The MSD Ignition - 'Exposed'

[Lorenfb]

A technical evaluation of the MSD capacitive discharge ignition (CDI) system
indicates a number of shortcomings compared to the original Bosch CDI used
by Porsche in the 911 up until the 911 3.2 with the inductive discharge ignition
system.

The MSD CDI's key marketing feature is its multiple sparks which produce
a maximum of nine sparks at about 100 RPMs while cranking to a single spark once
the RPMs reach 3000 RPMs. The time between sparks is about 1.5 milliseconds
which contributes basically nothing to enhance the combustion process as the
significant part of the fuel charge burn time is less than 1.5 milliseconds.
Additionally, the spark pulse width of 40 microseconds is half of the Bosch CDI
which results in less time to fully facilitate complete fuel charge ignition resulting in
potential mis-fires and greater emissions.

The MSD CDI produces a primary pulse voltage over 450 volts compared to the
Bosch CDI's pulse of about 300-350 at typical engine RPMs, thereby stressing the
ignition system wiring, the coil, and the rotor & cap. This further shortens the life of
these components beyond what the multiple spark effect does, i.e. at idle the MSD
is producing four sparks, versus one for the Bosch CDI, resulting in significantly
shortened ignition component life.

To implement the multiple spark system, the MSD CDI requires a significant number
of components compared to the Bosch CDI and thereby significantly reduces its
reliability compared to the Bosch CDI, as the reliability of any system is reduced
as its component number is increased. Additionally, the MSD CDI requires additional
wiring for power because of the its high current requirement to produce the multiple
sparks, whereas the Bosch CDI has a simple power and ground.

Since most all the torque is developed at RPMs greater than 3000 RPMs and the MSD
CDI produces just one spark there, it operates in a mode the same as the Bosch CDI
and thus has no real advantage in performance, even if the multiple spark had a
benefit. So given the insignificant benefit of the multiple spark effect with the
additional complexity and number of components, and a shorter initial spark, a less
desirable ignition system results with the MSD CDI than with the Bosch CDI ignition.

Included are images of the oscilloscope trace of the 9 sparks at 100 RPMs, the inside
of the MSD CDI showing it two ICs (PIC processor, & driver IC), 7 power semiconductors,
4 electrolytic capacitors (bad choice for high temps in an engine compartment), and the
extensive number of other components, and the very simple and reliable Bosch CDI.

[haasfogle]

excellent work - thank you.
confirms what i always expected....
again, thanks.
haas

[quattrorunner]

I lover the way my msd equiped car starts bertter than ever before. I have heard all the shortcomings about the msd ie. needs to have a more stout rotor, no gain over 3000rpm, opinion it looks too american hot rod, etc..
But it's pretty cheep, easy to install, it is reliable and you can aford to have another one in your tool kit along with all the other emergency parts we need to have on hand for roadside repairs (just in case).
I know your an expert Loren, and I apreciate your post. But your pictures of the insides make me more happy I use MSD. More modern and seemingly less likely to fault.
Other than these points, I'm super far from any expert on this subject.

[Lorenfb]

"More modern and seemingly less likely to fault."

Just like the Boxster & 996 engine, right?

[quattrorunner]

Well, we're not talking about mechanical are we? More like 30 year old wiring.

[RSTarga]

The original Cd box is still on my 73 so the Bosch unit is very reliable. I also have had an MSD on another car for 20 years and it has never failed either so although I would never switch the Bosch on my 911 it is a good alternative for those who don't want to spend the money on the Bosch.

[don gilbert]

I believe you can have the Bosch unit repaired by any radio geek pretty cheap.

[T77911S]

if you want to count components, the permatune i had years ago had FAR less than either one. as an electronics tech, i like the "less is more" less to break idea, although not a fan of permatune.

i like the higher voltage. i think the 450 volts stressing the ignition wiring is stressing the bias you seem to have for bosch. the MSD puts in new wires to the coil that replaces those old brittle wires of the 911 that will handle the 450v.
if anyone is concerned about "stressing" the coil, use the MSD coil, it is designed to work with the MSD. a lot of problems with coils comes from mounting them upside down.

i like the idea of the additional wiring. i think it is better to have a dedicated wire from the battery to supply power to the MSD versus running power through the ignition switch all the way to the back of the car. that could limit the current available to the bosch unit due to high resistance reducing its output.
i get power from the starter battery cable for my MSD.
MSD also has a dedicated ground from the coil back to the CD, thus eliminating any ground problems between engine and body.
you can also get MSD with a rev limiter. i have the rotor rev limiter and the MSD.

