The AUSubaru.com.au Forums

http://img209.imageshack.us/my.php?imag ... rolsz0.png

If all things work on this first attempt to stick up a non photo image shack helped image I will be able to ask the next question

OK. The link worked ( I was hopng for the diagram to show here, but)

I have it on good authority that spark control advance/retard on the EA82T relies on an AFM signal. I am not questioning that authority, but I was of the impression that from this diagram from the 3plug ECU flapper style AFM factory shop manuals that the knock control circuit that electronically controls the sprk advance retard is a completely independent circuit one stop shop having no ecu or afm input

This is where I thought I would have my spark control all sorted in my LPG conversion/adventure - needing different spark timing - it would look after itself and work fine without ecu (my theory anyway) EXPLANATION is that LPG loves more initial advance in lower revs and less advance at higher revs. THINKING is that this unit will just give it advance until knock sensor reduces it continually. THE SUGGESTED approach is just use/try normal EA82 dizzy (lucky I have an electronic one of unknown working condition) and use normal non turbo initial timing setting. THE BRAIN is just trying think out the 20 DBTDC @ idle of petrol turbo V's the 6-8 DBTDC of non turbo - wasting valuable thinking space again I am sure !!

comments?

In short,.. no

On a 3 plug ECU, spark advance/retard is controlled by the vac servo and the centrifugal weights in the distributor, much the same as on a carby engine.

The knock control unit simply retards the ignition in the presence of knock.

As you can see in the diagram, an ignition timing signal and a phase signal are sent from the ignitor in the distributor to the knock controller. If no knock is detected, the ignition signal is sent back to the coil unaltered through the ignition voltage wire.

If knock is detected, the knock controller adds a delay to the output of ignition voltage to retard the timing. Once the knock is eliminated, the ignition timing is gradually increased until it is back to normal.

Its the 4 plug ECU that calculates its ignition timing via inputs from throttle position, air flow and engine temperature sensors

Thanks Andy !

Gannon, throwng more light on the subject, thankyou also !

thanks to FUJI, I found this in the manual in fuel injection section 2-7 p13

Control Of Ignition Timing
(Australia TURBO Only)

The turbochargers functioning causes the compressing pressure to increase in the vicinity of full load, resulting in the knocking being prone to occur.. To prevent the intensive knocking which may damage the engine, the knocking control system is designed for detecting its occurence and performing automatic retardation of ignition timing.
Furthermore, the distributor of * positive pressure retard type is adopted which performs the timing retardation against an excessive supercharging pressure from the turbocharger.Thereby optimum ignition timing is maintained to enhance the power and the fuel efficiency as well.

* Good authority told me IN PLAIN ENGLISH this dizzies vac advance unit can retard beyond normal vacuum advance rest or neutral position on BOOST. clever dicks weren't they ? Guess we dub it the vacuum advance/ boost retard cannister

Knocking Control System
(Australia TURBO Only)

GENERAL

The knocking control system consists of a knock sensor, knock control unit, and distributor with built-in IC igniter. It optimises the ignition timing so that the engine may develop the maximum output with minmal fuel consumption but without causing a knock. When a knock occurs in the engine, the knock sensor detects its vibration and sends a signal to the knock control unit which determines the intencity and sends a timing retard signal to the IC igniter in the distributor. Thus, the ignition timing is retarded at the moment just before the knocking occurs.

KNOCK CONTROL UNIT

The KCU processes the knocking signalsent from the knock sensor and transmits retarding signals to the HIC igniter built into the distributor. On occurences of the knocking, the momentary retardationis performed. If there appears no knocking, the advance is perforemed. If the knockingis encountered in the course of advance, the retardation is made as in the former case

KNOCKING OCCURS > RETARD >KNOCKING ILLIMINATED > ADVANCE GRADUALLY > KNOCKING OCCURS and so lfe goes


In this fashion occurence of the knocking is supervised at all times to provide the optimum ignition timing against any changein the driving condition, gasoline characteristics (OOPS) and environment

That little lady (I laugh every time I see that Billboard too) has the right idea about low octane fuel .

Basically OEM turbo engines have a lower static CR because under boost the effective or dynamic CR is higher - boosted by the airpump .
High cylinder pressures usually create lower detonation thresholds so the easy way out is to lower the static CR and take steps to prevent knocking ie stipulate higher fuel octane/intercooling which drops the charge temperature/knock sensed ignition retard .

If you look at 80's era turbo cars done on the cheap you usually find "soft" CRs and no intercooling . Knock sensing and ignition retard is relatively cheap to do and usually ensures these vehicles survive the warranty period .
Examples I can think of are Nissan 280ZX Turbo , Bluebird Z18 T , Comode VL Turbo (Nissan RB30 ET) and Subie RX Turbos .

Most of todays turbo cars are intercooled and the CR is higher because these are what gives better throttle response and importantly better part throttle fuel consumption . It's to do with the dynamic (effective CR) being higher and the detonation threshold being higher .

I can see where you are trying to go with LPG Turbo/3 plug style knock control , not sure if you're going to have a win with low octane LPG and lack of an intercooler .
Assuming you can get the mixture control right I think your answer may be water injection though good kits with good ag pumps that can inject into a boosted environment can cost 600 plus .
Expensive yes but solves problems that the usual fixes struggle with . It's basically in cylinder intercooling that can be properly regulated (referenced) against boost pressure . Can even mix up to 50% ethanol/methanol in to help with the fueling side of things .

Cheers A .

still feel this problem is likely to be resolved with renewed valve springs ( I can see SCOOBIDOO smiling here) . With the new HVLA in and 20W50 oil at operating oil and water temp (60 km drive with no boost coz I am throttling off before it comes on) I can now get lifter pump up on backing off in gear then causing the intermittent miss to come and then go quite quickly.

Steve_rising_sun has recommended 10W30 oil for EA82T as has someone else in the past even though manuals indicate 20W50 is good for 0 to 32 degrees C . Taken Steves advice an got some to try -planning to come back in here with the news....

New genuine valve springs exJapan fixed the problem - happy to report five years later finishing off the thread as preferred, than leave it unfinished like I did !