Index Home About Blog
From: John De Armond
X-Source: The Hotrod Mailing list
Date: Mar 1992
Subject: Octane boosters

>What about these "Octane Booster" products that are supposed to enhance
>the octance rating of the fuel?
>
>I have a '63 Vette with a 11:0 compression ratio.  The best fuel I can
>get is the Super unleaded stuff which is usually around 91-92.  Do I
>need to boost the octane?  Also, is it alright to run unleaded in this
>engine which was designed for leaded gasoline?  What about these "Lead
>substitute products?"...

These generally fall into three catagories:  Tolulene or xylene based,
methanol based and organometallic based (typically tetraethyl iron.)
I'm, of course, discounting "Real Lead" which is nothing more than
very expensive leaded gasoline.  I stay away from lead in my work for
a number of reasons.

Most that I've looked at are toluene based.  They all work but I consider
them ripoffs nontheless.  Why?  The price, of course.  Technical grade
toluene in 5 gal quantities can be had for about $3 per gallon.  Recycled
solvent-grade toluene can be had for as little as $1.50 a gallon.  Buy
in 55 gal drum quantities and the price is even lower.  Check the yellow
pages.  Contrast this with the several dollars for a pint of "octane
enhancer."

The neat thing about all octane boosters is that the relationship between
quantity of octane boost and the amount added is very non-linear.
That means that adding a whole lot of booster does little more than
adding just a touch.  For example, pure toluene has an octane rating of
about 125.  Adding just a half gallon to 20 gallons of 92 octane gas
may boost it as high as 102 octane.

Here's my suggestion.  If you don't want to fool with running down
toluene from an industrial supplier, simply buy some Turbo Blue
racing fuel from your friendly local hotrod supplier (or mailorder).
Then experiment with mixing this in with your pump gas.  Since
Turbo blue is mostly toluene, you should be able to use it
in a similar manner as the octane boosters.  A quart to a 12 gallon
tank of gas should do it.  Experiment with your engine until you
discover the minimum amount of booster necessary to supress knock.

You need to use a knock sensor in order to determine this.  An engine
makes its best power when incipitent (silent) detonation is present.
This knock IS visible to a knock sensor.   There are two easy ways to
do this.  One is to buy a knock sensor from a commercial vendor.
The only one I know of right off the top of my head is the SafeGuard
system from J&S electronics (714) 534 6975.  This is actually a knock
control system that also displays knock activity.  It's expensive at
$425.  I've not evaluated it so I cannot comment on its effectiveness.

The second method is to install a knock sensor on your engine and
monitor it with an oscilloscope.  To get a sensor, simply visit your
local car parts emporium and look in their sensors catalog.  find
a sensor for an engine similar to yours that was equipped from
the factory.  The sensor is acousticly tuned to the knock frequency of
an engine type so it is important to match engine types.  In general,
I've found that all V8 sensors, all in-line six sensors and all
4 cylinder sensors, etc, are interoperable within the engine type.

The signal must be viewed with a scope triggered from ignition pulses.
This is because other engine noise can mimic knock.  What you're
interested in is looking for knock from about 5 degrees after ignition
firing to about 20 degrees after top dead center on the power stroke.
I have a Fluke Model 97 digital storage scope that makes this easy.
It can be done with an analog scope but will require an assistant to
run the scope.  With my digital scope, I can gate the sweep from
the ignition pulses and trigger the acquisition cycle on amplitude
of knock signal.  Then I just drive and when the scope triggers, stop
and look at it.  With the ignition signal on the other channel,
I can compute the RPM where the problem happens.

The neat thing about knock sensor is that they pick up resonances set up
in the block and head stimulated by the detonation process.  This means
that the same signal can be generated with a sharp hammer rap
on a solid spot on the block.  Thus you can calibrate your knock sensor
without having to run the engine.

While I'm at it, (boy did this wind around away from the article)
This fluke scope is an absolute essential for the high technology engine
hacker.  Take a handheld DMM, give it a shot of steroids so it grows
100% in all dimentions, add a large LCD screen, put in a two channel
50 mhz digitizer and make a handheld digital storage scope.  The scope
is double insulated and isolated to 1000 volts RMS which means you
can float it across the primary of even CDI and magneto systems.
It stores several setups and waveforms in memory.  It drives a fiber
optic interface that can either data dump or dump the screen to
eitehr an epson or thinkjet printer.  The scope is also a digital multimeter
that measures true RMS AC as well as DC, ohms, frequency, and diode test.
When in DMM mode, it displays a small scope screen below the large numbers
so you can still see the waveform.  It runs about forever on its
Ni-Cads or about 4 hours with the EL backlight on.  When the battery
runs down, you can pop it out and replace it with dry cells.
All this for only $1795 retail.

The only problem is you gotta kiss Fluke's ass to get 'em to sell you one.
They started national advertising last november but only shipped a
few weeks ago.  I've raised hell since early January, including calling
the Prez of Fluke and I only got mine about 3 weeks ago.

