What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Ford edis with megajolt

toothed wheel

OK, gotta ask---------is the toothed wheel a one off item, or something commercially available for purchase?

image%209%201000w_zpsboxcn1g1.jpg


Is there any reason the pickups cant be mounted at the case split line, thus removing all the heavy bar stock used to offset the pickups. Of course if the pickups are mounted at case center, the skip tooth must be offset to compensate.

Considering all the stuff in the prior pages about the drive speed of a six cyl mag/rotor assembly, I suspect your mag hole trigger may not be a viable option for me???

As I read these, your setup is actually running twice the need RPM at the pickup--------and most likely wasting a spark in the exhaust stroke??
 
OK, gotta ask---------is the toothed wheel a one off item, or something commercially available for purchase?

I can send a .dwg file, and you can take to a waterjet shop. Same thing really, and for you, it's free...the drawing, that is.

Is there any reason the pickups cant be mounted at the case split line, thus removing all the heavy bar stock used to offset the pickups. Of course if the pickups are mounted at case center, the skip tooth must be offset to compensate.

Mount 'em anywhere you like, as long as they are 60 degrees (for a 6-cyl) after the skip when the crank is at TDC. Don't discount the need for a stiff bracket.

Considering all the stuff in the prior pages about the drive speed of a six cyl mag/rotor assembly, I suspect your mag hole trigger may not be a viable option for me???

Not viable given a 540's mag drive ratio. You need a trigger wheel turning at crankshaft speed.

No particular need to use the big wheel. Heck, do something like the Electroair crankshaft wheel, in 36-1 teeth rather than 60-1. It clamps on the shaft just behind the prop hub.

Looks like this:

http://www.electroair.net/5C_installation.html

There are a great many VR's available, auto and industrial. I don't think the EDIS unit cares as long as the signal is a nice sharp sine wave.

http://www.phoenixamerica.com/technologies/variable-reluctance/default.html

http://www.spectecsensors.com/html_pages/magnetic_sensors.htm

As I read these, your setup is actually running twice the need RPM at the pickup--------and most likely wasting a spark in the exhaust stroke??

It is a very conventional wasted spark system.
 
Last edited:
A few words about reluctor polarity, for those who want to experiment.

It's not unusual for a new EDIS/Megajolt installation to make no sparks on the initial try at cranking. The first thing to check is reluctor polarity; swapping the two reluctor wires often cures the problem.

Back in the 90's we learned that an ND pickup with reversed polarity would trigger a GM module, but the result was (1) imprecise timing (the timing marks would look real fuzzy when viewed with a strobe), and (2) timing that retarded with increasing RPM

The reluctor choice, tooth shape, and RPM combine to shape the waveform. In general, all the VR-triggered ignitions fire the plug when the output sine wave crosses zero volts, moving positive to negative.

Here's the cranking-speed output from an ND reluctor wired correctly. At low speed the negative-to-positive slope is gentle, while the positive-to-negative is steep. Because the slope is steep, the time range around zero is short, so there is little variation in timing from event to event. The slopes will steepen as RPM rises (more RPM means more cycles in the same time period), a happy property which improves timing precision. (Note to geeks; about the o-scope display, yeah, it was a quick snapshot, and I forgot to center the wave on the scope's zero line. Here the 5-line will just have to substitute for the zero line, ok?)

ND%20Slow%20Waveform%20w%20Notes.jpg


Just to illustrate, next I'll flip the same photo to simulate what the waveform would look like if the reluctor were wired backwards. Now the negative-to-positive is steep, and the positive-to-negative is flattened, thus the time near zero crossing is a much wider range. As a result, spark delivery becomes imprecise at low RPM, and will move as the slope steepens with increasing RPM.

ND%20Slow%20Waveform%20Reversed%20Polarity.jpg


UPDATE: Stumbled across a good explanation (in the Bowling and Grippo Megasquit manual) of why the EDIS often refuses to fire if the pickup polarity is reversed:

Note that you must get the sensor polarity correct for the missing tooth code. This is even more important than with regular triggers, because if you get the signal polarity reversed you will get two (½n+1)*t missing tooth 'gaps', instead of the desired one (n+1)*t gap (and the engine might not run at all).

