Knock sensing,

[technicaltom]

When ever I read a build list for performance engines on the many forums and in magazines, There is allways a mention of detonation and ways to minimumise it,Quite often good engines are damaged by this same detonation,In a car with silencing and on a highway its possible to hear the clues such as pinging plus the tell tale puff of smoke from the tailpipe.
What im curious to know is it possible to use knock sensors that are now easily availalable in Car breakers yards from a great variety of cars these days, now what im thinking is there a way the signal could be converted and sent to a set of Hi FI type of headphones?so you could hear very clearly the knocking/pinging as you set up your engine ,especially a noisy engine ,another variation of this same theme would be a light that was clearly visible that would light up when the knock sensor senses detonation. Maybe this type of system is available commercially ?

[900HP]

do a search on J&S safeguard, I think that's more along the line of engine protection.

[tonyg64]

Tom type in "knock link" into google. Quite a few places sell them here is one:

http://www.rallitek.com/Knock-Link-Sensor/p-4544

Tony

[tonyg64]

Heres another one:

http://www.phormula.co.uk/KnockDetectio ... 4Qod3GvgHQ

Tony

[W. Tripp]

The Phormla units are just electronic amplifiers with headphones - solid, but basic, knock detection.

The J&S Safeguard Vampire and Interceptor units are actual knock control.
http://www.jandssafeguard.com/index.html

[Big Speed]

MSD has one also.We have it on our dyno.Bill C.

[dwilliams]

I have an ancient Carter EKE "Engine Knock Eliminator." It's probably 30 years old.

It's all analog, and some electron-heads have made fun of it over the years, but it has kept me from scattering a few engines over the years.

At one time I wanted to make a couple of copies of it (this was after they were out of production, and nobody else made anything like it yet), but already half of the components were no longer available.

[JohnP]

I reverse engineered the EKE in 1983. Very clever design.

[user-9613590]

Take some old ear protectors,drill holes to pass a hose and place the other end of hose somewhere around the intake manifold.You can hear all the knock and with practice can separate real knock from other engine noise.Cost practically nothing and get you a reputation on the "geek driving around with ear protectors!"

[technicaltom]

Take some old ear protectors,drill holes to pass a hose and place the other end of hose somewhere around the intake manifold.You can hear all the knock and with practice can separate real knock from other engine noise.Cost practically nothing and get you a reputation on the "geek driving around with ear protectors!"

Sounds good. but im 50% deaf , so I will sift through the Electronic types in the above links. the combined AF meter /knock sensor is the most appealing so far, but I also like the MSD item.
Many thanks for all the suggestions.

[rebelrouser]

MSD has one also.We have it on our dyno.Bill C.

I was thinking of adding that capability to my dyno. I looked and you are right MSD did make one, but it looks like now it is no longer available. Does your unit record on the data logger? Or do you just watch for the LED's to light up?

[In-Tech]

The Phormla units are just electronic amplifiers with headphones - solid, but basic, knock detection.

The J&S Safeguard Vampire and Interceptor units are actual knock control.
http://www.jandssafeguard.com/index.html

I have one of those MSD knock alert BS crappy I would love to sell if you want.
Otherwise... Johns stuff Not sure that link is valid since this is an old thread, he's a member here so read around

[vwchuck]

If you are doing this on a dyno I would suggest hooking up a knock sensor and TDC sensor to an Oscope. Knock is quite easy to see on a scope on the dyno. You just look right after TDC for noise. You will see the amplitude increase when it knocks.
All those silly boxes are just noise threshold boxes that have nothing to do with actual knock. They are just monitoring engine noise for the most part.

[JohnP]

The Carter EKE worked pretty well on a quiet engine. I reverse engineered it in 1983 and found the knock circuits are all analog, based on two op amps and a bunch of resistors, capacitors, and diodes. Same with the MSD Knock Alert. Both are what you call noise threshold boxes.

I tested the EKE on my turbo Fiat and found the engine made too much noise for it to reliably detect knock. I solved the problem with a micro processor and digital signal processing and released The SafeGuard in 1991. Of course, it does more than detect knock.

In 1995 I was contacted by a NASCAR engine builder. He asked if I could build a unit with eight knobs, to retard each cylinder. In response, I came out with this in 1996: http://www.jandssafeguard.com/knockfinder.html

[David Redszus]

The fundamental knock frequency is a function of exhaust gas composition,
combustion temperature, and combustion chamber size and shape.

Since the actual combustion temperature is difficult to measure, only
estimates can be used. The combustion temperature may vary by several
hundred degrees from cylinder to cylinder, and from cycle to cycle.
This is why knock can be difficult to predict and hard to measure.

The molecular weight of exhaust gas is not the same after burning as it is
before being burned. The molecular weight will change with the air/fuel ratio
of the gas in the cylinder. For the usual racing engines, a molecular weight
of 28 can be used.

Since knock measurement is the measurement of the frequency of pressure
waves in the chamber, the frequency will change with the shape of the
combustion chamber.

If a symmetrical or hemispherical combustion chamber is used, enter the bore
diameter. If extensive squish band width is used, enter the bowl diameter
in place of the bore diameter.

If an oblong or bathtub combustion chamber shape is used, or created by the
piston pop-up, enter the length and width of the chamber. This type of chamber
will have at least two different knock frequencies corresponding to the direction
in which the sound waves travel.

In addition, the vertical dimension of the combustion chamber also produces
a knock frequency. The distance from the piston crown to the combustion
chamber wall at 15 degrees after top center should be entered for chamber
height. This frequency will change, or decay, as the piston moves from the top.

Since each engine will have its own fundamental frequency (as well as
secondary frequencies), and some may be the same as noise generated by
engine operation, knock signals can easily be mistaken for engine noise.

Very high frequencies (10kHz+) tend to dissipate very quickly and may be
ignored. Lower frequencies, however, can do great damage. They typically have
greater amplitude and do not dissipate very quickly.

A knock senor that picks up all frequencies makes it difficult to denote which
signals are due to knock and which are merely engine noise. Advanced
engine management systems will contain pre-programmed algorithms which
will differentiate knock signal to engine noise.

An alternative method (for non-computerized engines) is to use a resonant
tuned knock sensor that only responds to certain frequencies and will ignore
engine noise unless at the same frequency. These are readily available but
it is necessary to know with close approximation, the knock frequency which
will be produced by a specific engine.

Many years ago, we produced a program can be used to predict which frequencies
will be produced by a specific engine under a range of conditions and therefore
assist in the proper selection of the correct knock sensor.

John P, can you believe how long ago we worked together on engine knock?
Think back almost 30 years. See how time flies when you are having fun.