency.

Most people know that the beauty of variable valve timing is the ability to have qualities of larger duration/ more aggressive cams at higher rpm, and lower duration / less aggressive cams at lower rpm. The reson for this is at low rpm, where the flow of air / fuel mixtures and the corresponding spent exhaust gas is slower, requiring less valve opening time to facilitate an efficient burn. However at higher rpm, with faster inlet, combustion and exhaust flow, valve opening needs to be longer/timed differently to foster effiency. Problem is the required timing of these events is different for low, medium and high rpm and load points of an engine's operation.

In a sohc engine such as the g series or 4z series, you are stuck with fixed valve opening and closing points.

Valve overlap is important to understand for this discussion, and overlap is the amount of time, measured in degrees of crankshaft rotation, which both intake and exhaust valves are simultaneously open at the end of the exhaust stroke and beginning of the intake stroke. You're talking milliseconds here. Literally.

For n/a motors, generally for balanced performance you want high lift, medium duration and a bit of valve overlap at high rpm to assist with scavenging. Wont go into scavenging discussion here, but for higher power at higher rpm range, this would need to be longer duration, higher valve lift and more overlap. This has been discussed to no end in engine camshaft and tuning papers for many many years. Difficulty however was that overlap in an n/a motor at low revs hurt dynamic compression, as by opening the exhaust valve earlier to overlap with the opening of the intake, or advancing the intake valve opening so it happens while exhaust is still in progress, then you're releasing some cylinder pressure which reduces the actual compression of air fuel mixture in the cylinder. Lower compression means a smaller bang, and less efficiency. At lower rpm this resulted in sluggish power, and thus why high duration high overlap cams were known to "push the power curve to the right of the rpm band", and result in a sluggish performance at low rpm.

Historically it was believed that for a turbo motor, the camshaft requirements were the opposite of an n/a motor. That is, the thought was you needed lower duration, less overlap, as valve overlap on a forced induction motor would "blow the compressed mixture out the exhaust valve and waste it" or an even better one was "exhaust chill" which purportedly reduced the exhaust temperature and affected turbo spool.

This is true for a supercharged engine. And the reason relates once again to my earlier point around pressure ratios. A supercharged engine generally has the same exhaust pressure ratio as an n/a motor, as the exhaust system is the same (headers out to exhaust). Obviously the intake is higher pressure though (compressed boosted air). This is pretty static and constant across the entire rev/ load range. So at 2000rpm and say 7psi of boos (21.7psi absolute), the intake pressure is 7psi, and exhaust pressure is at most, 0psi above atmosphere (or 14.7psi absolute ). A pressure ratio of 1.48:1 (21.7:14.7) If you recall my point earlier around fluid dynamics, the flow of air will therefore naturally gravitate toward the lowest pressure area, in this case the exhaust manifold. So overlap here will push boosted air straight out the exhaust without an opportunity for it to be compressed and then ignited in the power stroke. A 1.48:1 pressure ratio indicates a much lower pressure area in the exhaust and therefore the boosted air is going to face very little resistance in getting to the lower pressure area very quickly.

With a turbo however, the actual turbo hot side housing and turbine wheel create an exhaust restriction all the way back to the exhaust port on the head and the exhaust valve by extension. This restriction increases the amount of exhaust pressure between the exhaust valve and the turbo.

As stated earlier, with smaller turbine and housing like an oem turbo, this restriction is higher and thus creates more pressure. Typically your intake to exhaust pressure ratio on a factory turbo'd engine could be 1:1 at up to 7psi, 1:1.5 at 7-14psi, 1:2 at 14psi-20psi and 1:3 at 20+psi (for example)

In this instance if you run an n/a cam on a 4z or G series but add a turbo that is similar in size to a factory turbo, you could face problems. If we apply an aggressive cam profile on such a setup, where we increase duration of the intake and exhaust, thereby making the inlet valve open sooner and for longer, and the same with exhaust so you increase the overlap, then lets see what we come up with.

You would create more overlap, meaning your compressed air/ fuel intake mixture is being introduced while the exhaust valve is open. Using above example, at up to 0 psi (off boost), your pressure ratio is 1:1 so there is no bias to which direction the flow of compressed air and fuel mixture would go. It would just follow the natural path (ie through to the turbo). The more overlap you run here the more volumetric efficiency you create as you allow the engine to breathe better, but also allow some of that unburnt air:fuel mixture into the hot exhaust which will help spool the turbo more funnily
enough...

At 7psi-14psi, with a more restrictive exhaust and turbo setup, this pressure ratio rises. Lets assume 1:1.5 times. This now means flow will favour the lower pressure of the inlet should there be an opportunity for that to happen. So if we have both inlet and exhaust valves open simultaneously here, reversion can occur, where hot exhaust gas flows back into the chamber and even making its way up past the inlet port and diluting your nice cold air and fuel charge. This is obviously counter productive to power, and works as exhaust gas recirculation to assist in pollution control.

