Index Home About Blog
From: John De Armond
X-Source: The Hotrod Mailing list
Subject: Re: ignition questions
Date: Tuesday, Jun 09 1992 13:17:34
X-Sequence: 1216

>I have a fresh chevy 383 (400+hp) in my Chevelle and am thinking about
>upgrading the ignition system.  Currently I have a stock Chevy HEI
>w/Moroso's "performance" advance springs, Accel plugs and wires.  Can
>you give me an estimate on the performance increase of adding the MSD
>box, MSW wires, and whatever plugs go best w/this system?

Hard to say.  It will depend on other engine parameters.  The HEI is
not a bad system, though the stock module is a bit short on electronic
dwell.  In my research using motorcycle engines, I'd typically see
an improvement pretty much across the RPM band of around 5% when
moving from a stock-type system to a then-state of the art system.
(Gross generalization but I don't feel like typing in all the results -
if they even survived the fire.)

The big thing "long burn" ignition brings to the table is it covers
for other sins.  If you carburation or mixture distribution is a little
off, it does not hurt you as bad as with low energy ignition.

Whatever plugs you are using are just fine.  I open the gap as far as I
can before the arc starts flashing over to the shell.  This VASTLY
improves idle and low speed response with no degradation at high
speed even on turbocharged engines.


Newsgroups: wiz.hotrod
Subject: Re: ignition questions
Date: Tuesday, Jun 09 1992 15:59:41
From: John De Armond

>> Whatever plugs you are using are just fine.  I open the gap as far as I
>> can before the arc starts flashing over to the shell.  This VASTLY
>> improves idle and low speed response with no degradation at high
>> speed even on turbocharged engines.
>> John
>Neat idea! but how do you tell when the arc starts flashing to the
>shell?  I assume by grounding the plug to the header manifold and
>watch it while the engine is running on the other 7 cylinders,
>(that is, after taking a good long leak! :)  ?

Little more complicated than that.  The pressure of compression at the
time of ignition raises the dielectric strength of the air and thus
increases the necessary voltage.  You might remember the old sparkplug
tester/cleaners you used to see in the garages that had a pressure
chamber that you screwed the plug into and could view the arc under
shop air pressure through a glass window.  I have a homemade chamber
that lets me do the same thing.  I just put shop air on it while
exciting the plug with a neon sign transformer.  My shop air is 175 psi
so that's pretty close to cylinder conditions.  As a practical matter,
if you set the gap to about 4/5ths the distance between the base of
the center electrode where it emerges from the insulator and the shell,
you're in good shape.

You can verify the setting if you have a storage scope you can operate
in the car (#include <std plug for Fluke 97 Scopemeter).  Both the
peak voltage and the arc sustaining voltage vary directly with compression
and with plug gap.  At a given engine speed and throttle setting, the
waveform should be very stable.  If the waveform is jittery and/or the
voltage changes by several hundred volts on a particular cylinder, you
can pretty much bet flashover is taking place.  I have a 1000:1 voltage
divider I use to monitor the actual voltage instead of relying on
an inductive pickup.

>Also, is there a way to increase the electronic dwell on a stock
>chevy HEI?

Yes.  Several aftermarket modules are available with increased dwell.
I'm not really into GM hotrodding so perhaps someone else can give you
specifics.  I've seen the beasts advertised in the hotrod magazines.

I did note while thumbing throgh the Summit Catalog that MSD has something
called the "Super HEI kit" that consists of an MS-6a box, a "Blaster II"
coil and an adaptor to connect the coil where the HEI coil used to go
on the distributor.  $174.

>And since we're on the subject....  When purchasing a set of
>"performance" springs for the timing advance, there are usually 3
>sets of colored springs.  Any idea what the colors designate?  I know
>each spring is supposed to give you different performance and the
>package just says to try each and use your favorite, but....

I love that kind of advice.  The only way to know what each combination
does is to curve your distributor with each one attached.  Then you
have to try them to see what you engine likes.  By curving each
combination, you know what change will occur when you make a change.
To curve the distributor, you either take it to a distributor shop
that has a machine or you build your own.  I'm working on an article
right now for a machine you can build at home for under $200.
(#include std plug for my new magazine).

>Thanks very much for the help, I'm still young in the performance
>field and love to learn these types of things!

Believe me, this is a disease with no known cure.  The only thing that
mitigates the symptoms are frequent large injections of speed.


Newsgroups: wiz.hotrod
Subject: Re: ignition questions
Date: Tuesday, Jun 09 1992 17:47:16
From: John De Armond

> Any recomendations on what coil to use with an MSD system? I have an MSD
>for my 351W, but no coil yet....

Doesn't much matter.  In this application, the coil is simply a step-up
pulse transformer as opposed to Kettering (and electronic variations
thereof) ignition in which the coil is also the energy storage mechanism.
With the MSD box, a series of sharp 550 volt pulses are applied to
the coil.  The coil only has to transform these pulses to the appropriate
voltage.  Most all coils are 100:1 ratio so the open circuit voltage is
about 55 KV.

I've used conventional coils, the MSD Blaster (a conventional coil with
a fancy label), Accel's superCoil, Mallory's Hyfire coil and even
one of the little motorcycle coils with equal results.  The only
criteria is the coil must be able to handle the heat load.  The motorcycle
coil got hot real fast and is not recommended.  Indeed, MSD sells essentialy
this coil in a rugged case and labels it only for drag racing.

The other consideration is to make sure your distributor is phased properly.
That is, make sure the rotor tip is over the appropriate output post
when the spark happens.  Important for regular ignition, it is
vital for the MSD box because the high energy spark can erode the
terminals if it has to jump far and because the fast rise time
pulses can punch through the rotor.

The way to look at phasing is to take an old distributor cap and cut
away all the material below one of the terminals so the rotor is
visible.  Then install the cap, hook a timing light to that lead
and crank the motor.  Observe the rotor while illuminated with the
timing light.  It will show you almost exactly where the rotor tip
is when the spark fires.  Most lights have somewhere around a degree of
propagation delay so keep that in mind when looking at the tip.
Be sure to look at it over the whole range of the advance mechanism.

If the tip is out of time, such as when more than stock advance is needed,
the easiest way to correct the problem is to drill and tap new holddown
holes for the magnetic pickup.  It is possible in some distributors to
repin the reluctor star but that's a lot of work.


Newsgroups: wiz.hotrod
Subject: Re: Point ignitions
Date: Tuesday, Jul 14 1992 13:24:03
From: John De Armond

>Ok race fans, here's the dork question for the day. I'm going to install
>my old Accell Super coil on my '74 Chevy van that has points in the
>distributor. I plan to wire the coil dirrectly to the battery (through a
>high current relay of course) with 10 gadge wire but I'm not sure if I
>need a ballast resistor on this beast or not. Some people tell me leave
>it off and others say the points will burn up faster. Any one know what
>is correct and how does one determin if he can run with out one in
>certain set ups? BTW, yes I do plan to gap the plugs to about .070 when
>I get done.  Thanks guys!

Gotta  have a ballast on that coil.  More to the point, why not drop another
$25 and install a Chrysler electronic ignition box?  A little known
fact is the ignition box is also triggerable by points.  I've used that
box on many motorcycles where it was impossible to fit a magnetic pickup.

With straight Kettering (points) ignition, you cannot run a gap that
wide.  The problem is not voltage but the rate of rise of the pulse.
Kettering ignition has a rather long rise time and that allows
the energy to leak off via other paths.


Newsgroups: wiz.hotrod
Subject: Re: Point ignitions
Date: Thursday, Jul 16 1992 15:10:44
From: John De Armond

> You mean I could do this to an HEI module as well?

