Louie928

Hi All,
I've been using a product since June and I'm quite pleased with it. It is a J&S Electronics Vampire knock senser timing retard unit. I believe I've seen a few others here that use it too. As I've run my GT more and taken longer trips, I found that the fuel octane changed enough depending on where I travelled to cause concern with detonation. Some locations have 93, 92 is all I can get where I live, and recent trips South only had 91. Go to a really high altitude loacation and could be all you find is 89 (at premium prices). That 89 will still be in your tank when you get back to near sea level. The compression ratio in my GT is 11.4:1 and with timing set right at peak torque, I don't have much margin for lower grade fuel before I could detect audible detonation even at less than WOT. Boosted engines will have an effective CR close to, or even higher than mine NA. I didn't want to set the timing back for a larger safety margin so I had to find a solution. The J&S was it. Now, I know my situation doesn't affect very many 928 owners. The big exception is the pre '87 cars with superchargers. Specifically the '85/'86. John Speake now has his Sharktuner available to tune the EZ-F on those cars as well as the LH. That means you can set the timing so it doesn't detonate and also has the most torque available. You can also set the A/F mixture for good power without using excess fuel for detonation control and having that excess fuel wash down the cylinder walls and dilute the oil. The Sharktuner is a good thing and we all appreciate John's efforts to bring it about.

The unknown part is when you get fuel of less octane than when the car was tuned, or the engine is hotter, or has a few more miles on it and more carbon has built up to make it more prone to detonate. Then what? You could use the low octane fuel jumper and take that performance hit, but who wants to do that. You could throw in a can of octane boost. Might help. You could tune timing very conservatively and give up 10 - 20 of those hard won boosted HP. Instead, I'd suggest the J&S Vampire. This unit will automatically retard your timing to take out detonation. It works better and is more precise than the EZ-K on '87 and later 928s. It has a "gated" listening period so only listens for detonation when it could be occuring which makes it less susceptible to retard on continual engine background noise. It also retards each cylinder individually and works with the dual coil ignition on the '85/'86. It'll also work on the single coil pre '85 928s to give them automatic retard when detonation is detected on an individual cylinder basis. It has a manifold pressure sensor and controls to adjust retard depending on boost. This would be very useful to the boosted pre '85 928s that don't have the luxury of the benefits of Sharktuning. Ther is also a separate knock monitor gauge with LEDs to show how many cylinders knocking and how much retard is being used. It's a useful tuning tool for older than '87 cars without Sharktuner knock sense input as well as on those few 928s with aftermarket ECUs.

Here is a link for information on the J&S Vampire.
http://www.jandssafeguard.com/Vampir...S_Vampire.html
It's relatively expensive, but for me cheap insurance.

I contributed some J&S testing information on another (tuners) forum and I thought I'd bring it over here.

Here is a dyno chart showing what happens when using the J&S to protect your engine. The black line chart is the baseline run with timing set as close to optimum as I can reasonably expect to do. It's right at maximum torque and on this engine pretty close to the detonation threshold. I was showing a few knock indications on the J&S monitor, but only 1 or 2 cyl had infrequent detonation which would have caused a short duration (one or two firing cycles) of 1 degree retard. Not really enough to show up as a dip on the dyno. The brown line is after I advanced the timing by 10%. That would be between about 2 and 2.7 degrees in this RPM range. No other changes. The J&S monitor showed quite heavy knock retard. The knock retard was a different amount throughout the range and went from a few cylinders being retarded to all cylinders retarded. I don't know for sure how much retard, or exactly which cylinders, but from the J&S knock monitor, it would have been from 1 to 4 degrees. You can see where the greatest retard occured, by the power loss.

http://members.rennlist.com/louie928...nced%2010%.JPG

Now, you may think you don't want to have this slight bit of power loss. The point is that the J&S Vampire was protecting the engine from sure detonation with the timing I had during this run. You wouldn't normally have the timing set for this much advance. It does illustrate what would happen with my baseline timing and if I had picked up a tank of less than 92 octane fuel. Even 1 octane number makes a significant difference in both power that is possible and detonation protection. To show that, I added 2.5% Xylene, 117 octane, to the 92 octane I had in the tank. That should have made the mix 92.6 octane. With the more advanced timing, I showed only a few knock retards rather than the massive ones I saw with straight 92, and saw about 8.5 more hp and 10 more lb ft of torque. Going back to my normal baseline timing and making a run, I saw no knock retards on the K&S monitor at all with the 92.6 mix. Having a Vampire on the car will make timing tuning easier too because, with the monitor, you can see when knock retard is happening, how many cylinders, or how much retard, and can adjust timing to minimize those.

Here is a link to a dyno chart showing the effects of changing octene from 92 to 92.6 with Xylene. Toluene will do the same thing. Not much of an octane change, but made quite a difference in power produced and knock observed on the J&S knock monitor.
http://members.rennlist.com/louie928...ne%20added.JPG

Here is a formula I came across that gives the effective compression ratio when boosted. I don't think it's totally accurate because it doesn't take into account the different intake temperatures NA or boosted.

sqrt((boost+14.7)/14.7) * CR = ECR

sqrt = square root
boost = psi of boost
CR = static compression ratio of the motor
ECR = effective compression ratio

John Speake

Good thread Louie..... a cheaper possibility for the 85/86 928s could be the knocklite
http://www.turboxs.com/more_info.php?ID=212.

