What's new
Van's Air Force

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

P-Mag "B" Curve Advance Map

Brantel

Well Known Member
I decided today was the day to put the P-Mag monitor that I built a couple years ago to the test and use it to map out the advance curve on the P-Mag. (Dan Horton suggested this in another thread)

This is a P-Mag set to the factory settings using the "B" curve and the manifold pressure line is connected.

Hope this helps settle some dust around the mysterious P-Mag advance curve.

One thing that I noticed is how the P-Mag is advancing the timing even at high power settings by about 7°. Based on the info David (RV10inOz) has been providing I am definitely going back to the "A" curve and may make more drastic changes with the software..... I have always had to limit climbs due to hitting my personal limit of 410°, this may help.
 

Attachments

  • Pmag B Curve Map.png
    Pmag B Curve Map.png
    51.1 KB · Views: 105
Last edited:
I am just an observer in regards to EI, but I thought that the bottom right of the chart was supposed to be populated with 25 degree timing for either A or B curves.
Thanks for the data Brian
 
I am just an observer in regards to EI, but I thought that the bottom right of the chart was supposed to be populated with 25 degree timing for either A or B curves.
Thanks for the data Brian

The B curve is offset no matter what by the amount entered into "adv shift/constellation shift" field inside EICAD or with Bill's EICommander.

I chose an RPM and MP range that a typical RV may use. I have no idea if the advance would go lower at RPMs and MP off of my chart but I really am not interested in those areas since I never operate there.

I can tell you that I was shocked to see it advancing over 7 degrees on a full power climbout at my field altitude but I should not have been. The manual spells this out pretty clear. If I had noticed this before, I would not have been using the B curve all along. My mistake for not reading the manual well enough.
 
Last edited:
Holy cow!

There's no good reason to have ANY advance at full power.

I agree. These EI makers need to make their default timing MAPs available so that people don't have to go thru this to understand what their ignition is doing.

I have no idea why I waited so long to do this after installing the CS prop, I should have done it sooner. It would have been real hard to impossible to do with the FP prop.
 
Oh and I can tell you the engine does not like being forced to 1800 RPM at 23-27" MP. At least it did not feel comfortable to me!
 
Say Brian, would you happen to also have a similar chart for the "A" curve? Or perhaps Emag makes that available somewhere and I've just missed it... there is a chart printed in the install manual, but it says something like "for illustration only, does not represent actual timing curve".

mcb
 
Not yet Matt but I plan to map out the "A" curve next flight.

Emagair has never made these maps public to the best of my knowledge.
 
The manual states the B curve simply shifts the entire curve 5 degrees

The manual contradicts itself and is inconsistent in several places as well as does not match the real world data.

One reason I decided just to map the curve myself.
 
Just to clarify my comment, there should be no advance at full power beyond the stock 20/25 degree stock mag timing depending on the engine.

If their timing is relative to TDC (zero degrees) then 32.2 is still far advanced...7.2 degrees more than stock.
 
Just to clarify my comment, there should be no advance at full power beyond the stock 20/25 degree stock mag timing depending on the engine.

If their timing is relative to TDC (zero degrees) then 32.2 is still far advanced...7.2 degrees more than stock.

Yeah, we knew what you meant, Bob:D.

FWIW, I am sticking with the stock Slick mags for now. Timing on air-cooled engines is not usually perfect for most efficient combustion but a compromise with other things, like TO perf, cruise perf, starting, piston temperatures and peak cylinder pressures (PCP) (and more).

What "warranty" do the EI suppliers offer for engine damage resulting from advanced timing?
 
I am using the program curve to reduce the advance...

After quite a bit of testing... first with two mags, then with one P-Mag, then with two P-mags... I use the program curve to reduce the advance on both P-Mags by around 5 degrees (currently I think that have them set for -2.8). I also set the timing with the prop 2-3 degrees past TDC to help with reduce potential reverse firing with my lightweight prop.

The windows program that they provide to program the settings is pretty crude but it works...

I have tried to get the base curve from Brad several times without success. Pretty irritating that they have been so resistant to providing the table; even a generalized representation would be okay but it appears that the detail that is provided may be inaccurate.

Glad to see some empirical reports.
 
Last edited:
...

The windows program that they provide to program the settings is pretty crude but it works...

...
You can also use the EICommander to adjust the curve and monitor the advance of both P-mags.