Ab-sol-tee-lutely!  In fact, look on page 110 of Summit's July-august
catalog, up in the upper left corner and you'll see an Accel brand
conventional distributor with an HEI module just sorta stuck on the side.

> This is great stuff!  I have a couple of point-type vehicles I need to
>convert.  From prior experience with the HyFire, the points generally
>last for years without trouble as long as they're just used as a switch
>for the ignition module.  Sure, it ain't "real" electronic ignition, but
>you don't see those much for a 235 Chevy where the entire distributor
>rotates for vacuum advance...  <grin>

Yep.  Same here.  I put FOUR of these things along with 4 coils on a
Honda Goldwing and triggered each pair from one of the two sets of
stock points.  Never again touched the ignition system.  I polished
the point cam for less wear and lubricated it with silicone grease
and then forgot about it.

> I saw a guy with a Jeep with the AMC-built Buick V6.  His aftermarket
>intake manifold wouldn't clear an HEI distributor, so he'd replaced the
>point cam with the reluctor piece from an oddfire GM V6, drilled and
>tapped the breaker plate for the pickup, and mounted the HEI module
>underneath the distributor.  Worked like gangbusters.

Even better, consider that by using standard GM parts, you can have electronic
ignition with knock sensor-based spark retard.  Consider this article
form the Great De Armond/PE Archives :-)
>From emory!!dogface!posms!rick  Sat Dec 14 14:36:31 1991
Date: Sat, 14 Dec 91 12:23:54 CST

Regarding the GM EST (Electronic Spark Timing) system, it was used from
about 83 to 87 on non computer controlled pickup trucks, Suburbans and
some light to medium truck engines.  They are identifiable by the EST
logo on the tailgate.  This distributor has standard vacuum and mechanical
advance plus a 4 wire connector connecting the distributor and a EST
controller.  The EST controller, wiring, knock sensor and ignition module
plus a power wire are the complete system.  The easy way to identify the
system is that the distributor has a vacuum advance AND a 4 wire connector.
No other GM system does.  A Motor, Mitchell or Chilton manual will give you
the wiring diagram.  If you adapt this system, you also get variable dwell
control, current limiting and the strong GM HEI coil.

Rick Kirchhof   Austin, Texas                   |
Domain:                   |      Someday...
Bang path: ...!!dogface!posms!rick |


Rick's not at that address anymore, I don't think.  The EST box is
a spark retard box that connects to the 5 wire HEI module, listens to
a knock sensor and retards the timing when it hears knock.  I've tried
it and it works.  Got the connection diagram right out of a Chilton
manual.  If you change major engine family (to a v-6 or inline or whatever),
you'll have to experiment to find a knock sensor tuned to the proper
frequency.  Or you could subscribe to PE and read all about it :-)
(forgive me, Lord, I can't help putting these plugs in :-)


Newsgroups: wiz.hotrod
Subject: Re: Point ignitions
Date: Wednesday, Jul 22 1992 16:23:54
From: John De Armond

>>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.
>Whoa! That sounds way cool...  A bit like the J&S knock controller, only
>better... and much more reasonable. Will it be possible to detect an
>incipient knock signal profile without complex tweaking for different
>engine blocks and sensor types? Perhaps the knock signal waveform isn't
>that much different between different engines... you got me thinking here.

The first version will not do knock control.  This will be an evolutionary
project that lasts probably well over a year.  I'm designing it in
a modular fashion so added functionality can be added as it is developed.
The feature list will also track Dallas's clock speed increases on
the DS-5000 :-)  The really whippy thing about the DS-5000, other than
it contains a whole computer, is that it is programmed by sending
Intel Hex records to the serial pin.  No programmers, no special
fixtures and it can be programmed in-situ.  That means whole new
firmware can simply be downloaded from the PC.  The car hacker's dream!

>>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.
>I was enticed by that also... Although I HATE buying magazines that are
>sealed in plastic so you can't check out the articles first :-)
>When I saw that his DC-DC converter put out something like 700 volts,
>it worried me a bit. I wonder how many OEM coils on modern
>cars can handle 700 volts on their primary. The coil(s) on my car
>look a bit scrawny to handle that kinda of voltage. I've heard that some
>of the higher powered CDI systems can destory a standard ignition coil.
>There was no mention of this in the article... Comments?

Other way around.  In a Kettering system (and electronic versions thereof),
the coil itself stores the spark energy in its magnetic core.  That's
what is happening during the dwell.  If the current is looked at on
a scope, it is observed to rise exponentially according to the LR time-
constant formed by the coil inductance, the coil resistance and the
impedance of the supply including the ballast resistor.  When
the points are opened, the dI/dt induced the output voltage in the
secondary.  The amount of energy depends on the inductance of the primary
and the voltage depends on the peak primary dV/dt and the ratio of
primary to secondary, usually 1:100.  Thus the coil is both an energy
storage inductor and a transformer all in one neat package.

The CDI system is completely different.  Here a capacitor stores the spark
energy.  In the PE system, that is a 1 microfarad cap.  When the unit
is triggered, an SCR dumps the charge into the coil.  The coil now acts
as only a transformer.  The 400 to 700 volts is stepped up by the
100:1 ratio to 40,000 to 70,000 volts.  The inductance of the coil
merely serves to slow the discharge of the capacitor, a desirable
feature.  Since the actual current pulse is very brief, practically
any size coil will do.  Even the little motorcycle coils will work
if the ratio is correct.

>>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.
>I started thinking about multiple sparks too... I could reduce the pri
>DC voltage somewhat, to maybe 100 volts or so, and then trigger a
>rapid string of multiple sparks. The coil might be able to live with
>that a bit easier.... I'd hate to fry my coil cause I'd probably have
>to replace it with 2 conventional coils. (My engine is a 4 cyl that is
>directly fired by a small dual coil that uses no distributor)

No.  No problem with the voltage.  The voltage must be there in order
to achieve the desired secondary voltage.  If for some reason the
coil WERE to be sensitive to the energy dump, the proper thing is to
reduce the size of the storage capacitor.


Newsgroups: wiz.hotrod
Subject: Re: Point ignitions
Date: Wednesday, Jul 22 1992 16:28:12
From: John De Armond

I said...

>The first version will not do knock control.  This will be an evolutionary
>project that lasts probably well over a year.  I'm designing it in
>a modular fashion so added functionality can be added as it is developed.
>The feature list will also track Dallas's clock speed increases on
>the DS-5000 :-)  The really whippy thing about the DS-5000, other than
>it contains a whole computer, is that it is programmed by sending
>Intel Hex records to the serial pin.  No programmers, no special
>fixtures and it can be programmed in-situ.  That means whole new
>firmware can simply be downloaded from the PC.  The car hacker's dream!

I forgot.  One of the best part is my entire development system package
cost less than $150.  I'm using pseudosam's free 8052 assembler,
Pseudosam's simulator ($90) and Micro-C, a $40 shareware table driven
compiler that can generate code for most of the single chip micros
and the PC.  I can even make everything except the simulator available
on the net to accompany the firmware. :-)


Newsgroups: wiz.hotrod
Subject: Re: Curious about electronic ignition dwell
Date: Monday, Jul 27 1992 11:34:44
From: John De Armond

> I have a question about the "normal" dwell angle for an electronic
>ignition distributor.  I know this question may be somewhat
>inappropriate for the "hotrod" list, but I really need the advice
>from a group of automotive experts.  I've come upon this question
>while trying to track down the source of a rough running engine.

