CPI timing optimization ?

I have two CPI units factory programmed for conservative settings. When it is time to optimize them per Toolbuilder's CHT/MP observation protocol, is it OK to make changes on one unit in flight, and then modify the second unit to match it?
My reason to do this is space limitations suggest a temporary mounting to access the CPU for experimentation, and then permanently mount it with the second unit in the avionics bay . I don't anticipate fiddling with the units once they are reasonably and conservatively set for my engine and fuel preferences.

Larry DeCamp said:

I have two CPI units factory programmed for conservative settings. When it is time to optimize them per Toolbuilder's CHT/MP observation protocol, is it OK to make changes on one unit in flight, and then modify the second unit to match it?
My reason to do this is space limitations suggest a temporary mounting to access the CPU for experimentation, and then permanently mount it with the second unit in the avionics bay . I don't anticipate fiddling with the units once they are reasonably and conservatively set for my engine and fuel preferences.

Click to expand...

Not really. The closer you get the peak pressure to occur at the optimum angle , the more optimized the power output is. Peak pressure angle is modified via the combination of both sparkplugs firing position, relative to TDC.

You can leave one ignition fixed a some timing and only modify/optimize the other, but you can't modify one and then apply those settings to the other. The results would change from the experimentation you conducted.

Generally on a twin plug set up, you want both plugs to be advanced the same. Two equal flame fronts meeting in the center is optimal for many unformly shaped cylinder combustion chamber, like the lyc. That said, it's doubtful you would notice any significant difference in performance if they are different by a few degrees.

Larry

Larry,

If you have a parallel valve engine, we did a bunch of flight testing recently on Les Kearney's RV-10 with an EM-5 system looking at timing vs. mixture vs. TAS which might give you some ideas: https://www.youtube.com/watch?v=7YvXmq7vwzY

The easiest way to find optimal timing with a CPI is to just reset the LOP advance amount in both units, flick the switch and watch the TAS.

We had perfectly smooth air and the autopilot holding altitude for us.

Optimal timing changes noticeably with mixture, especially LOP.

ROP at high MAP, somewhere around 24-25 degrees seems good. At lower MAP (15-20 inches) maybe 1-2 degrees more than that. LOP, 30-32 depending on MAP and how far LOP you are.

We were up at 16,000 for this test so MAP was fairly low.

There is another video where we went to FL200 and did some testing too: https://www.youtube.com/watch?v=FgXH-1VJl44

I installed the CPI unit in Bill Beatons HRII with the cabling & vac line long enough so he could hold the control head in his lap while programming. After which, the unit was just velcro'd out of the way when programming was complete.

I installed another CPI in a RV-7A that hinged from the bottom edge of the instrument panel. When programming was complete, the unit rotated forward and latched out of sight, out of mind.

I would suggest you set both CPIs to engine factory set (25 adv. or what ever) for initial runs and flights to confirm proper operation before introducing custom curves.

If you make a modification to one unit, you have to make the same to the other. Remember the engine will be running on the unit that sparks first (most advance), the second unit would just be along for the ride at that point. Ross's suggestion of using the LOP switch setting to progressively test advance settings is a good way to manage your progressive customization.

Its been said before but bears repeating. You are looking for the "optimal" PCP/crank angle. This will show up as the best speed (most mechanical work available). The PCP/crank angle is a result of the initiation of the combustion event. The timing of the combustion event in a dual plug ignition is a composite of the timing of the two plugs. If one plug is later than optimal, it drags the composite to retarded; if one is early, it drags the composite to advanced.

So the short answer is no, you cant optimize the advance on one system and apply that setting to the other system without over advancing the resulting composite.

Thank much to all for input.

I was just hoping to avoid longer wires and hoses. I have the hinged panel under mountings fabricated and thought these responses would be informative to others as well. The switch is a great idea to safely sneak up on optimum.

Given your circumstances Larry, I'd consider locking one ignition out at 25 degrees (or whatever your data plate value is), then play with one to arrive at your optimal value. Once found (let's say it's 34 degrees), remove 4, then apply that to both ignitions. If your engine likes 34 on a single, then it should be pretty close on 30 with both. It's not ideal, but it will quickly get you in the ballpark and will save you many steps of iteration.

rv6ejguy said:

Larry,

If you have a parallel valve engine, we did a bunch of flight testing recently on Les Kearney's RV-10 with an EM-5 system looking at timing vs. mixture vs. TAS which might give you some ideas: https://www.youtube.com/watch?v=7YvXmq7vwzY

ROP at high MAP, somewhere around 24-25 degrees seems good. At lower MAP (15-20 inches) maybe 1-2 degrees more than that. LOP, 30-32 depending on MAP and how far LOP you are.