Truth is you are not really adjusting the curve as much as you are adjusting the "configuration". Meaning where the "standard curve" starts and ends.

The A curve starts at 26* BTC and the B curve "Advance shift" starts at 31* BTC, per their manual.
 
Last edited:
I can't remember anyone ever claiming or documenting any benefit to running the B curve and I have even less reason to try it now. Quite happy using the A curve though.

Erich
 
Just so no one gets the wrong idea....

So far I love the P-Mag for its simplicity, very easy starts in all conditions, smooth idle, cheap plugs, smoother running and further ability to run LOP (my carb induction is the limiter in my airplane).

I just think that based on this timing map, that I want to go back to at least the "A" curve and may take further steps to control the timing.
 
Here's an interesting little graphic to contemplate. It is one of many interesting bits found in the 2009 Swift Enterprises fuel test report, available at the FAA tech library site. Subject is an non-turbo angle valve IO-540-K, thus I believe much of the data is applicable to those of us with IO-360 and IO-390 power.

16c8202.jpg


The graphic plots the point of peak pressure ATDC for each cylinder, in degrees, in this case at 2700 RPM and full throttle. The original graphic included peak pressure for Swift fuel at three timing settings. I've removed the Swift plots for clarity, leaving only the 100LL plot. Air/fuel ratio is not precisely quantified. The notes say only "at takeoff power", so I assume mixture is ballpark 200-100 ROP. As you would expect, power plots (2700/FT) show the peak around 100 ROP, with detonation onset at 30~40 ROP and 20% intensity.

The point of interest? Note the base timing, 20 BTDC, and the resulting peak pressure point, averaging roughly 7 ATDC. Base timing set at 32 BTDC does not appear to be a good idea with this cylinder configuration.

I say "with this cylinder configuration" because (1) I suspect the angle valve head has a faster burn rate than the parallel valve head, and (2) the rate probably changes with compression ratio. A low compression parallel valve cylinder would probably record a later peak pressure under the same conditions.

The A curve starts at 26* BTC and the B curve "Advance shift" starts at 31* BTC, per their manual.

Don't think I'd want 26 degrees for my 390.

Anyway, Bill, do you have any idea how the map values were established? And, with due respect for your position, would you say Brian's posted B-figures are "reasonably accurate"?

BTW, like Brian, I have no rancor for E-Mag or any other EI vendor. Like many here, I'm interested in going to an electronic ignition at some point. However, I'm not at all interested in bolting on something which produces actual ignition timing unknown to me. I'm too much of a gearhead, and 390 cylinders are way too expensive.
 
Last edited:
And, with due respect for your position, would you say Brian's posted B-figures are "reasonably accurate"?

I can tell you this Dan which leads me to believe that my numbers are close.

1. My data comes straight out of the PMag using the data stream protocol provided by EMagAir. The Advance variables are pretty simple to decode.

2. I have another data map provided by someone else (very reliable source but they may want to remain anonymous) that used the EICAD software provided by EmagAir to map the curves and his data matches mine more or less. There is a margin for error on all of these test because it is hard to get RPM/MP dead nuts on and log numbers and fly at the same time while your pass records numbers. There is also the variable of different MP sensors and their accuracy.

3. The Pmag uses low resolution 1 byte integers for a ton of these variables. For instance 1.4° is the minimum resolution for an advance variable in the map. EICAD also rounds numbers and there are certain values that it will never display because of the rounding. My monitor uses the raw number and displays it as a floating point number with a decimal point. This gives a little more accuracy over EICAD.

4. Tom from EmagAir confirmed my method for decoding their Advance variable and said it was correct.

If my numbers are incorrect, then the variable spit out by the Pmag that is used by EICAD and my monitor to display the advance is incorrect.

Like you, my goal with all of this is personal satisfaction and knowledge of how the Pmag works.
 
Last edited:
actual EI improvement over mags

As I contemplate my overdue for inspection Slick mags.(850hrs), I wish I could get a real world comparison that I could trust. I am struggling with ROI in EI.

For my situation My RV8 has dual Slick impulse mags, and Avstar Fuel injection.
I can achieve 168 KTAS at 10K to 12K altitude. Thats at 60% pwr running 25 degrees LOP. I burn 7.5 ACTUAL gph at this alt. I can lean further with smoothness to another 1 gph less but at a substantial speed loss, down to 160 KTAS.
My engine starts easily still. Timed at 25 degrees, mag drop is identical at 50 rpm at 1750 rpm run up.