i like my MSD and it has worked fine for me.
for a stock engine, you arent going to see any improvements between the 2 units, although some say their cars start better with the MSD.

one other thing, their are way to mnay race cars, non-porsche too, that run MSD for it not to be a good unit.

dont get me wrong. i am a fan of OEM/bosch. i would have put a bosch unit back in my car but the price was just too good for the MSD, plus, i can buy an MSD at any speed shop if it goes bad, along with a coil, although i carry a spare permatune and coil with me.

that is my first glance inside the MSD. even though the bosch unit is a pain to work on, those wire lugs are a pain to deal with, the bosch unit is more "repairable" than the MSD, in that i can fix a bosch, i dont think i would try to fix and MSD.

[Bob Kontak]

Originally Posted by quattrorunner

More modern and seemingly less likely to fault.

We went to the moon in the 1960's with similar (more elementary?) circuit boards to the Bosch. Their computer system had one tenth of a megabyte of CPU memory.

I like my MSD, even with my plug gaps set at only 30 thousandths, but I am not picking up a "seemingly less likely to fault" from the look of the Bosch innards.

I fixed a set of Advent speaker cross-overs once (with a friend helping me) so I have a firm basis for making these statements. :-)

Originally Posted by quattrorunner

Well, we're not talking about mechanical are we? More like 30 year old wiring.

This makes more sense to me as a layman vs. the MSD looks modern and is therefore tougher.

[wwest]

Originally Posted by Lorenfb

A technical evaluation of the MSD capacitive discharge ignition (CDI) system
indicates a number of shortcomings compared to the original Bosch CDI used
by Porsche in the 911 up until the 911 3.2 with the inductive discharge ignition
system.

The MSD CDI's key marketing feature is its multiple sparks which produce
a maximum of nine sparks at about 100 RPMs while cranking to a single spark once
the RPMs reach 3000 RPMs. The time between sparks is about 1.5 milliseconds
which contributes basically nothing to enhance the combustion process as the
significant part of the fuel charge burn time is less than 1.5 milliseconds.

Additionally, the spark pulse width of 40 microseconds is half of the Bosch CDI
which results in less time to fully facilitate complete fuel charge ignition resulting in
potential mis-fires and greater emissions.

But isn't that shorter spark duration more than made up for by the 50% voltage overdrive of the coil primary? 450V vs 300V increases the potential for firing a fouled plug significantly......

The MSD CDI produces a primary pulse voltage over 450 volts compared to the
Bosch CDI's pulse of about 300-350 at typical engine RPMs, thereby stressing the
ignition system wiring, the coil, and the rotor & cap. This further shortens the life of
these components beyond what the multiple spark effect does, i.e. at idle the MSD
is producing four sparks, versus one for the Bosch CDI, resulting in significantly
shortened ignition component life.

That largely depends on the risetime of the MSD voltage peak vs the CDI, a faster risetime, AND to a 50% higher voltage results in quicker firing of the plug thereby lowering the "wear and tear" on the other components. Once the plug fire the voltage coil voltage falls significantly....<200 Volts...?

To implement the multiple spark system, the MSD CDI requires a significant number
of components compared to the Bosch CDI and thereby significantly reduces its
reliability compared to the Bosch CDI, as the reliability of any system is reduced
as its component number is increased.

Not so in this case when comparing a design going back >30 years rather than the more modern, and SIGNIFICANTLY more reliable, components used in the MSD. My educated guess would be that the computed MTBF of the MSD will be at least 3 times that of the factory CDI.

Additionally, the MSD CDI requires additional
wiring for power because of the its high current requirement to produce the multiple
sparks, whereas the Bosch CDI has a simple power and ground.

Simply NOT true....the MSD's current draw is no higher than the CDI's at full engine RPM.

Since most all the torque is developed at RPMs greater than 3000 RPMs and the MSD
CDI produces just one spark there, it operates in a mode the same as the Bosch CDI
and thus has no real advantage in performance, even if the multiple spark had a
benefit. So given the insignificant benefit of the multiple spark effect with the
additional complexity and number of components, and a shorter initial spark, a less
desirable ignition system results with the MSD CDI than with the Bosch CDI ignition.

