What's new
Van's Air Force

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

Fuel Stain around Injector #2...

bjdecker

Well Known Member
Ambassador
...and a possible solution...

On the RV-7s with IO-360-A1B6 (200HP/Angle Valve) that I've built and operated, I've always seen a residual blue stain around the injector in the #2 cylinder (Pilot side, front).

The fuel injectors are "clocked" correctly (per Lycoming SI 1275C), so there is a bit of a mystery as to what's going on. I recall seeing a discussion on VAF about the incoming airflow messing up the injector breather dynamics - but couldn't find it.

During the course of building an RV-14, I noticed that Van's has changed the the layout/shape of the #2 air dam/temperature riser. So I figured what the heck -- couldn't hurt. (picture attached)

Question for the hive mind - has anyone else seen this issue? Has anyone changed the shape of the riser to deflect air around/over the injector body? Scott McD -- was this reason for change in the RV-14?

Cheers!
 

Attachments

  • IMG_4243.jpg
    IMG_4243.jpg
    410 KB · Views: 292
For those not up to speed regarding nozzles and restrictor theory...

Fuel flows through the restrictor and across an air gap into a larger diameter passage in the base of the nozzle. The nozzle has an air bleed hole. It's under the external shroud and dirt screen. During most of 4 stroke cycle (720 degrees of crank rotation), the flowing fuel entrains bleed air supplied to an annular space around the tip of the restrictor. The result entering the intake port is a fuel/air froth, in particular at low flow rates.

Note I wrote "during most of the cycle". Due to wave dynamics in the intake tract and the effect of the closing intake valve, there are short periods during which pressure inside the intake port is higher than external pressure in the plenum above the engine. The result may be some very minor fuel flow out through the bleed passage. Theory says it's also possible to temporarily lower the external pressure in the vicinity of the bleed shroud, due to flow effects, blade passing, or turbulence. This may increase the duration of reverse flow. Some users have installed turbo nozzles, in which the bleed air source is piped to the nozzle, thus local pressure in the immediate vicinity of the shroud has no bearing.

To be strict, another cause can be a partially blocked nozzle, (not restrictor) although that's pretty rare.

I had not noticed the additional tab in front of the #2 nozzle (Brian's photo), so I too am curious how it came about.

Back to regular programming...
.
 

Attachments

  • Nozzle Detail.jpg
    Nozzle Detail.jpg
    97 KB · Views: 216
  • Nozzle Diagram.jpg
    Nozzle Diagram.jpg
    28.6 KB · Views: 164
  • Turbo Shroud.jpg
    Turbo Shroud.jpg
    116 KB · Views: 224
Last edited:
I have contemplated a dam like the one pictured to avoid massive bug strike to injector/tube.
 
I don't believe the dam will prevent this as I have seen fuel stain on all my four cylinder. Airflow performance suggest to put the bleed hole towards the bottom , if possible, but clocking it that way is not easy especially that you might break it by torquing it too much and that will become a major issue.
 
Service Instruction

I don't believe the dam will prevent this as I have seen fuel stain on all my four cylinder. Airflow performance suggest to put the bleed hole towards the bottom , if possible, but clocking it that way is not easy especially that you might break it by torquing it too much and that will become a major issue.

Here's the link to the Lycoming SI: https://www.lycoming.com/content/service-instruction-no-1275c

...here's the excerpt that describes the proper clocking of the injector:

LW-18265: In normally aspirated engines where the nozzles, P/N LW*18265 (see
Figure 2), are installed horizontally, particular attention must be paid to the
identification marks stamped on one of the hex flats on the nozzle body. This mark
is located 180° from the air bleed hole and must appear in the lower side of the
nozzle to assure that the air*bleed hole is on top in order to reduce fuel bleeding
from this opening just after shutdown. To ensure nozzle is correctly torqued,
tighten the nozzle to 60 in.*lbs. torque. Then continue to tighten until the letter or
number stamped on the hex of the nozzle body points downward.


From the pictures in DanH's post - I can see how airflow directly across the air bleed hole could create a low pressure zone and draw fuel out of the injector. It would seem to explain the absence of fuel stains on #1,3 & 4.
We'll see how the air dam changes this.

Scott McDaniels (rvbuilder2002) -- any insight into the design changes for this between the RV-7/8, RV-10 and the RV-14?

Cheers!
 
From the pictures in DanH's post - I can see how airflow directly across the air bleed hole could create a low pressure zone and draw fuel out of the injector. It would seem to explain the absence of fuel stains on #1,3 & 4.
We'll see how the air dam changes this.

Cheers!

