3T GTE Tech

[YelloRolla]

Given the ongoing discussions in GW's thread about different (and not necessarily on topic) ideas and theories related to 3T GTE tuning, I thought it a good idea to start this thread.

The idea here is to throw around what you have done and what gains etc. It is probably best to note when posting whether or not something was gleaned form personal experience, is hearsay, or is theory. It would be good to note the source of any theory.

There are a few people on these forums who have extensive experience with the 3T GTE. Kojab for one had the first 3T GTE I ever seen. I think that it was 1992 and Dick's car ran mid/early 13's, this car was inspirational for my build up.

Toyman_75 has lots of experience with T series engines in general.

Let's see where it goes (I will add more content tonight as I am quite busy now). Please start posting - I have got to go.....

editors note - I am NOT an authority on the 3T GTE, nor am I trying to be. I have a fairly well developed engine, but there is more development left (maybe more than has already been done).

Head tech? head type and why?
There are few hybrids getting around. I for one use the 2T G head. I use this particular head because it has a nicer combustion chamber design (IMO). Sure it flows more than the 3T GTE head, but I feel that the 2T G head does not impart as much mechanical motion of the mixture. I believe that the 3T head achieves better mechanical mixture motion by the turn in the port, but offers poor squish. That is home grown theory though and will be subject to further learning.
Spark - twin, single or mixture of both?
A lot has been said about this. I have used various set ups and I have tried different things on the dyno. I tried switching off one of the coils in an attempt to reduce the knock sensitivity of the engine, I tried firing on the front plugs, the rear plugs and combinations in between with no change to how knock prone the engine was. Only changes were to the emissions. These experiments lead to my tuner suggesting that we leave it the way that it was. Admittedly these tests were done in the days before I fitted an intercooler and the car was making ~150rwhp. Shortly thereafter I went to a direct fire coil set up with wasted spark, with an intercooler installed and a revised plenum/intake manifold power went up to ~250+rwhp. This power was made at 24psi with a CT26, the boost would go up to 27-28psi and bleed back to 24psi (cheap and nasty boost controller).

What have you tried? How did it work?
Intake manifold design
text text text - to be added later
Exhaust manifold design
text text text - to be added later
Combustion chambers
some info above - more text later
Turbo choice
I have tried - CT20, CT26, Garrett GT3076, HKS GT2540.
Cam profile
My car has always used the stock 3T GTE cams, dialled to where Toyota said they should be.

What cams do you have? How well do they work?

can you add to this list?

[w810sc]

cam profile maybe?

[kojab]

Kojab - the rotor button is fixed, that is there is no way to
change the event of one spark relative to the other.

If I understand you correctly you say that there is no way to change the event of the front spark plug relative to the rear spark plug in the same cylinder because the rotor button is fixed. Although the rotor is certainly fixed we do in fact have two individual distributors using the same rotor. What this means is, if the the rear coil is fired several degrees after the front coil we will have a separation in spark events.

Take a look at the Knock Control circuit diagram here.

http://users.chariot.net.au/~stmezz/knock.htm

The front and rear coils have a common 12v supply but the primary triggers on both coils are not joined but separated as seen in the circuit. What the ignitor is doing is not known by me and others. If the two primary triggers are in fact triggered at the same time under all operating conditions (RPM Boost) then of course both plugs would certainly fire at the same time and conversely if there is any timing variation between these coil primary triggers the two plugs would fire at different times.

So this is the 60 million dollar question. What is going on inside the 3T-GTE
Ignitor

The maximum boost that I ran with the twin plug head set up was ~28psi. This was
with the CT 26 and JMR intake manifold. 28psi was recorded early in the rpm
band (around 4000-4500 rpm) which is incidentally where these engines are most
detonation prone.

I am going to start another thread titled 3T GTE tech.

Thank you YelloRolla for starting a new thread.

Dick

[w810sc]

I thought:
twin spark engines fire one after the other on the power stroke causing higher compression as the flame front travels across the combustion chamber. The second spark fires in the extra compressed mix, ensuring a cleaner burn and less unburnt fuel in the exhaust as well as a higher thermodynamic efficiency.
And uses
"wasted spark" system, where - between the exhaust and inlet stroke of each pistons "partner" - a useless spark occurs, this may improve emissions very slightly, but the intention is to reduce cost.

Great thread btw.

[Hulk]

Finally, I though this thread would never start. Subscribed.

[thechuckster]

fwiw: the 3TGTE in Simon Konig's TA does not use a trailing spark table (Haltech ECU) - both igniters fire at the same time.

w810sc: the second spark acting as you would describe would behave like a waste-spark system - it would put a significant load on the coil firing late as a spark firing into post-combustion gasses is nowhere near as powerful as a normal spark and there is much higher resistance across the spark gap. Given that both coils are charged up to the same amount of energy, it's unlikley that they'd use a system that deliberately stresses one coil.

