Ignition testing

[bradrs]

Yes the hotwire can be seen as a 'band aid' rather than a full on fix for the lost voltage, but then why would you want to tear a harness apart and spend hours/days tracing something that may or may not be found when the hot wire costs 10 bucks, takes an hour tops and WORKS without a doubt.

Basically by adding another 1-2V, you are probably adding another 10-30% spark energy, and a bit more voltage to start the spark. It sounds like that has been enough for most people here. No add on boxes either. Seems like a no brainer to try this first. That, and the slight reprogram of the ECU both are good, simple fixes.

The DLI simply increases "dwell", the time the coil primary side ground is closed and the associated field on the secondary can build. This greater time for current to flow through the primary side of the coils results in greater voltage and energy available on the secondary side to fire the plug under marginal conditions which need the greater energy to fire.

The DLI is supposed to work not by increasing dwell, but by laying a CDI style spark on top of the standard inductive spark that is being generated by the stock ignition.

The inductive spark builds up the current(dwell time) in the coil primary, and that is where the energy is stored. By shutting off dwell, the magnetic field in the coil primary collapses, and that makes the secondary voltage for the spark.

The CDI stores the energy in a large capacitor in the module. When it fires, it dumps all the energy at once into the coil primary which immediately creates the voltage in the secondary.

The DLI is supposed to work by sensing the stock signal, and firing its CDI style spark at the same time, laying it on top of the inductive spark.

How this increase in dwell is made to happen is just one of the great mysteries of electrical/electronic magic. If this were a 1969 Road Runner with a distributor, you increase dwell by adjusting the points to stay closed longer. The DLI? only God and his specially selected EE wizards know. And even if they are telling I'm not understanding. Not one of the elect.

On your Roadrunner, instead of just trying to increase dwell, you'd be better off by using something that regulates the dwell time to get a fixed current. When you increase dwell angle with points to get more current at high RPM, now you have a really long dwell at low RPM. A module that adjusts the dwell time to give a consistent dwell current is going to give a better result. And it will allow you to use a low resistance coil that charges even quicker, for better high RPM performance. Of course it isn't OE, so there is that downside.

[anyonebutme]

They make like 18V batteries with a 12V tap. Run the 18V tap to the coils and starter and everything else 12v. Don't know how long the coils will like running on 18 volts, but hey, as Clarkson would say: POWWWWWWAAAAAAAAAAAAAAA.

[Unlogic]

Basically by adding another 1-2V, you are probably adding another 10-30% spark energy, and a bit more voltage to start the spark. It sounds like that has been enough for most people here. No add on boxes either. Seems like a no brainer to try this first. That, and the slight reprogram of the ECU both are good, simple fixes.

I have to agree. Doing the simple coil hotwire and increasing the dwell cured all my spark problems at almost no cost. I'm running 8+ year old spark plugs and wires at upwards of 18 psi without any problems.

I'm sure the DLI works and is a great device but before one shells out with the money for one try the simple stuff first.

[Jimvr4]

Basically by adding another 1-2V, you are probably adding another 10-30% spark energy, and a bit more voltage to start the spark. It sounds like that has been enough for most people here. No add on boxes either. Seems like a no brainer to try this first. That, and the slight reprogram of the ECU both are good, simple fixes.

It's definitely more than 10-30%. At idle it's possible to increase the current from 5A to nearly 8A, while at 7500 rpm nearly 5A can be maintained where the stock system drops to 3.6A. That's nearly 40% more current at 7500 rpm.

The DLI is supposed to work not by increasing dwell, but by laying a CDI style spark on top of the standard inductive spark that is being generated by the stock ignition.

The inductive spark builds up the current(dwell time) in the coil primary, and that is where the energy is stored. By shutting off dwell, the magnetic field in the coil primary collapses, and that makes the secondary voltage for the spark.

The CDI stores the energy in a large capacitor in the module. When it fires, it dumps all the energy at once into the coil primary which immediately creates the voltage in the secondary.

The DLI is supposed to work by sensing the stock signal, and firing its CDI style spark at the same time, laying it on top of the inductive spark.

This is also what I thought but deeper reading seems to disagree. The way the DLI hooks up across the coils means that it can actually provide an independent transistor control from what the ECU is doing with the PTU. Think of it as two separate switches that are controlling the coils and imagine that before the ECU turns off the PTU that the DLI transistor in parallel is also shorting the coil to ground. Then when the ECU turns off the PTU the dwell continues because the DLI is controlling it now instead of the ECU. The DLI uses the increased dwell time to store charge in the capacitor and then discharge it to get the ~400V flyback voltage in the primary side.

[Jimvr4]

Since this thread is about testing let's look at some more data.

15V powered ignition with 90% dwell

For this test here is the test setup. Note the voltage is set for 15.1. I'm using a home built adjustable power converter and I'm able to dial in any voltage up to around 16.


OK now I've rewired the primary side input to 15V. This represents about a 2.5V increase from the stock level at idle and possibly more at load if the battery voltage starts to sag.
This particular circuit maintains the voltage level across rpm. Will have to try and load test it as well to see if it holds up.