>Here goes:
>    vehicle - 1980 Dodge van, 225 slant six, auto., 125 k-miles

We'll make an exception for this pig one time :-)

>While measuring the dwell angle, at idle ( 600 rpm ) the dwell meter
>indicates 52 degrees of dwell.   As I increase the engine speed to
>greater than 1500 rpm the dwell  decreases to approx. 47 degrees.
>To me this seems a bit high, naively comparing to a point type ignition,
>and also I was under the impression that the dwell was to remain constant
>independent of rpm.  These numbers just seem to be too high for a
>6 cylinder distributor, what do  you guys think ?  Is my distributor
>okay or is it a cause for concern ?

Let's talk a moment about dwell and what it means.  Flashing back to an
earlier post of mine, you know that the coil in a conventional
ignition (point or electronic) is both a step-up transformer and an
energy storage device, the energy for the spark being stored in the
inductance of the primary winding.

Storing this energy takes a finite amount of time.  This is known as the
LR time-constant.  It is the same as the more familiar RC time constant
for capacitors.  In the case of inductors, when a voltage is applied,
the current rises exponentially according to the time constant established
by the inductance, the resistance of the inductor and the resistance of
the voltage source.  The current will eventually reach (after ~5 time
constants) a steady state determined by the applied voltage and the
total resistance.

In a points ignition system, some relatively long period of time, set by
mechanical restraints, after the points open, they close again.  This
starts the dwell time.  Dwell ends when the points again open.  Dwell
is important because of the following reasons:

*	Points can handle only relatively small currents so the total
	resistance is high, making the time constant long.

*	Because of this long time constant and the mechanical limit in how
	agressive the point cam can be, the available dwell at high speed
	approaches a significant portion of the electrical time constant.

*	Inadequately charging the coil inductance reduces spark energy
	at high RPM - exactly the wrong thing to do.

Setting dwell is a compromise between how fast the points can be operated
and the desire for the longest possible dwell for maximum spark energy.

Electronic ignitions are a whole different story.  Because a big honking
transistor (technical term :-) is switching the current, there is
no technical limit as to how much that current may be.  In practice,
lower resistance coils and frequently no ballasts are used which
at least doubles the current and sometimes triples or more.  The charging
time constant is accordingly shortened.

More importantly, there is no points inertia to deal with.  The actual
spark event takes only microseconds and immediately after the spark,
the current can be turned back on.  This is why you'll typically see a
"dwell" of within a degree or so of the spacing between cylinders.

Which brings up the next consideration.  A high enough current and a
low enough impedance to charge the coil at high RPM means at low
RPM, most of the dwell time is wasted and only serves to heat the
coil.  At idle, the coil may be charged within a degree or two.
Smarter boxes vary the dwell with RPM to minimize this problem.
Smarter boxes also turn off the current when the engine dies.
Lots less stress on the coil.

What all this boils down to is dwell in an electronic setting is relativly
meaningless.  The service manual may specify a "dwell" but this is only
to let you verify any dwell reduction circuitry is working.  The actual
value is pretty meaningless.


Newsgroups: wiz.hotrod
Subject: Re: Point ignitions
Date: Monday, Jul 27 1992 11:58:04
From: John De Armond

>I understand now... My only concern was that the primary might have
>inadequate dialectric or insulation properties to keep the 700 volt
>discharge pulse from finding an undesired path to ground.

No.  Remember the transformer ratio is the same in either case so in order
to get that 40,000 volt spark from a conventional ignition, the primary
voltage has to rise to the same voltage.  The dV/dt inductive kick does it.

>I must admit that this thought only occured to me after reading
>one of Mallorys' ads for their HyFIRE ignition system. Their
>explanation of how their unit actually worked was a bit vague.
>They seemed to imply that the increased voltage that a CD system
>dumps into the primary can "burn" the coil. It was further claimed
>that their system avoided this problem, but they provided very
>few clues as to HOW it actually did this.

yeah and Jacobs claims to bottle pure mountain spring magic reenforced
with PhD goo. When I can hang a scope on it and see something different,
THEN I'll believe the claims.  That's one reason I like MSD so much.
it does exactly what they claim.


Newsgroups: wiz.hotrod
Subject: Re: Ignition wires?
Date: Tuesday, Jul 28 1992 20:23:04
From: emory!!btree!hale (Bob Hale)

Greg raises some questions about ignition wires and their resistance.

Briefly, the resistance tends to be irrelevant because the coil puts
out a current.  I know, everybody raves about how many KV they get
from their coils but the voltage is only important *before* the
spark occurs.  Look at an ignition waveform on an engine scope
sometime and you'll see that the voltage rises quickly to maybe
20KV and then drops to some lower value, usually about 5KV, for the
duration of the spark.  These voltages will vary widely depending
on things like plug gap, engine temperature, cylinder pressure, etc.

Before the spark occurs there is negligible current flowing in the
spark plug; the only place for the current to go is into charging
the capacitance of the plug which isn't very much (anybody measured
a plug?  If not then I'll check one now that my curiousity is
aroused).  A resistive spark plug wire will slow this charging by
a few microseconds if the resistance is within reason, say less
than 10K ohms.  Once the capacitance has charged to a sufficient
voltage then the spark will occur.

Once the spark occurs then current does flow.  This current is
primarily limited by the impedance of the coil's secondary.  There's
a lot of inductance in that secondary as well as some DC resistance;
both factors conspire to limit the current flow.  Any resistance in
the spark plug wires only has a slight effect in reducing the
current; if you find that you get better engine performance by
running non-resistive wires then either your old wires were shot
or your ignition system was extremely marginal to start with and
needs to be fixed.  Notice that in the example above (20KV to 5KV)
the coil drops 15KV internally.  A few more volts drop in the plug
wires won't matter.

It is true that many factory plug wires didn't last very long when
a high voltage ignition system was installed.  This was due to the
voltage breaking down the dielectric of the old design spark plug
wires.  The old wires were designed to stand up to a conventional
Kettering ignition with about 20-30KV maximum potential and they
frequently weren't up to the stress produced by a high output ignition

Newer factory plug wires are pretty good; GM switched from 7mm to
8mm diameter when they went to the HEI.  And GM wouldn't spend the
extra nickel unless they had to.  Some aftermarket wires are as
large as 9mm but be careful: the diameter is only one factor in
how well the wires will last.  The major factor is the type of
insulation used (and there are some *bad* ones on the market).
Personally, I've had good results using Jacobs wires with an MSD
box in front of both Delco and MSD coils.

Bob Hale                                      ...!ucsd!btree!hale
...!btree!                       ...!ucsd!btree!

From: John De Armond
Newsgroups: wiz.hotrod
Subject: Re: More ignition stuff
Date: Wednesday, Jul 29 1992 12:38:13

>Putting economics aside, since that is always the opposite side of the
>coin with any production technology, what would we be aiming for in
>an ignition system ?

The fastest possible rise time to ensure fast spark lightoff even in the
face of fouling.  The slowest tolerable rise time to mitigate RFI (see
any compromises yet?). Sufficient voltage to lite the fire under all
conditions. The longest possible spark duration to increase the size of
the combustion nucleus which has a marked effect on emissions and detonation.
The highest energy possible to make the hottest spark. The lowest energy
tolerable for long plug life (more compromises.)

>Firing voltage for a fixed gap will vary over operating conditions dpending
>on what's happening in the chamber at that time. With that in mind,
>what are the physical limitations on spark gap ?
>A larger spark gap will require a higher voltage to strike the spark
>and then a higher voltage to maintain the current flow which will mean
>a shorter overall duration since coil energy content is fixed. It also
>puts higher requirements on insulation, etc. from the electrical point
>of view.
>A smaller gap fouls easier and provides less spark 'area' to ignite
>the mixture.
>What have I missed or misinterpreted ?

On the money.

>CDI and similar systems seem to be based on improving the voltage available
>for striking the spark, although I haven't examined them very closely.

