plehrke

plehrke

Well Known Member
Patron
I got my new SDS CPi EI system installed and still retained the LH mag firing the lower plugs. I have notice that the EI is much more sensitive to mixture at idle (800-1500 rpm) then when I had two mags. With two mags, if I was idling rich, the engine ran pretty smooth and not that much different then when I would aggressively lean during taxi. With the EI system it runs significantly smoother when I aggressively lean during taxi then if I am taxing with mixture full rich. It also seems to allow me to lean even more then when I had 2 mags. I seems to be able to taxi with a smooth running engine with mixture almost entirely at idle cutoff.

Another thing I have noticed is I can defiantly idle at a much lower rpm then when I had two mags where, when engine cold after just starting, I could never idle below 700-800 rpm. With the EI system I can now idle after start at 500-600 with no roughness but, like I mention above, I need to have the mixture pulled way out.

Is this common with EI or do I need to adjust my idle mixture differently with a EI system?
 
Probably due to the advance that you are getting at idle/low MP. What is your typical idle, taxi advance value?

With the EI advance, you will almost certainly have to turn down the idle stop to get back to your desired RPM, but the mixture should still be set to deliver the specified RPM rise on the lean cutoff test.
 
I see the RPM settings... Very similar to mine.

How about the MP settings? (These really drive the idle/taxi behavior)
 
Do you intend to run it like that from here on out? The MP settings is where most of the "magic" happens... The advance with higher altitudes, LOP, idle, etc.
 
currently just have the CPI system programmed to replicate a mag except down at idle. Once I get some flight time in I will be adjusting per your thread.

That thread does not give a description on how you go about setting up the MP map. So I guess now is a good place to ask how that gets done. Do you set up the rpm map and then use some guideline to phase in some retard when working the engine hard to stay below a detination point? One note is I have a fixed pitch prop if that makes a difference in setting the MP map.
 
Last edited:
The other threads I've done show what total advance the engine likes at various conditions- but you're right, building a map from that info was not specifically covered. The reason for that is that there are a few ways to derrive that value. For example - If you know through testing that your engine is making the best power at your normal cruise altitude, and lets just say that's 22 inches and 2500 RPM, perhaps the total advance is 31 degrees. Well, since the total advance is a composite of the RPM setting and the MP setting, there's plenty of options to add RPM and MP to equal timing. Since I have CS prop and MP is the primary indicator of engine load, I like to play with MP as the real workhorse of the tuning. So in my case RPM advance tops out at a conservative data plate value of 25 degrees, all in by 2000 RPM. since 2000 RPM covers virtually all "in flight" conditions, I can concentrate on the level of load the engine is under - and thats MP. So in our example above, I'd target the total advance value and "back into" the MP setting. Since I want 31, already have 25 (from the RPM), then my target MP setting is 6 degrees advanced at 22 inches. 25 degrees RPM + 6 degrees MAP = 31 degrees total. But one data point does not build an ignition curve, you say... Correct. So lets look at a few more points of interest. I know through testing that a max effort takeoff at sea level does not require even close to the data plate value of advance, so I target a negative (retard) value at 30 inches just to build some detonation margin. I've demonstrated how flat the advance curve is at 100% power, so my goal here is not to max out performance, but back off as much as possible without reducing performance. Lets call it 22 degrees for example. So on takeoff we are maxed out on RPM advance (25 degrees) AND have reached 30 inches... I need a value of -3 for this MP setting. 25-3= 22 degrees total. There's another data point on the curve. Another performance goal is a smooth idle, so lets hit that one up for another data point. With the engine running at idle and warm, its easy to punch up the advance (I'd use the LOP setting for this test) until the engine RPM peaks and otherwise "seems" happy. Theres a pretty good chance that you will be able to move the RPM several hundred just by advance alone. Pick an advance value that your engine likes (good idle, crisp throttle response), note the MP and total advance, then derrive the MP setting by subtracting the known RPM value from the total. Lets say this MP value is 14 inches and the advance is 15 degrees. There's one more point on your MP curve. So far we have:

14 inches = 15 degrees advance
22 inches = 6 degrees advance
30 inches = 3 degrees retard

These are 3 "hard" points on my curve, and the rest I simply filled in with a natural slope. The more data points you have, the more accurate the curve, obviously. But I'm really only concerned about these few conditions since this is where the engine spends 99% of its life. As I build time in other regions, I'll go back and refine the curve a bit.

In the interst of full disclosure, this is contrary to Ross' recommendations and there are some good reasons for that. If you have 25 degrees available with RPM as well as 15 degrees more with MP, then if you find yourself where both of those conditions are met, then you are going to see 25 + 15 = 40 degrees total. And that does happen in my airplane durring a descent into the pattern where the throttle is closed enough to pull the MP down to the max value and RPM is above 2000 (my "all in" value). There's no detonation danger in such a condition and its not going to hurt anything, but 40 degrees is completely unnecessary. Its a harmless, but annoying byproduct of my tuning scheme. However, it does show that you need to think things through before plugging in these values. Keep in mind that the 25 and 15 values are my particular maximum... You are free to plug in much more than that if you wanted to for some reason.

Another, more conservative option is to go big on the RPM setting -say, set 32 degrees max at 2200 and above, and then use the MP setting to retard only. This means that the most you will ever see is 32 degrees no matter what the MP is doing. But this also means you might not have an optimal idle total advance. No great loss, but illustrates a difference in philosophy.

