80/TA/LT1

iTrader: (2)

Hey guys, just got my car tuned and the tuner said no matter what he did i was getting knock above 10 psi. I'd like some opinions on how to fix this, i'm thinking my hotside is too small at 2 inches, but i'm new to turbos

my setup

5.3 stock other than pac 1218 springs

truck manifolds 2 in hotside into a t4 vs racing denmah kit gt 45 69mm

3 inch intercooler

walbro 450, deka 60s, 1-1 rate fpr

br7ef out of the box gap (.024 /.025 )

if i mised any info just ask, i'll add it.

Just want to head in the right direction before i throw money at dyno sessions that **** aint cheap lol

thanks in advance

80/TA/LT1

iTrader: (2)

Oh 93 octane pump gas also

stevieturbo

And exactly how was he determining "knock" ?

What compression, timing, AFR's etc etc ?

Surely your tuner had some input as to what might be the problem or need changed ?

80/TA/LT1

iTrader: (2)

I don't know, honestly. I am new to turbos and tuning so bear with me here. He was thinking hotside too or the turbo itself having too small of exhaust size. I have hp tuners and am trying to learn how to datalog so i can put some. I just thought something in my setup might be an immediate red flag to somebody....

stevieturbo

It doesnt really answer the question, the most important of which is exactly how was he determining knock ?

It seems far too many just blindly and naively watch knock sensors...which are simply microphones that hear all noises.

And if it's a turbo kit that works for others, that would rule out such thoughts.

80/TA/LT1

iTrader: (2)

I'll have to check with him and report back. And i agree, this is why i posted here

JoeNova

Nothing wrong with your hot side or turbo.

The biggest part of having someone else tune is that any info we give you is pointless. You don't know how much timing is being used, you don't have datalogs, you probably don't have a copy of the tune itself, know the knock sensor settings, AFR, IATs. And then any info we give you and you try to pass to your tuner is going to go in one ear and right out the other as soon as you say "I read on the internet".

forcd ind

iTrader: (7)

Not that it has anything to do with anything, but first thing I would do is ck the gap on the plugs-if you just took them out of the box and installed them, gaps could be all over the place. There could be something else causing the sensor to think there is a knock.
As said, what are the specs on the tune.

stevieturbo

Doesnt matter what the sensor says, it's up to the tuner to determine whether there is knock or not. The sensor when all calibrated correctly within whatever ecu they are using can be a helpful addition..or it cannot.
All basic tuning.

If the dyno time is expensive, you expect a competent operator.

ddnspider

iTrader: (26)

Stevie how do you typically check if knock on a sensor is real or not other than plug reading, especially when not on a dyno?

80/TA/LT1

iTrader: (2)

I do have a copy of the tune, just no datalogs yet, i just started learning hp tuners. I will double check plug gaps, thanks for the advice, so far and please keep it coming, i'm going to try and fix this myself with your guys help, i'm not just gonna run straight to the tuner,i'm broke lol.

stevieturbo

Simple...you listen.

Although on road, ideally from the passenger seat.

There are many electronic listening devices to hook up to the engine, although these really do vary in performance. And you still need to know what to listen for.

Eg and in no particular order

http://www.plex-tuning.com/products/plex-knock-monitor/

http://www.phormula.com/KnockAnalyserPro.aspx

http://theknockbox.com.au/

Or DIY electronic versions

http://www.autospeed.com/cms/article...-Part-1&A=0348

But...I find they really dont filter out background noise as well as they should, although the above are quite good ( use use the top two )

That said, I sometimes find the best...is the cheapest. Just google "diy det cans" and there will be dozens of options. Basically get good ear defenders, some tubing and a piece of metal/copper tubing to affix to the block to route the tubing between the block and headphones

eg

https://www.enginebasics.com/EFI%20T...et%20Cans.html

Overall this method can be the quietest with less background noise, but detonation can sometimes be a little harder to hear. But once you know what to listen for, you can easily pick out even the mildest onset of detonation.

But your ears really are a superb filter simply because you know what noises are regular, and irregular. ie all engine noises would be regular, detonation would not.

Trying to listen on a dyno or passenger seat is easy. I would not recommend trying to listen from the drivers seat.