CHT's in cruise are 300 to 320 at 25 degrees LOP. I can watch the CHT's move up as I richen to peak and watch them fall as I go leaner.

Replacing with dual p mags will cost me almost $4K.

2 new mags will be under $2K.

I fly 350 hrs a year with probably 200 hrs spent in cruise.
I have new fine wire plugs that will probably last until engine overhaul.
I can't believe I can achieve better than .5 gph improvement in cruise over what I have now, or can I ????
Seems like it would take me at least 3 years to recoup my investment.
 
Jon,

LSE II+ is only $200 more than a 4370 mag (or $130 more than 4371). With 0.5 gph saving, you'd pay it off in a few months at most. And long term, the savings would be huge (ie you won't be spending $900 every 500-850 hrs).
 
A & B Curve Clarification

The following is to help you understand how the P-mags work. I will not comment on the actual Emag timing map as that is considered proprietary to Emag Ignitions and due to our relationship with them we promised that we would not publish their timing map/curve. What Brian has documented is the MAP's contribution to the timing curve. As you can imagine, those numbers are not always used to calculate the firing angle. In Brian's defense, the numbers he published at the start of this thread are what is reported by Emags EICAD program.

The P-mag has, in effect, two memory locations, A & B. Commonly called the “A Curve” and “B Curve”.

To switch between the two, a builder installs a jumper between pins #2 & #3 on the P-mag. The default, no jumper installed, is to run on the B Curve. Once the jumper is installed, the Pmags switch to the A curve. (When our EICommander is installed, no jumper is required. I will explain that later.)

The B Curve is fully configurable via Emag’s EICAD program or our EICommander. This discussion assumes an unmodified B Curve.

The A Curve, jumper installed, starts the timing at around 26* BTC and will allow the timing to advance up to 34* BTC, depending on RPM and MAP.

The B Curve, jumper NOT installed, adds a five degree shift to the A Curve. Thus timing starts at around 31* BTC and advances up to 39* BTC.

That is all in the Pmag manual, page 14 and mimics what we have found on our test stand and in actual flight conditions.

The pilot has the ability to adjust the max advance and the advance shift via the EICAD program or the EICommander. For example, if your engine requires a timing of 20*BTC, then you can enter in a negative shift and reduce the timing to 20* BTC and the max advance will also be reduced.

With this in mind, it should not surprise anyone who has read the manual that when operating with the B Curve (No jumper installed) under high power settings, such as during takeoffs, that the Pmag sets the timing at 31* BTC. With the jumper installed, the Pmag will set the takeoff timing to 26* BTC.

During the development of the EICommander, we took an old engine and cut the case in half and used that to spin two Pmags on our test bench. We discovered that there is so much gear slop in our drive trains that even standard magnetos vary by as much as 1.5*. This is really noticeable when running a light, composite propeller.

The idea of setting the timing one or two degrees after top dead (TDC) center is suggested by Emag for those who run composite propellers so the spark fires after the prop has crested TDC and the spark helps spin the prop. This may be necessary because when below 200 RPM’s the Pmags fire at TDC, not after TDC, and a light prop may not have the inertia to get past the compression stroke and spark plug firing when starting. Metal props don’t seem to have this problem. (For the record, I run the A curve and my TDC is set at zero for my O-360 with a Catto prop.)

EICommander
The EICommander that we produce is simply a piggy back unit for the Pmags. It can monitor the Pmags and display the current advance numbers (not the MAP contribution), along with any divergence in timing between two Pmags, the condition of the ignition harnesses and a number of other parameters.

When first installing an EICommander, the Pmags will automatically default to the B Curve. Thus, we recommend our clients send the P-mags the A Curve, if they want to run the A Curve. (New engines, or newly rebuilt engines should run the A curve for the first 10 to 15 hours and cap the advance to 26* BTC either via a custom EICommander configuration or by plugging the MAP lines to the Pmags.)

We tell clients to think of the Pmag as having two memory locations, A & B. The A location is fixed and cannot be changed. The B location is fully programmable and when the EICommander sends the A configuration to the P-mags it is actually stored in the B memory location. Thus, the A & B configurations are exactly the same at this point but because there is no jumper installed (or required), the Pmags will act like they are running the A Curve.

