Mailing List flyrotary@lancaironline.net Message #11010
From: Al Gietzen <ALVentures@cox.net>
Subject: RE: [FlyRotary] Re: MAP filter
Date: Wed, 1 Sep 2004 17:45:55 -0700
To: 'Rotary motors in aircraft' <flyrotary@lancaironline.net>

>>

>

> Al, been there, doing that with the RC thing - .04"/1mm and a cheap

> generic plastic fuel filter work very well at all RPM and load ranges. I

> can send you some datalogs if you wish. If the orifice is too small, the

> ECU has to wait a few tenths of a second for an accurate MAP signal.

> This causes a short period of enleanment and a momentary stumble. It's

> actually the flow in and out of the capacitance that you're restricting,

> not the extremely miniscule flow that the sensor diaphragm generates, if

> any. I think most MAP sensors are strain gauges. You want

> Manifold->R->C->MAP sensor as your sequence of parts. I've spent many

> hours getting this right by trial and error and have extremely smooth

> sensor response even at .01 second data logging intervals at 800 rpm idle.

> -Mike

>

Can’t argue with success. Yes, you would not want a very small

 restrictor upstream from the accumulator. My logic was telling me that

 the most effective pulse dampening with the least effect on response

 time would be a smaller restrictor on the controller side. But logic

 doesn’t always match reality, and I suppose that option was one of

 your trials.

 

What application and ECU are you using? Do you suppose it is

independent of those things?

>

I'm running a streetported 12A rotary in an '83 RX-7 using the

MegaSquirt fuel only ECU, Weber Alpha 2x50mm bodies (very similar to the

TWM bodies) w/ an approximately 19" tract from the rotor face in each

port to the end of the TB intake bell. I'm no EE but from what I can

tell, most pressure sensors up to the job of reporting MAP are pretty

much the same as far as operating theory. A Motorola MPX4250 is the

sensor used in my app. Another thing I'd like to mention is that I also

have approximately 6' of 1/4" line from the manifold to the RC/MAP

sensor combo so there's definitely some extra volume from the hose run.

 

This could be quite a difference – ¼” i.d that long is about 3.5 cu. in. upstream from the restrictor.  How big is the filter?

 

I believe that each application/install has it's variables and no one

solution will be perfect for everybody though we can all learn from each

other.  Well said.

 

My setup had a short run of about 3’ if 1/8” i.d. line, and the running was rough.  So far the best results I’ve had was on the first quick try where I teed together the A & B controller lines with a 5/32 plastic T, then added in about 18” of 5/16” i.d. vinyl tube from there to the sensor.  The T is a restrictor of sorts because it has an i.d. of about .065.  So it was line, restrictor, volume (approx. 1.4 cu in.).  Unfortunately, the test was very short because I was also measuring oil drain from the redrive, and the engine had not warmed up.  But I had about 15 seconds of completely smooth running at 1600 rpm before shutting down.  I was thrilled because up that point I had nothing but roughness.

 

I bought a small plastic fuel filter, made an orifice in one end by filling the inlet with about ¾” of epoxy microglass, and drilling it out to .063 (smallest drill I had). The filter has roughly 1 cu. in. of volume.  I tried this in place of the 18” of vinyl tube with the orifice toward the manifold.  This was better than no filter, but still rough.  I then reversed it placing the orifice on the end toward the sensor.  Seemed to run a little smoother, but at little lean of optimum mixture it got into kind of a slow surging behavior which I had not experienced before.  Still no joy.

 

This may all be a bit meaningless since during these runs the engine was warming up.  When I went back to the first 5/16” tube setup I could not get complete smoothness at 1600; maybe because this time the engine was warm.

 

It’s a guessing game, and what is needed is an oscilloscope on the MAP sensor output.

 

Al

Subscribe (FEED) Subscribe (DIGEST) Subscribe (INDEX) Unsubscribe Mail to Listmaster