Very fast rise time and high voltage.  Unfortunately unless mitigating
actions are taken, the spark is also extremely short, too short for
most applications.

>What about MSD ? If it does what it's name suggests  (and from John's
>comments it does) it seems to be the only thing that concentrates on
>maintaining duration of spark.

Yes.  The MSD is actually a CDI box that fires multiple times.  I wish
I could post graphics here because I have a nifty scope screen of
the MSD pulse from a MS-6AL box.  It is a fast (<100 ns) rise, 580 volt
pulse that lasts only about 70 microseconds.  It is very evident from
the trace where the spark lights off at about 20 microseconds.  This
pulse is repeated up to about 12 times within a few milliseconds.
Hmm.  If anyone is really interested, My FlukeScope dumps in PCL
format so anyone with a laserjet or thinkjet can print it.

>How long do we want a spark to last for ?

Can't really say because I've never had the equipment necessary to
mantain an arbitrarily long spark.  I can say that when I was doing
a research project on the subject, the more spark energy input, the
more HP output.  Within a narrow range of power, of course.
You can't replace your Roots with a Tesla coil yet :-)

>As an aside, John. From the experiences I've been having with so-called
>experts lately, an article on interpretation of Ignition-Analyser (Analyzer
>in American, I think) scope displays would be worthwhile. There seems
>to be a lot of information in them that I'm (and many others) are not
>aware of.

Great idea.  I've never used an auto ignition scope enough to know a lot
about it.  I've always just used a lab scope.  Anyone with a lot of
experience want to take a shot at an article?  If not, maybe I can get
Sun or someone to send me to school :-)


Newsgroups: wiz.hotrod
Subject: Re: More ignition...
Date: Wednesday, Jul 29 1992 21:09:58
From: John De Armond

>range. If for example, the spark gap has a firing resistance of 10K,
>and you have 1K ohm resistance in your plug wires, then 10% of spark
>energy would never make it to the plug gap and would be dissipated by
>the plug wires themsleves... OK, enough speculation on that for now.
>It would seem however that spark lead resistance might have some
>bearing on the spark duration... A higher resistance discharge path
>should maintain a longer duration, lower amperage discharge... ???

Yes.  The spark environment exhibits a negative resistance characteristic
like most all other gas discharge events.  The gap will exhibit some
resistance (obviously) but external factors such as the plug wire
resistance and the coil LR timeconstant are primary determinants.

>The MSD system must have a pretty stout DC output current, considering
>that it has to recycle, or recharge the discharge capacitor every few
>microseconds to provide its burst of output pulses. Does it actually
>recycle that quickly, or does it charge up a string of small capacitors
>all at once and then discharge them in quick succession?

Ask and ye shall receive.  compliments of my ever-handy flukeOscope
and my DC amp clamp, we now have an answer.  I'd never done this
test before.  I knew the average current to be pretty hefty but nothing
like what the scope shows.  The box pulls an average of 6 amps idling
BUT it pulls a peak current of over 30 amps after each pulse.  The
waveform is well shaped and obviously designed to minimize external
intereference.  Probably a rather large cap internal to the box to
smooth the draw.  I do know that the MS-7al comes with an external
electrolytic cap for added smoothing.  I get more impressed with the
MSD box every day.

I tried to get a picture of the primary and secondary current but
there was too much electrical activity flying about :-) to get a good
one.  I'm not sure I believe what I saw but it appears that the peak
primary current is around 200 amps or so.  At a coil ratio of 100:1,
that would set the peak spark current at about 2 amps +_ maybe 100%.
This is believable in light of the fact MSD warns strongly against
resistor plug wires.  I'll go back later and do a better test - maybe
actually put a current shunt in the primary lead.

I have a screen dump of this stuff available if anyone is interested.
I have both the PCL file and a hardcopy that I could fax.  (Please
don't totally blow over my fax phone bill :-)

Is there enough interest in this for me to post the image files?
They're not very big even uuencoded.

>How's their box work??? (can't clone it if I don't know how it works :-)

The box is riveted together but I'm planning on buying a spare so
I can open 'er up and see what's inside.


From: John De Armond
X-Source: The Hotrod Mailing list
Subject: Re: MSD box examination
Date: Friday, Jul 31 1992 12:42:49
X-Sequence: 1793

>I'm curious about how the MSD box works only in a fairly general
>sense. A detailed analysis of all the circuitry and components might
>get pretty involved! I wouldn't really want to build and exact clone
>anyway. The MSD boxes are so inexpensive there really wouldn't be
>much point in doing that (plus it takes all the fun out of it!).
>I'd me more interested in designing something that utilizes the same
>basic principles, and who knows, maybe even finding a way to make it
>betterments o( block-diagram level of understanding of the MSD box would
>certainly helpful as a starting point though.

A few things to think about based on my previous research.  Alternating
polarity sparks may give better plug life.  High voltage RF instead of
pulsing appeared to have promise.  the longer and the higher the arc
current the better, disregarding plug life. The faster the rise time,
the better.  An ideal spark would probably be a high voltage square wave.

>A quick question for anyone that has used one of the MSD units..
>The ignition system on my car (Mistubishi-Diamond Star) has an
>ignition system that closely resembles what is often used on
>motorcycles. It has no distributor, and fires the four plugs
>through 2 coils. The power transistors fire the coils alternately,
>and the output is send to the two cylinders that are both currently
>nearing TDC. Of course only one is really on the power stroke, so
>the other spark is wasted. (The appear to be able to get away with
>this because the engine has a fairly small amount of valve overlap)
>So... can I use an MSD box on this car? And, since I have two coils,
>will I need T-------------------c of them? If this is the case, then it
>*might* actually be cheaper to build my own.

You need two and even then it might not work if the ignition diagnoses
the coils.  It would take some experimenting.  Probably would not
gain you anything unless you start modifying the engine.


From: John De Armond
X-Source: The Hotrod Mailing list
Date: Jul 1992
Subject: Re: Point ignitions

>Dave, do you know what a typical ohm value is on one of these things?
>5 ohms sticks in my mind but I may be wrong. I have access to high
>wattage low value resitors and was considering using something around a
>5 ohm 20-50 watt resistor, will this work? Also will the inductive
>characteristics (sp?) of a wire wound resistor effect anything adversly or
>is a ballast resistor typically a wire wound device?

Ignition resistors have a fairly high positive tempco that is designed to
reduce the current through the coil when it is hot.  Use a regular
ballast resistor.  Cheap enough.


From: John De Armond
Newsgroups: wiz.hotrod
Subject: The MSD Ignition scope dump
Date: Sunday, Aug 02 1992 02:19:22

Ok, here's the first scope screen dump.  This was an excellect exercize in
getting data from the scope to the publishing system.  I think it's
STILL faster to make hardcopy from the ThinkJet and then scan it.  Ah,
the wonderful world of file formats.

This is a GIF file.  I debated as to what format to post.  A TIF file
is half the size but I was not sure a TIF viewer would be available
to everyone.  GIF viewers seem to be everywhere.  If TIFs will work
for everyone, I'll use that next time.  It would cut the traffic in half.

There are two screens in this file.  Both show the pulse applied to the
coil and the current drawn by the MSD box.  Only the time scales are
different.  Note that the second trace still only shows one ignition
cycle.  The two pulses are two of the approximately six pulses fired
per ignition cycle at the test speed.  The box and the scope were
triggered by a pulse generator so a stable display could be obtained.
Note also that the pictures are averaged over 32 sweeps.  This
box is NOISY!

Newsgroups: wiz.hotrod
Subject: Big spark plug gaps
Date: Wednesday, Sep 09 1992 14:02:22
From: emory!!dave.williams (Dave Williams)

-> I know this subject was discussed recently, but what is the advantage
-> of a bigger-than-stock spark plug gap?  Is there any particular set
-> of conditions or RPM range in which it's especially good or bad?

