restart Mk2 Pi ignition fail

[RobPearce]

As usual, the instructions are too simplistic and only cover the most minimal case. Your last drawing is still the correct one. Accuspark have fitting instructions online with a ballast diagram at the bottom, but that's still not right because it doesn't account for your car's current sensed tacho.

[frenchiemk2]

Rob, is it possible that i might connect tis white/grey before ballast to red E.I?

Seems like on fig 3 on the link you gave

[RobPearce]

No, you shouldn't, because it's not directly connected to the ignition switch. If you look closely at the WSM diagram, and note my comment about not getting misled by the way it's drawn, the rev counter sits between. I've updated your drawing from earlier:

Connecting the white/grey to the EI red might work for testing, but so does direct to battery.

[frenchiemk2]

OK clear, 

I sent this diagram to the seller's shop where I bought the new E.I kit (they're pro electric car detailers and builders).

 

[frenchiemk2]

Got an answer today : remove the ballast, check positive feeding from neiman to coil + [wires from E.I] red wire to positive coil and black to negative coil

 

 

Edited January 23, 2023 by frenchiemk2

[RobPearce]

1 hour ago, frenchiemk2 said:

Got an answer today : remove the ballast

Err... what?

Sorry, the people who answered your query do not know what they're talking about. That second diagram - the "Accuspark recommendation", if I understand you right - will cause your rev counter not to work.

Stick with the drawing on the left, which is the same as my earlier one.

[Nick Jones]

4 hours ago, RobPearce said:

, if I understand you right - will cause your rev counter not to work.

 

Why? All it’s doing is connecting the white/grey directly to coil+ve, which will bypass the ballast resistor but the rev counter will still work as the coil positive feed still passes through it. Possibly not while cranking though due to the temporary additional supply, unless the white/yellow from the solenoid is removed. 
 

12v coil is needed if wired like this though.  I’d say it has merit even if just for initial test purposes.

Nick

Edited January 23, 2023 by Nick Jones

[RobPearce]

26 minutes ago, Nick Jones said:

Why? All it’s doing is connecting the white/grey directly to coil+ve,

Actually it's also adding a direct wire from the white circuit to the coil +ve (the red feed to the EI) which not only bypasses the ballast but also the tacho. I don't know whether That's a drawing that Frenchie did from their wording, or a drawing they sent back to him, but either way it's wrong. The one on the left, with the ballast still in, is right.

[Howard]

Rob,

I am following with interest but am somewhat confused. The circuit on my car is shown below (which is non standard and includes interface relays !)

When the ignition switch is in the run position, 12v is supplied to the electronic ignition, but the coil is fed via the ballast resistor and only sees about 6v.

In the start position, the coil has 12v applied to give a better spark

The tacho connection is fed from the wire between the coil & the electronic ignition module. The ignition module breaks the current causing a collapse of the magnetic field in the coil and a high voltage pulse to the spark plugs + a low voltage pulse to the tacho which is different to the sketches I have seen.

Am I missing something?

Cheers

 

 

Edited January 23, 2023 by Howard

[thescrapman]

Howard, your rev counter is going to be capacitive rather than inductive I think.

[RobPearce]

9 hours ago, Howard said:

Am I missing something?

As thescrapman said, you have the later (and far more common) "voltage sensing" rev counter. The very early Mk2 PI had a "current sensing" rev counter, which has to be wired in series with the supply to the coil. Those were common at one point but rapidly dropped out of favour as the later type came in. So much so that many people, Accuspark included, it seems, have never heard of them.

Incidentally:

9 hours ago, Howard said:

+ a low voltage pulse to the tacho

it's not as low voltage as you may think. If you have a pure booster EI (the type that retains the points but reduces the current in them) then you can't move the tacho to before that, because it would then only see 12V and that's not enough for it to reliably work. They are designed to sense the spike you see on the LT side of the coil, which is a few hundred volts.

[Howard]

Thanks for explanation, but would query the size of the spike on the LV side of the coil. Without a 'scope I have no way of checking, but when the Piranha units were made power transistors capable withstanding a few hundred volts were few and far between.  I would not expect to see a peak of more than a hundred volts.

[frenchiemk2]

"put an electronical ignition, it will make the job easier" they said... 🤣

Thanks guys.

As soon as i get my HP fuel pump, coil wires and spark testers, i"ll try this as i don't know if my actual E.I is not burnt or else.

Edited January 24, 2023 by frenchiemk2

[Howard]

When I was tracing a fault in my electronic ignition, all I needed was the ignition module wired up and to be able to turn the engine over using the starter. Simply unplug the HT cable from the coil to the distributor at the distributor end,  and insert a spare plug into the end of the cable. With the plug body touching the block, turn the engine over using the ignition key and if the ignition is working you will see a bright spark.

Its seriously bad idea to touch the plug while doing this as you could have a very nasty electrical shock!

 

[frenchiemk2]

39 minutes ago, Howard said:

Its seriously bad idea to touch the plug while doing this as you could have a very nasty electrical shock!

 

Yes, nobody would do this...

[RobPearce]

1 hour ago, Howard said:

when the Piranha units were made power transistors capable withstanding a few hundred volts were few and far between.

When I built a Maplin EI (booster) kit back in the early 1990s, the power transistor in that was rated for 400V, and was protected by a limiter. That's the type of device I'd expect to see. The ones fitted in the Rover MEMS units of the time were better.

Standard reckoning within the industry is that 300V is typical, but HT faults can increase it significantly. It will depend on the coil, and actually may be a little lower on a ballasted system.

[Howard]

Just a few years earlier in '76 my final year project at uni. was the construction of a three phase variable frequency inverter for driving a.c. motors using transistors. The biggest design problem was the power transistors. At the time they suffered from Second Order Breakdown (SOAR?) when switched rapidly, forcing my prototype to be tested at 240v, despite using transistors with nominal rating of significantly greater than 415v.

I lost track of power transistors for many years after that, but would not have expected that generation to have survived for long with voltages of more than a couple of hundred volts being applied.

[RobPearce]

1 hour ago, Howard said:

would not have expected that generation to have survived for long with voltages of more than a couple of hundred volts being applied.

That's possibly why early EI units had short lifespans 😁

A typical ignition coil has a turns ratio of 100:1 and puts out 30kV HT. While the usual "transformer equations" are inapplicable to ignition coils, that one does apply in the specific case. The magnetic energy built up during the dwell phase is released to both coils, and the voltages track to the turns ratio. If the spark path is good, it may spark (and hence conduct the energy release away) before the HT hits 20kV, in which case the LT side won't reach 200V. If, however, the king lead has been removed, it can go way past that. Ballasted coils may be on the high side for turns ratio and probably don't give as much HT either.

[frenchiemk2]

Great news today and it's all thanks to you mates and this very nice forum ; 1st time in my life the car starts and i listen to her.

Guys you helped me so well. Thanks. And back to the main thread.

 

[frenchiemk2]

For later users, I share my diagram.