Gemini tuning thread

General Gemini Chat
Forum for all gemini based chat & questions
Post Reply
DR_GEM
Long Term Member
Posts: 890
Joined: Mon May 30, 2005 7:47 pm
Location: 2196

Gemini tuning thread

Post by DR_GEM »

Thought I'd start a new topic to share ideas on tuning for all engine types.

Think of this as a database to share tuning philosophies / queries, and gain insight into tuning methods that work

This will mostly relate to ecu controlled fuel and spark (injected), although feel free to chip in with carby tuning and mechanical distributor ideas.

I have been playing with tuning for a few years now albeit limited to my Subaru liberty GT (2004 model, oem ecu with open source tunability). Runs a 2.0 litre turbo engine that is pretty high tech. This engine combo has made 412kw at all four wheels through a 5 speed auto box (heavily modified and built).

Have learnt a heap load along the way albeit there is always more to learn.

I do this for fun and as a result of not really finding anyone willing to spend the time to extract the best balance between power, reliability and safety for my combo. Being a very modern, variable valve timing (across all four cams), alloy boxer engine, with 2200cc injectors and a 650hp turbo, it is tricky to maintain this balance.

Happy to share my learning and knowledge with the forum and also hope to learn from anyone else with experience.

Mick
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: Gemini tuning thread

Post by Jonno »

Subscribed and cause Micks a top c*#t and I think this will be max useful stickied for life
75 TX Coupe
GemiNats > https://www.facebook.com/GeminiNationals
Image
#TurboFamily #Geminats #Bucketship
YT: https://www.youtube.com/channel/UCPKSHb ... akJ04sOYjw
Turbocoupe75
Long Term Member
Posts: 1238
Joined: Sun May 13, 2007 9:34 pm
Location: Adelaide

Re: Gemini tuning thread

Post by Turbocoupe75 »

X2.... Love to tune my Tribeca... but will have to win Lotto first.
#Turbocoupe75
arrow
The Grand Master
Posts: 44
Joined: Wed Dec 15, 2010 6:10 am
Location: Canberra

Re: Gemini tuning thread

Post by arrow »

Mick was wondering if you have noticed power gains by changing the cam timing with different boost levels. I have only ever played with cam timing at modest boost levels. found the best overall power at say 15psi then wound more boost in from there. Do you think there is a ruff rule of thumb of say for every 10 psi you retard cam timing by ??.
Has any one else played with this???

Scooter
DR_GEM
Long Term Member
Posts: 890
Joined: Mon May 30, 2005 7:47 pm
Location: 2196

Re: Gemini tuning thread

Post by DR_GEM »

Thanks Jonno

Hopefully it will form a database for people to refer to in future

Turbocoupe75 wrote:X2.... Love to tune my Tribeca... but will have to win Lotto first.
Happy to help you out to start off with on the subaru - wont cost much to get going. Tactrix cable is about $220 delivered from the states and opensource software is cheap.
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: Gemini tuning thread

Post by DR_GEM »

arrow wrote:Mick was wondering if you have noticed power gains by changing the cam timing with different boost levels. I have only ever played with cam timing at modest boost levels. found the best overall power at say 15psi then wound more boost in from there. Do you think there is a ruff rule of thumb of say for every 10 psi you retard cam timing by ??.
Has any one else played with this???

Scooter

Cam timing changes volumetric efficiency as you would know. (Changing timing of valve opening and closing changes peak engine airflow)

The optimum VE of any engine is dynamic and changes according to conditions. However if we assuming static boost (take an N/A engine for example), then the engine will only have ONE set of inlet and exhaust cam timing values, and ONE ignition timing advance value for given load and rev point.

Therefore theoretically speaking, boost in and of itself should not change VE / peak efficiency of an engine relative to cam and ignition timing.

The one variable however is inlet to exhaust back pressure ratio. Before I go further we should clarify that boost is a measurement of inlet restriction (from pre turbo air filter to combustion chamber). So the ratio of this restriction to the exhaust manifold pre turbo restriction is what I'm referring to when I say inlet: exhaust PR (pressure ratio)

I'm going to go into a big spiel here so for those who know this already (scooter this is most likely you), forgive the detail but I think its important for those new to this.

Old school theory used to dictate turbo engines don't like valve overlap (ie inlet and exhaust valves open at the same time for a brief period). This was because we didnt understand inlet to exhaust pressure ratios and everyone was s*#t scared of lag, so went with the smallest exhaust housing required to bring boost on faster. So what you had was a tiny ass turbine housing choking the engine, and once boost was raised to a certain level, exhaust manifold to turbine back pressure increased exponentially, and therefore airflow was "bottlenecked" in the echaist manifold and inlet to exhaust pressure ratios were sitting somewhere like 1:3 as an example.