 Theoretically, even the tiniest spark will suffice to initiate
combustion.  In actuality, the mixture isn't consistent throughout the
chamber, and the plug may be in an excessively lean or rich area.  This
is even worse at idle.  Port injected engines are slightly more
consistent at idle than carbureted or throttle body injected engines,
but port injection often isn't real great either.  You can sometimes see
fuel wash on the piston top or chamber.

 In my experience, the bigger the gap, the better.  That's assuming your
coil and wires will take it, of course.  After a certain point, some
ignition systems will try to crossfire or arc within the cap instead of
at the plug.  Old-time point ignitions usually had weak blue sparks.
Fat white sparks that go "ZAP ZAP ZAP" are much better.  <grin>

Newsgroups: wiz.hotrod
Subject: Re: Big spark plug gaps
Date: Wednesday, Sep 09 1992 17:11:52
From: John De Armond

> In my experience, the bigger the gap, the better.  That's assuming your
>coil and wires will take it, of course.  After a certain point, some
>ignition systems will try to crossfire or arc within the cap instead of
>at the plug.  Old-time point ignitions usually had weak blue sparks.
>Fat white sparks that go "ZAP ZAP ZAP" are much better.  <grin>

Absolutely.  The only limit is usually the point where the arc starts arcing
over to the shell or down the insulator.

About 20 years ago I did some very extensive research on the effects of
spark intensity, spark energy and combustion chamber design.  This
work included making a pyrex head for the engine, some 10,000 frame
per second photography and a homemade dyno.  Though the work was on a
two stroke engine, I have no reason to believe the results would be
any different for a 4 stroke.

In general, the wider the gap, the more energy delivered and the longer
the duration, the better the ignition and over a narrow range, the more
power produced.  A minimal spark, one barely able to ignite the mixture,
forms a nucleus of flame inside the plug gap.  This nucleus expands
geometrically outward at a very non-linear rate, most likely driven
by the increasing cylinder pressure.  The growth curve is essentially
exponential so that at some point the combustion process appears to
suddenly occupy the whole chamber.

A hot, long duration spark, on the other hand, ignites a rather large
ball of fire. If the spark lasts a long time, what looks like a stream
of fire develops on the down wind (turbulence) side of the gap.  The
difference in effect is similar to that where you try to burn a field
by setting one fire in the center vs setting multiple fires all over
the place.  You'll notice a marked decrease in the amount of spark lead
necessary (a good thing.)

In general, the practical limit to the amount of spark energy is set by
external influences such as distributor arc-over, coil heating and so on.

BTW, one way to juice up an electronic, non-CD ignition is to parallel
the coil with additional inductance.  I've used the secondary of a
radio shack 12 volt, 4 amp filament transformer with good effect.
If you try this, you need to scope the coil current at max RPM to make
sure the coil is still getting saturated during the electronic dwell.
If not, a control box capable of more current and a smaller ballast
resistor is called for.


From: emory!!gt7038d (MCCLENDON,WILLIAM THOMAS)
X-Source: The Hotrod Mailing list
Date: Feb 1993
Subject: Re: Electronic Ignition

Someone talked about hooking up a Chrysler or GM HEI module to the Duraspark.
Are these modules any better than the duraspark module?  Would it just be
better to run MSD 6A?  Or do I need both?  Also, what does anyone know about
the MSD 5 module?  They seem to be a whole lot cheaper than the 6A.  Sorry for
all the questions.  Thanks.
Tom McClendon

[Hey, don't worry about the questions.  That's what this list is for.  I'll
have to let someone else address the Duraspark but I can comment on the
generic chrysler module.  I've used the generic chrysler module for
over a decade for everything from racing motorcycle engines up.
First a little background.

For electronically switched ignitions (as opposed to capacitor discharge)
where the coil is the energy storage element, there are only two
major and two minor considerations.  The major ones are a) how much current
the pass transistor will flow and b) the dwell or how soon after the trigger
the ignition turns the current back on.  Just about every ignition system
I've looked at will pass more current than any commercially made coil
will withstand so that is not really a consideration.  The sooner the
ignition switches back on after the trigger the better.  That is because
the inductance of the coil takes a finite amount of time to charge.
At high RPM on multi-cylinder engines, that charge time can approach
the amount of time available between sparks.  (A reason why coil-per-coil
has become so popular)  There are "high performance" chrysler-style
ignition modules sold for huge premiums (in the Summit catalog, Mopar
modules range in price from $39 to $79 compared to the $19 I pay
for mine) that claim more spark, stable dwell (who cares) and so on.
My measurements here in the lab shows that the generic module Rockhill
brand LX-101 (the cheapest I could find) turns the current back on
literally while the secondary ringing is still going on.

How much energy is delivered in the spark from an electronically switched
ignition is strictly dependent on the amount of current and the coil
inductance.  About the only effect the module can have on this is to
switch the (-) terminal closer to ground.  The Rockhill module goes to
within a few hundred millivolts of ground which is about as good as one can
expect.  BTW, if there is an indication that there is insufficient
spark energy in a drag racing or other short term environment, one
can increase the energy by reducing or eliminating the ballast resistor.
The coil will likely overheat over the long term so don't do it
on the street.

The minor considerations are whether the module reduces the current at
idle and stop to protect the coil and whether the module actually
manages the charge in the coil.  Some brands of modules do this
and some don't.  The original chrysler module DID.
There is now a chip from Motorola available that does this and manages the
charge in the coil. That is, it figures out how much energy was used in the
last spark and how much dwell is needed to fully recharge the coil
and applies power accordingly.  I'm not aware of which modules use
this chip.

The GM HEI-style module works equally well but is more trouble to use
because it must be heat sunk.  On the other hand, there is a HEI module
combination used on some 84-87 GM trucks that listen to a knock sensor
and implement knock control.

Regarding MSD modules, I don't consider the MSD-5 to be any improvement
over the factory modules.  They both deliver about the same energy.
The MSD-6 series is the all round best module for street and gasoline
fueled racing.  Lots of energy, fairly tolerant of low supply voltage
and reliable.  Note that the coil size does not matter with CDI ignition
because the coil functions only as a transformer and does not store
energy.  The only size consideration is in regards to cooling.

The MSD modules will trigger from points, variable reluctor triggers
or the Mallory Unilite points substitute.

One thing to watch with all these modules is to correctly phase the variable
reluctor pickup.  The waveform looks like this:

     /   \
----/     \x      /------
           \     /
            \   /
             \ /

The ignition should trigger at the zero crossing, point "x".  that is the
most stable point vs RPM.  If the polarity of the trigger is reversed,
the ignition will likely trigger on the rising edge, the position of
which changes with RPM.

Checking this is easy.  Take an old distributor cap and cut away some
of the side so that the trigger can be viewed but all leads are intact.
Install the cap and start the motor.  Hook a timing light to the center
lead and shine it on the trigger.  The image should show the teeth of
the rotor exactly aligned with the reluctor pole.  If it is advanced
or retarded from the pole, the phasing is wrong.  You should also see
the timing change with RPM if the phasing is wrong.  This is also
a good time to check the rotor phasing, that is, making sure the
rotor aligns with a terminal when the spark triggers. connect the timing
light to the lead  nearest the hole and observe the rotor while the
motor is running.  The rotor should be aligned with the terminal.
Check it across the whole advance curve.  If the timing is off, it is
correctable by changing the position of the trigger vs the rotor.
Usually done with slotted holes in the pickup.


From: emory!!tdrury
X-Source: The Hotrod Mailing list
Date: Mar 1993
Subject: MSD Ignition - I'm impressed
X-Sequence: 4345

Does anyone have test data to support MSD-style ignition over
single spark ignitions?  I'm curious, but I dont see where
multispark improves anything.