The rotary system uses trailing spark as the flame front moved up the side of the chamber following the same rotation/direction as the rotor - and the distance between the two plugs (plus the indentation in the housing) makes what is in effect conjoined combustion chambers.

In the case of the 3T (and the early Alfa twin-spark systems), there isn't a huge gap between the two plugs - but there is potential in the head design for the flame front not to move across the piston head evenly - so by having two ignition sources, the combustable charge is more likely to evenly burn across the head/piston surface rather than travel in one direction and halfway thru the burn, make the remaining gasses spontaneously ignite resulting in detonation.

my gut feeling is that the twin spark reduces the likelihood of detonation - combined with the anti-knock system - making the engine more reliable and less likely to require onging/significant maintenance. It wasn't released at a time when emissions standards were excessively strict

[YelloRolla]

Kojab, I didn't use the knock control computer OR the boost references. I simply grounded the wiring that was supposed to hook up to the pressure switches when I had the stock ecu. IIRC there was no feedback from the coil/ignitor to the ecu, and in the diagram that you linked, the feedback from the ignitor goes only back to the knock control unit.
When my car was using the factory ecu I originally used factory 3T GTE ignitor and then afterwards I used 2 x Bosch (BIM 024) ignitors - these were wired so as to fire simultaneously. Maybe this is why I never saw any difference with ignition events?

I understand this process similarly to how thechuckster explained it. I see that if you have the flame ignited from both sides of chamber, then you can mask/compensate for poor mixture motion and reduce the shance of secondary pocketing within the combustion chamber. In simplest terms, you are doubling the spark energy! Would you want to reduce it in any way? (rhetorical question aimed at nobody).

The more ignition advance that an engine needs, the less efficient the combustion chamber (I think that I read that from Smokey Yunick). This tells me that we want the flame to propagate throughout the combustion space as quickly and controllably as possible. Everything we do (I do; incase we are not following the same performance path) is trying to make it happen faster; better fuel (control), more boost (faster flame propagation), better squish (control), better air/fuel homogenisation (faster flame propagation). When tuning (that is when I watch Alex tune my car), I want to see an engine that will tolerate a lot of advance, but I want the engine not to need the advance (that is it makes no more power if we keep winding timing in). /end ramble.
With that all in mind, I cannot see any benefit from delaying a spark. Only benefit in burning everything faster, and faster to me means that when the mixture is all squished and ready to burn, not before and not later.

[The Witzl]

I'll chirp in here and add what little i know about the stock ignition system when i played with Riceburna's 3TGTE..... the ignitor its pretty stupid in reality, here's my lamens explanation of how it works:

• VR sensor picks up the little signal from the dissy's star wheel, and sends signal to ignitor
• ignitor recognises that signal, and directly uses that signal to trigger its two coil outputs.
• there is no "staging" of the coil outputs - they are both fired simultaneously as they are triggered by the same event (the VR pickup in the dissy)

We are talking about electronics from 1982 here guys - it was pretty basic back then!

For reference, when we then put a Microtech LT4 (or LT8) onto Riceys 3TGTE, the stock twinspark ignition with distributor was maintained - but ignition advance control was handed to the Microtech.
This was done by welding the advance plate of the dissy and feeding the VR pickups signal to the Microtech, which then applies the necessary ignition advance and provides a SINGLE output trigger for a dumb ignitor.
THe ignitor we used was a dual output number, which had both outputs triggered from the same signal - just like the stock system.

That make 150rwkW from memory, with a CT26 on the stock manifold with about 15psi (limited by factory injectors)
.

I find these ideas intriguing, i wish to subscribe to this newsletter.

[The Real Roadrunner]

the easiest way to end the leading/trailing arguement is to put a std 3tgte on the dyno.

and it must run std ecu and ignitor/dizzy setup

1 do a run with both plugs firing
2 next run disconnect 1 coil
3 then disconect the other coil and reconnect the orig coil, do a run and then compare results.

if 1 plug is fired b4 the other then there will be vastly different power output comparing the single coil runs.


cheers
linden

ps if someone donates a car ill get the test done for nicks.

[thechuckster]

i think you'll have power output changing simply by virtue of having half the spark the head is designed to run with.

you could test the theory without an engine - just drive the dizzi with a drill or lathe and put a 2-channel 'scope on both igniter outputs from the ECU

[The Real Roadrunner]

i think you'll have power output changing simply by virtue of having half the spark the head is designed to run with.

semi true, but if one sparkplug is retarded then it should lose some power over the one that is advanced therefore you would be able to see it on the dyno (unless other factors like crap mixture distribution come into play but i havn't seen the chamber design so im not sure)

cheers
linden

[kugzs]

Linden, im happy to let my car as the stock 3tgte to test on.

