REFLECTOR: Balanced Injectors

[Chuck Jensen]

A few weeks ago, I posted a request for a recommendation on getting injectors in an IO-540 balanced.  I related that I had done the GAMI routine, but after a year, the lowest spread achieved was 1.5 gph.  One of the best, if still not great, settings after multiple GAMI injectors change outs was to run #3 and #6 LOP and the other four ROP.  Not ideal but it kind of worked.  When pressed about why the tuned injectors weren't tuned after innumerable lean test runs and injector change outs, GAMI said "huh, that's odd" but never solved the problem.
 
Following the suggestion of Ronnie and a couple others, I contacted Air Flow Performance in Spartanburg, SC.  After a couple missed appointments due to schedule and weather, we hooked up on Saturday (yes, they volunteered to come in on Saturday to accommodate my schedule).  From the baseline flight of 1.5 gph spread, the next set of restrictors dropped it to 1.2, then 0.8, then less than 0.4--done!  
 
In 4-5 hours, I'd resolved the issue that I'd been piddling with for over a year.  Now, in all fairness to GAMI, they were very patient and persistent.  Over the year, they sent a LOT of injectors.  Of course, each required cowl removal, injector change out, then a new set of lean tests, just to find out they really didn't change anything.
 
After the second flight at Airflow, a 5 minute explanation finally made sense of why the GAMI injector change outs never changed anything.  It turns out the Flow Divider opening pressure is around 1.5 PSI.  Since our test points were around 12 GPH that resulted in a nozzle back pressure of around 1.7 PSI.  What was happening was that even though a larger nozzle was being installed in cylinder #6 the Flow Divider was influencing the division of flow to the nozzles.  The solution was counter intuitive, but eminently sensible when understood.  To make #6 rich, the other cylinders were 'cinched down' with smaller restrictors, which caused the nozzle back pressure to increase driving the Flow Divider more open so that it would not influence the division of flow to the nozzles.  With more fuel being pushed to #6 and they dialed right in.
 
Now, I assume GAMI is probably that smart also, but the advantages of hands-on problem analysis and immediate retesting are inestimable, something that GAMI, due to geography couldn't do.
 
More important than the 0.5-0.8 gph fuel savings was the smoothness factor.  The canard, which is an excellent tell-tale for vibration, was dead calm throughout the ROP area and displayed only a small vibration when I went into the LOP region.  That small vibration vibration is still less than anything I'd achieved with the GAMI injectors in either ROP/LOP operations.
 
Finally, I note there is a difference between GAMI and Airflow for making injector adjustments.  GAMI changes out the whole injector whereas Airflow uses an injector body, then they just swaps out the restrictor inserts, which only requires removal of the fuel line nut, relatively easy to do even when we have to work around/through the Velo plenum.  With the GAMIs, I was never enthused about screwing multiple injectors into the cylinders. The odds of cross-threading and screwing up the injector or cylinder is low, but still there.
 
Whether you bring the plane to Spartanburg (SPA) or do it remotely, I'm guessing you'll come away smarter and with as good or better results than from some of their competitors.  The office walls are plastered with all the show pilots they tune, including Patti Wagstaff, Tucker, et al.
 
And the best part; the total cost for the Airflow injector balancing, including new injectors, was about ½ the cost of the GAMIs...and I didn't have to do all the work.
 
So, as you might guess, I'm sold.  Without reservation on cost, service or results, Airflow gets my vote.  
 
Chuck
 
Disclosure:  I don't own or have an interest in Airflow...but I wish I did.
  
  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.tvbf.org/mailman/private/reflector/attachments/20050815/e09cc253/attachment.htm