John

From: John De Armond
X-Source: The Hotrod Mailing list
Date: Mar 1992
Subject: Re: Octane boosters

>Is there anyway I can determine if detonation is occuring
>without a scope?  I am not sure what the output of the
>knock sensor is?
>
>I don't have the $$$ to buy a scope now, but I just passed my
>amateur radio test, and I do need to buy a good DMM, will
>this do?  If so, can you recommend one?  I would like a Fluke.

No.  You've got about 4 choices.  I've given two of them and the third
is to find some other product I'm not aware of that does gated monitoring.
The last is to build a monitor.

I've spent a whole lot of time and a moderate amount of money trying
to build a simple non-gated analyzer.  I finally built up a data
acquisition system, digitized the output of the sensor and did a
fast fourier transform analysis on the data.  I was hopeing I could
find some characteristic that distinguishes engine noise from knock.
No luck.  The only way to make this reliably work is to gate the
monitoring circuit with ignition.  All the literature I've read says
the same thing.

An analog circuit can be built but it is complicated by the fact that the
gating window varies in the time domain with RPM.  Much easier to simply
precision rectify the output of the sensor then apply it to the analog
input of an 8751 or equiv programmed with an appropriate algorithm.

This is a problem I've been addressing for several years now.  You see,
I have this crazy propensity to want to run 30psi of turbocharging
on street engines.  One little knock == instant piston kit :-(

>I'm not sure you answered my question about lead substitues.
>Is there a relation between lead additives and octane boosters?
>
>Your reply seemed to suggest that lead is used as a anti knock
>agent.  Is this correct?

Tetraethyl lead (TEL) is the traditional octane booster used by gasoline
refiners from about 1920 until catalytic converters came on the scene.
TEL is really nifty because it is very cheap and can make even so-called
"crack gasoline" - gas that has an octane of from 20 to 60 - into
premium with an octane of 100 or better.  The negatives of TEL include
making econazis wring their hands and spin in place, poisons cat converters
and lambda sensors, craps up spark plugs and exhaust, and causes
exhaust systems to corrode faster.  I can post some references if someone
is interested in reading LOTS of heavy text on the subject.

Another booster is Tetraethyl Iron (TEI).  Unfortunately, the breakdown
product of TEI is iron oxide, a very abrasive material.  The breakdown
product of TEL is lead oxide, a substance that melts in the combustion
chamber and makes nice soft little ball-ettes.  iron oxide is also
electrically conductive so it rapidly fouls plugs.

Other boosters include oxygenated hydrocarbons such as MTBE, Iron Carbonyl,
nickel carbonyl, aniline, ethyl alcohol, methyl cyclopentadienyl
manganese tricarbonyl (AK-33x), ethylene dicloride, ethylene dibromide and
a variety of proprietary goops.  Then you have to consider extenders or
synergists.  These chemicals enhance the action of TEL and certain other
primary anti-knock agents.  A typical synergist is Tetra-butyl acetate
(TBAc).  It breaks down during compression to yield, among other things,
acetic acid.  Acetic acid greatly enhances the effectiveness of TEL and
reduces the sensitivity of the fuel.

It is safe to say that the only boosters available to us hotrodders are
toluene and xylene.

As an aside, I saw an interesting note in some literature.  It said that
during the 30s, gasoline dealers stocked little glass vials of pure
TEL.  The dealer was supposed to dump the TEL into the customer's
tank in order to make regular into premium when the customer paid for
premium.  Can you see the safety nazis spinning about that? :-)

John

From: John De Armond
X-Source: The Hotrod Mailing list
Date: Apr 1992
Subject: Knock sensors

>> I have learned some things about the MSD 8972 that may be of interest
>> to this group.

>I have been thinking about purchasing the MSD 6A setup for our
>motor...

Get the -6T or the one with the rev limiter built in (6L?)  The-6T
comes with a very nice rubber vibration isolation kit for mounting.

>does the above work along with the 6A?

Yes.

>what would be involved in
>installing a knock sensor into an engine not originally equipped with
>one?  what does the knock sensor do?  How does it work?

I posted a monologue on knock sensors a couple of weeks ago.  Part is
included below.  Physically installing the sensor is no problem - just
find a good solid point on the head or block, drill a hole if one does
not already exist, tap and screw the thing in.  Getting it to work with
the ignition is another trick alltogether.

I've bought several so-called knock controllers over the years.  None
worked.  They all responded to the gross signal from the knock sensor
which meant they are always actuated at high RPM.  For it to work, the
controller MUST gate the knock signal to crank position.  You have basically
four choices.

1.	Certain GM vehicles used an external knock controller that interfaces
	with a special HEI unit.  I believe it could be adapted to any system.
	A note describing the system set to me by a friend is enclosed below.

2.	The SafeGuard system by J & S electronics.  This is a digital aftermarket
	system that supposedly controls knock on a cylinder-by-cylinder basis.
	I've not used the system but I looks OK.  Expensive at about $425.

3.	The Electromtive direct ignition system.  This system uses a crank
	trigger, direct ignition (no distributor), is PC programmable and
	listens to a knock sensor.  The system has gotten some bad press on
	the net but my experience is that it works as advertised.  Mounting
	the crank trigger is CRITICAL and requires a machine shop.