Go here, near the bottom of the page: http://www.megamanual.com/ms2/wheel.htm

Excellent stuff on reluctors and Hall effect sensors here: http://www.megamanual.com/ms2/pickups.htm

If you're a long time Contact subscriber, you'll find the ND experiments written up by Paul Messinger, circa....ummm, 1996 or 97.

 
Last edited:
Not viable given a 540's mag drive ratio. You need a trigger wheel turning at crankshaft speed.


It is a very conventional wasted spark system.

OK, just a bit of mental gymnastics here.

Does the system really need a trigger wheel turning at crankshaft speed, or does it simply need to see the correct number of pulses per rev??

A 24 tooth wheel turning 1.5 crank speed gives the required pulse count.

As long as the wasted spark is not happening during something important, this should work---------I hope.
 
Last edited:
In the course of this thread there have been several mentions of toothed wheels in 36-1 and 60-1 teeth. With all due respect, I would like to point out that the 60-1 tooth wheels are in fact 60-2 tooth wheels.

Just didn't want somebody making one for themselves and then not have it work.
 
OK, just a bit of mental gymnastics here.

Does the system really need a trigger wheel turning at crankshaft speed, or does it simply need to see the correct number of pulses per rev??

A 48 tooth wheel turning 1.5 crank speed gives the required pulse count.

As long as the wasted spark is not happening during something important, this should work---------I hope.

pulses per rev is a software operation separate from determining the revolution and a revolution is 360 degrees

If the software is set up for crank speed, it needs cranks speed. Some software can be set up to run at cam speed or .5X crankspeed, but then need a crank synch of some sort. In my auto experience, I know of none that operate off a timing signal that is 1.5 X crankspeed though the software can do most anything if designed for it. I would guess that anyone doing a mag-based trigger on a 540 would be doing it in software, though it could also be done via gearing.

It is not the number of teeth that is the issue, it is the missing tooth/teeth. The missing tooth tells the software when you are at TDC or technically at some unique position. As Dan mentioned, the s/w uses an offset value to relate this to TDC. The s/w expects to see it once per revolution. The teeth themselves help to provide a more stable RPM indication (more teeth = more reliable RPM indication) and allow the software to determine the actual crank position at any point in time (s/w can then fire sparks at specific crank angles). The software will also count the teeth to predict TDC if it somehow loses the missing tooth position. The software re-synchronizes the cycle at the missing tooth each revolution. It also compares against the prior revolution to look for anomalies so it can post errors when appropriate.

Larry
 
Last edited:
pulses per rev is a software operation separate from determining the revolution and a revolution is 360 degrees

Yep, understand that.

The MegaJolt is designed to use a 36-1 so that is the formula I was trying to achieve when dealing with the 1.5 drive ratio. Ten degrees of crank rotation per tooth, to make the existing software happy.

By the way, I had a bit of a brain fart earlier, it should be a 24 tooth wheel, not a 48-----------at least that is what my brain now seems to think. And, yes i did not mention the missing tooth, figured that was a given in all this.

And, I suspect there is yet another factor to be considered-------that is tooth size/shape. Too small a tooth, or too narrow of a tooth would most likely not trigger the pickup correctly.
 
By the way, I had a bit of a brain fart earlier, it should be a 24 tooth wheel, not a 48-----------at least that is what my brain now seems to think. And, yes i did not mention the missing tooth, figured that was a given in all this.
Problem is, no matter what you do the index pulse will not occur once per revolution of the crank.
 
Yep, understand that.

The MegaJolt is designed to use a 36-1 so that is the formula I was trying to achieve when dealing with the 1.5 drive ratio. Ten degrees of crank rotation per tooth, to make the existing software happy.

By the way, I had a bit of a brain fart earlier, it should be a 24 tooth wheel, not a 48-----------at least that is what my brain now seems to think. And, yes i did not mention the missing tooth, figured that was a given in all this.

And, I suspect there is yet another factor to be considered-------that is tooth size/shape. Too small a tooth, or too narrow of a tooth would most likely not trigger the pickup correctly.


It has nothing to do with the number of teeth, only the rotation synch, which is achieved with the missing teeth . The software is designed to see a missing tooth once per crank revolution. It is expecting a one to one relationship between the synch event and one full engine revolution. Everything else if based upon this, including the # of degrees between spark events.