Included are images of the oscilloscope trace of the 9 sparks at 100 RPMs, the inside
of the MSD CDI showing it two ICs (PIC processor, & driver IC), 7 power semiconductors,
and the extensive number of other components, and the very simple and reliable Bosch CDI.

Given the proven, WELL proven, propensity of the factory CDI's SCR to "self-fire" with the slightest over-voltage applied, and a design that permits that over-voltage, I don't see how any modern CDI design would out perform the factory CDI overall. How many of us have had to send the OEM CDI away for repair simply because that SCR's "breakover" rating isn't sufficient for the application? It doesn't take many "self-firing" episodes before the SCR overheats and fails permanently.

Mine now has a Shunt voltage limiter at the input to SINK input voltages above 14 volts.

[wwest]

Originally Posted by Lorenfb

A technical evaluation of the MSD capacitive discharge ignition (CDI) system
indicates a number of shortcomings compared to the original Bosch CDI used
by Porsche in the 911 up until the 911 3.2 with the inductive discharge ignition
system.

The MSD CDI's key marketing feature is its multiple sparks which produce
a maximum of nine sparks at about 100 RPMs while cranking to a single spark once
the RPMs reach 3000 RPMs. The time between sparks is about 1.5 milliseconds
which contributes basically nothing to enhance the combustion process as the
significant part of the fuel charge burn time is less than 1.5 milliseconds.
Additionally, the spark pulse width of 40 microseconds is half of the Bosch CDI
which results in less time to fully facilitate complete fuel charge ignition resulting in
potential mis-fires and greater emissions.

The MSD CDI produces a primary pulse voltage over 450 volts compared to the
Bosch CDI's pulse of about 300-350 at typical engine RPMs, thereby stressing the
ignition system wiring, the coil, and the rotor & cap. This further shortens the life of
these components beyond what the multiple spark effect does, i.e. at idle the MSD
is producing four sparks, versus one for the Bosch CDI, resulting in significantly
shortened ignition component life.

To implement the multiple spark system, the MSD CDI requires a significant number
of components compared to the Bosch CDI and thereby significantly reduces its
reliability compared to the Bosch CDI, as the reliability of any system is reduced
as its component number is increased. Additionally, the MSD CDI requires additional
wiring for power because of the its high current requirement to produce the multiple
sparks, whereas the Bosch CDI has a simple power and ground.

Since most all the torque is developed at RPMs greater than 3000 RPMs and the MSD
CDI produces just one spark there, it operates in a mode the same as the Bosch CDI
and thus has no real advantage in performance, even if the multiple spark had a
benefit. So given the insignificant benefit of the multiple spark effect with the
additional complexity and number of components, and a shorter initial spark, a less
desirable ignition system results with the MSD CDI than with the Bosch CDI ignition.

Included are images of the oscilloscope trace of the 9 sparks at 100 RPMs, the inside
of the MSD CDI showing it two ICs (PIC processor, & driver IC), 7 power semiconductors,
4 electrolytic capacitors (bad choice for high temps in an engine compartment), and the
extensive number of other components, and the very simple and reliable Bosch CDI.

Well...Wow, you convinced me.

The next time my OEM CDI fails, as unlikely as that might be, instead of sending it to you for repair again I'll simply replace it with the more more design, and thus more reliable, MSD.

[T77911S]

Originally Posted by Lorenfb

The time between sparks is about 1.5 milliseconds
which contributes basically nothing

i dont know what your oscope is set to, but based on what your claim, "40us pw and 1.5 ms spacing of pulses or time between sparks", that is not what you are showing.
if the last pulse, or spark, on the scope is about 40us wide, your oscope setting for time must be .2ms per devision.(?) the distance from the leading edge of the last pulse to the pulse before is one full devision, which is .2 ms or 200us, not 1.5ms as you said. the entire multi spark duration is 1.5ms not the time between pulses as you said.

[mb911]

I have btdt and would never suggest changing out a good working bosch unit for an MSD. though I would consider replacing a junk bosch unit simply for rev limiter option. I had 2 msd boxes on my last car and had repeated failures with one of them and had the wiring done correctly I did get 2 6AL's under warranty as They were faulty from the factory and the car would just shut off going down the road like you shut the key off. both boxes were mounted on an aluminum plate under the seat so heat was not an issue from the engine.

[Brando]

What MSD unit was used for this comparison?

Just asking because there are differences between manufactures..

[RWebb]

The one detriment you identify with the MSD box is a "spark pulse width of 40 us" which is half of the Bosch CDI" spark pulse width.