4.	Roll your own.  My choice.

John
-----------------------------------------------------------
[[see above for article on knock sensors]]


------------------------------------------------------------------------
From: emory!posms.cactus.org!rick (Rick Kirchhof)   [not there anymore jgd]
Subject: Re:  EFI design ideas

As to the GM ESC system, this is exactly identical in appearance to any
1975 or later HEI system.  It has coil in cap, vacuum and centrifugal
advance in the conventional manner.  It is connected to a black plastic
box by 4 to 6 wires.  The box is about 1 X 4 X 6 inches and is USUALLY
mounted with its length along the length of the truck and parallel
to the ground.  It is often found near the power brake booster or A/C
evaporator housing.

The best way locate a suspect vehicle is to look for a non-computer
controlled engine in a pickup or surburban.  These are most easily
identified by the vacuum advance can on the distributor.  NO GM vehicle
ever had electronic spark timing AND a vacuum advance can except the HEI
ESC units.  Once you have a non-computer engine, you must look at the
distributor.  If it has only one wire (usually 12 gauge, pink, this is
power) that attaches it to the engine harness, you have a plain HEI.
Sometimes, there is a second smaller wire for a tach or cost down
solenoid if manual transmission.  It is smaller than the power wire and
just beside it.  If it is as above, plus there are four other wires
leaving the distibutor, you have your prize.  Don't count the wires that
join the cap to the distributor base, these don't connect to anything
external.

If you need to spec it out, try a 1983-85 C-10 (half ton pickup) with
a 5.0L carbureted engine.  If this gets you into on-board computer
control, try it as a C-30 (heavy duty emissions package).  The GM HEI
ESC unit is the opposite of a standard GM electronic advance.  It is
fully advanced all the time, and looks for knock, then retards all
cylinders equally so long as there is knock.  The retard increases so
long as there the knock continues.  It waits for knock to stop, then
adds timing back to normal settings.  The distributor can be junked,
and the module mounted and heat sinked elsewhere.  Then all you have to
do is to wire the knock controller, 12V supply, trigger coil from the
distributor and knock sensor to it.  You can use a late model divorced
GM coil, and there you have it.

[end]

Newsgroups: wiz.hotrod
Subject: Re: Point ignitions
Date: Wednesday, Jul 22 1992 03:30:38
From: John De Armond

>-> You want one?  I have one I'll give you if you'll pay shipping. Junk
>-> compared to the new stuff.  The thing is not synced to spark so async
>-> mechanical noise triggers the thing.
>
>  I'll take it!  Where should I send the check?

John De Armond
1631 Whitlock Road
Marietta, Ga 30066

>        10456017  ESG Knock Sensor $ 44.82
>        16042121  ESG Module       $164.34
>        1894308   ign. module      $ 90.85
>        --------  ---------------- -------
>                      total:       $300.01

> The dealer offered a fat 15% discount, but I decided not to buy. <sigh>
>I could probably crossbreed a knock sensor from something, and it might
>be possible to find the module in a junkyard, but it looks like it just
>moved 'way down on my priority list.   If it doesn't rain today I might
>cruise by a couple of the junkyards and look around, though.

Illegitimo non carborundum (don't let the bastards wear you down!)

That's crazy.  The knock sensor is $14 from the local NAPA store.
I found an EGS module in about 10 minutes at our only local tiny
junque yard (yuppy haven, remember?)  The ig module should be about
$30 from NAPA.  I'll check when I get a chance.  When I get a chance
I'll write up how to use it with the 4 wire $19 module :-)

> Is it really that important to synchronize the sensor with the
>ignition? You can set the sensitivity of the Carter anywhere you want;
>my Yamaha and most of the knock sensor equipped GM cars I've driven
>have audible detonation before the box cuts the spark lead.

Yes.  There'll be scope screen pictures in the magazine that illustrate
why.  what appears to be valve clatter generates roughly the same
level signal as incipient knock.  The best power is made when the engine
is in incipient ("silent") knock.  That is the first problem.  The second
problem is the timing should be backed off only to the point where incipient
knock exists.  The carter unit whams the timing back to the setpoint and
then lets it drift back up until audible knock (or engine noise) whap
it back again.  If your engine really needs knock control - as in it will
self-destruct if not controlled - what you get is a cyclic surging
that is very unpleasant.  The next problem is only the cylinder(s) that
are knocking should be retarded.  Each cylinder may require a different
retard.  The carter whaps them all in sync.  I believe but am not sure
that the EGS unit does cylinder by cylinder control.  Lastly,
the really intelligent controllers actually analyze the waveform 
to distinguish the rather distinctive knock sound from all the other noise.
EGS does not do that.

> What we need in PE is a good article on "How to build your own engine
>management system with a $3 Radio Shack soldering iron and parts
>scavenged from your Cuisinart."

