No problem Bob, this is understood.
I just try to say that if I use the knocklink device and the EGT gauge as reference (comparison) point, I can have partly an indication about whether my engine runs ok or not. Of course must know the duty cycle and fuelling, but from the other I heard that the RRoad analyser isn't so accurate cause the air flow is much more on the road.
May be I am wrong!
JIM

 

Still waiting for your comments Mr Bob!
Stelios where are you to clarify what is happen with your car and the above modifications?
Or am I the only person with such silly questions?
JIM

 

Not silly questions. I hear what Bob is saying. But if the ECU advances to the limit of knock and I have loads of ignition advance all over the range and my top end is fuelling 9% CO on such a lightly modded car I really am not convinced of the worth of a Knocklink yet.

 

Come on Bob, we are still waiting with bated breath...

Matt

 

back to the top

 

PPP+hks manifold+magnex d/p +catless mid+ pdrive back box
driven with stethoscope on block + MoTeC wide band lamda (bosch)and gauge, ambient approx 10 celcious, still rich ! and with rich acceleration enrichment boost never got over 1.05 bar.
APS induction dont think it leans the flow its just a kn filter after all. Its just the long tubing that messes the maf sensor undereading flow so at steady throttle the ecu leans the mix as gi-go in computing jargon (garbage in -garbage out)I think it was Bob that said that and it was done in an earlier thread.

 

Curious as to what Bob was planning to say. Has anyone seen a Knock Link picking up good going detonation and the standard setup (ie not some aftermarket ECU with less sophisticated knock detection/correction) not attempted to retard the ignition assuming it had some more retard available?

ie assuming I have plenty of spare retard, can I trust my lightly boosted engine to the PPP ECU and OEM knock sensor?

I reckon unless someone has seen the above situation I am safe enough. Please correct me if wrong and I buy a Knock Link. Otherwise seems pointless if it will not help me - what is the point in buying a box of flashing lights if I have an ECU developed at the cost of Yen+++++ to do the job of watching it and correcting it? If my ECU could only retard a tiny bit then maybe?

I am not taking the **** just trying to get some answers.


[Edited by john banks - 1/8/2002 12:57:12 PM]

 

I have emailed Bob to find out what he was going to say....don't leave us in suspence Bob, its killing us!

 

Sorry folks but been a bit busy since last Friday.

First off apologies to those that are up to speed with this.

Since the knock sensor is just a microphone of sorts its out put is proportional to "noise", noise manifests itself in many ways, for example an engine with high cam clearances will generate noise, loose heatshield anyone, piston slap ? and so on, the oem ecu contains filter circuits which are designed to stop the extraneous noises from making it act, in practice the oe knock sensor and the Bosch sensor that is used with the knocklink are probably pretty good, the Bosch has more output afaik, that means its not really suitable to use as a replacement with the factory ecu. Knocklink receives signal and translates into led's, it has a sensitivity pot adjustment somyou can set it to be "quiet" according to the level of background noise, for example I was in a car on Sun that had relatively high piston clearances and high lift modified cams (no it was not a standard engine) and the knocklink was hitting the red light all the time due to this, no det at all though, adjusted the pot and set the level so that knock would then show up as a spike above background, without the trim it would be impossible to tell.

Det cans should be used for all medium/high boost work, the correlation between det cans and the knocklink is pretty good and its easy for the "educated ear" to discern when an engine starts to det and when its just generating background noise with the cans on. With the right design you can hear injectors, lifters, turbo bearings etc etc. On link the knock border is adjustable so that when the ecu reads above a set voltage it assumes knock and retards itself in the zone that it registered the transition. However you can't use the same setting for all, cars of the same type and model have different background noise levels which means that det cans are needed to determine the set point. On the 44s link the knock trip level is also programmable to increase with revs ... logical eh. can I hear det that the ecu doesn't react to ? Yes indeed, when my car was running standard management with just an exhaust and filter change it was boosting to 1.3 bar and holding 1.2, above 5800 it detted to a noticeable level using cans, not audible in the car, even on 97 and booster yet the ecu did absolutely nothing about it.

