J&S "Safeguard" individual knock control system...

Question, how fast of a data logger would we need to log cam sync? I been doing some math and it looks about 60 fps. The fast dash logger can log at 100 fps, would this be fast enough to see the retard and how would we determine which cylinder it is? Would we need a faster logging rate for this?


7500 rpm/2=3500 rpm/60 rev seconds=58.8 cam sync events per second?
 
IMO, logging which cylinder is knocking would be critical. Then you could fix the problem, instead of just having that cylinder knock all the time and using this device to remove the knock. In other words, be proactive and fix the problem instead of being reactive AFTER it knocks. I can certainly see the value of this device for those who run their car on the edge of detonation all the time.
 
The gauge in the video posted earlier displays knock retard plus the voltage from a narrow band O2 sensor.

The top row of LED's show two cylinders are being retarded, one more than the other.

The bottom row of LED's show the voltage from the narrow band O2 sensor. In the "Power" mode (as shown), the display reads from 0.81v to 0.90v. In the "Cruise" mode, the gauge reads from 0.35v to 0.96v.

It correlates pretty well with his wide band display. Could use a little more filtering, though.
 
John can you tell us about the Son of knock finder device, and the story behind it, and will they ever be available again?
 
Actually, Son of Knock-Finder came first, in late '95. I made one, then got a call from a NasCar type engine builder in North Carolina. When I told him about it, he asked if I could add eight knobs so he could manually adjust timing in each cylinder. He said he would buy one if I could get it ready before Daytona. I added more features and put it into a bigger case, but missed the date a couple weeks. Never heard from him again.

I showed them both to KennyD. He helped get us in a couple issues of Hot Rod in '96, so I offered him a Knock-Finder, he said he would rather have the little one. We called it Son of Knock-Finder because it was smaller.

Both were designed for distributor ignition. I used a timing light pickup to synchronize the unit to the engine.

I intended to price Son of Knock-Finder at $1500. The Knock-Finder was $2495, sold six of them before I stopped making them in 2001. I FedEx'ed a thick package of documents to Circle Track magazine after KennyD tested it, but they didn't return my calls. Guess I'll eat some worms. :)

I designed the Vampire version in '07, but didn't realize I could show the retarded cylinder until a year later, so the knock-finder option is an add-on daughter board.

The Interceptor version came later, so it has a built in knock-finder. It shows which coils are being retarded, so if you run waste spark it would only show which pair of cylinders are being retarded.

Link to old web page:
http://www.jandssafeguard.com/knockfinder.html
 
Thanks for the explanation. So if we wanted to get technical, the only way to find out which cylinder is knocking you would have to data log the spark event and the cam sync at the same time. So we could take raw data from the crank sensor and cam sensor. If we have 3 crank events per revolution at 7500 rpm the logging rate would have to be 375 frames per second and the cam event at 58 frames per second or am i looking at this incorrectly. I don;t see how we could figure out each which cylinder is knocking with just the cam reference?
 
So we just have to log the EST signal, not the crank sensor directly, makes sense since the crank signal would not give us the timing , just the reference point. What would be the logging rate for the est signal we would need for 7500 rpm?
 
So we just have to log the EST signal, not the crank sensor directly, makes sense since the crank signal would not give us the timing , just the reference point. What would be the logging rate for the est signal we would need for 7500 rpm?
Norb looks like top row on the scope shows gage voltages in firing order??
 
Norb looks like top row on the scope shows gage voltages in firing order??
Its hard to tell, the video is blurry, it shows knock, but I'm not sure if its in the firing order or not.
 
So do you agree to see individual spark events on the est trace, we will need to log at a frame rate of 375 frames per second at lets say 7500 rpm max, or we won;t be able to get fast enough data. I can upgrade my pcs dash to 1000 frames a second if needed.
 
Yes, to log the data from a six cylinder at 7500 RPM you would need 375 samples per second on the clock and retard channels.

Note that the retard voltage is updated after the knock window closes, and is stable for 120°. If you sample at the trigger point (EST goes low), the retard voltage will be from the previous cylinder.

Refer to the diagram near the bottom of this page for a timing diagram:
http://www.jandssafeguard.com/tech.html
 
Yes, to log the data from a six cylinder at 7500 RPM you would need 375 samples per second on the clock and retard channels.

Note that the retard voltage is updated after the knock window closes, and is stable for 120°. If you sample at the trigger point (EST goes low), the retard voltage will be from the previous cylinder.

Refer to the diagram near the bottom of this page for a timing diagram:
http://www.jandssafeguard.com/tech.html
John thanks for all the clarification you have provided.This is a great product and will extend engine longevity.I will be shortly orderring a vampire to go along with my fast.
 
John thanks for all the clarification you have provided.This is a great product and will extend engine longevity.I will be shortly orderring a vampire to go along with my fast.
You better ask john on which knock sensor he can supply with the unit, not sure if you can use the GM sensor with the newer units, they use a bosch sensor. I have an older unit and I use the fast knock retard at the same time as the JS since I have the GM sensor.
 
The gauge in the video posted earlier displays knock retard plus the voltage from a narrow band O2 sensor.

The top row of LED's show two cylinders are being retarded, one more than the other.

The bottom row of LED's show the voltage from the narrow band O2 sensor. In the "Power" mode (as shown), the display reads from 0.81v to 0.90v. In the "Cruise" mode, the gauge reads from 0.35v to 0.96v.

It correlates pretty well with his wide band display. Could use a little more filtering, though.
John would you know if the OE GM knock sensor can be used or can the vampire be adjusted to use the GM sensor input? Also which knock sensor comes with the vampire? Most OE's are using pairs of knock sensors.How would this relate to the vampire?
 
This is a great product and will extend engine longevity

not if the first knock event takes out your head gaskets, i dont care how fast it is, unless it can detect knock before it happens there can be signifocant damage already done.

but anyway, im throwing in the towel on this one.
 
not if the first knock event takes out your head gaskets, i dont care how fast it is, unless it can detect knock before it happens there can be signifocant damage already done.

but anyway, im throwing in the towel on this one.

Sorry that you feel that way. Bye.
 
John would you know if the OE GM knock sensor can be used or can the vampire be adjusted to use the GM sensor input? Also which knock sensor comes with the vampire? Most OE's are using pairs of knock sensors.How would this relate to the vampire?

Yes, there is an internal adjustment that allows for high output GM single wire or low output two wire sensors.

One sensor is all that's required, though you could add another sensor if you want.
 
Top