andrew phillips

Well Known Member
Airflow inj cable routing solution

It seems a few builders are stuggling with the AFP cable system on forward facing sumps. Instead of longer cables or a bell crank system I opted to modify the length of the control arms on the metering valve body to prevent interference then it was a simple matter of fabing a bracket to the block and having a straight run of cable from just below the heater box in the cabin straight to the fuel injection system. If anybody sees anything wrong with this approach I would appreciate hearing about it.


 
Last edited:
Andrew,
Looks like you shortened the arms a little. It will shorten knob travel at the panel and make movement a bit more sensitive, but no big deal.

Details. Flip the AN3's over so the nuts are on top. In this application the bolt is a cantilevered beam. The highest stress is at the base of the beam, in this case at the control arms. You never want bending stress at or near bolt threads. Yes, forces are low and 1000's are probably flying the way you have it now, but it never hurts to pay attention to structure fundamentals.

Add a 970 washer above the rod end; it retains the rod end even if the ball retention fails.

When you have the bolts loose in the arms, retract the throttle cable and make sure you're not running out of available angle where the push-pull rod comes out of the sheath at the bracket.

The cable bracket badly lacks edge distance at the cable holes; throw it away and start over. Personal opinion; I think engine control brackets should be steel and somewhat overbuilt. The bracket supports the cable mass, and is subject to high-cycle vibration. Aluminum has no knee in it's S-N curve, so we should always be cautious about using it in an application where the total number of cycles builds very quickly. Assume 2400 RPM x 2 shakes per rev and you have 288,000 cycles per flight hour. The arbitrary fatigue limit for aluminum is usually set at 10,000,000 cycles; you're there in 35 hours and fatigue strength is still falling.