If I crank up the voltage to your clock and make it spin twice as fast, the big hand will do a full revolution in 30 minutes. You now need to re-wire your brain to convert that full cycle to mean something different (48 hours in a day). This is the same concept of teaching the EI computer to process differently.

I just don't think you can get there from here with your current idea.

Larry
 
Last edited:
Problem is, no matter what you do the index pulse will not occur once per revolution of the crank.

Sure it will. Possibly twice now and then??

But---I think I just realized where I have been getting things wrong.

With the setup I have asked about, the timing skip tooth will be 180* out of position on every other rev of the crank.

Or something like that.

Bottom line is is that I suspect it aint gonna work as I have described.

My brain hurts
 
Last edited:
It is expecting a one to one relationship between the synch event and one full engine revolution.

Yep, finally got that through my thick skull a bit ago.

In addition the sync pulse should always be at the radial distance from TDC----which I was not considering at the time.

I just don't think you can get there from here with your current idea.

Larry

Pretty sure I have arrived at the same conclusion--------although the wasted spark aspect of all this is still nagging at me.

If there was a way to make sure the wasted spark happens on the out of sync pulse---------and the wasted spark was not going on during intake stroke or such, and the in sync pulse was only in the correct timing in relation to the four stroke needs of every other rev---------OK, babbling now.

Crank fired seems to be the best solution for the 6 cyl folks like me.
 
Last edited:
Stepping back to the configuration with no megajolt, and 2 reluctors spaced to give you two different timings, switching between reluctors.....I wonder what the EDIS thinks when it sees this switch event. The switch could cause a momentary signal dead spot, or could provide 2 TDC indications within one revolution, or could cause a voltage spike if the signals combine, etc. The switch event may have to be carefully scoped and tested, and maybe buffered out with some simple circuitry.
 
Sure it will. Possibly twice now and then??

But---I think I just realized where I have been getting things wrong.

With the setup I have asked about, the timing skip tooth will be 180* out of position on every other rev of the crank.

Let me be more specific. The index pulse will not occur at the correct point in the crankshaft revolution, each and every time. For some reason piston engines seem to be really picky about getting the spark delivered at consistent times.
 
Electroair

Stepping back to the configuration with no megajolt, and 2 reluctors spaced to give you two different timings, switching between reluctors.....I wonder what the EDIS thinks when it sees this switch event. The switch could cause a momentary signal dead spot, or could provide 2 TDC indications within one revolution, or could cause a voltage spike if the signals combine, etc. The switch event may have to be carefully scoped and tested, and maybe buffered out with some simple circuitry.

For what it's worth, Electroair has a certified version of a 6 cylinder mag-hole trigger http://www.electroair.net/press_releases.html. I Suspect that theirs uses some sort of software trickery or the unit has an internal gear reduction unit to maintain a 1:1 ratio with the crank.

Skylor
 
Skylor thanks but my question does not pertain to the 6 cylinder ratio conversation. My question/comment is expanding upon shipchief's post #51 and DanH's comments on kirks dual reluctor setup.
 
Yep, finally got that through my thick skull a bit ago.

In addition the sync pulse should always be at the radial distance from TDC----which I was not considering at the time.



Pretty sure I have arrived at the same conclusion--------although the wasted spark aspect of all this is still nagging at me.

If there was a way to make sure the wasted spark happens on the out of sync pulse---------and the wasted spark was not going on during intake stroke or such, and the in sync pulse was only in the correct timing in relation to the four stroke needs of every other rev---------OK, babbling now.

Crank fired seems to be the best solution for the 6 cyl folks like me.

In a wasted spark system, you are eliminating the burden of sequential cylinder firing, which require an additional synch pulse that is .5 crank speed, usually coming from a distributer or cam position sensor. Without this, you pair the two cylinders that will be at TDC at the same time and fire both of them. One will be at compression and the other will be at exhaust. The spark in the cyl in the exh stroke doesn't do anything and is considered "wasted," hence wasted spark.

Larry
 
For what it's worth, Electroair has a certified version of a 6 cylinder mag-hole trigger http://www.electroair.net/press_releases.html. I Suspect that theirs uses some sort of software trickery or the unit has an internal gear reduction unit to maintain a 1:1 ratio with the crank.