There are several ways to skin this cat, and one has to be aware of the potential pitfalls. But the rewards are great with this much adjustability.
 
Last edited:
Toobuilder said:
The other threads I've done show what total advance the engine likes at various conditions- but you're right, building a map from that info was not specifically covered. The reason for that is that there are a few ways to derrive that value. For example - If you know through testing that your engine is making the best power at your normal cruise altitude, and lets just say that's 22 inches and 2500 RPM, perhaps the total advance is 31 degrees. Well, since the total advance is a composite of the RPM setting and the MP setting, there's plenty of options to add RPM and MP to equal timing. Personally, I like to play with MP as the real workhorse of the tuning so my RPM advance tops out at a conservative data plate value of 25 degrees, all in by 2000 RPM. With the RPM value established as 25 degrees at all realistic flight RPM settings, I can concentrate on the level of load the engine is under - and thats MP. So in our example above, I'd target the total advance value and "back into" the MP setting. Since I want 31, already have 25 (from the RPM), then my target MP setting is 6 degrees advanced at 22 inches. 25 degrees RPM + 6 degrees MAP = 31 degrees total. But one data point does not build an ignition curve, you say... Correct. So lets look at a few more points of interest. I know through testing that a max effort takeoff at sea level does not require even close to the data plate value of advance, so I target a negative (retard) value at 30 inches. Lets call it -3 for example. So on takeoff we are maxed out on RPM advance (25 degrees) AND have reached 30 inches (-3 degrees). 25-3= 22 degrees total. There's another data point on the curve. Another performance goal is a smooth idle, so lets hit that one up for another data point. With the engine running at idle and warm, its easy to punch up the advance (I'd use the LOP setting for this test) until the engine RPM peaks and otherwise "seems" happy. Theres a pretty good chance that you will be able to move the RPM several hundred just by advance alone. Pick a condition you like, note the MP and total advance, then derrive the MP setting by subtracting the known RPM value from the total. Lets say this MP value is 14 inches and the advance is 15 degrees. There's one more point on your MP curve. So far we have:

14 inches = 15 degrees advance
22 inches = 6 degrees advance
30 inches = 3 degrees retard

These are 3 "hard" points on my curve, and the rest I simply filled in with a natural slope. The more data points you have, the more accurate the curve, obviously. But I'm really only concerned about these few conditions since this is where the engine spends 99% of its life. As I build time in other regions, I'll go back and refine the curve a bit.

In the interst of full disclosure, this is contrary to Ross' recommendations and there are some good reasons for that. If you have 25 degrees available with RPM as well as 15 degrees with MP, then if you find yourself where both of those conditions are met, then you are going to see 25 + 15 = 40 degrees total. And that does happen in my airplane durring a descent into the pattern where the throttle is closed enough to pull the MP down to the max value and RPM is above 2000 (my "all in" value). There's no detonation danger in such a condition and its not going to hurt anything, but 40 degrees is completely unnecessary. Its a harmless, but annoying byproduct of my tuning scheme. However, it does show that you need to think things through before plugging in these values.
Click to expand...

Those of us still running 1 mag will get ignition no later than data plate value, assuming it's set correctly. I've been studying your advance map Mike, and Ross's recommend map, I believe they both assume youre running two EI's, but with one mag there is no reason to limit advance at less than data plate spec, 25 on my engine. For this reason I'm considering resetting my magneto to 20 BTDC, this may help with takeoff CHT's
Lots to consider when playing with advance maps.
Tim Andres

PS, Mike did you get your jig back? I forgot to mention I included some spare magnets I purchased, should you ever need some.
 
Tim, got the jig back - thanks!

Concerning the single vs dual issue. It does not hurt to pull the magneto back a little to help with detonation margin on takeoff, but also remember that 25 degrees with two magnetos is not the same as 25 on a magneto and 22 on an EI because the combustion event is a composite of the two firing events. Remember that we have dual plugs in an aircraft engine to make sure the flame front makes it across that big piston in time to do meaningful work. 25 degrees (or whatever) is a derrived value from the real goal- peak cylinder pressure at the optimal crank angle. Flame speed is variable based upon mixture, and the variable RPM means the optimal crank angle for peak pressure is also a moving target.

So what that boils down to is that an ignition timing mismatch such as 25 on one plug and 20 on another is going to be a composite of the two. In your scenario, even if you keep the magneto at 25 and you set the EI at 20 for detonation margin, you still will see an effective timing of less than 25.

That said, the 100% power setting is a very readily combustable condition (which is why it is so insensitive to perfect ignition timing), so your effective retard composite is going to be much closer to the 25 of the magneto than the 20 of the EI.

With all of that out of the way, when I switched from the "optimized" split system (CPI/mag) to an optimized dual CPI system, I found that I had to retard 3 degrees across the board for the cruise conditions.

So long story short- the answer is 3. ;)

Its noteworthy that I kept my magneto at the data plate value of 25 degrees however, so if you pull yours back to 20 you will likely have to advance your EI a bit more to compensate. If you are trying to interpret my dual CPI map, I'd bet that your optimum will likely be my map +4. The good news is that my map, used as is, will be even more conservative.
 
Last edited:
You must log in or register to reply here.