This appears to have a 5v output signal when knock is detected. This could be coupled to the EZ-F octane loop to give instantaneous retard. At present this would be limited to all cylinder retard of 3 degrees.

I haven't had time yet to mount a Bosch knock sensor and test the system, although I have bought a knocklite unit. They are widely available.

AO

Louie... That's quite a post. Obviously you've put a lot of time into this, but it's people like you that help make the advances for the rest of us - it also shows me how much I have yet to learn.

For grins I decided to try your little equation.

for 8.0 psi of boost with a 10.0:1 CR engine my effective CR would be 12.4:1, but I guess I got lost on what that tells me. Propensity for knock? Sorry for being so dense, I'm just trying to relate this to my setup.

AO

I decided to search the interweb and found a couple of ECR calulators. They all put my motor at 15.4:1 with 8.0 psi and stock CR of 10:1.

Here's an intersting read I found on Hondaswap.com

Glen McCartney

Louie, great write up. Good to know that there is something out there that can be easily added to the 85/86, especially as boost is added and work on individual cylinder detection and individual cylinder retard. Is it the T3 2/3/4 channel that is in their product listing that you have used?

SwayBar

Nor does it take into consideration the most important aspect, how long after BDC the camshaft holds open the intake valve which determines an engine's dynamic compression ratio, which is more important than it's static compression ratio.

This thread just re-surfaced earlier today discussing that:

https://rennlist.com/forums/928-forum/114563-gear-heads-altering-compression-ratio-another-way.html

Does that guy even know what he's talking about?

SteveG

This is of great interest. Thanks Louie. Re: newer cars: But why do I have knock sensors then? Is it correct to assume this is a far more sensitive system that makes corrections the stock computer doesn't handle?

Shane

Thanks Louie!! Excellent timing on this!!

Lizard928

Hey Brendan, you should get this with VEMS (btw I have a local VEMS dealer ) and install that into your car.

Louie928

Hi John,
Good idea. That's a good application for one of the simpler systems and I'd think it would help. When researching the different knock sensing devices available, I soon found that simply listening for an acoustic event isn't too effective. You can set the gain so at some point in operation, with detonation, there will be an indication of it happening. For me, that wasn't good enough. The way modern knock sensing systems work is to only listen at the time of firing to a few degrees after firing. That helps eliminate much of the unwanted noise and concentrates on just the critical period. Also, the background noise and knock threshold level, at the sensor(s) is different for each cylinder. A method to store the non-detonating sensor output level for each cylinder is desireable so when a knock happens. Each cylinder has it's own non detonating level stored. That way when a knock event happens it's able to be captured using that cylinders non-knock level as a threshold. Very precise method of detecting detonation. I couldn't find any other knock detecting system that worked like that.

Louie928

That formula is only for some sort of comparison and I can't think of much real world use. It's the dynamic compression ratio that is important and that involves the volumetric efficiency of the engine which changes with RPM. The way virtually all boosted methods measure their worth seems to be by measuring intake pressure. That is not a measure of the amount of air flowing through the engine. It is a measure of how much air the compressor is trying to force through an engine, but being impeded by the intake manifold, ports, valves, and finally exhaust. Open up an engines ability to breathe and boost will go down. So, having boost as one element of the equation with nothing else to account for all the impediments along the way can't give a measure of the dynamic compression ratio.

Louie928

Hi Glen,
I used the T3. I have 4 coils. I explained what I had to John at J&S and he sent me a unit with the wiring harness already made including the cables to connect to the two knock sensors, and coils. Later, I decided to get the 2001 knock sensor monitor so I could better see what was happening and to make tuning easier. The Vampire unit does have a single LED which shows knocks. The configuration of DIP switches on the unit determines what both the single LED or the multi LED display of the monitor will show. Where I mounted my Vampire unit makes monitoring the single LED on the unit near impossible while driving. That LED can easily be remoted and John will tell you how.

Louie928

Yes, he does. It's that without the ability to change cam timing while the engine is running under different conditions of load and RPM, it all doesn't matter much for our ancient 928s as it relates to detonation control. You could change static cam timing and affect either the low end or the high end. The middle RPM performance stays pretty much as it is.

Louie928

I'm not suggesting this for '87 or later 928s unless they have an aftermarket ECU. The normal Bosch EZ-K knock sensor system works well on those and can be sharktuned for safe operation with boost or higher than stock compression, or with different cams. Even stockish earlier than '87 928's wouldn't need it. When you get to pushing the engine with boost where detonation will be a problem, then you need to do something to guard against detonation.

Knock sensors have become the norm on newer cars because that's the only way, at this time, to be able to run the engine near the threshold of detonation. Operating there wil give best power as well as take advantage of wider range of fuels. Other methods of detonation detection are being used, but are way expensive now.

A few aftermarket ECUs have built in knock sensing systems. I think Motec has that, but way out of my price range to even consider and I don't like their way of charging extra (after you've bought the unit at high cost) for each and every additional feature you realise you need later. All the stuff is in there, but you have to pay more for the unlock code to use it. The VEMS has knock sensor inputs too. Two I think. The VEMS does use the gated method of listening for knocks. I don't know if it has automatic individual cylinder knock threshold.

Rick Carter

Thanks for your research and sharing.