What's new
Van's Air Force

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

Engine Sweet Spot

Look up the difference between standard Bendix-style nozzles and turbo nozzles. Short version...

As Mark noted, a small amount of air is entrained in the fuel flow inside the nozzle. The air comes from a small hole in the side of the nozzle body, under a shroud and screen.

Standard nozzles take bleed air from the surrounding atmosphere, meaning from the upper cooling plenum. The deltaP across the nozzle is thus subject to cooling air pressure recovery, an area where some folks do better than others (see CR 3405).

Add a turbocharger, and MP becomes higher than upper plenum pressure. So, turbo injectors plumb all the nozzle air bleeds to an air supply rail (generally a steel tube) which is fed by a pressure tap just downstream of the compressor outlet. That makes bleed pressure again higher than MP at the intake port, as the induction system suffers some pressure loss between the compressor and the intake valves.
Dan,
Much appreciate the explanation, the light bulb is going on and getting brighter and brighter. I am really enjoying and learning from this topic and conversation, sadly don't have much to contribute.
 
I look at that drain every time I?m under the plane and think the same thing.

I wonder if this is a real problem - if so, put the drain in clear air and point it forward. I would think it would ?leak? less that way?

I am tweaking that exact area now - maybe I can do something to let that outlet see clear air...

Yes, a pitot tube on the outlet would solve the problem, wouldn't it.
 
Dan,
Much appreciate the explanation, the light bulb is going on and getting brighter and brighter.

This should fire up the 'ole rotating beacon:

https://airflowperformance.com/wp-content/uploads/2016/11/INJECTOR-NOZZLES.pdf

Lots of ram air rail photos in Appendix F:

https://airflowperformance.com/wp-content/uploads/2018/12/API-Installation-Service-Manual-9003.pdf

Dan, can you clarify and expand on this a little? I have a direct (unfiltered) ram-air intake to a horizontal induction sump. I have roughly zero GAMI spread (certainly within 0.15 GPH) , and at my typical 9.5K-10.5K cruise altitudes, I can lean quite a lot (more than 50F LOP, often70--80F LOP) and stay smooth. I do get a lot of fuel staining. Would I benefit from a turbo-type pressure rail system?

...and I wrote:

I think you would, but the difference will be small. A real factor here is not easily measurable....perception of smoothness, and the point at which it is unacceptable. Perception and acceptance seem to vary a lot.

Yesterday I had 6.5 stick time with nothing to do (MGM to OLV, OLV to MGM, MGM to CXY), so I played with alternate settings, as well as verifying advance values for the 390 at 50 LOP. Also thought about your question. I should try to illustrate further.

For me, there two kinds of LOP roughness. The first might be described as "not quite as smooth as best power". There's nothing really objectionable about it. I'm not claiming to be some sort of engine whisperer, but I've sat behind the same set of spinning parts for 850 hours, and there is a difference. How significant? Here's the telltale. I had another pilot in the back seat for the OLV-MGM leg. After settling into cruise and allowing time for boredom to set in, I transitioned from peak to 50 LOP...then asked my buddy if he had noticed any engine change during the previous five minutes. He said no, none.

I'll describe the other kind as a hiccup, a little miss, a shiver in the seat, variable in severity. With mixture set at some reasonable point LOP (let's say 50 for example), it is largely random, a millisecond shake in an otherwise normal string of combustion events. Frequency might be an irregular series, or average once in 30 seconds, or perhaps just a a few in 10 minutes. I don't like a miss, because I think it's hard on pendulum bushings, prop hubs, etc.

Down low (like 5K or less) we typically run with the throttle plate partially closed. Nozzle bleed deltaP is high. Make the deltaP very high (1000 feet, half throttle loafing, for example), and we can pull to a very lean mixture without real roughness; nozzle bleed is doing a fine job of entraining air into the little pencil stream of fuel squirting from the nozzle.

WOT at altitude is a different story; deltaP is low. However, measured with ram air feeding turbo nozzles, it is positive throughout 720 degrees of crank rotation, with one short exception which I don't think results in reverse flow. The key point is that scanning across quite a few cycles, the deltaP is consistent; the plots don't change.

