Modify Velleman Kit - Transistor Assisited Ignition, kit no. K2543

I have done a quick analysis and, by using the Flame Thrower, 40KV (1.5 Ohm) coil , only the following modification to the Wellerman ignition switch would be required:
(1) Remove both 150V Zener diodes
(2) Replace output power Darlington BJT with a 600V (min), 22A (min) power MOSFET. (I have found many suitable NMOSFETs: cost ~ $6US)
(3) Possibly make a few modifications to the circuit between the points and the MOSFET (this is not a big issue)

As far as I can tell that would be it.

spec

The MOSFET would need to be a 'logic level' type to ensure that it is switched on fully with a gate voltage of only 6V or less. That narrows the choice.
What voltage does the magneto put out at very low idle speed and kick-start speed?

alec_t said:

The MOSFET would need to be a 'logic level' type to ensure that it is switched on fully with a gate voltage of only 6V or less. That narrows the choice.
What voltage does the magneto put out at very low idle speed and kick-start speed?

Click to expand...

Hello Alec,

I think the bike has a 6V battery

Yes, that is true: you have to provide sufficient gate drive voltage.

In addition to Vgth, a MOSFET would have to have many parameters matched to the application.

One of the biggies in DV/DT and DI/DT in an application like this

The insulator between the heatsink would also need to be checked that it would handle 500V without ark-over.

The SOAR is another worry with such a high voltage and current.

And there is the basic maximum junction temperate and the implications of the thermal resistance, Tjunc/case with potentially 30W dissipation in the coil.

All good fun

spec

POST ISSUE 2 of 2017_01_15 (to include CDI option (4))

Hi again

As far as I can determine, there are three main approaches to using the Flame Thrower II, 45KV (0.6R Ohm) coil with the Wellerman ignition switch:
(1) Ballast resistor in series with the coil
(2) Fit a constant current generator in series with the coil (small signal BJT and power PMOSFET on a substantial heat sink)
(3) Use constant dwell (micro controller probably the easiest)
(4) Use capacitor discharge (CD)

To give you a feel for what is involved, consider option (1) above
(1) Remove both 150V Zener diodes
(2) Replace output power Darlington BJT with a 600V (min), 22A (min) power MOSFET. (I have found many suitable NMOSFETs: cost ~ $6US)
(3) Possibly make a few modifications to the circuit between the points and the MOSFET (this is not a big issue)
(4) Fit a 0.47 Ohm 50W power resistor in series with the coil

spec

Spec,
I'm inclined to go with the Flame Thrower (FT) 1.5 ohms coil, as that is the easiest to fit into the existing circuit and the Velleman kit. Particularly since everyone on the board is "across the water," making it hard for me to run to them when smoke starts pouring out of the circuit.
However, would the FT 1.5 ohm coil even be an upgrade at all? Recall that this coil is 1.5 to 9k ohms, with 1:100 turn ratio. The original coil is 2.0 to 6.8k ohms, unk on the turn ratio.
See the FT canister coil specs here:
https://www.pertronix.com/catalogs/pdf/ptx/2015/ptx2015_coils.pdf

alec_t said:

The MOSFET would need to be a 'logic level' type to ensure that it is switched on fully with a gate voltage of only 6V or less. That narrows the choice.
What voltage does the magneto put out at very low idle speed and kick-start speed?

Click to expand...

Alec,
This 1969 Italian bike has an odd ignition. A transitional style from magneto to battery+ coil. You kick it to turn over the engine, but the spark is still provided by the small 6 volt 8 amp hr battery + coil.

Additionally, it has a Bosch mechanical voltage regulator with an usual feature. If the battery's dead you can flip a switch, changing over to magneto ignition and push starting the bike.

Ok. So we can assume there is a minimum of 6V to drive the MOSFET gate. That is still in 'logic-level' territory.

alec_t said:

Ok. So we can assume there is a minimum of 6V to drive the MOSFET gate. That is still in 'logic-level' territory.