Skylor

It's not so much trickery. In the auto world, there is no such thig as a device spinning at 1.5 X the crankspeed, so no one bothers to program a solution for it. If I were to venture a guess, there is a gearing solution in the mag unit, bringing the toothed wheel back to crankspeed. It would be a very difficult assignment to deal with this in the software, as you have no obvious way to recover from an error and get back to "home base," so to say. A good chunk of the code would seemingly have to be re-written.
 
Skylor thanks but my question does not pertain to the 6 cylinder ratio conversation. My question/comment is expanding upon shipchief's post #51 and DanH's comments on kirks dual reluctor setup.

I don't want to speak for Dan, but it seemed that the two reluctors were feeding two different EDIS units and megajolts. You could swictch between two reluctors with one unit. You will lose spark for a few revolutions, but the EDIS unit will pickup the signal in probably two revolutions after seeing the signal. All electronic ignitions are designed to pickup a lost signal gracefully. All of the pickups are low voltage with circuits and s/w designed to deal with all sorts of problems. The most common electrical issue in a car or plane is loose wiring, causing the signal to drop out and come back. ANY system must deal with this.

The problem with Dan's set up is the offset pickups. If you swithc pickups, the software needs to modify the offset that I referred to in an earlier post and I can't think of a way that the s/w could know which sensor it is reading and therefore which offet to use. Remeber, the EDIS uint is doing some work and the megajolt is changing things on top of that. Redundant pickups are very doable, but they are always placed in the same redial plane as one another and not offset like Dan's picture.

Larry
 
Last edited:
Stepping back to the configuration with no megajolt, and 2 reluctors spaced to give you two different timings, switching between reluctors.....I wonder what the EDIS thinks when it sees this switch event. The switch could cause a momentary signal dead spot, or could provide 2 TDC indications within one revolution, or could cause a voltage spike if the signals combine, etc. The switch event may have to be carefully scoped and tested, and maybe buffered out with some simple circuitry.

This is my first post to this forum. May take a while to get things right. This is Kirk Harrell.

The second reluctor is for ignition #2.

Ignition #1, the reluctor is clocked to provide TDC cranking fire (but doesn't really; more like 3 degrees BTDC) and EDIS only 10 Degrees advance limp home timing without the controller. Ign #1 is married to the Megajolt and provides rpm and manifold pressure controlled advance.

Ignition #2 is not energized for engine cranking. The reluctor is clocked to fire the plugs 15 degrees BTDC at cranking speeds. The EDIS advances 10 at about 300 rpm and remains fix at this setting. This will provide duplicate spark timing like a normal 25 degree fixed mag with the advantages of hot spark, wide plug gaps of EI.

Switching the electronic ignition off. I have done this before every flight with the standard mag check. About 75% of the time the engine just gets happy again when Ign#1 comes back on. Another 24% of the time the engine stumbles slightly, like the controller takes a nanosecond to find itself. 1% of the time it seems to take just slightly longer. Same thing in flight.

The mag check goes from left ign to right and back. If the EI takes a split second to find itself it will sometimes reward you with a nice bark! Have you ever accidentally killed both mags, then switched back on real quick? Yeah, like that.
 
Last edited:
Hi Kirk,
I guess I had somewhat misunderstood your system. But at the same time that got me thinking of switching between reluctors in order to achieve a 4° ATDC like pmag does for starting, using no megajolt. This would be on both ignitions, so everything is a duplicate and redundant. So my worry was about how the edis would handle the switch between reluctors. The fact that you say your mag check can cause some popping and crackling makes me think something similar would happen with a system as I describe. I am trying to get the easy starting for a lightweight prop, without using any megajolts anywhere in the system. Its just a thought experiment for now.

Lets pretend we have a pair of identical mag hole triggers, each with reluctor #1 at 4° ATDC and reluctor #2 at 15° BTDC. You start up on reluctor #1 which gives you 4° atdc below 300 rpm. Once the engine starts and rpm increases the ignitions advance +10 to 6° btdc. Now you flip the switch to the #2 reluctors and the ignitions settle out at 25° btdc.
 