The same can be said for remaining boost levels as pressure ratio increases, only that the flow "back" to the inlet will happen alot faster as the pressure ratio rises due once again to the flow characteristics of air/ fluid.

It can be seen that valve overlap here is definitely not of benefit.

Taking this further, it is clear that at higher rpm / load / flow, where exhaust back pressure is highest, then the same applies - ie you don't want much/ any overlap at all. The biggest difference here between a turbo and n/a engine is the backpressure, which is minimal in an n/a engine and results in a pressure ratio of closer to 1:1 than a turbo engine.

This has multiple problems also, in that a more restrictive exhaust pressure doesn't allow as much exhaust flow even after the inlet valve closes, and this limits the efficiency of the engine to pump out the burnt exhaust gas. Remember an engine is like a big air pump so the slower this burnt gas is able to be pumped out the lower the power/ torque. It also then slows down your combustion process (killing torque) which reduces your ability to advance timing (further killing torque). See how its spiraling??

However with a free flowing turbine housing and a larger more efficient turbine wheel, the pressure ratio is maintained at 1:1 for alot longer. This means more aggressive cam and ignition timing can be used to promote volumetric efficiency which is actually beneficial to engine efficiency and therefore turbo spool.

If you take a 2.3 litre and stuff it with a 46mm turbine wheel and 0.48 a/r rear housing, you are bumping that Inlet to exh pressure ratio up quite substantially.

Let's assume the 2.3 is a 4zd1 and has a ported head that flows well. The head is the only restriction in an n/a engine so we can do some math to calculate peak airflow of the head. Let's assume 80% efficiency, therefore engine flow is calculated using the equation:

Engine capacity (in Litres) x Volumetric Efficiency (%) x Pressure Ratio (number) x Peak Engine RPM
Flow (in cfm) = -------------------------------------------------------------------------------------------
5660

2.3 x 80% x 1 (atmo remember) x 7500rpm (for e.g.)
Flow =. ----------------------------
5660

Peak Engine flow in cfm is 243cfm (n/a) - this is a s*#t hot engine and assuming optimized ports and valves etc.

To convert to lb/ minute, you divide cfm by 14.27.

Therefore the peak theoretical flow of the 2.3 in n/a form is 17 lb/ minute.

There is some complex maths involved in converting this to actual corrected peak exhaust flow (based on air fuel ratio and exhaust gas temperature), but to optimize the rear housing you would ideally size it at the peak n/a engine flow in lb/ min

To match the appropriate turbine wheel and exhaust housing you would need to look at the turbine pressure map. Garrett publishes theirs and gives us a baseline to work from.

I'll use the turbine corrected flow map for the gt28 ball bearing series turbo - from the website this turbo in the 0.57 housing has a corrected turbine flow of 15lb/ min.

With that housing, you can assume the 2.3 engine will "choke" once corrected exhaust flow exceeds 15lb/min - it is hard to determine exactly when this may occur due to the variables of exhaust temperature, afr etc, but fair to say it would be before the 7500 redline I've used above.

So if you use an ever smaller housing such as what comes on a standard t28 from say an s14 Silvia (0.48 a/r I believe), then this is going to choke even earlier. And this choke will increase backpressure which will increase reversion and blah blah per above.

There is even more to this such as the ideal intake valve opening and closing points relative to crank rotation and piston location in the cylinder, and again for exhaust valve opening and closing.

Generally, opening the inlet valve earlier (relative to piston top dead centre before the inlet stroke) allows for the valve to be out of the way so the air/fuel mixture is aiding flame propagation as the piston comes down the bore 15 degrees after top dead centre. This allows a more complete and efficient burn.

Keeping the exhaust valve closed longer allows for higher cylinder pressures and more torque, but at a trade off of high egt and scavenging (if the exhaust pressure is low enough to allow for it).

A complex balance and mix to get right depending on your goals.

Mick
I wish i had something cool and witty in my signature
Turbocoupe75
Long Term Member
Posts: 1238
Joined: Sun May 13, 2007 9:34 pm
Location: Adelaide

Re: Race Car Update...

Post by Turbocoupe75 »

Thanks Mick

Awesome thank you very much for the insight / technical wording. I had a reasonable understanding of the issues of a incorrectly sized turbo but that will immensely help me understand the limits of my set up and what upgrades will be of best benefit later on.

Originally it was going on a g200, but the E1/d1 is the way of the future.. so with a standard cam and keeping the rev range down And running low boost 8-12psi I may get away with that turbo to get me started. This would still be a huge power upgrade from the 90rwhp from my G180/200 set up.

Yep and looking around at modern turbos the T28 is a dinosaur.. be it very unused...

Once I get the turbo to water air intercooler bend sorted and my manifold spacer made I will build the mechanicals of the motor / head, source ecu and injectors.

Aim of the game is budget but do it right, I would prefer to be down on HP and build a reliable motor than break components by pushing them past their limit.
#Turbocoupe75
Post Reply