I don't see that happening. The fuel is coming out of the restrictor at a minimum of 2 PSI and therefore at a decent velocity. I don't know the physics calculations, but it would seem like a pretty significant vacuum would be required to suck fuel out of the bleed hole. I believe that fuel only comes out of the bleed hole at shut down, as discussed in the details you quoted.

At shut down, the fuel in the 1/8 lines almost immediately starts boiling and the pressure caused by this pushes the fuel past the restrictor, as it can't go in the other diretion. This time there is no pressure, so it goes where gravity takes it.

Also, you won't get much air flow in that area due to the wall behind it. It will be mostly swirling, turbulent air and not a typical laminar flow over a smooth surface that could create a low pressure.

Larry
 
Last edited:
I don't see that happening. The fuel is coming out of the restrictor at a minimum of 2 PSI and therefore at a decent velocity.

The flow divider spring may be calibrated for 2 psi. However, given standard 0.028" restrictors, nozzle pressure doesn't reach 2 psi until roughly 8 gallons per hour. See the Lycoming chart.

I don't know the physics calculations, but it would seem like a pretty significant vacuum would be required to suck fuel out of the bleed hole. I believe that fuel only comes out of the bleed hole at shut down, as discussed in the details you quoted.

Generally true, although I have seen a video of a Continental spraying fuel from the bleed, in flight. Probably had a plugged nozzle.

At shut down, the fuel in the 1/8 lines almost immediately starts boiling and the pressure caused by this pushes the fuel past the restrictor, as it can't go in the other diretion. This time there is no pressure, so it goes where gravity takes it.

Which is why we point the bleed hole up.

Also, you won't get much air flow in that area due to the wall behind it. It will be mostly swirling, turbulent air and not a typical laminar flow over a smooth surface that could create a low pressure.

There are variations in local pressure at the four (or six) nozzle locations. We're all familiar with the idea that cylinders don't necessarily get the same cooling mass flow, and that's all about local pressure. I placed a bubble rock next to each nozzle and recorded some numbers a while back. Note that #2 was actually the highest local pressure.

Each bubble rock value is an average over a time period well in excess of a 720 degree cycle. Instantaneous pressure is highly variable, so deltaP across the nozzle bleed varies continuously during the cycle. I have those measurements too. Requires a fast differential pressure sensor and laptop recording.
 

Attachments

  • Plumbing.jpg
    Plumbing.jpg
    203.2 KB · Views: 236
  • Pressure Tap Data.jpg
    Pressure Tap Data.jpg
    22 KB · Views: 130
  • Scan_0001.jpg
    Scan_0001.jpg
    371.9 KB · Views: 153
Last edited:
... I placed a bubble rock next to each nozzle and recorded some numbers a while back. Note that #2 was actually the highest local pressure.

Each bubble rock value is an average over a time period well in excess of a 720 degree cycle. Instantaneous pressure is highly variable, so deltaP across the nozzle bleed varies continuously during the cycle. I have those measurements too. Requires a fast differential pressure sensor and laptop recording.

"Bubble rock" took me a while but after staring at the photo I figured it out - very clever, Dan!
 
Here's the link to the Lycoming SI: https://www.lycoming.com/content/service-instruction-no-1275c

...here's the excerpt that describes the proper clocking of the injector:

LW-18265: In normally aspirated engines where the nozzles, P/N LW*18265 (see
Figure 2), are installed horizontally, particular attention must be paid to the
identification marks stamped on one of the hex flats on the nozzle body. This mark
is located 180° from the air bleed hole and must appear in the lower side of the
nozzle to assure that the air*bleed hole is on top in order to reduce fuel bleeding
from this opening just after shutdown. To ensure nozzle is correctly torqued,
tighten the nozzle to 60 in.*lbs. torque. Then continue to tighten until the letter or
number stamped on the hex of the nozzle body points downward.


From the pictures in DanH's post - I can see how airflow directly across the air bleed hole could create a low pressure zone and draw fuel out of the injector. It would seem to explain the absence of fuel stains on #1,3 & 4.
We'll see how the air dam changes this.

Scott McDaniels (rvbuilder2002) -- any insight into the design changes for this between the RV-7/8, RV-10 and the RV-14?

Cheers!

The tab was added to the left front dam to shield the injector body from the dynamic flow it was exposed to, during early flight testing of the prototype RV-14, in order to alleviate fuel staining like you have seen.
It seemed to work so it was incorporated into the new generation baffle kit for the RV-14.
No other detailed pressure testing similar to what Dan has done was done, because it provided the desired result.
I think it was never rolled back into any of the other baffle kits because the RV-14 was the only model we had ever noticed this on any of our prototypes.
 
Back
Top