The other thing that should be considered is that the knock sensor only picks up one type of det, there are actually several different manifestations which sound totally different, only det cans will differentiate.

Its worth mentioning that audible det is major, in an enclosed space you can hear "tinkling" which is usually when elevated charge temps are inducing light pinking (rolling roads normally only time this is heard) and if you can hear it when driving then its very serious.

Usefullness of the knocklink ? The oe ecu can only retard so much, so what is going to happen when you fill up with that particularly nasty tank of really crap fuel ... and we all get this from time to time, happens to me maybe 4 times every 6 months. Without some sort of warning indication the car will det, yes it will retard itself but only so far, and then you drive round wondering why it feels like a slug. Also knock sensors are known to fail ... again if that happens then no retard at all ... bad tank and woops.

Early pre MY98 ecu's retard but don't advance very well, thats when they will benefit from an ecu reset, MY99-00 are better and more dynamic, they don't benefit from resets in general terms as they do adust according to the knock sensor but not "real time" there are still limits once again, the MY01 on uses OBD ecu technology and is real time dynamic (just like Bosch Motronic only they were doing this in 1993, Motronic can even differentiate as to which cylinder is suffering and only retard that one) so it "lives" on the knock sensor output advancing and retarding real time constantly.

Just because you have upped the fueling to compensate for mods does not mean there will be no det, trying to control with charge cooling from overfueling is a classic method but no substitute for correct fueling with proper control of charge temperatures allowing best ignition advance. The oe ecu on MY97 on is very overfueled as standard in most cases and will tolerate mild exhaust and induction mods, MY99-00 are more susceptible as they use a different MAF design which is very sensitive to air flow patterns. Thats why some owners report "hesitation" etc with these cars, the MY00 also suffers in that the ecu fuel map appears to be a bit borderline in a couple of areas wrt leanness. MY96 back, though, are not generally as overfueled and ran with bigger turbos and slightly lower boost, the boost cut is also lower. this means they are more likely to run lean as a result of exhaust and induction mods.

A PPP has been tweaked to improve mid range performance, Prodrive have sensibly played safe and not pushed the top end much especially bearing in mind that they had to cope with owners running the PPP on 95 as well as 97 ron fuel. A boost controller on a standard ecu will/should be able to produce as good a performance as a PPP without, a PPP and boost controller means that the top end can be lifted and you then end up with similar cars. Its not fair to try and compare two different cars in the two conditions, the cars themselves will be different and contribute to the result even ignoring the ecu's, same car with the two set ups is the only comparitive way. Regarding ignition advance the thing that seems to make the most difference to the way the oe ecu runs is compression, it vary's quite a bit in my experience, this will allow different cars to seem better or worse than a conterpart.

 

Thanks very much Bob - a very helpful explanation.

 

Yep - cheers Bob

So a qestion leading on from that is, if you buy a Knocklink, how do you know where to set the adjuster?

Matt

 

... And also Bob, have you ever examined a ppp with decat dp and seen det? Basically, in your experience am I safe to fit a downpipe without having to continually monitor for det?

 

The only time I have seen det on a UIK car with eihther standard or PPP is when the car had poor fuel and/or something just "wrong". You decide where to set the sensitivity using det cans, you can, however, just leave it set at max sensitivity, you "may" get a steady bottom bright green (or even a second) but if its steady then it should not be det.

Paul, I would seriously consider one if you are going to use 95ron fuel but you should not need one if running on 97. Thats for a UK car, jap spec should have one irrespective.

BTW the select monitor can certainly be used to monitor the lambda output and shows voltage. Its also possible to datalog using it with the right interface.

 

Cheers for that Bob

Ta,

Matt

 

Cheers Bob,

I always run on su as recomended for use with PPP. If this is your experience, and from what I have seen so far I think i'm gonna be o.k. I might get one anyway just to alert me of bad fuel.