Patience, my friend, patience.  The engine management system I'm working
on for the 2nd issue uses the Dallas Semiconductor DS5000 microcomputer-on-
a chip.  This module (cost ~$80), an A/d converter and a few switching
transistors make up the whole thing.  I will make the chip itself,
a hard-to-find parts and PCB kit and a complete kit available from the
magazine.  I'd guess the complete kit will cost less than $300.

Meanwhile I'm working on the $100 distributor curving machine (first or 2nd
issue), the Fuel Injector tester (sample and first issues), finishing 
typesetting the sample, calming my blood pressure after trying for FOUR
EFFING DAYS to get my $%^&%&%^& Linotype service bureau to make the 
first frigging sheet of film!  Oh, and the new task. Buying a linotype 
machine ($10,000 used if I'm lucky.)  This is fun :-)  Gawd, I gotta
hire a co-op student.....  Had a benevolent supporter donate an 
HP ScanJet to the project so now I don't have to redraw submitted 
artwork.  This is really getting to be fun...

Oh, back to hotrodding.  Pick up September issue of Popular Electronics.
The cover article is a CDI ignition box you can build for about $30.
The author really does not know much about ignition but the circuit is
fairly sound.  My 5 second analysis says the spark will be too brief 
but that's easily fixed.  About as cheap a CDI box as can be had.
A single 555 timer chip would make it equivalent to the MSD box.

Headed out next week to get my hands on a RacePack data acquisition system
on which to do a technical review.  This is the DAC all the top fuel and
Pro Stock guys are using to do those fancy printouts after each run.  
We'll see what makes 'er tick and see how good it is.  Amazin' how 
opportunities fall in one's lap once one calls himself "publisher" :-)

As of the AM, I'm off to the Datsun Z-car national convention in 
Knoxville, TN.  Runs wednesday through Sunday.  It's at the 
World's Fair Holiday Inn.  Just come to the city on I-40 and look for
the big golden Elephant's nut towering above the city.  Head toward
that thing and you're within a block of the motel.  Anyone in the vicinity
should drop in.  I'll be wearing the very first of the Performance
Engineering Tee shirts. :-)   I'll probably be piddling with the 
radar gun around the autocross and slaloms...

Later doods,
John

From: emory!chaos.lrk.ar.us!dave.williams (Dave Williams)
X-Source: The Hotrod Mailing list
Date: Apr 1993
Subject: Re:  Stalling in a '79 Mustang
X-Sequence: 4861

-> I read last year in HRM about a company that makes small block ford
-> distributers that are GM HEI on top.  No more Dura-no-spark problems.

 Hey, I did that in 1981!  Hacksawed the GM top off and had it welded
onto the Ford shaft.  Cleared the intake manifold (just barely) and
worked fiiiine.  Now that I have a lathe I could do it again, but after
listening to John talk about triggering the HEI or Chrysler modules with
points, I'll probably just use a Dura-Suck distributor housing and
pickup and a Chrysler module.  My buddy Doug is cloning a few Carter
Knock Eliminator boxes for it.  If it weren't for that and I had to
recurve the distributor, I'd certainly want the GM HEI top - you can get
zillions of weights, springs, and cams to control the advance, and
they're right there on the top.  Fords like to hide them underneath
where you have to disassemble the distributor to get to them.

[Oh my Gawd, what have I done?  (I gave Dave a Carter Knock Eliminator)
You're gonna actually create more of those things?  Carter's idea
of eliminating knock is to puke a piston through the bottom of the block!

All joking aside, Dave, why don't you adapt one of the GM knock controllers
instead?  It at least has a chance of working.  The problem with any of
these devices that don't do signature analysis is at high RPM where knocking
does the most damage, it is impossible in a generic manner, to
discriminate between mechanical noise and knock.  The digital controllers
such as the J&S unit, can distinguish between knock and noise but that
unit is kinda expensive.  JGD]

From: emory!chaos.lrk.ar.us!dave.williams (Dave Williams)
X-Source: The Hotrod Mailing list
Date: Apr 1993
Subject: Re:  Stalling in a '79 Mustang
X-Sequence: 4876

-> [Oh my Gawd, what have I done?  (I gave Dave a Carter Knock
-> Eliminator) You're gonna actually create more of those things?
-> Carter's idea of eliminating knock is to puke a piston through the
-> bottom of the block!

 The one on Monzilla has been in operation about ten years, still works
fine, and was on the car when we took first place at the Divisionals.
It beats the snot out of playing Advance Curve Man any day of the week.

[Depends on what you call "work".  I know from instrumenting the one I gave
you that at medium and high engine speed it hears the valve train rattle
and slams the timing back more or less continuously.  Admittedly a
Datsun OHC engine and the BMW OHC engine (I ran it on both) have more valve
noise than a V-8 with hydraulic lifters but I doubt the box's response to a
bit less valve noise would be much different.  It DOES do a good job of
getting rid of low speed rattle but that's mainly an esthethics issue. ]

-> All joking aside, Dave, why don't you adapt one of the GM knock
-> controllers instead?  It at least has a chance of working.

 You *are* joking.  I have been looking for one in the junkyard for
eight or nine months.  GM didn't make many of them.  My friendly Chevy
dealer wants THREE HUNDRED DOLLARS for the box, weirdo ignition module,
and sensor.  I ain't gonna pay it.