Click to expand...

Hi Alec,

Don't worry too much about the gate drive voltage.

The gate threshold for the NMOSFETs with 500 VDS, and over, is 4V to 6V, but the gates will need to be hit with around 10V ideally, at 7A drain current, to ensure a low RDss.

The primary voltage with the Flame Thrower (1.5 Ohms) will be 400V max and with the Flame Thrower II (0.6 Ohms) the primary voltage will be 450V max.

This means, with voltage catching, the Flame Thrower (1.5 Ohms) would need a 450V switching device, and the Flame Thrower II (0.6 Ohms) would require a 500V switching device.

Without voltage catching, a 600V device would probably be needed.

The only devices I can find that will handle that sort of voltage, provide enough current gain, limit the loss, conduct 7A, are MOSFETS- they are also cheaper because MOSFETS are so widely used.

spec

sign216 said:

I'm inclined to go with the Flame Thrower (FT) 1.5 ohms coil, as that is the easiest to fit into the existing circuit and the Velleman kit.

Click to expand...

I now find that it would make little difference to the complexity of the circuit which coil you use. So the latest feeling is go for the Flame Thrower 2. That is what you wanted in the first place and it will give you another 5KV, not that it will make much difference, I shouldn't think.

sign216 said:

However, would the FT 1.5 ohm coil even be an upgrade at all? Recall that this coil is 1.5 to 9k ohms, with 1:100 turn ratio. The original coil is 2.0 to 6.8k ohms, unk on the turn ratio.

Click to expand...

I'm pretty sure that a FT (1.5 Ohm) would make a vast difference, like 17KV to 40KV at a guess. And at high engine RPM the difference would probably be greater.

The ratio of the primary resistance to the secondary resistance is not really significant when comparing coils.

sign216 said:

the spark is still provided by the small 6 volt 8 amp hr battery

Click to expand...

Is there any scope to fit a battery with more AH?

spec

spec said:

I now find that it would make little difference to the complexity of the circuit which coil you use. So the latest feeling is go for the Flame Thrower 2. That is what you wanted in the first place and it will give you another 5KV, not that it will make much difference, I shouldn't think.

Click to expand...

Then yes, if the circuits are similar let's use the best coil for the application. More spark is more better!

spec said:

Is there any scope to fit a battery with more AH?

spec

Click to expand...

There's quite a bit of room for a bigger battery. If that makes it easier for other variables then I can put in whatever you need.
But I thought the amps were the problem? That they might overheat the coil.
Educate me, please.

Corrected 2017_01_15

sign216 said:

Then yes, if the circuits are similar let's use the best coil for the application. More spark is more better!

Click to expand...

Fine, but if I discover, as the design evolves, that the 40KV coil is more suitable I will let you know.

sign216 said:

There's quite a bit of room for a bigger battery. If that makes it easier for other variables then I can put in whatever you need.

Click to expand...

Excellent

sign216 said:

But I thought the amps were the problem? That they might overheat the coil.

Click to expand...

What you say is sort of true but has no relevance to the design of the ignition system.

The reasoning is this.

There are two extreme situations with a vehicle:

(1) Engine not running and thus battery not charging: battery voltage 6.25V . In this situation the coil current will be lowest

(2) Engine running and battery on full charge: battery voltage 7.5V. In this situation the coil current will be maximum (in fact this will be the normal situation).

You may be thinking that a smaller battery would produce less current, but this is not true: the battery will supply whatever current it is asked to supply, but a bigger battery will supply the current for longer before it gets flat.

From this you can see that whatever battery you fit the current through the coil will be the same.

Of course, like all basic explanations, this is a gross simplification, but the fundamental principle is correct and, hopefully, gives you an idea about how a vehicle electrical supply works.

So now you are asking, well if the capacity of the battery makes no difference why the hell fit a bigger battery.