Should a client wish to create a custom “C” configuration (Say with a 20* BTC starting point), they simply send that to the Pmags and it will be stored in the “B” memory location. In other words, with the EICommander installed, the Pmag never use the “A” memory location, regardless of what configuration is sent the Pmags.
 
Last edited:
The B Curve, jumper NOT installed, adds a five degree shift to the A Curve. Thus timing starts at around 31* BTC and advances up to 39* BTC.

That is all in the Pmag manual, page 14 and mimics what we have found on our test stand and in actual flight conditions.

With this in mind, it should not surprise anyone who has read the manual that when operating with the B Curve (No jumper installed) under high power settings, such as during takeoffs, that the Pmag sets the timing at 31* BTC.

My P-Mag out of the box had the following settings when I connect it to EICAD, they have not been modified:

ADV Shift = 5.6° not 5 (correction, it was actually 4.2°)

Max Adv = 40.6° not 39

The stock A curve minimum cannot be 26, it is either 25.2° or 26.6° due to the 1 byte resolution of the P-Mag processor and mine seems to be 26.6° (actually I think it is more like 28° based on my findings)

When running on the B curve, my minimum that I see is 32.2° which is the same as 28° (Base curve Min) + 4.2° (B Curve Constellation shift) That being said, my minimum advance data matches what I would expect it to based on the variables involved.

When running on the B curve, my maximum that I see is 40.6° which is the same as what EICAD says my Max Adv is set to.

Like I have said before, the manual is full of these kinds of inconsistencies due to rounding up and down and the fact that the processor has a low resolution to work with and what appears to be different defaults than what is in the book.

I don't understand why we you are recommending that we disregard the effect of the MAP sensor on the advance. Since the MAP sensor in the Pmag does effect the ultimate timing advance, why would we ignore that variable?

I stand by my chart....and based on the default settings and your second sentence above, your data also backs mine up within degree or so due to rounding/small errors.

My only point is this:

If one operates their P-Mag with the default settings on the "B" curve,
Their Max Advance is going to be 40.6° and it comes in pretty steep based on my charted data.
Their Minimum Advance is going to be 32.2° even under full power which is 7.2° more advance than the 25° standard from Lycoming. 12.2° more for those with engines recommending 20°.
 
Last edited:
As I contemplate my overdue for inspection Slick mags.(850hrs), I wish I could get a real world comparison that I could trust. I am struggling with ROI in EI.

Mutha... or er, I mean, (scratch head) what's that new call sign again? Badger....:p

I wouldn't wait long on that Slick inspection- especially if that time above is more than SOH. Lost the slick after 1200 hours new and after a 600 hr inspection calling both slicks good. Pushed it too far.

Operating the 114 P-Mag in service now for more than 400 hours(sometimes in PHX extreme hot environments). A-curve operation only. Data shows a marginal fuel efficiency(carburated Parallel valve 180) .5 gal hr but with some ability to run LOP if I must. Smother running engine with cheap plugs also. Yes the ROI is probably over a longer haul with the FI you already have.
My engine is hotter, even with the A-curve and the #3Cyl baffle mod. So I'm real interested in Brantel's next test.
Thanks for data!:D
 
What Brian has documented is the MAP's contribution to the timing curve. As you can imagine, those numbers are not always used to calculate the firing angle. In Brian's defense, the numbers he published at the start of this thread are what is reported by Emags EICAD program.

I think you're saying is the values reported by Brian are in fact the actual firing angles with the manifold pressure line connected. Correct?

The manual (page 15) says

2. Test with (near) standard magneto timing. With the ignition indexed
at TDC and using the ?A? curve, you can disconnect the MAP tube from
the ignition(s). In this configuration, the maximum advance will be near
standard mag timing (26 degrees).


Does this mean that when operated with the manifold pressure line disconnected, timing will be fixed throughout the RPM range, just as a mag is fixed?
 
I think you're saying is the values reported by Brian are in fact the actual firing angles with the manifold pressure line connected. Correct?

The manual (page 15) says

2. Test with (near) standard magneto timing. With the ignition indexed
at TDC and using the “A” curve, you can disconnect the MAP tube from
the ignition(s). In this configuration, the maximum advance will be near
standard mag timing (26 degrees).


Does this mean that when operated with the manifold pressure line disconnected, timing will be fixed throughout the RPM range, just as a mag is fixed?