[Whot, did Slick Willie take all the good junk yards with him when he
left? :-)  First little podunk 10 acre junkyard I walked into had
exactly what I wanted.  ]

-> discriminate between mechanical noise and knock.  The digital
-> controllers such as the J&S unit, can distinguish between knock and
-> noise but that unit is kinda expensive.  JGD]

 Isn't that one about $500?  From their propaganda sheet, it looked like
they expected you to already have a late-model, computer equipped car,
to which you attached their box.  I don't need an "engine management
system", I just want to keep from holing a piston at an embarrassing
moment.

[It will work with any conventional distributor ignition.  I'm not
sure about points.  I did see one in use on an old Porsche 912 at the
Porsche club race at Road Atlanta two weeks ago.  I didn't get a chance
to talk to him because I was too busy helping another fellow work
out his Electromotive crank trigger problem (Jon, don't say a WORD!)
I'm going to see if I can get one of those to evaluage for the mag.]

 You keep mentioning all these wonderful systems, but I don't see
anything that will (a) work on a non-electronic motor, and (b) doesn't
cost the proverbial arm and testicle.  I can get the spark curve pretty
damned close with an ordinary distributor.  Is the last, ultimate degree
of timing that critical?  Not in my experience.  My $500 would be much
better spent somewhere else.

[That's a personal choice, of course.  I'd spend the $500 as insurance
against having a bad batch of gas or a mis-read timing mark trashing
my engine.  $500 is a lot of money but in the context of other
Hi-Po parts, not terribly bad.

For drag racing power, I agree that a simple distributor advance curve
will do the job.  No concern with midrange power or economy.  For
street hotrod motors - my main interest - I DO care about the best possible
midrange torque and the best possible mileage if for no other reason that
I want to skip a gas station here and there.  Electronic timing control
makes a remarkable difference in a street environment.  JGD]

From: emory!lynx.unm.edu.unm.edu!chama.eece.unm.edu!mwalker (Mark Walker)
X-Source: The Hotrod Mailing list
Date: Apr 1993
Subject: Knock sensor signature analysis
X-Sequence: 4907

The discussions about knock sensors got me to wondering what the signal
signature for these devices looks like.  Since I work in a EE
department big on DSP (Digital Signal Processing), I got to wondering
about the possibility of smart (well, smarter than the OEM and less
money than other existing solutions) use of the signal the
sensor provides.  Is this information readily available for these
sensors?

[See "Automotive Electrical/Electronic Systems" from Bosch.  Page 133
has some good scope snapshots taken from their narrow band sensor.  I also
have some scopemeter screens stored away around here somewhere if you're
interested in GIF files. The OEMs will supply information on their
sensors if you find the right person.  I have some Bosch info here.]

Oh, it also seems like I saw an article somewhere that mentioned use
of a more common device (a microphone or something) as the sensor.
Ring a bell with anyone?

[There are three basic types of sensors in use.  The resonant sensor,
the narrow band sensor and the wide band sensor.  A resonant sensor is
tuned to the block's mechanical knock resonance and provides some
mechanical selectivity.  Not commonly used because so many things can
change the block's resonant frequency.  Narrow band sensors are
also resonant but with a much lower Q.  These are typical of OEM sensors.
Wideband sensors, also called accelerometers or microphones, are sometimes
used but the demand on signal processing is much greater.

Do I hear a volunteer to do a DSP knock controller?  :-)  JGD]

--
 "There's no such thing as fast enough!"  | Mark Walker
     (Or something to that effect...)     | mwalker@chama.eece.unm.edu
                James Taylor              | 505/277-3688  (home 899-0644)
                _Two_Lane_Blacktop_       | Albuquerque, NM

From: emory!chaos.lrk.ar.us!dave.williams (Dave Williams)
X-Source: The Hotrod Mailing list
Date: Apr 1993
Subject: Re:  Stalling in a '79 Mustang
X-Sequence: 4912

-> [Whot, did Slick Willie take all the good junk yards with him when he
-> left? :-)  First little podunk 10 acre junkyard I walked into had
-> exactly what I wanted.  ]

 We don't have junkyards around here.  They all got together and formed
a union, or something.  75% of dealer list price, take it or leave it.
I try to find what I need at the scrap iron places first, but they
usually don't have much.


-> I want to skip a gas station here and there.  Electronic timing
-> control makes a remarkable difference in a street environment.  JGD]

I don't want to sound like a broken record here, but who makes one of
these miracle boxes that'll work with my non-computer engine?  All I
need is a damned knock sensor; most of these places either sell boxes
that plug into your existing (or non-existing, in my case) EEC-IV or TPI
unit, or complete crank-trigger, coilless wonderjobbies that cost almost
as much as a whole motor.

