Turbo - Static MAP vs. in-flight MAP?

[NWade]

Got the catch-can welded back together. Got all of the hoses and fasteners tightened up. Added the coolant and tried to run the pump so I could get the air bubbles out of the system aaaaaaand....... the water pump won't turn. Looks like the unit I bought back in March was DOA (I never tested it as I didn't want to run it dry). Also, since its been a few months, I can't return it; I have to buy another one. Argh!

I swear my Sonex has a fear of flying, and is trying everything it can to delay things. :-P

--Noel
Sonex #1339

[John Monnett]

Reverse the wires on the pump. It will run.

[NWade]

Reverse the wires on the pump. It will run.

Thanks, John.

So I tried the wires coming from the automotive-style plug both ways last night, and got no flow. I also tried a couple of bare wire-ends and got no flow, but this morning I realized I only tried that one way.

I re-examined the automotive plugs I got off of Amazon this morning and realized that the "latch" on the male plug interferes more on the pump receptacle than I realized. It was preventing the plug from seating *quite* all the way down. So as an experiment I snapped the latch mechanism off of the plug (I have spares), and was able to get the plug to fit just a wee bit further down.

Voila! One functioning pump. :-) Not the issue you identified, but still a good nudge to get me to look at it with fresh eyes. Thank you, sir!

--Noel

[NWade]

Work has been busy so my time in the shop has been very sporadic, but I finally got the remainder of the Turbo cooling system in & running. I spent some time buttoning up all of the electrical wiring and tubing again, then took the airplane out of the workshop to give it a static run in the driveway tonight.

The engine started and ran smoothly - just a touch of light gray smoke out of the exhaust during the early warmup, then a nice steady idle at about 900 rpm. Then a test-run up to 1800 RPM and a quick mag-check: 70 RPM drop on one system, 40 RPM drop on the other. Not perfect; but pretty good. Back down to idle, another scan of the gauges, and then a smooth steady power up. I used caution as I approached full throttle - just in case I didn't assemble the waste-gate actuator properly or something (and because the wastegate area was what the Turbo company had re-machined).

And the result? NO CHANGE. If anything, its a little bit worse now! Instead of getting ~38.5" The MAP stopped rising at ~37.8" and RPMs topped out around 3070. Arrrghhhh!

I really, really am not sure what the next step is (beyond calling Sonex again and asking them for their thoughts)... *sigh*

--Noel

[John Monnett]

Your wastegate actuator push rod needs to be adjusted (tightened) to "pop" off at a desired higher pressure.
John

[NWade]

Your wastegate actuator push rod needs to be adjusted (tightened) to "pop" off at a desired higher pressure.
John

I previously had it as short as possible, and observed no difference in behavior no matter what setting I used. But I will double-check it tomorrow and post video. At this point I would be HAPPY for someone to point to something I've set up incorrectly - anything that'll resolve this issue!

For anyone curious or willing to find fault with my setup, I have videos and photos posted here. They're in reverse chronological order and cover my entire build - engine run videos, turbo & cooling assembly, wiring, MGL engine log graphs, etc.

--Noel

[NWade]

Brief update:

The actuator is as short as possible. Photo here (attachment is rotated for visibility in the photograph): https://flic.kr/p/NnTddi

Recent exchanges with Sonex point out the similarities in RPM between my tests and theirs, as well as some other data-points they have from their testing. My own _personal_ interpretation is that they think the engine RPMs and MAP are high enough to produce strong power, that the engine is flyable; and my issue may simply be a sensor issue, air-balance line issue (between the turbo and the actuator) or some other small variance in my system.

On the one hand, I've attempted to cross-check the MAP sensor with a mechanical gauge and found the two read similarly. I've spliced that same mechanical gauge into the air balance line in the past, and it shows a lower amount of boost than I'd expect when the MAP plateaus (when I believe the wastegate is opening). Again, the boost level of the air-balance line cross-checks pretty well with the MAP sensor. Furthermore, I'd expect any leak in the air-balance line to result in the actuator not getting _enough_ pressure to open the wastegate (which I think would result in over-boost, not the early wastegate opening I believe is happening). And, as a reminder, I can wire the wastegate shut and get at least 40" of MAP (I do not go full WOT with the wastegate in this state, for obvious reasons - I just carefully take it up to 40" to prove the engine can make more power, then back it down to idle).

I also want to point out that the factory has taken my turbo back to get the wastegate area re-machined. They also shipped me a second wastegate actuator to test against. And they've looked through several emails and photos and videos I've sent them.

...In the end, I really want to get to the bottom of this!

On the other hand, I will state that the engine runs strong and smooth, and I'm definitely getting MAP readings several inches above the ambient atmospheric pressure. So the engine is definitely getting _some_ boost. And since resolving my high oil pressure issue, the engine has had zero leaks or signs of trouble in ~2 hours of ground running (in careful ~10 minute increments). By all visible accounts, its healthy and strong. So I _am_ tempted to just finish off the airplane and go fly her...