I image one initial spark combusts the mixture in the immediate
vicinity of the electrode and the combustion starts to spread...
Any further sparking of the electrode is sparking into already
combusted material.  Or is the swirling action of the cylinder
always bringing new fuel/air to the plug.

In any event, what sort of power increase can be achieved?  I dont
really care to hear what MSD claims it to be.

"doubting" tim

[  THe MSD system reverts to single spark at some RPM (actually it drops
a spark per cycle gradually until it is down to 1) so the high speed power
gains are strictly from the increased energy in the spark.

I did some research as part of a science fair project (tells you how long
ago that was :-) on the effects of ignition energy.  I built a dyno
for the small motorcycle test engine and a pyrex head (two stroke).
I found a direct correlation between power and spark energy over a very
wide range of energy.  And the longer the duration the spark, the more
the power.  The difference is not large but it IS there.  Even better,
the part throttle running is MUCH better.  Looking at the combustion
process through the pyrex head with the aid of a homemade high speed film
camera, one could see the nucleus of combustion change markedly as
the spark duration increases.  While the conventional ignition makes
a nucleus of fire around the plug, that flamefront turns into a streak
with long duration as the turbulence of the mix blows fresh mixture into
the spark.

The effect on off-peak power and driveability is marked.  My brother
was racing a Hodaka 100 cc motocross bike (another indication of how long
ago this was)  It was the fastest bike on the track (ahem :-) but with
the stock magneto ignition it literally had no power band and its tendency
to foul its plug was so bad that we carried around a set of rollers
for warming the engine before we went to the line and we bought plugs
by the case.  As some of you other old pfarts might remember, the Nippon
Denso magneto had a flywheel with three coils inside.  One coil was the
magneto coil and the other two were for lighting and battery charging.
Based on the research above, I redesigned that magneto by rewinding all
three coils to provide magneto energy  and replaced the tiny coil with
a conventional automotive coil.  The change was marked.  The damn thing
would even idle.  Midrange power appeared.  Lap times went down markedly,
teaching me the importance of maximizing the area under the power curve.
I got to change points after every event :-) but I fixed that later with
a homemade transistorized ignition.

I have seen similar results on automotive engines so this was not just a
2 stroke fluke.  JGD]

From: emory!!!mwalker (Mark Walker)
X-Source: The Hotrod Mailing list
Date: Mar 1993
Subject: Re: MSD Ignition - I'm impressed
X-Sequence: 4356

In article <>, (The Hotrod List) writes:
> Does anyone have test data to support MSD-style ignition over
> single spark ignitions?  I'm curious, but I dont see where
> multispark improves anything.


> I did some research as part of a science fair project (tells you how long
> ago that was :-) on the effects of ignition energy.  I built a dyno
> for the small motorcycle test engine and a pyrex head (two stroke).

W O W  John, cool science fair project!

[ Lots of fun.  I got my first introduction to scientific politics on
that project.  I won the regional fair with it but at the state level I
lost to a kid who flew and photographed Estes rockets!  His dad turned
out to be the boss of two of the judging engineers.  That was an erie
awards ceremony.  I'd won awards from the Army, Navy and Air Force
(a gold medal, a leather briefcase and a scholarship respectively)
They announced the grand prize winner and... no one clapped.  Silence.
I didn't understand what was going on, being a nerd novice in politics
but my parents had grins from ear to ear :-)  OK, back to hotrodding.  JGD]


> by the case.  As some of you other old pfarts might remember, the Nippon
> Denso magneto had a flywheel with three coils inside.  One coil was the
> magneto coil and the other two were for lighting and battery charging.

Yeah, they were kind of weak all the way around.

> Based on the research above, I redesigned that magneto by rewinding all
> three coils to provide magneto energy  and replaced the tiny coil with
> a conventional automotive coil.  The change was marked.  The damn thing


And I thought I was having fun in high school rodding my Suzuki, Impala
and mid-engineed Corvair around.  Now I wish I had attempted some of
the "techier" things that rolled across my brain from time to time.

> ----------
> Posted by: emory!!tdrury

Mark Walker

From: emory!!gallant (Robert Gallant)
X-Source: The Hotrod Mailing list
Date: Mar 1993
Subject: Re: Wired
X-Sequence: 4438

On 9 Mar 93 04:35, hotrod wrote:

>For what it's worth, the $20 Autozone Lifetime Warranty Maganetic
>Suppresion wires that I've got on my Jeep are 1.4 ohms per foot.
>MSD's magnetic supression wires should be similar, but should have
>better insulation aand cost more :).

Boy that is low!!

I called MSD.  Their wires used to be 150 ohms per foot but are now
850 ohms per foot.  The Jacobs are 150 per foot.


[I'm beginning to wonder if we're all talking about the same thing.  Real
MSW is a spiral of metal wire wrapped around a fiberglass or similar
core.  It should have minimal DC resistance.  Any resistance wastes energy
with CDI ignition.  I've measured a peak current of >1amp from the coil
of an MSD system.  Even a little resistance throws away energy.  JGD]

From: "Chris Kent Kantarjiev" <emory!!cak>
X-Source: The Hotrod Mailing list
Date: Mar 1993
Subject: MicroDynamics Ignition control?
X-Sequence: 4603

Anyone have experience with these guys? They have a very slick looking
rev limiter, but say that it won't work with a CD ignition (like MSD).
Instead, they are pushing their own ignition control, which they call
"exponential discharge" - "has a fast rise-time output that works with
the coil, not against it".

Before I jump into unknown waters here, I'd like to hear from anyone
that has info on the boxes. Thanks.

[You know, there's a whole discipline called electrical engineering
that has a formally defined and widely recognized vocabulary for
describing electrical devices and phenomena.  Any company that tries
to sell a product using homemade terms and/or abused standard terms
for common phenomena immediately goes on my bullsh*t list.
Sounds like this company needs to join Jacobs and their "Energy Team"
on that list.

I have no experience with that box but their hype has my interest approaching
zero.  JGD]

From: emory!!lusky
X-Source: The Hotrod Mailing list
Date: Mar 1993
Subject: Coils for CDI
X-Sequence: 4648

A few weeks ago I called Adrenaline for some info on their CDI
box (I like our MSD 6AL just fine, but my faculty advisor insisted
we try the Adrenaline).  The Adrenline guy asked what I was currently
using for ignition (GM computer controlled HEI dist, GMCM, MSD 6AL,
Mallory Promaster, Mallory spiral core 800ohm/foot wires), and commented
that we should be using the stock HEI coil (external) because the windings
ratio was too high in the Promaster, resulting in decreased efficiency and
less current to the plugs.  I was under the impression that the
energy in a spark varied proprotial to the square of voltage, and that
higher voltage was very desireable.  Whats the correct answer?

Jon Lusky <<== preferred address

 79 Rx-7 12A/Holley 4bbl
 67 Camaro/350/4spd (under construction)
 89 Jeep Wrangler/258 (currently self destructing)

[Quick look at spark phenomena.  First the Kettering (inductive) style.
When the points/pass transistor break, the inductive energy stored
in the coil builds voltage in accordance with the familiar L*dI/dT
relationship.  When the voltage reaches the value required to ignite a
spark across the gap, the spark discharge starts.  This may be as low as
8000 volts at idle or as much as 40,000 volts under worst case conditions
of compression, electrode erosion and distributor cap gap.  Once the
arc ignites, the voltage drops to the a much lower maintenance voltage.
This voltage is typically 3,000 to 10,000 volts, again primarily depending
on combustion chamber conditions.  The arc continues at this constant
voltage until the energy stored in the coil is depleted.  Since the
established arc behaves like a classic negative resistance device,
the current is dependent primarily on external factors such as the resistance
of the coil secondary, the resistance of the plug wires and so on.
The characteristic of the arc is almost identical to that of the common
glow discharge voltage regulator tube that hams may be familiar with.
Also quite similar to the behavior of a fluorescent lamp.