When you disconnect the MAP line, the PMag assumes that the engine is at max power regardless of the RPM. The max advance will be:

26.6° when running the "A" curve (actually I think this is more like 28° based on my findings)
28° + "Adv Shift" (Constellation Shift) default of 4.2° = 32.2° when running on a default "B" curve

You can also set the "Max Adv" to limit the Pmag from advancing above that variable.
 
Last edited:
I think you're saying is the values reported by Brian are in fact the actual firing angles with the manifold pressure line connected. Correct?
No, not correct. If you take Brian's numbers, the engine would be firing at very high firing angles when under low power. At higher power, he is close. The numbers he presented, as read from the EICAD program, are simply used to calculate the firing angle, they are not the actual firing angle.

The manual (page 15) says

2. Test with (near) standard magneto timing. With the ignition indexed
at TDC and using the ?A? curve, you can disconnect the MAP tube from
the ignition(s). In this configuration, the maximum advance will be near
standard mag timing (26 degrees).


Does this mean that when operated with the manifold pressure line disconnected, timing will be fixed throughout the RPM range, just as a mag is fixed?
Correct, it fixes the timing at the lower setting. In Brian's example, should the MAP line come disconnected, the timing will be fixed at 32* BTC.

That is why I recommend people with new or newly overhauled engines to set them to the A curve and to plug the MAP line for the first 10 to 15 hours. That will fix the timing at 26* BTC and force the P-mag to act like a standard mag until the rings are set. This will help remove the P-mags from any high CHT issues the builder might have and hopefully point them towards their baffles.
 
My P-Mag out of the box had the following settings when I connect it to EICAD, they have not been modified:

ADV Shift = 5.6? not 5

Max Adv = 40.6? not 39

The stock A curve minimum cannot be 26, it is either 25.2? or 26.6? due to the 1 byte resolution of the P-Mag processor and mine seems to be 26.6?

When running on the B curve, my minimum that I see is 32.2? which is the same as 26.6? (A curve Min) + 5.6? (B Curve Constellation shift) That being said, my minimum advance data matches what I would expect it to based on the variables involved.

When running on the B curve, my maximum that I see is 40.6? which is the same as what EICAD says my Max Adv is set to.

Like I have said before, the manual is full of these kinds of inconsistencies due to rounding up and down and the fact that the processor has a low resolution to work with and what appears to be different defaults than what is in the book.

I don't understand why we you are recommending that we disregard the effect of the MAP sensor on the advance. Since the MAP sensor in the Pmag does effect the ultimate timing advance, why would we ignore that variable?

I stand by my chart....and based on the default settings and your second sentence above, your data also backs mine up within degree or so due to rounding/small errors.

My only point is this:

If one operates their P-Mag with the default settings on the "B" curve,
Their Max Advance is going to be 40.6? and it comes in pretty steep based on my charted data.
Their Minimum Advance is going to be 32.2? even under full power which is 7.2? more advance than the 25? standard from Lycoming. 12.2? more for those with engines recommending 20?.
You are correct in that the manual rounds out the numbers. The numbers I shared came right from their manual and very closely match what we have seen (rounded, of course). As I said, I will not divulge their timing curve.

At lower power settings, the standard "B Curve" will never see 40.6* BTC because they set the max advance number.
 
The numbers he presented, as read from the EICAD program, are simply used to calculate the firing angle, they are not the actual firing angle.

Hmm. Put me down in the "sure would like to know what the full-power timing on my expensive engine is actually going to be" column. :)

That is why I recommend people with new or newly overhauled engines to set them to the A curve and to plug the MAP line for the first 10 to 15 hours. That will fix the timing at 26* BTC and force the P-mag to act like a standard mag until the rings are set.

Bill, just checking, did you really mean plug the MAP line, or did you mean just disconnect it as stated in the manual? Seems like it could mean two different things, assuming the P-mag's sensor is sealed up internally.

thanks,
mcb
 
Last edited:
Hmm. Put me down in the "sure would like to know what the full-power timing on my expensive engine is actually going to be" column. :)
At full power; down low with high MAP, you will run the minimum advance as configured in your P-mag. (A vs. B curve) The P-mags, under these conditions, does not add anything to the firing angle. The reason, as you can imagine, is to keep from damaging your engine.