You then give your expectations that this will cause incomplete ignition of the fuel charge, thereby resulting in "potential mis-fires and greater emissions."

Do you expect this to be worse for larger cylinders?

Do you expect this to be worse at idle, near idle, mid-range rpms or high rpms?

[winders]

Originally Posted by T77911S

one other thing, their are way to many race cars, non-porsche too, that run MSD for it not to be a good unit.

Way too many.............unless race car builders are just plain stupid.

Scott

[brads911sc]

I am no expert like you Loren. But I can buy 9 MSD's at $225 for the price of 1 Bosch at $2,002 and thats for a rebuilt one (Thats the Pelican Parts Price). I have no issues with my MSD. Use their blaster coil, their coil wire, Magnacore wires, and have it located on the firewall away from vibrations. Dont think id go back to the Bosch ever based on cost alone. cant claim to notice a difference in the performance. But the cost has to be factored into the equation somewhere.

[Bob Ashlock]

This might be a little like debating politics or religion, but I think there is very little performance difference to be gained going from the stock CDI to the Multi-Spark CDI. I do think the multiple sparks might help under certain low-rpm conditions. You can imagine a not-so-well-tuned car having some mis-firing at low-speed loads. Having a multiple spark situation could help make sure the charge gets lit and thereby smooth out an otherwise rough engine. My first Porsche (many moons ago) was a '66 911 with standard Kettering (points/condenser/coil) ignition and Solex carbs. New plugs didn't last that long, and within a few thousand miles or so, the car became harder to start and missed under load. In that era, we didn't even have MSD available, so I built a simple CDI unit (based on something I'd seen in Popular Mechanics I think) ...very similar to the Porsche 3-pin unit. THAT thing made all the difference in the world as far as easier starting, low-speed driveability and long plug life. My point is that CDI is a massive improvement over Kettering, but the difference between stock CDI and MSD multi-spark is small at best. I have tried a multi-discharge unit on my current 911 2.2T engine and I can't feel any difference between it and a stock 3-pin Porsche CDI. Admittedly, that may be the case because my old T engine is pretty docile compared to a higher power engine.

[brads911sc]

One other factor that should be in the equation...

I can have one go bad, run to most decent auto parts stores or speed shops and be back on the road in an hour.

I do think that arguments about the technology are valid. However, even if the technology is equal at best, these other factors make it a viable choice for those not needing originality.

[Lorenfb]

"But I can buy 9 MSD's at $225 for the price of 1 Bosch at $2,002 and thats for a rebuilt one (Thats the Pelican Parts Price)."

Wrong! You need to call Pelican, i.e. You know there still exists a phone service
to clarify potential pricing issues.

"I do think the multiple sparks might help under certain low-rpm conditions. You can imagine a not-so-well-tuned car having some mis-firing at low-speed loads."

You think? As stated, the ignition process has basically completed before the next
spark, i.e. 1.5 milliseconds after the 1st.

"i dont know what your oscope is set to, but based on what your claim, "40us pw and 1.5 ms spacing of pulses or time between sparks", that is not what you are showing."

The scope was set to show 9 sparks about 1.5 ms apart at 100 RPMs. Obviously you
can't show a 40 us pulse AND show 9 X 1.5 ms on the same trace clearly. If one
looks closely one can see the negative pulses of the spark signal. The positive portion
of the pulse is the MSD re-charge time of the capacitor and NOT the actual spark pulse
which is negative. The positive portion is only about 30 volts.

"these other factors make it a viable choice for those not needing originality."

Like what factors?

"You then give your expectations that this will cause incomplete ignition of the fuel charge, thereby resulting in "potential mis-fires and greater emissions.""

That's why all OEM vehicles now use an inductive discharge ignition, e.g. coil on plug, because
of the longer burn time of the spark vs the very short CDI spark time.

"Way too many.............unless race car builders are just plain stupid."

Bosch doesn't sell after market ignition systems and they wouldn't give-away systems to gain
free sponsorship and many NASCAR sponsors do. Besides, when was the last time you saw
a MSD on a Porsche Cup Car?

"How many of us have had to send the OEM CDI away for repair simply because that SCR's "breakover" rating isn't sufficient for the application?"

Few if any, as the OEM SCR is one of the least failing components in the Bosch unit.
It's only because of the circuit protection components' failures that cause excessive voltage.
And it's only when someone mis-wires a CDI, same for a MSD, does the SCR fail.