[YelloRolla]

the easiest way to end the leading/trailing arguement is to put a std 3tgte on the dyno.

and it must run std ecu and ignitor/dizzy setup

1 do a run with both plugs firing
2 next run disconnect 1 coil
3 then disconect the other coil and reconnect the orig coil, do a run and then compare results.

if 1 plug is fired b4 the other then there will be vastly different power output comparing the single coil runs.


cheers
linden

ps if someone donates a car ill get the test done for nicks.

I did this - with the exception of the knock computer. The point that the power outputs will be vastly different enhances my point that "why would you want a spark later?" So one plug has the potential to make less power?

[TheToyman75]

I did this - with the exception of the knock computer. The point that the power outputs will be vastly different enhances my point that "why would you want a spark later?" So one plug has the potential to make less power?

Work is a little off tap this week getting cars out before Xmas but I think there are actually 2 basic points on the go here. The first is what does a 3T-GTE stock Factory ignition system do ? We have already established its NOT run by the ECU and runs a simple twin post rotor button inthe dizzy. I haven't run a stock 3T for a good few years now BUT my research way back when supported that both Spark plugs fire Simutaneously. Jasons Personal experience yeilded the same conclusion and Karl's investigation as to how the ignitor works (For riceburners car) all points to the same thing - They fire at the same F#cking time !

As chuckster pointed out - Its easy as to check, just spin the Dizzy up in a cordless drill with the ignitor wiring etc dummied up and check the outputs. My dizzy is modified to work with the SM4 or I would do it now myself.


The 2nd point of discussion seems to be if firing 2 spark plugs at the same time is a good idea ? As I wrote in GW's Dyno thread:

Installing 2 spark plugs per cylinder increases combustion efficiency and reduces the time it takes for complete combustion to occur, This means ignition timing advance can be reduced because the spark can be fired closer to top dead centre.

reduced ignition timing (And still allowing for a complete burn) provides a performance gain due to the reduced amount of the combustion process happening BTDC. (Ie less power pushing back on the piston as it approaches TDC)

The Chuckster clarified it nicely with this:

It short-circuits the "flame front taking too long to propagate" by starting the fire in two locations.

a better defintion? ... Having a long flame travel in a less-than-satisfactory combustion chamber shape (e.g. poor squish) makes it prone to detonation, thus the twin flame fronts reduces the risk of the charge furthest from the spark/flame-front detonating before the flame front reaches it.

And Jason has numerously pointed out this same basic concept and described it best here:

I understand this process similarly to how thechuckster explained it. I see that if you have the flame ignited from both sides of chamber, then you can mask/compensate for poor mixture motion and reduce the shance of secondary pocketing within the combustion chamber. In simplest terms, you are doubling the spark energy! Would you want to reduce it in any way? (rhetorical question aimed at nobody).

The more ignition advance that an engine needs, the less efficient the combustion chamber (I think that I read that from Smokey Yunick). This tells me that we want the flame to propagate throughout the combustion space as quickly and controllably as possible. Everything we do (I do; incase we are not following the same performance path) is trying to make it happen faster; better fuel (control), more boost (faster flame propagation), better squish (control), better air/fuel homogenisation (faster flame propagation). When tuning (that is when I watch Alex tune my car), I want to see an engine that will tolerate a lot of advance, but I want the engine not to need the advance (that is it makes no more power if we keep winding timing in). /end ramble.
With that all in mind, I cannot see any benefit from delaying a spark. Only benefit in burning everything faster, and faster to me means that when the mixture is all squished and ready to burn, not before and not later.

With all that in mind WHY would anyone want to only run one plug per cylinder and reduce there engine's performance ? Surely its about as good an idea as using 91 Octane fuel instead of 98 ?

[kojab]

Kojab, I didn't use the knock control computer OR the boost references. I simply grounded the wiring that was supposed to hook up to the pressure switches when I had the stock ecu. IIRC there was no feedback from the coil/ignitor to the ecu, and in the diagram that you linked, the feedback from the ignitor goes only back to the knock control unit.
When my car was using the factory ecu I originally used factory 3T GTE ignitor and then afterwards I used 2 x Bosch (BIM 024) ignitors - these were wired so as to fire simultaneously. Maybe this is why I never saw any difference with ignition events?

Yes this makes perfect sense.

Unfortunately I could not get my hands on the factory ignitor or the knock control computer so had no chance to do any testing. My previous engine before the 3T-GTE was a Turboed 2T-GEU running Delco so when the 3T-GTE went in the Delco was retained and re-tuned. I did initially run both plugs using two 2TG ignitors with them both firing together but for some reason and I am racking my brain to remember (it was 15 years ago) why I ended up firing one plug per cylinder.

My 3T-GTE engine is still unopened, not even had the rocker cover off. My aim was then and still is to see what the maximum amount of power I can extract from this unopened engine.. At the last (a few years back) Toyota Dyno Day I made 166 RWKW firing on one plug so if I can safely implement simultaneously plug firing again it will be interesting to see what the power difference would be.