Bill, just checking, did you really mean plug the MAP line, or did you mean just disconnect it as stated in the manual? Seems like it could two different things, assuming the P-mag's sensor is sealed up internally.

thanks,
mcb
Yes, plug your MAP lines; both the line that goes to your P-mag and the one going to your MAP sensor. This assumes they are staying low for the first 10 - 15 hours. You could leave the P-mag MAP line open. (We did have one client that took his plane up above 6,000' with the P-mag MAP line plugged and it worked like a balloon and he saw his advance change.) The goal is to trick the P-mags into acting like a standard mag and not advance.

After you get 10 to 15 hours on your engine, then play around with the B curve, if you want. Just remember, it starts out at 32* BTC.
 
Last edited:
At lower power settings, the standard "B Curve" will never see 40.6* BTC because they set the max advance number.

On mine out of the box EICAD said my max advance was 40.6°. Are you saying that there is some sort of hidden limit other than the max advance angle setting?

You say that 40.6° is a high firing angle but if you take the rounding out, it looks really close to what the manual shows in the graph to expect at low power-upper RPM range on the "B" curve with default values.

Just trying to understand what is so different about my data vs the published data in the manual when corrected for rounding and the low resolution of the processor. I ain't seeing it.
 
Last edited:
On mine out of the box EICAD said my max advance was 40.6°. Are you saying that there is some sort of hidden limit other than the max advance angle setting?

No, I'm saying the numbers you are getting out of the EICAD program are not the actual firing angle. They are just used to calculate the angle and may not be used at all.
 
No, I'm saying the numbers you are getting out of the EICAD program are not the actual firing angle. They are just used to calculate the angle and may not be used at all.

Bill,

You tested them on a test cell, I did not. What you are proposing makes absolutely no sense to me personally but I am not in a position to argue since I have nothing other than in flight data using what the Pmag spits out as a reference.

Since I cannot confirm or deny your data without building my own test stand, I will be forced to take the numbers from the Pmag at face value as that is what EmagAir says they are. After all, they are the ones that designed and make the things.
 
Last edited:
Bill,

You tested them on a test cell, I did not. What you are proposing makes absolutely no sense to me personally but I am not in a position to argue since I have nothing other than in flight data using what the Pmag spits out as a reference.

Since I cannot confirm or deny your data without building my own test stand, I will be forced to take the numbers from the Pmag at face value as that is what EmagAir says they are. After all, they are the ones that designed and make the things.

I hear you Brian, the numbers just don't make sense and had us going in circles at first, same as you.
 
Increase the transparency with an Electronic Ignition Commander ? EIC

Like Brian et al I had a lot of concerns about the lack of transparency with the dual Pmag maps on my IO-390. I succumbed to my doubts and bought the EIC*. I feel about that device like I feel about my engine monitor. I wouldn?t consider doing without it. It has a zillion data screens but the two I use are the preflight ignition check (you can read about it on their site) and the Active stream mode screen that allows you to view your rpm, advance, voltage, temp and ignition harness status and so on while in flight. You can even change your advance settings ?on the fly? but I?ve never done it.

*I have no financial relationship with EIC.
 
3. The Pmag uses low resolution 1 byte integers for a ton of these variables. For instance 1.4? is the minimum resolution for an advance variable in the map. EICAD also rounds numbers and there are certain values that it will never display because of the rounding. My monitor uses the raw number and displays it as a floating point number with a decimal point. This gives a little more accuracy over EICAD.

Just out of curiosity is it encoded in BCD format?
 
Does this mean that when operated with the manifold pressure line disconnected, timing will be fixed throughout the RPM range, just as a mag is fixed?

Hi Dan,
This is my understand from reading the manual and talking to Brad at EMAG. This is the fail safe that it has, without the manifold pressure line, timing would be at a fix point near the 25 dBDC just as a slick mag. At least this was my understanding of the system.
 
Just out of curiosity is it encoded in BCD format?


Bob,

The Pmag will stream Hex or binary depending on what command you send it to request the stream.

In my display the ADV variable is streamed as 2 ASCII characters representing 2 HEX numbers that get converted to decimal. That value is then scaled using the 1.4 deg multiplier.
 
Hi Dan,
This is my understand from reading the manual and talking to Brad at EMAG. This is the fail safe that it has, without the manifold pressure line, timing would be at a fix point near the 25 dBDC just as a slick mag. At least this was my understanding of the system.