[maybe I won't sound like a broken record either.  J&S's Digital
Safeguard box can be triggered from most anything that can fire a coil.
From the cover letter on their literature package:


	Dear Sirs:

	Thank you for your interest in the Digital SafeGuard.  We have
	included all available information, including our dealer pricing

	The enclosed four page brochure on the SafeGuard was written
	for our earlier analog unit, which did not have the rev limiter
	or built-in magnetic pick-up adapter.  It does, however, give a
	good idea of a basic knock control system, so we have included
	it in the package.

	With the new system, you can set an aggresive timing curve,
	then let the SafeGuard fine tune it for maximum power in each
	cylinder, under all conditions, such as boost pressure, nitrous
	oxide injection, varying octane, altitude, humidity,
	intercooler efficiency, etc.

	The system is capable of retarding the knocking cylinder up to
	6 degree before it comes around to fire again.  Total retard
	capability is 20 degrees, in steps as small as 2 degrees.

	Blown engines without intercooles encunter pre-ignition, which
	is not the same as detonation.  This can trick the computer
	into retarding the wrong cylinder.  In this case, a user
	programmable switch can be set to retard all cylinders by the
	same amount.

	An easily set digital rev control is also included.  At the
	selected limit, program control allows two cylinders to fire,
	then skip one, fire two more, skip another, etc.  For six
	cylinder engines, the program fires three and skips two.  This
	rotates the misfiring cylinders throughout the firing order,
	and allows a large drop in power without stumbling or
	backfiring.  Note that this is not a hard limit, but about a
	one third drop in power.  We use it more as a "shift alert", as
	it gently reminds the driver to get his foot off the throttle.

	The unit also features a built in high energy coil driver,
	delivering at least seven amps of coil current to a high energy
	coil.  This compares to five and a half amps in the GM HEI.  No
	ballast resistor is required, simplifying the installation.

	The unit can be triggered from points, OEM electronic, or
	directly from magnetic pick-up.

	Sincerely,
	John and Shirley
	J&S Electronics

I should note that I've never personally evaluated this unit.  However,
everyone I've talked to says it works as advertised.  Retail on this
box is $450.  Kinda high but compared to say, an MSD box, rev limiter
and "timing computer" or boost retard, it's not bad.

The GM unit is another system that will trigger from any kind of
variable reluctor distributor.  It will probably trigger from points if
you really insist.  There's one of the largest junk yards (Longs) in the
south just this side of Nashville on I-24 so I know you're within, what,
maybe 3 hours of a source.

Sorry, I just can't come up with a source of free units. But the available
units can be had for prices that are quite in line with other high
performance parts.  JGD]


From: emory!chaos.lrk.ar.us!dave.williams (Dave Williams)
X-Source: The Hotrod Mailing list
Date: Apr 1993
Subject: Electromotive DFI
X-Sequence: 4919

-> > dealer wants THREE HUNDRED DOLLARS for the box, weirdo ignition
-> module, > and sensor.  I ain't gonna pay it.
>
> How bout using the sensor and conditioning module from a late model
> car, and building a seperate box to handle retarding the timing
> (taking its input from the conditioning module).

 Nope.  The system John and I are talking about is the weirdo "Computer
Controlled Spark" from the early '80s, which appeared on assorted
pickups and a random sampling of cars.  It's a plain old non-computer
HEI with a knock sensor box and sensor.  The box is like $160, the
special HEI module is $90, and the knock sensor another $100, though you
could probably use any knock sensor.  John can evidently waltz into his
friendly junkyard and get them for peanuts.  The only way I can get one
is to steal it off a parked car.

[ Actually I thought the junk yard situation here in Atlanta was pretty
dismal.  Bill Southern's little yard is nice and he's a rodder to boot
but it's small and it is only one yard.  Where I came from (Cleveland,
TN, Odometer rollback capital of the US), there's a 100 acre junkyard on
every major highway.  There's also something like 300 licensed used car
dealers in this little 20,000 population berg.

So guys.  What IS the junkyard situation elsewhere besides Atlanta and
Little Rock?

I've not priced the box but the 5 wire HEI module is about $30
from NAPA and the knock sensor is $12.  No fair quoting dealer
prices. No one pays that much..... Do they?  JGD]

From: emory!chaos.lrk.ar.us!dave.williams (Dave Williams)
X-Source: The Hotrod Mailing list
Date: Apr 1993
Subject: Electromotive DFI
X-Sequence: 4940

-> I've not priced the box but the 5 wire HEI module is about $30
-> from NAPA and the knock sensor is $12.  No fair quoting dealer
-> prices. No one pays that much..... Do they?  JGD]

 Unh... I'm not *that* dumb, John.  I tried pricing the module - which,
BTW, the dealer says is a seven wire part - at NAPA and other places,
but none of them have a listing for it.   I'd like to know what the $12
knock sensor is for; the lowest price I could find was for a Chrysler
for $38.  And you still need the brain box, which is still gonna be $160
from the dealer.

