FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 
Back to site

Det hearing system
Goto page Previous  1, 2, 3
 
Post new topic   Reply to topic    EFI University Forum Index -> Tuning Tips
View previous topic :: View next topic  
Author Message
mrx



Joined: 05 Nov 2005
Posts: 521

PostPosted: Thu Nov 22, 2007 12:40 am    Post subject: Reply with quote

F1 uses nearly pumpgas from what i know. maybe with some new additives. but it's not wonder fuel.

first i think the engines are perfectly tuned and second they have rpm on their side... at +10.000rpm it's allmost impossible to get knock because of the rapid combustion process... so at high rpm's it's allmost possible to run at MBT without knock...

also i think the combustiuon chambers and intakt manifolds / exhaust backpressure is very optimized to reduce the change of knock.
Back to top
View user's profile Send private message
sohc_mshue



Joined: 04 Mar 2006
Posts: 372

PostPosted: Thu Nov 22, 2007 1:32 am    Post subject: Reply with quote

Louie928 wrote:
sohc_mshue wrote:
i wish i had some pics of my new knock listening device i made. It uses a piezo mounted on an alligator clip, who's signal is fed into an amplifier circuit powered by a 9V battery and ran into an audio jack for plugging up headphones. It also has an adjustable volume control knob using a simple 10k pot. Works pretty effectively, but i'll need to do some testing to compare it to my old copper pipe with tubing knock listener i've been using haha. I was easily able to pick up some knock on my friend's subaru.

Are you really sure the piezo on the alligator clip works as intended? In my experimenting with knock sensor location, I found that the sensor had to be bolted solidly to the cylinder block. Bolted to the head wasn't good, and neither was bolting the sensor to something else that was isolated from the block by a gasket (such as intake manifold). Any acoustic isolation cut down on the effectiveness.

Yes i have found the same as you. I mainly tune hondas and i typically bolt my copper pipe onto the front of the block around the middle of the height of it. When i've tested the new device on hondas i've clamped the aligator clip onto a bolt and tightened the bolt snug until it was fairly tight on the clamp. You are right though on the subaru it was just a quick test we did right after making it to see how it performed. We slapped it on the intake mani, which i know isn't the best place, but it did pick up knock. His was loud enough that you could somewhat hear it without a listening device though...with the listening device it was pretty loud. Sounded just like knock and was always occuring around the same rpm. I was suprised though as his engine is pretty damn stock with the exception of a few exhaust upgrades. I thought the subaru ecus had a better learning mode than that to control long term knock.
_________________
www.shuetuning.net
efi101/102 graduate
Back to top
View user's profile Send private message
Louie928



Joined: 31 Dec 2006
Posts: 26
Location: Mosier, Oregon USA

PostPosted: Thu Nov 22, 2007 1:40 am    Post subject: Reply with quote

Flr Power wrote:
Yes I agree. So according to what you wrote, we can say your engine seems to make maximum power near the detonation zone. So obviously, using the J&S device is great under these circumstances. Low octane fuel has to be the worst enemy. I have known the J&S for quite a while but never took the time to fully try and test it.

I just wonder what the Ferrari F1 uses for knock defense, or do they really have the engine "perfectly" tuned. Anyone knows if F1 team uses knock sensors?

The timing was such that when set for minimum advance for peak torque with 92 octane, it would have some detonation with 89 octane. The J&S fixed that.
Back to top
View user's profile Send private message Visit poster's website
John at J&S



Joined: 12 Nov 2005
Posts: 291
Location: GARDEN GROVE, CA

PostPosted: Thu Nov 22, 2007 2:07 am    Post subject: Reply with quote

Subaru and Honda used to be a big market for us.

There was a big thread about the TurboXS Knocklite a while back on Nasioc. Somewhere around post #76 a guy said he had trouble setting his up until he borrowed a J&S.

Louie: Do you have mode switches 1 and 2 both down? How much timing does it remove when running 89? Observe knock amplitude (trace knock, hopefully) on your scope when it's retarding.

Do you think it's retarding about the right amount, or more than necessary? Some say knock controllers retard more than necessary, due to knock hysteresis. The theory is that it takes more retard to make knock go away than it would have taken to prevent it in the first place.