Since power is proportional to volts * amps, given that the voltage stabilizes
at whatever value required to maintain the arc, it becomes intuitive that
the more current available, the hotter the arc.  Since the total amount of
power stored in the coil inductance is finite, increasing the current
by lowering the resistance shortens the arc and vice versa.  At least
over a reasonable amount of time.  Flux leakage in the coil also consumes
energy so you can't stretch things indefinitely.

Since the total power delivered over time is finite, it is common to
express this amount of power in terms of Joules (1 joule == 1 watt-second).
The total energy available with a Kettering system is proportional to
the peak current in the primary and the coil inductance.  This is why
heavy current and large coils work better.  Also why electronic ignition
works better.  It permits more primary current than can be handled by

Capacitive discharge ignition works a little differently.  The energy
storage device is a capacitor charged to ~500 volts.  An SCR discharges
this capacitor into the coil which acts as a simple transformer.  Since
a typical coil transformer ratio is 100:1, that 500 volts will result in
a peak secondary voltage of 50,000 volts.  When the SCR fires, the
500 volts rather rapidly applied across the terminals of the coil primary.
This voltage rather rapidly starts a current flowing that is proportional
to the impedance of the coil.  As the current builds in the primary,
voltage rises in the secondary.  This is the dI/dT relationship
as before except that the current is RISING in this instance.
A higher inductance primary, desirable for Kettering ignition, is actually a
disadvantage because it slows the voltage rise.

Once the arc ignites, the voltage and thus the impedance of the arc
drops to the sustaining voltage as before.  This impedance drop is
reflected back through the coil to the capacitor.  The transformer
impedance ratio is the square of the turns ratio so for a 100:1 turns
ratio, the impedance ratio is 10,000:1.  Thus if the arc impedance is
8,000 ohms (reasonable, assuming 8000 volts and 1 amp current),
this is reflected back as 0.8 ohms to the capacitor.  Needless to say,
this causes the capacitor to discharge rather rapidly.  That is why the
spark of a CDI system is so brief.  What may not be intuitive is that
any stray impedance in the primary circuit greatly diminishes the spark
energy.  This includes wiring impedance, the coil resistance, the
impedance of the SCR and the internal impedance of the capacitor.

In order to minimize the proportion of power consumed in the stray
impedances, it is desirable to minimize the peak primary current.  To that
end, the coil (transformer) should have the LOWEST turns ratio necessary
to generate the necessary peak voltage to ignite the arc.  this is a
simple matter of impedance matching.  The super high voltage coils
which have higher turns ratios and/or higher impedance, are actually
undesirable.  The only parameters involved in the pulse transformer (coil)
are the ability to handle the peak current without magnetically saturating
and the ability to handle the peak current.  And, of course, the necessary
secondary insulation to prevent arcing.

Anyone ever notice the little MSD ProPower #8201 drag racing coil?  This
tiny coil, not much larger than a motorcyclecoil , is designed to work with
the MSD-7 and -8 units.  The -7 and -8 boxes are extremely high energy
boxes.  Yet this tiny coil handles the power.  If you look at MSD's
advertising, you will note that they claim "33% more current at the plug
than competing coils".  This indicates a lower turns ratio transformer
with larger wire, necessary to handle the peak primary current.
Note that it is NOT suitable for continuous duty due to heat buildup
from the high primary current.

On the other side of the equation, I've measured an Accel SuperCoil
vs an MSD Blaster coil (low ratio but larger in order to handle continuous
duty) on an MSD-7 box.  The Accel coil is capable of a gazillion volts
but does not deliver the current once the arc is started.  The Blaster
delivers over twice the current to the plug.

Now the real world question of how to select the coil.  In the absence of
the capability to test, use the coil recommended by the CDI box mfr.
The HEI coil is probably a good choice for most non-nitro applications.
It is built to withstand extreme heat so it should do well in the

I have a rig I've built to collect data for a future magazine article on
ignition which lets me measure the peak current and voltage across a
plug while subjecting the plug gap to varying pressure.  The rig is pretty
simple.  The pressure chamber consists of a large refrigeration sight
glass with a 14mm sparkplug hole drilled and tapped into the side.
I can view and photograph the arc through the sight glass while varying
pressure applied through the refrigerant port.  A Fluke high voltage
probe is used to scale the spark voltage and a 10 ohm resistor
from the pressure chamber to ground lets me measure the arc current.
A very simple RC filter clips off all the RF above 20 khz.  The
signals are monitored and recorded on my Fluke 97 scopemeter.
A simple el-cheapo function generator drives the Box Under Test (BUT :-)
at various "RPM".  Maybe if I can get some time, I'll capture some
waveforms and GIF 'em for posting to the archive.

Damn that got long-winded... JGD]

From: emory!!Steve_Baldwin (Steve Baldwin)
X-Source: The Hotrod Mailing list
Date: Apr 1993
Subject: Glass heads
X-Sequence: 4872


You mentioned a while back that you did some work with a Pyrex cylinder
head. I may be in total dream mode here but what are the chances of
that as a practical proposition ?
I'm thinking of my sidevalve motor here. A running motor with a pyrex
head would sure make those finned aluminium Edelbrock heads look real


PS. When I say practical, I mean in it in the same way that a T-bucket
is practical.

[Do-able but probably not too practical.  The head I made was for a small
2 stroke single cylinder with a simple hemi combustion chamber.  This
shape was ground using fairly conventional lens-making techniques.
Grinding an arbitrary combustion chamber shape could be done with
a sandblaster but polishing it optically smooth would be a real chore.

More importantly from a practicality perspective is the fact that
the glass does not conduct very much heat and therefore does not
cool.  My runs were only for seconds at a time.  In order to seal
the head to the cylinder, I had to fit an ordinary O-ring to the cylinder
and put rubber washers under the mounting bolts.  This limited the amount
of heat tolerable.  The spark plug was mounted in a metal sleeve epoxyed
into a hole drilled through the glass.  This also would withstand only
minor heat.

Last is the cost.  My experiment became a reality when I found an approx
4" thick pyrex mirror blank in the school's optical lab.  This thing
was pretty expensive.  Getting a thick slab of glass large enough
to cover 4 heads would probably increase the national debt.  JGD]

From: emory!!prg
X-Source: The Hotrod Mailing list
Date: Aug 1993
Subject: Re:  Jacob's Ignition System
X-Sequence: 6049

Joe asks, and John comments:

>Has anyone had any experience with a Jacob's Energy Team ignition system?
>I'd also appreciate any info on any of their products.
>Customer satisfaction is also something I'd like to hear about.
>Thanx - Joe
>[Executive summary:  Overpriced, grossly over-hyped, garden-variety CDI
>ignition that does nothing extraordinary and brings to the table all the
>problems associated with single spark CDI.  This is an FAQ (if
>we actually had an FAQ) so see the archives for details.  If you don't
>want to go the cost of an MSD system, a high output coil (Accel, Mallory,
>etc) with a high capacity OEM-style electronic box works very well
>and is MUCH cheaper.  I've found that the stock Chrysler box that
>can be had at Walmart for $16 is just fine for up to about 6000 RPM on
>a V-8.  JGD]
>Posted by: emory!!parys

John, am I to understand that since the last time you and I discussed this
topic you've had the chance to look at one of these units closely and disprove
what the doctor is claiming, or am I assuming too much?  Just curious!