You will remember that a combustion engine is just like an air pump, aided by fuel and spark to continue the four stroke cycle. The more air you can pump in and out of the combustion chamber, and the faster and more efficiently you can do this, the more power an engine will make.

Now lets also remember that fluid dynamics laws dictate that pressure favours or flows toward the lower pressure area. Sometimes this is referred to as an engine "sucking" in air which is technically not correct. What is actually happening os the higher pressure is flowing to the lower pressure area. When a piston moves down in a cylinder bore, this creates a low pressurre cushion in the combustion chamber. The higher pressure air in the atmosphere / inlet tract therefore "pushes" or "coerces" the inlet charge mixture toward the lower pressure point in the chamber. Then the 4 stroke cycle does its thing to create torque on the crank etc etc.

Boost itself does not "force" more air into the combustion chamber. This is a false principal. Boost merely increases the density of the inlet charge, allowing more oxygen atoms to be crammer into every square inch of air. Therefore the same amount of airflow is ingested, but a higher oxygen content per square inch is ingested by the "pressurized" or boosted inlet charge. Therefore just like bumping up static compression, you have more compressed air and therefore will generally create a bigger bang to make the above 4 stroke cycle happen faster and with more force, thus makimg more power

A turbo engine uses spent exhaust gas PRESSURE (not exhaust gas flow) to spool the turbine in the turbo. As above, in old school turbo setups, they were afraid of turbo lag caused by a restriction being placed in the exhaust (the turbo itself), and thus overcame this by using a smaller housing and turbine wheel that required a lower amount of exhaust pressure to spool up sooner. This however created high back pressure between exhaust valve and turbine wheel / housing.

When they increased inlet manifold boost, exhaust backpressure rose more, and therefore the inlet/ combustion chamber had a lower presure (even when boosted) than the exhaust. As I stated above, air / fluid will want to flow toward the area with lower pressure. So if your inlet and exhaust valves are open at the same time (overlap), and your inlet pressure is lower than your exhaust pressure, the gases / mixture in the exhaust will want to flow back toward the inlet. This is called reversion where the exhaust gases want to push back against the way they're supposed to be flowing (out of the engine and out the exhaust system).

So mistakenly, engineers and modifiers thought that turbo engines don't like valve overlap. What they didnt consider was the exhaust restriction.

With modern technology, turbo engineering and turbine housings have become alot more efficienct. So these days its not uncommon for a proper turbo setup to have larger turbine wheels and turbine housings. This results in lower exhaust valve to turbine wheel backpressure.

Therefore so long as your exhaust back pressure from exhaust valve to turbine housing is less than or equal to your inlet boost pressure (ie 1:1 ratio), then the lower exhaust manifold pressure will help "suck" or scavenge some of the unburnt air/fuel mixture into the exhaust, which will aid in VOlumetric efficiency sooner in the rev range and assist In spooling the turbo sooner. As stated above, this is only possible with a well thought out turbo setup.

Lets take 15psi positive manifold pressure. This translates to 29.7psi absolute pressure at sea level (14.7 ambient air pressure at sea level plus 15psi boost).

So if for example @ 15psi inlet manifold boost your exhaust valve to turbine housing back pressure is also 15psi, this is a ratio of 1:1 (29.7:29.7). This would allow a higher amount of valve overlap to assist in spooling the turbo early (opening and closing exhaust valve earlier in relation to piston position during/before exhaust stroke in crank degrees before bottom dead centre or after top dead centre).

By doing so, and maintaining a pressure ratio of up to 1:1, you are allowing boosted air to "clear" the combustipm chamber of spent exhaust gases early (ie sometimes incorrectly referred to as "blowing boosted air out the exhaust").

However lets say you up the boost to 20psi on the exact same setup. If your exhaust is now providing more of a restriction, and exhaist manifold pressure climbs to 36psi absolute, this results in a pressure ratio of ([14.7+20] : 36) or 1:1.05. What this tells us is if your inlet and exhaust valve are open at the same time (overlap), then your higher exhaust back pressure will send hot, spent exhaust gases back into your combustion chamber (otherwise known as reversion as stated earlier). This is bad for engine effiency. Very bad. (Great as an egr mechanism though).

