Delboy2

I sent John at J&S a couple of e-mails regarding the Safeguard etc but also received no response The Snet software then changed so received no thread activity either When the thread re-appeared the closing date had passed.

Cheers

Dont take the mic

I understand about not matter about 2.0 or 2.5 but your car is heavily modified? so how have you not used j&s anymore if ecu is not working correctly?
why does ecu not work?

john banks

Exhaust manifold, bigger turbo, higher boost are some of the factors that send the ECU deaf to det.

I still have the J&S in but when I remove it to send to the person that is buying it I will add a bit more methanol or retard the ignition.

The code for the ECU knock correction has been unfathomable so far especially on the 99/00 to work out why it goes off with modifications.

Dont take the mic

the impreza knock sensor could be giving false readings if we tune our cars depending on power level?
I see many peoples in here with tuned cars by exhaust, turbo and other peices but never come across this being mentioned before. Is that maybes why some engines could be blowing up because factory sensor has become deaf?

If there is an issue with std knock sensor how come you selling j&s?

nom

Most people don't fiddle around with their cars to the point that they lose knock control - most just change the back box to something that's noisier & the BOV to something that goes 'woosh'... neither are going to have an effect on knock!
The most common way of making a mess of knock control is with a very small number of exhaust manifolds AFAIK... and of course if you e.g. change the dimensions of the cylinder you are likely to start to change the knock frequency, which means the ECU is then sitting there listening to knock & carefully ignoring it . But at that point you are unlikely to be using the standard ECU - I think John is pretty much on his own here & is first finding all the limitations (and advantages) of the standard ECU!

IMO, the older cars are going to make more use of the J&S than the new (the newer ones - as is typical with newer ECUs - adjust/advance far faster after the onset of knock than the old ECUs). As long as the car is set up correctly, there are advantages to having an advance that doesn't take 3 tanks of fuel to return...
The biggest advantage is probably the 'per cylinder' knock control. I don't think any of the standard Subaru ECUs can do this (John?). Normally, if one cylinder dets, they are all retarded; the J&S only retards the single cylinder (which may be overheating/have a slightly different mixture, etc.) meaning less power is lost to retard in the same situation.

john banks

No, not aware of any Subaru ECUs that do individual cylinder control.

If you didn't know, my car is presently running just over double its standard power on a remapped standard ECU. Give the poor ECU a break it is still doing a fantastic job considering and has been reprogrammed well over 100 times

It is not an issue with the knock sensor, it is getting access to obscure and hidden parts of the workings of the OEM ECU to recalibrate them.

And the aftermarket ECUs often have no adequate knock control either, or it is very primitive unless you pay major wedge.

Dont take the mic

Thx nom for the reply.
This seems to be can of worms because you seem to say on one side we have John exploring boundaries of tuning using std ecu but hold on for a moment and lets think about tuning the engines.
Correct me if i am wrong but an ecu mapper normally maps the customer car by listening to the noise of the engine with his headphones and can make judgements on car timings through this. Correct?

I said in earlier post, why would you want to go to extremes of timing for a car which is used on a daily basis on the road unless you are out to try and gain more customers with your results or wanted to be seen as having most powerful car out there.
My thought only of course but i would want my car to be consistant with my power and use it every day without worrying about this noise and that noise that might or might not be happening inside my engine.

What about all these peoples buying 2.5 engines in other group buy? Are you saying they may blow up because the frequency of the knock sensor is not compatible?

Is a 2.5 knock sensor the thing to buy then or is it more complicated than that? Please excuse my lack of knowledge, i willing to learn as i said earlier. Sorry for all questions but good debate so far with your helpfulness

john banks

You can't just map a small turbocharged engine to ridiculous specific power outputs and not worry about knock and other engine parameters.

If you want that buy a PPP and leave it at that.

If you want to turn the car into something that for example does 60 in the 3s and 100 in the 8s in the wet like mine presently is you have to put a lot of effort in

As the US guys say, you play, you pay. You pay in terms of fickleness, expense, increased maintenance, risk of breaking things. Then you play with bikes, Porsches, Ferraris and TVRs instead of Hot Hatches. Decide what you want.

There have been 2.5s already blown up here and in the US likely because of detonation. Sort the det and the door is open to serious power. On Subarus at least you can't just rely on the ECU to sort out all knock going. Factory knock algorithms on Subarus are only listening at certain loads and RPM ranges. The factory maps are conservative to start with. The boost drops like a stone at high RPM. Start tuning them seriously and you lose these safety margins.

Porsche and Audi don't use miniscule boost on their big power motors for no reason... they drive more like NA and are extremely reliable, very tolerant to poor servicing, detonation resistant etc.

Dont take the mic

Thx John
So the knock sensor is okay. It is the ecu which has the problem? Okay. I understand.
Best to look at something which has adequate knock control for the car if going to absolute limit.

You obviously have spent a lot of time tuning your car John and learned from it what works and what doesn't but there is no getting away from facts that if you have adequate air going in and fuel and timings, the engine should not be detting. The limits of engine lie within the strength of the block/pistons/rods/bearings.

What i am saying here is why do you want to run your car (or anybodys else) with timings on the absolute edge of detonation? Does not make sense to me for a car used daily.
You seem to come across that you run your car on the edge all the time. I may be wrong but if right, why do you do that?

john banks

I am actually very cautious in the way I tune even my own car, and when I was doing others I was far more cautious, possibly overly so.

The whole reason for active knock control systems is that you can run the engine just before the onset of detonation to produce maximum torque and do it safely for thousands of miles.

Otherwise you run a fixed ignition map which can be compensated by all the variables you can think of, but detonation can take you by surprise and cannot be accurately predicted in the real world by commonly available ECUs without using knock control. When running off a fixed ignition map I do indeed pull something out for safety, or tune it on lower octane than I intend to run it on.

The only things of importance that have broken on my car have been gearboxes, one because it was Subaru's rather limited 5 speed, the other was not up to the spec it was sold for.

A J&S Safeguard used sensibly will allow you to run the base ignition map more to what you can run 90% of the time, and then it will look after the rest by retarding up to a couple of degrees and have some spare for surprises.

If there is det that gets through when you are using a Safeguard it is quickly controlled as the unit can retard up to 20 degrees depending on how you set it up.

If a fuel injector block, the fuel pump slows down, you get bad fuel, suddenly lose coolant etc it could easily save your engine or reduce the damage considerably.

Detonation for a few cycles is bad if it is severe, but if it goes on for too long pre-ignition can result which is invariably terminal.