[Yes that is true.  Jeff Deifik sent me one along with a "sooper dooper
energy coil".  I don't have a copy of his patent so I don't know what I'm
*supposed* to observe but what I saw is this.  The unit fires what is
essentially a signle distorted sinusiod of high voltage into the coil.
According to The Doctor, this is supposed to cause the high voltage secondary
to reverse polarity and "replate eroded plug material" and a bunch of other
voo-doo.  The secondary voltage does reverse but only incrementally more
than with a regular system.  The energy delivered is unremarkable and since
it is a single pulse CDI, it suffers from the same problems associated with
short sparks as any other CDI.  I ran several tests to try and detect any
"replating of eroded plug material".  This included firing a plug at
10,000 RPM equivalent for a couple of weeks, firing a gap composed of very
thin copper wires for the same interval and a gap composed of iron wires.
An MSD box fired side by side with it.  I could detect NO difference in
gap erosion.  Essentially none on the plug, some on the copper and a lot
with the iron.  I could detect NO difference in the coil.  I looked at
inductance, ratio, saturation (thinking it might be a swinging transformer)
and resistance.  Looks like just another coil to me.

If you want to get a rather ordinary CDI box and at the same time fund
an amazing hype and telemarketing operation (I had a hell of a time getting
their telemarketers off my phone after I called for a catalog), then the
Jacobs is your box.  Otherwise look elsewhere.

I should note that Jeff's box is an older model.  Their current box is in a
famcier box but I have no reason to believe the performance is any
different.  JGD]

From: emory!!dave.williams (Dave Williams)
X-Source: The Hotrod Mailing list
Date: Mar 1994
Subject: BS alert
X-Sequence: 7933

 I'm flipping through a May 1982 issue of "Custom Rodder" magazine.
They have a two page article by Christopher Jacobs, of Jacobs ignitions.
It's the most amazing example of fact and fiction I've seen since I
visited a friend in the mental hospital in 1990.

 Jacobs claims his "adaptive spark" computerized ignition can
automatically make "33 changes in spark duration, intensity, and phase
angle - within 1-1/2 degrees of engine rotation at 6000 RPM."

 The changing the duration part doesn't even make sense - when you stop
the spark, it's over.  Q.E.D.  The intensity thing... as far as I know
that's determined by the coil's input amperage and the number of
windings, though I'll admit what I know about coils could be written in
large print on my thumbnail.  Maybe the box does something.  And phase
angle?  What the hell is phase angle on a DC spark?

[When you read the fine print (or as I did, X-ray one of his little
potted boxes) and see that his "computer" is all-analog; eg, opamps
and stuff, you realize the full scale of his bullshit.  I've come
to the conclusion over the years that anyone who completely pots
something (as opposed to simple conformal coating) is hiding something.
I'm damn sure in the case of Jacobs.

His box does not drive the coil with a pulse like other CDI systems.
He drives it with one cycle of a 450 volt pseudo-sine wave.
The interesting thing is, by the time the coil inductance and secondary
capacitance gets through with it, the actual waveform that appears
across the coil primary looks almost identical to the one produced
by, say, a single pulse of the MSD box.

Stripping all the BS aside, it is a competent CDI box, comparable
in price with other simple CDI boxes but much inferior to other
multiple discharge boxes.  The main reason NOT to buy it, other
than the BS, is that it is potted and thus cannot be repaired
when broken.  JGD]

 Jacobs also claimed a typical 1982 car misfired 6 to 12 percent of the
time at cruise, and that his miracle box would give an "average
improvement of slightly over 16 percent" in gas mileage.  As Arnold
would say, "boul-sheit."

 Talking about lean mixtures, he says, "...the leanness is indirectly
set by the limits of what the ignition system can ignite and not by what
the engine could most efficiently use.  It is further complicated by the
fact that a spark that would be ideal to get a cold engine started would
lead to bad fuel economy and vice-versa, 'cavitating', etc."

 Cavitating?  What the fuck is cavitating?  As for his claim that a
spark for a cold engine should be different than a running engine...
bullshit.  Complete and total bullshit.  Somewhere I have three or four
issues of Cycle Magazine from back in the 1970s, when Gordon Jennings
published his several-part study of ignition systems.  Jennings may not
have a doctorate like "Doctor" Jacobs, but he has a degree in
engineering.  Furthermore, he knew what he was talking about, while
Jacobs is blowing hot air.  I modified my RD350's ignition system per
Jennings' instructions and picked up over a second in the quarter mile
and 10mpg; that's facts you can get a handle on.  Jennings' conclusion
of his study was something like, "we tested rise time, we tested
voltage, we tested everything.  These were all important, but only to
ignitions that were weak.  Sparks like flicking a Bic in a dark hangar
are bad, sparks that blast the combustion chamber like Thor, god of
thunder are good."  Jennings may have simplified things a bit, but I bet
Jacobs never kicked himself breathless on a balky RD's starter lever.

[He's come up with this zany theory that too much spark energy can
somehow blow the fireball away from the gap.  He wastes almost a chapter
on this in his book. (you really wanna puke, borrow a copy of
this tome.)

Right after Jenning's article came out, I decided to duplicate his
work as a science fair project.  I built a dyno out of an Opel disk
brake and used a 100 CC Hodaka racing engine as the test mule.
Guess what?  I found the same thing Gordon did.  I went so far
as to build some apparatus where I could switch high voltage
from a robust source to the plug.  All the way up until I started
melting the electrodes, I saw a small but definite increase in
power.  More importantly, a spark that is only competent and not
outstanding completely eliminated the stereotypical plug fouling
problem 2 strokes of that era were known for.  Following this
project, I rewound the combination lighting/magneto coils on our
Hodaka racing bike to produce as much energy as I could get.
Completely transformed the bike.  I've been a power spark evangelist
ever since.  JGD]

 Jacobs says, "The computer method is to continually 'seek' the ideal
spark to assure ignition for that particular cylinder for each cycle.
For the next cylinder, which may be richer or leaner, an entirely
different spark may be 'ideal.'"

 "Ideal?"  Maybe it's pink instead of blue?  What's an ideal spark?

 Jacobs continues, "Finally, nobody is exactly sure how an electric
spark ignites gasoline in the first place."  Gee whiz, I was never aware
there was any question.  Maybe it's orgone energy, or animal magnetism,
or Clintonomics.  He doesn't say.

From: John De Armond
Subject: Re: transistor ignition schematic
Date: Tue, 21 Sep 1999 16:11:05 EDT

NANS wrote:

> Dear Jens,
> There is a good diagram that you can consult here.However there is
> another diagram I have which incorporates a rev.limiter and points
> bounce control. The only adjustment would be for wear of the heel fibre
> that contacts the cam which is wears realy very slow.The points
> themselves never burn and you can add a switch to convert back to pure
> Kettering(points alone) if ou need to in an emergency.

Interesting to note that the generic chrysler ignition modules
(chrome/black stock unit plus all mopar performance hipo modules)
will work just fine with points.  The module fires when one of the
pickup leads (don't remember which one - can't damage the unit by
trying either) is grounded.  Simply hook the pickup lead to the
points, remove any condenser present, set the points for minimum
dwell and away you go.  Note that the coil fires when the points
CLOSE so there will be a rotor button phase error unless the dwell
is set as low as possible. Removing the points cam and polishing it
to a mirror finish makes the rubbing block on the points last
essentially forever.

Among the vehicles I've used this on is a 79 model Honda Goldwing.
Two modules, each firing two coils in parallel (high power versions
of the waste spark ignition system.)  There was not enough room in
the stock points housing to replace the points with a solid state
pickup so I just went with the points.  With the polished cam, the
points were still going strong when I sold the bike in 87.


Index Home About Blog