So long way of me saying in my view, playing with cam timing will only make a difference if you're still within a 1:1 inlet to exhaust Pressure Ratio. Soon as you go over 1:1, you will not see any benefits in altering cam timing. Given you can't change inlet to exhaust valve opening timing (only in reference to crank degrees before or after top dead centre), then its hard to determine whether changing static cam timing for higher boost will aid anything UNTIL YOU MEASURE EXHAUST MANIFOLD BACKPRESSURE at different load and rev ranges. Lets say you run 50psi and inlet to exhaust PR is still within 1:1, them you can retard your cam more to assist high rpm power (at the expense of low rpm spool and power).

If your turbo is effienct all the way to redline and you're making more power up there. Then do it I say :D

Interesting question would be what optimum cam timing is for your engine (given static inlet to exhaust cam opening and closing due to sohc which does not allow independent adjustment obviously like a dohc engine).

So lets say at 35psi boost your inlet to exhaust PR is still under or equal to 1:1, then at 50psi its at 1:1.5, do you set cam timing for overlap to occur at peak torque or peak revs.

Let me think about this one..

Mick
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: Gemini tuning thread

Post by DR_GEM »

I wanted to clarify what I said above about valve timing in a sohc being static.

Obviously our sohc Isuzu engines are sohc and don't have any variable valve timing systems, even with adjustable cam gear being used. What this means is inlet and exhaust valve opening RELATIVE TO EACH OTHER is not adjustable. Ever.

In a dohc engine, if you have adjustable cam gears on each cam, (even if you don't have any variable valve timing system), you are able to adjust the difference between and timing of inlet and exhaust valves opening and closing independently of each other.

With a sohc turbo engine, I would measure (or check the cam card) for lsa (lobe separation angle), to determine effective overlap.

For example a cam with 250 degrees advertised inlet duration and 250 degrees advertised exhaust duration, with a lsa of 110, gives the following:

(250/2) - 110 lsa = 15
(250/2) - 110 lsa = 15

You will see I've halved the 250 cam duration in the above calc. This is because 250 camshaft degrees is 125 crankshaft degrees as there are two camshaft revolutions for every one crankshaft revolution, so to get timing in crank degrees you divide by 2

Total overlap is therwfore 30 degrees in above example.

The crucial question to tune around is: at what boost level does your inlet to exhaust PR start exceeding 1:1. (Ie higher exhaust manifold pressure than inlet). And where in relation to your boost curve/ power band does this happen.

For example, lets assume peak boost is 30psi, and you achieve this peak boost on your g180 by 4000rpm and it holds all the way to redline (no boost taper).

Lets say your engine revs to 8000rpm, meanin your peak boost is available from 4000-8000rpm or half your rev range.

Lets then assume that @ 30psi and up to 6000rpm, your inlet to exhaust PR is 1:1, however by 6500rpm your inlet to exh PR climbs to 1:1.5. This means that scavenging / spool can be aided with overlap up until 6000rpm, and beyond that you're getting reversion.

So peak boost and almost peak torque would be available for 50% of the rev range, and your boost is not a restriction for 80% of your power band.

Because inlet and exhaust valve opening, closing and duration is fixed in a sohc g series, and we dont have any mechanism to vary valve timing, lift and duration during operation, then there is no way to alter the engine's VE and valve overlap at 6500 (where Inlet to exh PR goes above 1:1), to close up the overlap and therefore stop exhaust reversion.

So the solution is either to remove manifold backpressure (larger turbo housing), or reduce overlap (new cam), or limit revs to 6500 rpm (not worth it).
Last edited by DR_GEM on Mon Sep 07, 2015 6:18 pm, edited 1 time in total.
I wish i had something cool and witty in my signature
User avatar
moley
Long Term Member
Posts: 803
Joined: Fri Apr 02, 2010 11:40 pm
Location: Banjo Country

Re: Gemini tuning thread

Post by moley »

Its actually good to understand why we had results with our cars after different r and d. I never actually thought about the backpressure issue in regard to overlap
Thanks for taking the time to add the details
Coupe rebuild number 6..
viewtopic.php?t=44564
DR_GEM
Long Term Member
Posts: 890
Joined: Mon May 30, 2005 7:47 pm
Location: 2196

Re: Gemini tuning thread

Post by DR_GEM »

Thanks for that -Happy to provide thoughts for discussion

Would be great to hear from others on this topic and their approach / views / learnings

Mick
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: Gemini tuning thread

Post by DR_GEM »

Have picked up some spelling mistakes in my post above so I will edit as in some instances it does change what I'm trying to say
I wish i had something cool and witty in my signature
Post Reply