Last edited:
Lets pretend we have a pair of identical mag hole triggers, each with reluctor #1 at 4? ATDC and reluctor #2 at 15? BTDC. You start up on reluctor #1 which gives you 4? atdc below 300 rpm. Once the engine starts and rpm increases the ignitions advance +10 to 6? btdc. Now you flip the switch to the #2 reluctors and the ignitions settle out at 25? btdc.[/QUOTE]
 
The EDIS counts Nine teeth after the skipped tooth and initiates ignition. I can't think of a scenario where switching reluctors would cause it to fire abnormally early. Maybe a little late, but not too early.

IMO the greater risk would be forgetting to switch back to the "starting"reluctor. This could be defeated with a DPDT relay that switches pick ups when the starter is energized.
 
Last edited:
Switching the electronic ignition off. I have done this before every flight with the standard mag check. About 75% of the time the engine just gets happy again when Ign#1 comes back on. Another 24% of the time the engine stumbles slightly, like the controller takes a nanosecond to find itself. 1% of the time it seems to take just slightly longer. Same thing in flight.

Kirk, I don't recall...for a mag check, when you kill the EI, are you killing coil power, or EDIS module power? Either way, it's not an interruption of the reluctor signal, which seems to the subject here.

I kill power to the Megajolt/EDIS/coil, as a group (wiring diagram in post #4). No stumble or oddity when switched back on.
 
Yes, same here. All power off. Where we are different is switching. You are using two toggle switches. I am using the standard "left, right, both, start" ignition switch. When you mag check you go from 'both on' to 'one on' back to 'both on' every time. When you re-energize the EI the other ignition is still firing away. Any split second delay while the EI finds itself is masked by to other ignition.

When I mag check the switch I am using goes from left to right. The right ignition kills as the left ignition energizes, and visa versa. Any split second reorientation by the EI is not masked by the other ignition. And sometimes, no fire for just a split second fires any unburned fuel that got past the exhaust valve. Pop! Again, this is maybe 1% of the time.
 
Last edited:
gear

Kirk do you think the ring gear could just as easy with all the screws be split to go on and off easy?
Bob
 
Dan and I discussed yesterday mods to his early print from which I cut the ring. A split and a mounting hole at every tooth would serve several improvements. Lighter, easier on and off and more flexibility for clocking on the flywheel to aid in crank sensor mounting.

One thing to remember is the missing tooth needs to be balanced. I hung the assembled unit on a sensitive prop balancer and added longer bolts and washers where needed. I suppose one could static balance the ring on a mockup plate and transfer balancing hardware to appropriate flywheel locations. But with a multitude of mount holes to choose from, you would have to figure out which holes you are going to use first.

Alternative, dynamic balance ASAP after installation.

I must say that Dan is a great resource. **** near brilliant, he questions everything. If he doesn't know the answer he learns the answer. He questioned my pickup application. So, hold on before making a ring from the drawing he first shared. I think he has trigger wheel ver 1.1 in the works.
 
Last edited:
You could swictch between two reluctors with one unit. You will lose spark for a few revolutions, but the EDIS unit will pickup the signal in probably two revolutions after seeing the signal. All electronic ignitions are designed to pickup a lost signal gracefully. All of the pickups are low voltage with circuits and s/w designed to deal with all sorts of problems. The most common electrical issue in a car or plane is loose wiring, causing the signal to drop out and come back. ANY system must deal with this.

Hi Kirk,
Lets pretend we have a pair of identical mag hole triggers, each with reluctor #1 at 4? ATDC and reluctor #2 at 15? BTDC. You start up on reluctor #1 which gives you 4? atdc below 300 rpm. Once the engine starts and rpm increases the ignitions advance +10 to 6? btdc. Now you flip the switch to the #2 reluctors and the ignitions settle out at 25? btdc....all with no megajolt in the system

I guess I need to decide if it makes sense to create this electronic "mag replacement" ignition, that does not have a timing advance map.

pros:
-less moving parts
-cheaper parts and plugs (not necessarily lower quality)
-I can say I have EI
-the ford parts have some proven history

cons:
-why bother replacing a mag with a mag?
-the switch flip for the starting advance is more workload, but I envision a momentary-on switch, bet then again low rpm ground ops might benefit from the 6? advance.

questions:
-would I expect lower fuel burn?
-would I expect better starts and smoother running?
-how much would low rpm ground ops benefit from 6? btdc?
-my described system does not have it, but if added, what real world advantage would a timing advance map give me and is it enough to care about?

this all comes up as I am nearing that point of slick inspection and have been thinking about a switch to EI instead of servicing slicks. pmag seemed cool, but this edis stuff lets me tinker.
 