To test, run 89 with the sensitivity fully CCW (J&S deaf), but retard the timing until the knock amplitude on your scope is the same as when the J&S was in control. Measure timing, and compare to J&S retard amount.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
sohc_mshue



Joined: 04 Mar 2006
Posts: 372

PostPosted: Thu Nov 22, 2007 2:12 am    Post subject: Reply with quote

John i'm thinking about purchasing a safeguard for my own personal car. Do you think I'd be better off getting the universal setup or to just go ahead and get the honda unit? I dont use a stock knock sensor on my motor and i'd like to be able to use the included GM sensor for where i want to mount it on the front of the block. I havent really messed with the stock sensor on any cars as i've heard it is way too sensitive and really isn't worth a damn or do you think thats mainly just the stock ecu's way of interpreting the signals that makes it not very effective? Also i dont really need any fancy stuff like timing retard/psi. Basically i just want the bear minimum where i can monitor knock on a laptop and have the safeguard pull timing on an individual cyilnder basis.
_________________
www.shuetuning.net
efi101/102 graduate
Back to top
View user's profile Send private message
John at J&S



Joined: 12 Nov 2005
Posts: 291
Location: GARDEN GROVE, CA

PostPosted: Thu Nov 22, 2007 2:37 am    Post subject: Reply with quote

How soon do you need it? The processors we have been using are hard to get, so everything is in redesign.

I can get more if you can't wait. Might be another two months for the new version, but I hope to have test units in a couple weeks.

I'm working on a new eight channel version to replace the current four channel as well as the 1ch/2ch we sold for Honda/Subaru.

The Universal unit is also in redesign, but it's farther down the list.

All new designs will be using the 9S12 processor, as used in the Vampire.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
baldur



Joined: 12 Oct 2007
Posts: 623
Location: Reykjavik, Iceland

PostPosted: Thu Nov 22, 2007 4:33 am    Post subject: Reply with quote

Flr Power wrote:
Yes I agree. So according to what you wrote, we can say your engine seems to make maximum power near the detonation zone. So obviously, using the J&S device is great under these circumstances. Low octane fuel has to be the worst enemy. I have known the J&S for quite a while but never took the time to fully try and test it.

I just wonder what the Ferrari F1 uses for knock defense, or do they really have the engine "perfectly" tuned. Anyone knows if F1 team uses knock sensors?


They do use knock sensors. In fact the Pectel T10 which was used by some teams several years ago had inputs for 10 knock sensors.
Back to top
View user's profile Send private message Visit poster's website
sohc_mshue



Joined: 04 Mar 2006
Posts: 372

PostPosted: Thu Nov 22, 2007 9:24 am    Post subject: Reply with quote

John at J&S wrote:
How soon do you need it? The processors we have been using are hard to get, so everything is in redesign.

I can get more if you can't wait. Might be another two months for the new version, but I hope to have test units in a couple weeks.

I'm working on a new eight channel version to replace the current four channel as well as the 1ch/2ch we sold for Honda/Subaru.

The Universal unit is also in redesign, but it's farther down the list.

All new designs will be using the 9S12 processor, as used in the Vampire.

if i got it then it would probabl be in a month or so, but i'm still tossing around the idea. I've gotta make sure i have the funds first Smile
_________________
www.shuetuning.net
efi101/102 graduate
Back to top
View user's profile Send private message
Louie928



Joined: 31 Dec 2006
Posts: 26
Location: Mosier, Oregon USA

PostPosted: Tue Nov 27, 2007 1:45 am    Post subject: Reply with quote

John at J&S wrote:
<snip>

Louie: Do you have mode switches 1 and 2 both down? How much timing does it remove when running 89? Observe knock amplitude (trace knock, hopefully) on your scope when it's retarding.

Do you think it's retarding about the right amount, or more than necessary? Some say knock controllers retard more than necessary, due to knock hysteresis. The theory is that it takes more retard to make knock go away than it would have taken to prevent it in the first place.

To test, run 89 with the sensitivity fully CCW (J&S deaf), but retard the timing until the knock amplitude on your scope is the same as when the J&S was in control. Measure timing, and compare to J&S retard amount.


John,
I strapped the car to the dyno and did some tests today. As I re-read your question, I see I didn't do exactly as you requested, but I think close enough to get some useful information.

First, I disconnected one knock sensor and observed an increase in low load knocks on the monitor. This was with S1 & 2 both up for 20 deg and retard all cyl. Connecting up the other sensor returned the monitor to showing no, or very few, knocks at low load. This confirmed to me that connecting two Bosch sensors in parallel reduces the output of the sensors. The Vampire gain has to be near full CW to be set properly with both sensors connected.

Next, I did some runs and established a baseline. The Vampire S1 & S2 switches were both down for 10 deg, and individual cyl retard. I use wasted spark for 8 cyl so when one cylinder retards, it retards both cyls on that coil. I saw at most two LEDs flicker intermittently during the run.
I advanced the timing by 10%. That would be about 2.5 - 2.8 degrees at 4500 to 6500 RPM. The monitor showed quite a bit of knock sense activity with a few RPM areas having all 4 LEDs lit. I don't know how much retard was being used. I should have re-set S1 & S2 down, but didn't think of it at the time. You can see areas of the power curve where the knock retard reduced power slightly. I don't think it's a significant amount. Probably less than 1% decrease. Here is a link to that dyno chart.
http://members.rennlist.com/louie928/Ign%20advanced%2010%.JPG