Click to expand...

G'day Ross, I enjoyed watching the video, but one observation I made is that just about every ignition system on the market would probably been running maximum advance (~14-18*) at 17 In Hg. Your testing seems to indicate that about 5 degrees advance is optimal for that particular aircraft.

rwtalbot said:

G'day Ross, I enjoyed watching the video, but one observation I made is that just about every ignition system on the market would probably been running maximum advance (~14-18*) at 17 In Hg. Your testing seems to indicate that about 5 degrees advance is optimal for that particular aircraft.

Click to expand...

If you're talking total advance, no.

No Lycoming will be making any power at that figure. If you're talking MAP advance plus base RPM (25 + 17= 42 total), this would be WAY too much and lead to a loss of power and high CHTs not to mention likely detonation.

Several of the popular systems running canned curves run too much advance to be optimal and we often see reports of high CHTs..

You saw in the video, there was no speed gain advancing to 32 total at 16,000, even LOP. Running ROP, flame speed is higher, requiring less advance as the video explained.

The optimal ignition timing doesn't vary with airframe type, only the engine type. AV engines need even less total timing than this 9 to 1 PV engine.

rv6ejguy said:

If you're talking total advance, no.

Several of the popular systems running canned curves run too much advance to be optimal and we often see reports of high CHTs.

Click to expand...

Yes its interesting. The "standard" approach, common across most of the vendors I have looked at seems to be to start at about 25-24 inches MP and advance 2 degrees over base engine timing for every inch of MP reduction until they reach 14-18 degrees.

The same basic approach was suggested by Nigel Speedy in his May 17 KITPLANES article.

Each manufacturer provides just enough of an "artist's impression" of their curve that you can almost figure out how they do it.

The maximum advance would have to be a function of RPM as well, so one would assume you would need to be turning 2700 RPM and very low MP to ever see maximum advance in practice.

I don't know, but I suspect no one has really run an engine on a dyno to optimise too much. To me that's part of the appeal of the SDS system. To be able to start with a known good setup and optimise the 2 or 3 common power settings and cruise heights I fly at. Even better if it comes preoptimized for each engine.

rwtalbot said:

Yes its interesting. The "standard" approach, common across most of the vendors I have looked at seems to be to start at about 25-24 inches MP and advance 2 degrees over base engine timing for every inch of MP reduction until they reach 14-18 degrees.

The same basic approach was suggested by Nigel Speedy in his May 17 KITPLANES article.

Click to expand...

Note Nigel's suggestion that essentially the same cruise performance at less CHT is available by sticking with less advance and simply climbing a few thousand feet higher. The interesting thing about the approach is the ease of climbing higher...due to not fighting CHT in climb.

I don't know, but I suspect no one has really run an engine on a dyno to optimize too much.

Click to expand...

It will be interesting to see the Lycoming-derived advance schedules in the Lyc-branded version of the Surefly.

My 540 is going to spend some time on the Ly-Con dyno in the near future and I hope to be able to play with the timing. In my case, I'm looking to quantify the 100% power figure and see how it compares to my in situ flight testing.

As for development of a high altitude curve - that's pretty tough to do unless your dyno room is also an altitude chamber. Short of an altitude compensated test cell, the in situ testing is the way to go. Fortunately, the SDS product architecture makes that drop dead simple.

DanH said:

Note Nigel's suggestion that essentially the same cruise performance at less CHT is available by sticking with less advance and simply climbing a few thousand feet higher. The interesting thing about the approach is the ease of climbing higher...due to not fighting CHT in climb.

Click to expand...

Climbing higher is a good option if operational issues don't prevent that.

My ECI engine has cam squirts, a cold air (hot oil) induction and super corrosion resistant (extra blow by, too hard for mere mortals to hone) barrels. I have never really had an issue with CHT. Oil temps ...

It will be interesting to see the Lycoming-derived advance schedules in the Lyc-branded version of the Surefly.

Click to expand...

It would be great if Lycoming actually publish those curves. I am not sure we really need them as there are not a lot of aircraft suffering engine damage. It does seem the older platforms are not sufficiently open to really facilitate a lot of experimentation.

The more I learn about it the more I'm not overly comfortable installing an ignition with "special sauce" inside. 10+ years of operational history has had me chasing around fixing things which were most likely the result of sub optimal components supported by excellent marketing. Just my opinion...