[My knock sensor is for a Dodge turbo Lazer.  Only because someone suggested
it to me several years ago.  I should note that NAPA carries two lines
of such parts.  One is "name brand" such as Standard Parts.  The other
is what the guy behind the counter calls "green box" parts because they
are in otherwise unmarked green and white boxes.  The green box parts are
less than half the price of "name brand".  I recently went to buy a
chrysler ignition box that I pay $19.50 for.  The guy laid a Standard Parts
unit on the counter and chirped "$55".  He almost had my plastic before
it sunk in.  After the roof settled and he got me a green box part, I
laid them side by side.  Same damn boxes including the various production
markings on the bottom.  The local Downeys carries the same lines so
this green box generic line must be pretty common.  JGD]

From: emory!chaos.lrk.ar.us!dave.williams (Dave Williams)
X-Source: The Hotrod Mailing list
Date: Jul 1993
Subject: Fuel Injected 2-stroke Racebikes
X-Sequence: 5931

-> The cylinder pressure transducer, BTW, is primarily to reliably
-> detect knock. There has been quite a bit written about this in the
-> last year in various SAE papers.  This is vastly more reliable than
-> acoustic knock sensors for high speed engines.

 How long do these sensors last in use?  The data would be fascinating.
Also, do you have the numbers of any of those SAE papers?

[The intent is life-of-the-car.  Suspect they're a little short of
that goal.  These are just piezo transducers so the sensor itself
shouldn't be affected by the conditions.  Heck, I have a piezo
accelerometer here that is intended to sit on the pressurizer relief
valve at Three Mile Island and monitor flow that will withstand aboug a
jillion RADs of radiation, 600 degree steam and various corrosive
chemicals all at the same time.  :-)  15 year old technology.  Of course
this little puppy cost about half a year's salary.  But I digress.

I don't have any papers handy at the moment but I'll see what I can
dig up.  I use the push-off-the-end-of-the-desk-into-a-box method
of filing so it might take awhile.  JGD]

From: emory!chaos.lrk.ar.us!dave.williams (Dave Williams)
X-Source: The Hotrod Mailing list
Date: Aug 1993
Subject: Fuel Injected 2-stroke Racebikes
X-Sequence: 5933

-> How long do these sensors last in use?  The data would be
-> fascinating. Also, do you have the numbers of any of those SAE
-> papers?
>
> [The intent is life-of-the-car.  Suspect they're a little short of
> that goal.  These are just piezo transducers so the sensor itself

 I'm curious about the size of the sensor, and whether it can be mounted
up in the valve cover where it'll see oil on top.  Boring a hole through
the water jackets and into the chamber, welding in a threaded sleeve,
and trying to clear the spark plugs, rocker gear, valvesprings, and
ports is gonna be fun.

[Kistler makes a combination spark plug and water cooled pressure
transducer.  Reportedly it works very well though it does require a
charge amplifier external.  Kistler also makes direct insertion
transducers designed to thread in a drilled hole to the combustion
chamber.  These are precision, expensive lab instruments.  Both Bosch
and Hitachi have production-type sensors under development.  I'm not
sure if they'd be available to others yet or not.  JGD]

 It always looked to me like a rotation speed sensor on the crank would
be a practical way of detecting detonation.  The crank flexes according
to combustion loads anyway; when it sees an unusual load the vibration
characteristics would change.  Of course, you'd have to have enough
detonation to make itself felt on the bottom end instead of just looking
directly, but it'd be more sensitive than an acoustic system, and if
you don't have enough detonation to load the bottom end, you probably
don't have enough to hurt anything anyway.

[I posted here about a year ago on a magnetostrictive torque sensor
designed to fit inside the rear main bearing cap that could
measure detonation torque.  The sensor is real simple, consisting
of a cruciform ferrite core with windings in quadrature.  One
winding is excited with AC.  The resultant field is applied
along the axis of the crank across an air gap.  Torque distorts
the field and causes voltage to be induced in the other coil proportional
to the instantaneous torque.  I don't think a commercial sensor is
available but the SAE paper made it look fairly easy to build.
An LVDT exciter/amp would probably read the thing.  JGD]

From: emory!chaos.lrk.ar.us!dave.williams (Dave Williams)
X-Source: The Hotrod Mailing list
Date: Aug 1993
Subject: Fuel Injected 2-stroke Racebikes
X-Sequence: 5936

-> [I posted here about a year ago on a magnetostrictive torque sensor
-> designed to fit inside the rear main bearing cap that could

 You sure it was only a year or so ago?  I'm sure I would have
remembered that one.


-> to the instantaneous torque.  I don't think a commercial sensor is
-> available but the SAE paper made it look fairly easy to build.

 WHICH SAE paper?