I had a near full tank of about 20 US gallons of 92 octane fuel and I didn't want to drain it to test with lower octane. I did the next best thing by increasing the octane by adding 2.5% of Xylene. Xylene is 117 octane and works pretty good as an octane boost and power enhancer. That small amount of Xylene only boosted the octane from 92 to about 92.6, but enough to demonstrate the difference. I left the timing at 10% advanced over my normal map and did runs with the Xylene mix. I saw a few knocks on the J&S monitor. About like it was with the base timing and 92 octane fuel. I gained torque throughout the RPM range and peak power increased by 8.4 hp to 557.6 and peak torque by 9.3 lb ft to 519.1. Next test was to put the timing back to the base map I normally use and see the result on the J&S monitor and the power difference. The Xylene was still in the tank so the fuel would be the same as with advanced timing. The power dropped back some, but not all the way back to the earlier baseline with plain 92 octane fuel. The J&S monitor showed no knock retard at all. No LEDs lit. The Xylene doped fuel helps in this engine even with the timing set to a very safe advance. It's not much, but makes a measurable difference. Toluene works about the same as Xylene in this engine. The dyno chart for advanced timing and Xylene laced fuel is here.
http://members.rennlist.com/louie928/2.5%%20xylene%20added.JPG

John asked if the Vampire retarded too much. I think the 1 deg steps may be too much for a NA engine. The 10 deg max retard is more than I'll ever use, and 0.5 deg steps would likely keep the power sag to a minimum when retard is required. 10 deg max retard may be necessary for a boosted motor.

I found I got better results by getting my base advance map as close as I could to max power with minimum advance so the J&S normally showed very few retard events rather than simply running the base timing up and letting the J&S handle all the retard. You can see that approach isn't too good on the dyno run with timing advanced 10%. It works, but not optimum.

Some particulars:
The engine is a Porsche 928. V8, 6.4L, DOHC, 11.4 CR, NA.
Power is measured at the rear wheels on an inertia dyno, SAE correction.
Back to top
View user's profile Send private message Visit poster's website
Flr Power



Joined: 05 Jun 2007
Posts: 126

PostPosted: Tue Nov 27, 2007 8:24 am    Post subject: Reply with quote

Thanks Louie for taking the time to write this. This test reinforced the fact that an engine should be tune for mbt. It also confirms that your engine makes peak power very close to detonation zone which make the j&S shine under these circumstances.
Back to top
View user's profile Send private message
Louie928



Joined: 31 Dec 2006
Posts: 26
Location: Mosier, Oregon USA

PostPosted: Tue Nov 27, 2007 11:05 am    Post subject: Reply with quote

Yes, I think my compression ratio may be a little on the high side for pump gas which leaves little room for timing error.
Back to top
View user's profile Send private message Visit poster's website
John at J&S



Joined: 12 Nov 2005
Posts: 291
Location: GARDEN GROVE, CA

PostPosted: Tue Nov 27, 2007 11:50 am    Post subject: Reply with quote

Louie:

Thanks for the test. One point about waste spark. You can say that the unit retards two cylinders at a time, but "retarding" a cylinder on its exhaust stroke has no effect.

When the unit detects knock, it knows that the knocking cylinder is the one that was just on its compression stroke, and won't be on its compression stroke again for two more revolutions. The non knocking cylinder of the pair will be on its compression stroke in one revolution, so software knows not to retard "the pair" then.

If you want to send it back for re-programming, I could cut the knock retard range in half and increase the signal level.
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Louie928



Joined: 31 Dec 2006
Posts: 26
Location: Mosier, Oregon USA

PostPosted: Tue Nov 27, 2007 12:43 pm    Post subject: Reply with quote

John at J&S wrote:
Louie:

Thanks for the test. One point about waste spark. You can say that the unit retards two cylinders at a time, but "retarding" a cylinder on its exhaust stroke has no effect.

When the unit detects knock, it knows that the knocking cylinder is the one that was just on its compression stroke, and won't be on its compression stroke again for two more revolutions. The non knocking cylinder of the pair will be on its compression stroke in one revolution, so software knows not to retard "the pair" then.

If you want to send it back for re-programming, I could cut the knock retard range in half and increase the signal level.


Hi John,
Thanks for that explanation on how your clever device sorts out the correct time to retard the spark. Also, thanks for offering to reprogram the unit. I will take you up on the option to cut the retard in half. The gain is ok. Just because the gain control has to be near full CW, it's ok. I'd just set it back to where it is now anyway.
Back to top
View user's profile Send private message Visit poster's website
Display posts from previous:   
Post new topic   Reply to topic    EFI University Forum Index -> Tuning Tips All times are GMT - 7 Hours
Goto page Previous  1, 2, 3
Page 3 of 3

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group

©2007 EFI University
Website designed and maintained by 3LizardsMedia.com