My current plan of action is to:
1) Set up some GoPro cameras to see if I can synchronously film my throttle movements and the wastegate actuator - just to "prove" that its moving when I think it is (and that it isn't some other seal or issue causing a loss of boost at a certain point).
2) Try to borrow a Sensenich prop, as the differences between my Prince Prop and the factory's setup makes it impossible to do a perfect apples-to-apples comparison.
3) I may try to purchase a Turbosmart (or other brand) adjustable wastegate actuator, and see if I get different results from the brand supplied by the factory.
4) Get cracking on the remaining bits I need to do (powder-coat parts, make nonstructural fairings, get the airworthiness inspection rolling, etc). I've let this issue stall me out on the whole project for ~6 months and I need to get back to making progress towards flight.

Take care,

--Noel

[Brett]

I have a Sensenich cruise prop and will see about 43" at full noise. I rarely take off over 35" and if I do happen to go over 38" it isn't for long generally. Note though, I have never tied the plane down and tried to see what Map I can get...ever.

I have taken off plenty if times at 30" as I have a long runway and it doesn't really matter if I'm on the ground for a few more metres.

I think Sonex is right. You have been chasing this issue for ages but don't have a Sensenich prop installed. Myself, I'd be flying it and see if that Prince prop unwinds a bit when you're climbing out at 90kts and 900fpm. I'd personally really like to see the results of the prince prop as when I'm over 8000ft my prop could do with some extra bite... I really happy with it though.. I think the cruise prop is the perfect pick for what I really want and I'm happy Sonex recommend two options.

[lutorm]

I mean, if you get full MAP at partial throttle with the wastegate wired shut, but not with the existing spring at full throttle, there *must* exist a point in between where a strong enough spring gives you the correct MAP, no? There's no magic involved here. Maybe your spring is the wrong spec and is too weak?

[NWade]

For the curious:

I am now 100% convinced that my MAP issues are related to the actuator and not a sign of anything wrong with the engine. Why, you might ask? Well, because today I (carefully) put the "Experimental" in "experimental aviation":

The Kinugawa actuator that is supplied with the Aerovee Turbo kit is available in many "flavors" and options - including different strengths of spring (which delay the wastegate opening until higher pressures are reached). Based on my research, the spring supplied in the kit was likely a 0.3 bar spring. Recall that earlier this year Sonex kindly loaned me another actuator to make sure mine wasn't defective; but I still observed the same results. This lead me to believe that my actuator wasn't defective; but it was opening too early. After exhasuting all other potential engine issues I could think of, this week I decided to purchase the next available strength spring - a 0.5 bar unit - and see what would happen. If I still observed low MAP at full throttle, then I pretty much _knew_ that I had an engine problem; whereas if I got a higher MAP result then I could be pretty sure that the actuator is the ultimate source of all of my consternation this year.

The blue spring arrived late in the week and installation was fairly easy (although care is needed in positioning the silicone diaphragm during re-assembly of the unit). I wheeled the aircraft outside the workshop, tied it down, started her up, and idled for a few minutes to get everything up to temp. A systems-check at 1800 RPM revealed no problems so I dropped to idle, went full rich, and began advancing the throttle.

Initially the engine responded just as before, and I was used to the boost rising slowly/linearly as I approached full throttle. So in this engine run I pushed the throttle smoothly towards the firewall and was shocked when the rising MAP accelerated as the boost began to wind up! My eyes got big and I quickly pulled the throttle back some - dropping the MAP to 32". I let out my breath and slowly inched the throttle forward, stopping as it climbed up to 41". Worried about the engine getting too hot, I pulled her smoothly back to idle and then shut her off after a couple of minutes.

You can see the RPM & MAP "double-peak" as a result of my actions in this screenshot (from the MGL data recorder):
https://flic.kr/p/29G2DU7

Given that I had throttle-travel left over, I am a little concerned that this 0.5 bar spring will over-boost the engine. I don't want to be "heads-down" and searching for the right throttle position at moments I'm looking for full power, so I'm considering a few courses of action:

1. Keep the stronger spring installed but reduce the pre-load to see if the MAP will plateau at a reasonable value (~42"). I had 1 turn of pre-load on the actuator arm for this run, but maybe zero preload is all I need to ensure that the gate begins moving and doesn't let the turbo shoot past a reasonable maximum MAP...
2. Put the original spring back in and find some appropriately-sized washers to shim the spring (pre-loading/compressing it more than the actuator arm allows, to delay its opening of the wastegate)
3. Purchasing an off-the-shelf "white" 0.3 bar spring from Kinugawa. The "stock" spring inside my actuator doesn't quite match the images of a Kinugawa 0.3 bar "white" spring so I'm curious to hold them side-by-side and see if there's any difference. Perhaps the 0.3 bar spring is actually strong than what comes by default in some of the actuators? Or perhaps there's some variance in the product-line from unit to unit?

Regardless of what I do, I feel a weight off my shoulders that's been a psychological block on my progress all year. I am getting back to the build and hope to be flying around the end of 2018!

NOTE: For anyone that follows in my footsteps, please keep in mind that what I am doing is non-standard and not endorsed by anyone. There are very good reasons not to put a stronger spring in your wastegate actuator, and given that a non-turbo Aerovee can power a Sonex no problem, the ~38" MAP I was seeing with the stock setup is more than enough power for a fine-flying aircraft.

Take care,

--Noel