[Geez, makes me get up out of my chair, actually walk over to the
book case... :-)  SAE Paper 900264 - "Magnetostrictive Torque Sensors -
Comparisons of branch, cross and solenoid Designs".  Published in
"Sensors and Actuators" 1990 SP-805.  He lists in the references two
other papers, SAE 890482 and 890483.  There was also an article in
"Automotive Engineering" on the topic with pictures of the little
sensor embedded in the main bearing cap.  JGD]

From: John De Armond
Newsgroups: rec.outdoors.rv-travel
Subject: Re: Octane
Date: Tue, 03 Aug 2004 03:27:40 -0400
Message-ID: <hq7ug0h2fa3aegglrbt8h7nb30k7c0iqp0@4ax.com>

On 2 Aug 2004 18:44:20 -0700, e79z@hotmail.com (midwesterner) wrote:

>"Yofuri" <yofuri@oakharbor.net> wrote in message news:<10gsljo1ql8d51b@corp.supernews.com>...
>
>>
>> The 4.3 V-6 has a ping detector microphone screwed into the block.  The
>> computer retards the timing to kill the ping.
>>
>> Rick
>
>This ping detonation method always bothers me. The ping occurs and
>then you limit the damage. It's like the movie Men in Tights where the
>guy says watch my back and the other guy says, you just got punched
>twice. You damage your engine, but the damage occurs slowly enough
>that you can get through the warranty period so GM can sell you
>another vehicle.

That method would bother me too, if that was the way it worked.  It doesn't.
Speaking of post-85is GMs, of which I'm the most familiar, the knock sensor is
tuned to listen to incipient knock that is inaudible.  These are precursor
reactions setting up for the audible ping.  Both the sensor and the filter box
(external until OBD-II, internal to the PCM after that) are tuned for the
particular engine.  This phenomena and the detection thereof is extremely well
documented in the literature.  SAE is the place to look.

There are several other algorithms used on various engine lines to augment the
knock sensor control.  One involves looking at minute variations in crank
speed using the crank encoder signal to detect torque instabilities that come
with incipient knock.  This is a quite sensitive method of detection and will
likely replace the knock sensor sometime in the near future.

As far back as the Buick Grand National engine, the knock detection was
precise enough to retard individual cylinders.  This permits each cylinder to
develop maximum torque despite differences in cooling and minorly, in mixture.

The knock count and the individual cylinder retard data is available in real
time on the diagnostic connector.  The early slow speed ALDL bus wasn't fast
enough to deliver true real time data but a bus accessible under the PCM cover
is.  OBD-II is more than fast enough.  All that is needed for an OBD-II
vehicle is a network gateway cable (what the OBD-II diagnostic device is
called) and some free PC-based software.  I use this guy's products:
http://www.obd-2.com/

I have the Tricom cable but am going to trade it in for the TriCan (all three
OBD-II protocols plus CANbus) as soon as he gets some in stock.  His software
includes a real time datalogger function.  This lets one start the software,
drive around while logging and then play back the session later.  One can
easily watch the anti-knock system work.

This is particularly useful for figuring out what grade gas the vehicle
requires.  This generally changes twice a year, one for winter and one for
summer mix.  For instance, I found that my mom's Lincoln LS will happily run
on regular with every third tank intermediate grade (always filling up at half
empty) in the winter while in the summer, premium is required.  Her
Vortec-powered motorhome is happy with regular summer and winter.  My
Caprice/LT1 will run on any grade but it makes considerably more power on
premium, where there are normally no accumulated knock counts.

John


From: John De Armond
Newsgroups: rec.outdoors.rv-travel
Subject: Re: Octane
Date: Mon, 02 Aug 2004 18:27:31 -0400
Message-ID: <79etg0ljcgt87n07muk0cj50c1pgrkomac@4ax.com>

On Mon, 02 Aug 2004 21:03:06 GMT, "LongCoolWoman" <dagny68@yallareahoot.com>
wrote:

>"bill horne" <redydog@rye.net> wrote in message
>news:410EAA35.21113F28@rye.net...
>
>> I'm primarily concerned about whether violating the manual - re octane -
>> on a computerized, fuel injected engine will cause damage or other
>> problems.
>
>Unless you've increased the compression ratio, use a blower, or have
>significantly advanced the timing, buying premium fuel is NOT needed.  Nor
>does it provide ANY benefits except for subsidizing big oil (in case you
>think they need even more money).

Such generalizations are always dangerous.  In this case, dead wrong.  whether
premium fuel is needed or not depends on the vehicle.  Many new vehicles, the
Lincoln LS for example, specify premium and generally require it.  Other
vehicles, those not designed primarily for high performance, get by with less.
The only way to know is to read the manual and/or experiment.

>Because premium fuel is associated with high compression, high performance
>engines (very rare anymore), many people think they will get more
>performance out of their low compression engine if they use fuel made for
>high compression engines.

Actually high compression engines are the norm now.  The high compression
contributes to higher efficiency (fuel mileage) and the modern engine
management systems make it possible.  Consider the new Ford F-150 5.4 liter 3
valve engine.  9.8:1 compression.  Yet it can burn regular fuel.  Other
engines such as the one in the Lincoln LS require better than regular.  My
mom's LS will run on mid-grade but makes less power.  It will not run
satisfactorily on regular, pinging heavily.

There is yet another aspect to consider, the matter of fuel quality.  See the
post I made earlier today regarding Top Tier gasoline.  Given that the EPA has
allowed refiners to reduce the level of detergents below the necessary
threshold, it may be necessary in some cases to buy a higher grade fuel just
to get the additive package.

The best generic advice I could give would be "follow the manufacturer's
recommendation."  They did, after all, design and build the engine.

John


Index Home About Blog