But all the above is just very very basic tuning, and would be how you would verify whether any knock sensor settings are tuned correctly for each application. If need be you could induce light knock on a single cell to confirm everything is working as you expect, or tune a single cell at a time that you always pass through around peak torque. That should present almost zero risk even if some knock occurred.

I would be shocked if any dyno was not using such things when tuning...in fact I'd say stay clear, it really is basic stuff !

JoeNova

I never gap my BR7s, they always come in perfectly straight out of the box for me.

Knock sensors aren't an exact thing. If they hear a noise outside of their sensitivity setting, it picks it up as knock. If you're making boost on stock sensitivity settings, its going to detect knock whether its actual knock or not.

ddnspider

iTrader: (26)

Interesting info. I've always found the LS sensors settings to not exactly be accurate and give quite a bit of false knock.

stevieturbo

And as I keep repeating.

A knock sensor is just a microphone. It listens, nothing more. What you choose to do with that information is what matters. A knock sensor does not give false knock, but whoever, whatever is listening to it could misinterpret what it hears as something else, if it is not all configured correctly.

That's like saying a digital boost gauge gives you a wrong reading because the display is upside down. No...you're just not using it correctly.

The knock sensor is wired to the ecu. The ecu has various configuration settings. It's up to the "tuner" to ensure all of these are correct as best they can be for the engine they are working on.
It would be like throwing a different map sensor onto an engine then blaming something else if it blows up. The map sensor was working fine, doing it's job fine...but the tuner still needs to configure the ecu etc to work with that item.

Obviously in the case of an OEM engine, wiring, ecu etc etc then what GM have done should work pretty damn well. But that may not always be the case, and every engine/install is probably not exactly as GM left it.

truckdoug

iTrader: (25)

reading the plugs is a pretty good way to go about it

stevieturbo

Also slow, and rather difficult to do in real time on a dyno and nigh on impossible when driving.

better to catch and prevent further at the instant it may be happening, instead of much later.

truckdoug

iTrader: (25)

yes but very reliable. as usual a combination of methods is a good bet.

off track a bit, i have chatting with an engineer buddy about 2.2 ecotec swaps. I learned they use some fancy computing to sense combustion results through the sparkplug itself...instead of knock sensors

roastin240

iTrader: (13)

The factory knock sensors on an otherwise stock engine works reasonably well...the knock sensors are tuned for the harmonics created by the factory engine rotating assembly and valvetrain vibrations...Also the engine mounts (engine rigidity) makes a difference as the chassis is its own harmonic system as well. I mention all this, because when a lot of these things change, sometimes some adjustments to the knock sensor settings need to be made. I have seen false knock on a few of my customers cars.

I would get ahold of the tune and some data logs and see where your operating. If he has 25 degrees ignition timing in it at 10psi on 93 octane then ya it will probably actually knock and not just be false knock. If the sensor is reporting knock, and I pull timing out in that location and the knock does not resolve itself, you can usually assume its false knock. I desensitize it in various ways until it is gone, then introduce some timing until the knock sensors start to report knock in the same or different location...this gives me a kind of indication that the knock sensors are following the trend.

If I am on the ragged edge and not sure if the knock is real or not, I use a pair of knock Ears...they have allowed me to squeeze every last bit out of a setup when I thought it was done...it wasnt, the knock was being reported but not actually there.

For the most part I rely on the factory knock sensors, but as I said, they may need some adjustment. And without your "tuners" tuning parameters we cant really guess if its real knock or not.

BTW the 2 inch collectors are not the issue. Your turbine housing is large for your power and the collectors are not restrictive at that power. Meaning backpressure is not causing high EGT's, resulting in knock. Either his tune is too aggressive or the knock sensors need to be adjusted.

Hopefully that was a little more helpful and constructive than the previous responses

stevieturbo

Yea, them/Saab use ion sensing in some way via the plugs/coil pack.

I've heard of people retrofitting the ecu/hardware to other cars too, although never actually spoken to anyone who has done it.

Some other ecu makers use crank accel/decel to determine if knock is occurring. Doable on a 58x wheel, but not a 24x as it's resolution is *****.

A lot of modern diesels now have pressure transducers built into the glowplugs, so you'd think it's long overdue there should be cheap and foolproof methods of detecting knock and eliminating the risk of it for any build automatically. The technology is certainly out there.