Here is the reason: When the engine is just ticking over there is effectively no battery charge and the battery is supplying the current that your bike demands, mainly the coil current but, at night, also the lights. Bearing in mind that the high performance coil will be taking around twice the current, you will need a battery of at least twice the capacity to maintain the length of time the bike will run off the battery.

A bigger battery has other advantages, but I won't load you with too much information in one go.

spec

PS: you are not considering a change to a 12V system in the future?

I'll keep out of the electronics bit, however looking at the flamethrower 2 coil you linked us too you can see there is a dwell angle chart, at idle the coil is rated for around 15 degrees, if you exceed this the coil will overheat esp at idle.

Theres a project on one of my pc's that is for a electronic ignition for a trials bike, it doesnt have a battery, it runs from the bikes generator, during crank or kickstart the system is effectively out of circuit, turning into programmed Tci once running, if I find it I'll post it, I seem to remember its bipolar not fet.

spec said:

PS: you are not considering a change to a 12V system in the future?

Click to expand...

Spec,
I don't want to change to 12v because the Bosch mechanical 6v regulator has a nice feature, it allows for an Emergency Battery Bypass Switch. If the battery's dead you flip the switch, changing from battery+coil ignition over to magneto ignition and then push start the bike.

Actually, can you tell me if that feature is unique to the mechanical regulator, or if I can keep that feature and upgrade to 12v electrics?
Attached are wire diagrams of the bike and the regulator.

dr pepper said:

I'll keep out of the electronics bit, however looking at the flamethrower 2 coil you linked us too you can see there is a dwell angle chart, at idle the coil is rated for around 15 degrees, if you exceed this the coil will overheat esp at idle.

Theres a project on one of my pc's that is for a electronic ignition for a trials bike, it doesnt have a battery, it runs from the bikes generator, during crank or kickstart the system is effectively out of circuit, turning into programmed Tci once running, if I find it I'll post it, I seem to remember its bipolar not fet.

Click to expand...

Dr Pepper,
I'll work on finding out the cam's dwell angle. If it exceeds 15 deg does this mean the high performance coil is a no-go?
I'm also interested in the elec. ig circuit. Keep me posted please.

dr pepper said:

...looking at the flamethrower 2 coil you linked us too you can see there is a dwell angle chart, at idle the coil is rated for around 15 degrees, if you exceed this the coil will overheat esp at idle.

Click to expand...

Yes, the main concern is coil overheating without an adaptive dwell (ie constant dwell time with RPM) and a constant voltage.

Apart from their higher turns ratio and presumably better magnetics, these high performance coils have a very low primary resistance and inductance which means that the primary current ramps up rapidly which means that you get a full energy packet for each plug fire even at high revs with a nulticylinder engine, say 8 or 12 cylinders.

If you connected the FT2 primary coil (0.6V) across a 7.5V DC supply the current would quickly build to 7.5V/0.6V = 12.5A. Thus the power dissipation93.75 in the coil would be 7.5V * 12.5A = 93.75W which seems lot but nowhere on the data sheet does it say what the maximum power rating of the FT2 is.

The power rating of a traditional Kettering ignition system coil is around 12.6V * 4A = 50.4W. This would be at low revs or when the ignition were turned on but the engine were not running.

I would thus be surprised if the FT2 would not stand a similar power dissipation. But for the time being I have assumed a maximum dissipation in the primary of the FT2 of 30W which relates to a maximum current of 7A through the FT2 primary ( 7 * 7 * 0.6= 29.4W). This would seem to be perfectly safe.


And that is the theme of the above posts: the current through the FT2 primary coil can be limited to 7A or less by one of the three design approaches listed in post #24.

spec

Hi sign,

sign216 said:

Spec,
I don't want to change to 12v because the Bosch mechanical 6v regulator has a nice feature, it allows for an Emergency Battery Bypass Switch. If the battery's dead you flip the switch, changing from battery+coil ignition over to magneto ignition and then push start the bike.

Actually, can you tell me if that feature is unique to the mechanical regulator, or if I can keep that feature and upgrade to 12v electrics?
Attached are wire diagrams of the bike and the regulator.