Not exactly. It will be fixed at the low end of the curve you have selected. Thus, if you picked the "A Curve" (Jumper installed), it will be 26* BTC. 31* BTC, if you running an unmodified "B Curve" (No jumper).
 
Not exactly. It will be fixed at the low end of the curve you have selected. Thus, if you picked the "A Curve" (Jumper installed), it will be 26* BTC. 31* BTC, if you running an unmodified "B Curve" (No jumper).

You are correct and this is my understanding as well. I am currently running mine with the A curve.
 
I think this looks more reasonable.

I understand that the rpm range is way off and we dont really care about running less than say 17" MAP unless you want to go into the flight levels.

PC_Link_TC1.gif


This is from a newer Harley Davidson
 
I think this looks more reasonable.

I understand that the rpm range is way off and we dont really care about running less than say 17" MAP unless you want to go into the flight levels.

PC_Link_TC1.gif


This is from a newer Harley Davidson
Only one problem, their timing is much more accurate than any of ours and so is their fueling. When you combine the two, you can be much more aggressive. When I was racing we used to put the car up on a chassis dyno and tune it within a breath of its life. Not something we want to do with an airplane engine.

However, that graph shoes a fairly straightforward curve in the lower RPM ranges. Is that from their new, small bikes?
 
And how many spark plugs per cylinder in that Harley? We have two plugs and two flame fronts. Less advance in spark timing required.

Guys, has anyone got data on the actual spark timing? Not the timing mark set on the flywheel at 25BTDC, or reference TDC or the digital data that the EIC is reporting.

What the EI folk seem to not be addressing is that a piston aero engine operating at a relatively static RPM has a few "effective timing" adjustments already.

MP and RPM are all they look at. For any given MP, the only variables are then RPM and Mixture. We use in an airplane a static RPM…pick one between 2200-2500. The next is mixture.

All of this is focussed around achieving an optimum ThetaPP.

Unless the complete system is designed and tested around achieving this, and the actual spark timing is determined to achieve optimum ThetaPP, with changes in mixture, (remember not all cylinders have the same mixture anyway in many planes), then the whole idea is flawed.

Unless there is accurate Dyno testing of the system with respect to actual spark timing (not triggering or even the physical set up) and the resultant ICP and Theta PP, and all these parameters can be measured simultaneously and able to capture the individual combustion events (20 per second)……how will you know?

How many of the RV community and their EI suppliers have the equipment and ability to do this R&D. My gut feeling is not one us. I might come close as in terms of access, but I am not wanting the dyno offline for this experiment while something far more important to all of GA is going on.

I only know of one such EI and one such facility to test this in the world. And that EI is not yet on the market. It is a far more complex and sophisticated system indeed. Why? because unlike the auto industry, a lot of what they can do in an auto engine, can't be done in an aero engine.

Most of you know why I prefer two mags, I am not on an unlimited budget, but I am certainly quite capable of spending even large sums of money on something that was way better in terms of performance, safety and reliability. There is no amount of hard work or funding blocking my change over to an EI, but I choose to run two Slicks (would prefer two Bendix but that is another story).

If you are going to use an EI, my suggestion is to get the system to fire with an actual spark plug firing happening at something like 5-10 degrees later than they are now. This could mean setting the EI so that it is on the most conservative setting and instead of TDC 2-3 afterTDC as the reference point. It may be more? Depends on the lag in the system. The problem is you can't easily tell when that is unless you have very high speed capture of the spark event and an encoder on the crankshaft telling you the exact angle. Just think back to a recent post I made where the mags are set to 20 degrees and the actual spark firing was varying in the 13-15 degree range (from memory). Think about this in terms of the EI.

There is far more to this than many realise.
 
And how many spark plugs per cylinder in that Harley? We have two plugs and two flame fronts. Less advance in spark timing required.

Guys, has anyone got data on the actual spark timing? Not the timing mark set on the flywheel at 25BTDC, or reference TDC or the digital data that the EIC is reporting.

What the EI folk seem to not be addressing is that a piston aero engine operating at a relatively static RPM has a few "effective timing" adjustments already.

MP and RPM are all they look at. For any given MP, the only variables are then RPM and Mixture. We use in an airplane a static RPM?pick one between 2200-2500. The next is mixture.

All of this is focussed around achieving an optimum ThetaPP.