That's the thing...ram air supply is relatively constant. Without the rails, ambient pressure at each injector may vary a lot due to turbulent motion in the upper plenum; when I made measurements using a manometer and an aquarium rock strapped to standard nozzles, the indication was not steady. Highly variable bleed air supply pressure would make bleed air deltaP a lot more random. I suspect it's the root cause of the aforementioned random misfire, and why you might wish to try a set of turbo injectors and rails.

In summary, I'd say (subjectively!) the rails did not make a difference in my perception of the benign roughness, the "not as smooth as best power" kind which my passenger could not perceive. They did make a notable reduction in the frequency of those little misfires. Previously 50 LOP was not entirely clear of them, and now it 's good. Will it do the same for you? That I do not know, because I cannot predict turbulence in the vicinity of your nozzles. We have different cowls.

Does that help?
.
 
Last edited:
Dan,
Do you see the glowing light on my head!!!!
Thank you very much as your earlier explanation coupled with these document and pictures have really helped and now I can understand what I read here.
 
Yes, it helps, thanks.
I too notice both forms of 'roughness'. I don't mind the benign slight loss of smoothness. I do find the occasional random hiccup to be very disconcerting, and I richen just enough to make them go away completely. Usually that is still 50--60F LOP for me. But not always. And the 'not always' gives me pause - makes me start thinking about what could have changed. So this whole line of thought is much appreciated. I'm not sure I will go to the trouble of installing turbo injectors and rails, but I will think about it.

...and I wrote:

I think you would, but the difference will be small. A real factor here is not easily measurable....perception of smoothness, and the point at which it is unacceptable. Perception and acceptance seem to vary a lot.

Yesterday I had 6.5 stick time with nothing to do (MGM to OLV, OLV to MGM, MGM to CXY), so I played with alternate settings, as well as verifying advance values for the 390 at 50 LOP. Also thought about your question. I should try to illustrate further.

For me, there two kinds of LOP roughness. The first might be described as "not quite as smooth as best power". There's nothing really objectionable about it. I'm not claiming to be some sort of engine whisperer, but I've sat behind the same set of spinning parts for 850 hours, and there is a difference. How significant? Here's the telltale. I had another pilot in the back seat for the OLV-MGM leg. After settling into cruise and allowing time for boredom to set in, I transitioned from peak to 50 LOP...then asked my buddy if he had noticed any engine change during the previous five minutes. He said no, none.

I'll describe the other kind as a hiccup, a little miss, a shiver in the seat, variable in severity. With mixture set at some reasonable point LOP (let's say 50 for example), it is largely random, a millisecond shake in an otherwise normal string of combustion events. Frequency might be an irregular series, or average once in 30 seconds, or perhaps just a a few in 10 minutes. I don't like a miss, because I think it's hard on pendulum bushings, prop hubs, etc.

Down low (like 5K or less) we typically run with the throttle plate partially closed. Nozzle bleed deltaP is high. Make the deltaP very high (1000 feet, half throttle loafing, for example), and we can pull to a very lean mixture without real roughness; nozzle bleed is doing a fine job of entraining air into the little pencil stream of fuel squirting from the nozzle.

WOT at altitude is a different story; deltaP is low. However, measured with ram air feeding turbo nozzles, it is positive throughout 720 degrees of crank rotation, with one short exception which I don't think results in reverse flow. The key point is that scanning across quite a few cycles, the deltaP is consistent; the plots don't change.

That's the thing...ram air supply is relatively constant. Without the rails, ambient pressure at each injector may vary a lot due to turbulent motion in the upper plenum; when I made measurements using a manometer and an aquarium rock strapped to standard nozzles, the indication was not steady. Highly variable bleed air supply pressure would make bleed air deltaP a lot more random. I suspect it's the root cause of the aforementioned random misfire, and why you might wish to try a set of turbo injectors and rails.

In summary, I'd say (subjectively!) the rails did not make a difference in my perception of the benign roughness, the "not as smooth as best power" kind which my passenger could not perceive. They did make a notable reduction in the frequency of those little misfires. Previously 50 LOP was not entirely clear of them, and now it 's good. Will it do the same for you? That I do not know, because I cannot predict turbulence in the vicinity of your nozzles. We have different cowls.

Does that help?
.
 
Back
Top