Click to expand...

I was not suggesting that you do change to 12V electrics and I only asked in case that is your intention sometime in the future because if you were considering going to 12V it would impact the choice of coil.

I haven't checked in detail but I would not think that going to 12V would preclude the emergency start method provided you kept the magneto.

spec

sign216 said:

I'll work on finding out the cam's dwell angle.

Click to expand...

The dwell angle will be way over 15 degrees- more like 300 degrees if your bike is has a single cylinder and 240 degrees if it is a twin cylinder.

spec

spec said:

The power rating of a traditional Kettering ignition system coil is around 12.6V * 4A = 50.4W.

Click to expand...

I've been watching this thread and seen the mention of 12.6V on a Kettering type ignition. Don't know how it's done in the UK where your from, but here in the States the only time a Kettering(points and coil) ignition sees the full voltage is when the engine is cranking. Once it starts and the key is in the run position, there is a ballast resistor in the circuit, making the voltage to the coil ~8 volts. Don't know if this will have any bearing on the circuit your making.

shortbus= said:

I've been watching this thread and seen the mention of 12.6V on a Kettering type ignition. Don't know how it's done in the UK where your from, but here in the States the only time a Kettering(points and coil) ignition sees the full voltage is when the engine is cranking. Once it starts and the key is in the run position, there is a ballast resistor in the circuit, making the voltage to the coil ~8 volts. Don't know if this will have any bearing on the circuit your making.

Click to expand...

Hi SB,

Yes, on some of the later Kettering systems in the UK the automobiles are the same. Other, older cars, have no ballast resistor, but the important thing is the dissipation in the coil primary.

In the case you cited the dissipation in the coil would be around 8V * 4A say = 32W which is around what I figured to be a reasonably safe dissipation for a full sized coil. I suspect it is higher though because some coils get pretty hot (apart from engine heating).

The series resistor is the first approach of the three that I listed in post #24.

At this stage of the design development the absolute values are not that important- it is the architecture that counts and the actual values can be adjusted at any time as new information becomes available.

The two bits of informatin that are needed are:

(1) At what primary current does the coil acquire full charge? This is the fundamental parameter affecting the design.
(2) What is the maximum allowable dissipation in the coil.

The Rolls Royce ignition system for this bike, as I indicated earlier, is CDI. All the dwell angle problems go away then and you get a better spark and much lower power consumption.

Contactless points would also be good: timing stays constant as there is no points heel wear and there is much less timing scatter.

spec

Spec,
I'll ask Petronix on your questions. This is a holiday weekend in the US, so it's likely I won't get an answer until late Tues or Wed. Or maybe never; I've had poor luck getting technical answers from many manufacturers.

And....as you said I suspect the dwell on this single cylinder engine is way over 15 degrees.

Hi Sign,

sign216 said:

I'll ask Petronix on your questions.

Click to expand...

Would be a great help if we could get answers to those two questions.

sign216 said:

This is a holiday weekend in the US, so it's likely I won't get an answer until late Tues or Wed. Or maybe never; I've had poor luck getting technical answers from many manufacturers.

Click to expand...

Yes, I know what you mean about no help. Mind you its some of the enquirers that the public make that make manufacturers wary. I have been on both sides of the fence.

sign216 said:

And....as you said I suspect the dwell on this single cylinder engine is way over 15 degrees.

Click to expand...

Yes, you can get a rough idea of the number of degrees of dwell by looking at the contact breaker cam. The dwell is where the points are closed.

spec

Hi all,

There is another consideration about fitting a coil with a high current demand: will the bike generator provide the extra current.

The bikes generator is rated at 60W which means 60W/7.5V = 8A from Sign's post #8. I am suggesting running the FT or FT2 at 7A so, on a good day, that only leaves 1A for lights etc which is not enough.

Afraid this makes all bar constant dwell or CD, a show stopper with the FT coils.

spec