The Ford EDIS and Ford coil pack and attendant reluctors have no moving parts, won't leak oil, won't need contact point maintenance, or need retiming. Spark plugs fire a wider gap and automotive plugs are inexpensive. You can buy everything you need for less than the price of an overhaul for one mag. Some custom parts are required, but are not impossible to acquire. Will not need routine maintenance, except occasional fresh plugs.

Operationally, my engine will still run smooth at leaner settings. Starts great, hot or cold. Doesn't act hot during post flight taxi. Purrs all the way to the hangar. My only experience is with Megajolt mapped ignition advance. I think Dan may have more information on fixed timing.

The two reluctor start and run option is a great idea. Fixed timing ignition will run the engine at least as well as mags have for incalculable hours.

And then there is the "tinker" factor. Gotta love that.
 
megajolt

Anybody know how to combine a display to the Megajolt to show advance and program in plane.
 
I guess I need to decide if it makes sense to create this electronic "mag replacement" ignition, that does not have a timing advance map.

pros:
-less moving parts
-cheaper parts and plugs (not necessarily lower quality)
-I can say I have EI
-the ford parts have some proven history

cons
-why bother replacing a mag with a mag?
-the switch flip for the starting advance is more workload, but I envision a momentary-on switch, bet then again low rpm ground ops might benefit from the 6° advance.

questions:
-would I expect lower fuel burn?
-would I expect better starts and smoother running?
-how much would low rpm ground ops benefit from 6° btdc?
-my described system does not have it, but if added, what real world advantage would a timing advance map give me and is it enough to care about?

this all comes up as I am nearing that point of slick inspection and have been thinking about a switch to EI instead of servicing slicks. pmag seemed cool, but this edis stuff lets me tinker.

You left off a significant con. The mags greatest assest is that it runs without electrical input. Not so big a deal if you only replace one. replacing both shifts the balance scale in my opinion.

questions:
-would I expect lower fuel burn?
I don't think so. The benefits here stem from proper timing. As long as a spark can light a fire, having a stronger one does little for you. At altitude, where the mag's weak spark can't light the fire consistently, you'll see improvement. Same holds for aggressive LOP operation.

-would I expect better starts and smoother running?
Same as above. Mag's have a weak spark at very low RPMs and the coil will improve consistency. 4 additional degrees retarded won't add a lot to starting performance in my opinion, beyond a slightly reduced risk of kickback.

-how much would low rpm ground ops benefit from 6° btdc?

Running your engine at any RPM at 6 BTDC would degrade performance and is not recommended. IDLE operation is optimized somewhere in the teen's to upper teen's depending upon conditions. ignition should advance in a somewhat linear fashion with RPM until you reach the max or peak advance in the 3000 range. Lower idles advance reduces heat and was used in the 70's to reduce emission ratings.

-my described system does not have it, but if added, what real world advantage would a timing advance map give me and is it enough to care about?

considerable and yes. The vast majority of benefits you hear about EI are related to variable timing. There is a reason cars don't use static timing. It is worse than archaic. Heck, even the Model T had variable timing albeit manual (a stalk on the steering column). However, it is not unheard of. Many drag engines use a static timing arrangement to eliminate failure and improve consistency. However, those engines only run at Idle and WOT.

Larry
 
Last edited:
thanks Dan, thanks Larry.

toolbuilder's post is interesting (saw his cool setup at Inyokern last weekend). The response to mixture setting is not able to be included in a timing map is it?

It also makes me think I could never pinpoint timing so precisely in my fixed pitch, wood propped, efis-less, autopilot-less RV-4. Maybe I could use my 2 reluctor setup for a WOT full rich climb setting and a lean cruise setting. And I am not against using a megajolt, which could be added to my proposed system at any point, and then throw a switch to advance the entire map!