Unless the complete system is designed and tested around achieving this, and the actual spark timing is determined to achieve optimum ThetaPP, with changes in mixture, (remember not all cylinders have the same mixture anyway in many planes), then the whole idea is flawed.

Unless there is accurate Dyno testing of the system with respect to actual spark timing (not triggering or even the physical set up) and the resultant ICP and Theta PP, and all these parameters can be measured simultaneously and able to capture the individual combustion events (20 per second)??how will you know?

How many of the RV community and their EI suppliers have the equipment and ability to do this R&D. My gut feeling is not one us. I might come close as in terms of access, but I am not wanting the dyno offline for this experiment while something far more important to all of GA is going on.

I only know of one such EI and one such facility to test this in the world. And that EI is not yet on the market. It is a far more complex and sophisticated system indeed. Why? because unlike the auto industry, a lot of what they can do in an auto engine, can't be done in an aero engine.

Most of you know why I prefer two mags, I am not on an unlimited budget, but I am certainly quite capable of spending even large sums of money on something that was way better in terms of performance, safety and reliability. There is no amount of hard work or funding blocking my change over to an EI, but I choose to run two Slicks (would prefer two Bendix but that is another story).

If you are going to use an EI, my suggestion is to get the system to fire with an actual spark plug firing happening at something like 5-10 degrees later than they are now. This could mean setting the EI so that it is on the most conservative setting and instead of TDC 2-3 afterTDC as the reference point. It may be more? Depends on the lag in the system. The problem is you can't easily tell when that is unless you have very high speed capture of the spark event and an encoder on the crankshaft telling you the exact angle. Just think back to a recent post I made where the mags are set to 20 degrees and the actual spark firing was varying in the 13-15 degree range (from memory). Think about this in terms of the EI.

There is far more to this than many realise.

Thanks David!

I think I'll just overhaul my mags, or better yet maybe Rocket Bob will help me do it myself!:D
 
David, I think I understand pretty much what you are saying and agree that ICP should be the ultimate determinant factor. But there is one thing that you have discussed often that I do not understand, and that is your contention that the actual spark event is often happening way after we think it is in a mechanical system (conventional mags).

When you check timing using a dynamic, inductive timing light, are'nt you seeing the timing of the actual spark event??

I can understand that is there is a lot of mechanical lead/lag in the gear train and mag mechanisms and that there is less in crank sensor fired electronic ignition, but isn't the sum product of all that going to be reflected in the actual current being sent down the wire that is indicated by an inductive timing light??

If the timing light says that the spark event is happening at say 25 degrees BTDC, are you saying that comprehensive engine instrumentation will indicate that it is happening later?

Just trying to understand. Thanks
 
Last edited:
In a former life I worked in R&D electronics in Detroit. One area of focus was engine control which included ignition and injection timing. I can say without question that what Brantel's experiments are teaching him is common. The numbers out of a box, without knowing what they are used for, are fairly useless. We actually took steps to confuse anyone trying to decipher our curve by looking at raw data. They figure it out anyway, but only if they really know what they are doing....and often only by testing real world, certainly not by looking at the raw data coming out. I'm not saying this is what Pmag has done...just saying proprietary info is often guarded.

This discussion is certainly worth having, but what struck me as odd from the beginning is why would a manufacturer call their curve proprietary if they were going to let it be spit out as raw data? I suspected it was calculated...and if I'm hearing Bill correctly...it is. They protect it as proprietary so that a bright guy with a homemade box won't take their curve and how they achieve it...put it on the Internet for someone else to put in another box to package and sell for 20% less than Pmags. Easy enough to do if you don't have to do any of the engineering or testing.
 
Last edited:
They protect it as proprietary so that a bright guy with a homemade box won't take their curve and how they achieve it...put it on the Internet for someone else to put in another box to package and sell for 20% less than Pmags. Easy enough to do if you don't have to do any of the engineering or testing.

It doesn't take a bright guy with a box to get good timing curves. LSE and Electroair systems have an analog output for a voltmeter so one can see at a glance how many degrees of advance the system is operating at. In one hour of flying I could have a complete advance matrix populated with a timing curve that has years of good results.

Its clear in this case that the one manufacturer that chooses to obfuscate this information uses a curve that one has to wonder what in they heck they were thinking. And if you don't like it, just crank the timing retarded past TDC a few degrees. Mmmmkayyy.
 
Back
Top