The particular circuit I'm using has some noise but you can see at idle there is a very large increase in charging current to about 8A now!


At 4347 rpm there is still nearly 7A of charging current


Finally at 7500 rpm we can stll see a strong 5A of charging current


So, what did we accomplish? The primary side currents have gone up at least 1A from below 4A to above 5A. This is at least a 25% increase in charging current at max rpm
and probably higher as the previous tests did not reach 7500 rpm. The percentage increase in the primary current is even higher at idle and lower rpms where the currents were
as high as 8 Amps at idle. The percentage increases there are 60%.

Further, this is proof that there is no "limit" in the current imposed by the PTU per say. Yes, the average power across the PTU has increased with the higher currents but there
is certainly no evidence that the PTU ever comes out of saturation. In fact it never will. What will happen with increased primary voltage is the currents will continue to rise to 9, 10,
even 15 amps possible and be limited only by the ignition coil core saturating.

Yes, the ignition coils would eventually saturate and the charging rate of them would flatten out. The fact is we don't have enough primary side voltage to find that operating point
of the coils yet. Maybe it's above 18-20 volts - who knows?

My circuit has a possible advantage over the simple relay as it should maintain the voltage level across load. This won't be the case with the relay since the alternator isn't able to keep
the system voltage from sagging under load. Those boxes Jester found should also be able to maintain the level at load and would be great to test!

I doubt the DLI can provide additional benefit over powering the coils through one of those boxes.

[tom93vr4]

Interesting measurements Jim. Since you seem to have everything set up, can you measure the PTU hi-side drive voltage to the coil primary with your 15V setup? I ask because my measurements suggest the PTU is where a current limit function is - however I never made any measurements with higher supply voltage.

[Jimvr4]

Hi Tom, welcome to my thread and thanks for all that you did in testing this way back

I can set up that measurement as well but it shouldn't be any different than what I presented here. The supply current has nowhere to go other than through the PTU to ground.

[tom93vr4]

That is correct. But my measurements showed PTU hi-side voltage rising to about 4V at the end of dwell (stock supply voltage). A typical NPN power transistor driver with an emitter series resistor would behave this way and thus produce an effective current limit given that the ECU drive voltage (to the PTU) is probably 5V (didn't actually check this).

Kind of interesting to see this thread here now - I was just looking at my ignition drive stuff a few weeks ago due to a different problem.

[bradrs]

This is also what I thought but deeper reading seems to disagree. The way the DLI hooks up across the coils means that it can actually provide an independent transistor control from what the ECU is doing with the PTU. Think of it as two separate switches that are controlling the coils and imagine that before the ECU turns off the PTU that the DLI transistor in parallel is also shorting the coil to ground. Then when the ECU turns off the PTU the dwell continues because the DLI is controlling it now instead of the ECU. The DLI uses the increased dwell time to store charge in the capacitor and then discharge it to get the ~400V flyback voltage in the primary side.

That would mean the DLI retards ignition timing when you install it. The ECU doesn't know to compensate when you install the DLI, so it would still be ending the dwell when it wants the coils to fire. If the DLI lengthened the dwell past that, it would retard the timing.

And you are thinking it uses that short additional dwell period to charge the capacitor? A standard CDI will take (very roughly) 10 amps for a millisecond to charge the capacitor. At just 3000 rpm, that is 20 degrees of ignition timing. If the DLI retarded ignition timing by 20 degrees, people would complain. Very loudly.

I haven't disassembled the DLI, but I do know similar products. And they actually do have a circuit exactly as I described. A seperate 12V DC-DC converter that charges the capacitor, which is triggered on the firing of the stock inductive ignition. You wouldn't want to try to charge the capacitor by a lengthened dwell, the time it takes to charge the capacitor to 400V isn't something you could spare.

[bradrs]

That is correct. But my measurements showed PTU hi-side voltage rising to about 4V at the end of dwell (stock supply voltage). A typical NPN power transistor driver with an emitter series resistor would behave this way and thus produce an effective current limit given that the ECU drive voltage (to the PTU) is probably 5V (didn't actually check this).

That is the exact same sort of thing I was seeing on the PTU coil - signal. It would climb, and climb quickly once you hit current limits. It could be a series resistor as you describe, but I am thinking it is more along the lines of the gain of the transistor. Here is a datasheet of a darlington transistor that is used for ignitions. This is probably not exactly the same as was used, but is typical for something from that time frame:

http://www.jetav8r.com/Vision/TCIReb...E5740Specs.pdf

Looking at the specs, you see the gain drop WAY off as it approaches 8A, by a factor of 10 or more. Which means you'd need to drive it with that much stronger of a signal to keep it saturated.

And by reading the voltage where we were reading it, that makes it clear it is NOT the same as saturated coils.

Jim, it looks like you are at a 10mV/A scaling? If so, if you average out the noise, it looks like about a 7A peak at idle. That looks promising. Have you looked at the coil - signal? Make sure the scope is able to handle those 400+V flyback voltages, I've seen too many people not carefully check out the scope parameters and blow a channel doing that.