I also read this amazingly long thread last night http://www.vansairforce.com/community/showthread.php?t=108079
 
If my ongoing testing is any indication, the timing is relativily insensitive to mixtures richer than peak. Yes, it would be neat if the timing map could follow the jump to LOP ops, but I suspect that people who operate LOP have one primary setting they go to, so a discrete switch which adds the preselected advance gets you there with very little effort. That said, you do need to do the testing to see what that advance value is for your particular engine. The ability to plug in a custom number (like the CPI does) is a huge benefit.
 
If my ongoing testing is any indication, the timing is relativily insensitive to mixtures richer than peak. Yes, it would be neat if the timing map could follow the jump to LOP ops, but I suspect that people who operate LOP have one primary setting they go to, so a discrete switch which adds the preselected advance gets you there with very little effort. That said, you do need to do the testing to see what that advance value is for your particular engine. The ability to plug in a custom number (like the CPI does) is a huge benefit.

Michael,

Your testing and results have been VERY helpful for me in setting up my initial advance map and I thank you for your generous sharing. I intend to run increased advance starting in the 24-25" MAP range. In my case, with an FP prop, much of my climb is in the 24-25" range with power levels in the 80-85% range. I will run ROP here and want a lower advance (am planning high 20's or 30). However, when I hit cruise, I am turning 2700 and still pulling 24". In this case, I am at 70-75% power and want 35* of advance as I will run LOP. In this case, the table switching is very convenient, as I would otherwise have to have some fine resolution in the RPM side of table to produce a 75% line and use it for increasing advance. Although, as I am thinking about this now that may be a better approach. I'll have to play with this.

Compiling your results across your tests, it was good to see a confirmation that the desired advance reduced as you went to dual ignition. It seemed that your optimum advance dropped by almost 5 degrees when you went from a single to dual setup, which is in the range that I was expecting. I mention this for the benefit of others leveraging that data. Optimum advance figures for a dual setup are not the same as those for a single EI setup. With advance, we are really try to optimize the timing of peak cyl pressure and with dual spark, that will be based upon a blend of two different spark events and there subsequent advance settings.

I hope to finish my install on Saturday and will try to post results after testing.

Larry
 
I did mention that the timing is "insensitive" if ROP - but thinking about it more I don't mean to imply there is no benefit to having the "correct" timing. Variable timing is a big benefit over a fixed magneto even if you operate at the relatively rich and relatively low altitude "sweet spot" for magnetos. As we've known for decades, the ABILITY to light off a fat mixture in thick air is essentially the same for mags or EI, but the TIMING of that event is where EI's should have the advantage. I say "should" because as my testing indicates, there is a big disparity between engine characteristics and yes, whether one EI is used or two. It is clear to me that one size does NOT fit all. EI's with fixed or limited adjustable maps certainly have their appeal, but their limitations are exposed after testing the infinitely adjustable CPI product.

That said, I think that we foot stomp one more time that "more" advance is not "better" - you are looking for the "right" advance. It is noteworthy that based on my 100% power testing that I have tweaked my map and am now RETARDING timing at high MP. I'm pulling out 3 degrees from the data plate timing (and remember, this is a low compression, bone stock 260) because the performance is the same, and it will buy me more detonation margin.

Also keep in mind that the above is discussing ROP ops and thick air. The other discussion is when LOP and high - where the power of the hot EI spark positively influences the "ability" to light the fire (and yes, the variable timing) really blows the magnetos away.
 
Last edited:
Heads up for anyone considering using a flywheel mounted trigger wheel like in the photo below.

image%209%201000w_zpsboxcn1g1.jpg


There are at least two flywheel designs with different diameters of "V" belt pulleys out there.

If the "V" belt pulley is too big, there is no room to mount the pickup as the teeth are hidden behind the pulley.

P1010002.JPG


No, I do not know the right size, I do however know the wrong size......
 
Last edited:
Yep, good catch. That "small" pulley is the early version... Getting kind of rare.

This is exactly the opposite situation with the CPI pickup - no design yet for this small pulley.
 
That small pulley would work great with an automotive alternator.... These were found on 150 hp Warrior's
 
Yep, good catch.

Too bad I did not catch it before having the water jet folks make the wheel.:mad:

I just wanted to let others out there who might be pondering this setup know that there could be an issue with the large pulley.
 
Back
Top