Return-Path: Sender: "Marvin Kaye" To: lml@lancaironline.net Date: Tue, 23 Nov 2004 14:27:50 -0500 Message-ID: X-Original-Return-Path: Received: from imo-m19.mx.aol.com ([64.12.137.11] verified) by logan.com (CommuniGate Pro SMTP 4.2.5) with ESMTP id 543386 for lml@lancaironline.net; Tue, 23 Nov 2004 13:32:05 -0500 Received-SPF: pass receiver=logan.com; client-ip=64.12.137.11; envelope-from=Sky2high@aol.com Received: from Sky2high@aol.com by imo-m19.mx.aol.com (mail_out_v37_r3.8.) id q.9a.1a450096 (3972); Tue, 23 Nov 2004 13:31:27 -0500 (EST) From: Sky2high@aol.com X-Original-Message-ID: <9a.1a450096.2ed4dbff@aol.com> X-Original-Date: Tue, 23 Nov 2004 13:31:27 EST Subject: Conclusion to S-Tec 50 Auto-Pilot Pitch Problems X-Original-To: lml@lancaironline.net X-Original-CC: efoxfield@sbcglobal.net MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="-----------------------------1101234687" X-Mailer: AOL Strauss Beta Client sub 1180 -------------------------------1101234687 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit You may remember: <<<<<<<<<<<>>>>>>>>>>>>>>>> (the originating message is at the end of this missive) More importantly, the sucking and blowing produced voltages of the same sign at the servo. So......................... With agreement from S-Tec, I sent my AP computer in once again. After 10 days and two calls later, I was informed that S-Tec tests showed the computer to be operating properly. In a state of exasperation, voice moving up an octave and several Db in volume, I went thru all of the problems over the last few years, including the prior $200 S-Tec ck that said it was OK and our recent tests while the tech was on the phone with me (it is important to stay with the same person). I said, "don't bother to send it back without it being fixed," as I reminded him of his participation in the tests with me. After having his memory jogged, he said that he would have the pitch control boards replaced and would do so under warranty. Yesterday I had a chance to install and test fly the AP with the new pitch boards. After ensuring I had fine tuned the roll computer, I was off into the mild blue yonder. All navigation functions were first checked and then Alt Hold was engaged. Remember that elevator neutral now had the servo set with the push-rod connector at 12 o'clock - max control sensitivity. One minor oscillation and the altitude was held within 50 feet. The stick was bumped, pushed/pulled and the proper trim lights came on, afterwards always returning to the "held" altitude. I even mis-trimmed the elevator and engaged hold resulting in the AP settling down after a few minor oscillations to the desired altitude. This is the best that altitude hold has ever worked. I will need no adjustments to the "sensitivity" unless the AP gets weird in turbulence, a function yet to be checked. Conclusions: 1. There are serious problems with S-Tec's ability to check the proper performance of their autopilots. They have been informed of this and whether anything comes of it remains to be seen. In this case, they were unable to detect or note the sign (controls motor direction) of the voltage sent to the pitch servo! 2. I believe that the pitch function was deteriorating over time - eventually destroying the original servo motor ($600 to replace). This was probably caused by randomly signed short burst low voltages being sent to the motor even though "hold" was not engaged. Note that the motor can be operated independently of hold having been selected. Heaven knows what the computer was sending during the periods when hold was activated but controlled the airplane poorly. Hmmmmm, now that I think about it, maybe the motor wasn't destroyed -- If only the boards had been fixed before I replaced the motor. Of course, I had to disassemble the motor and gears - a curious mind must sometimes de-construct to learn. 3. The old board (circa 1992) may have had inherent pitch sensitivity problems. S-Tec noted that there have been several re-designs of that function. 4. Persistence pays off (to whom I don't yet know). $1600, 3 months, numerous test flights, many "service" hours in the hangar, numerous phone calls, special test equipment and an understanding tech finally got this thing operational. Or, at least ready for a sale so I can get a Tru-Trak. 5. Note that I have a amateur-built break-out box suitable for testing 50-pin D-sub connections (9, 15 and 25 too). Fly-in sometime and we can check your broken electrical stuff. Scott Krueger AKA Grayhawk N92EX IO320 Aurora, IL (KARR) Some Assembly Required Using Common Hand Tools UPS and a Telephone. <<<<<<<<<<<<<>>>>>>>>>>>>>>>>>>>>>>>> -------------------------------1101234687 Content-Type: text/html; charset="US-ASCII" Content-Transfer-Encoding: quoted-printable
You may remember:
 
<<<<<<<<<<<<sent 10-22-04
 
A conversation with S-Tec while sitting in the plane at the hangar will= =20 result in my performing some ground tests - but, note this -
 
Engaging hold on the ground sometimes resulted the stick moving forward= .=20 While talking, I saw the trim lights (randomly) come on and a stick bump wou= ld=20 turn them off. 
 
Sucking and blowing on the sensor generally resulted in down elevator=20 movement???????????
 
More tests today.
 
I will mention the "sensitivity" fix to S-Tec tech.  This is proba= bly=20 a good solution to alt hold oscillations, but there is something seriously w= rong=20 with my A/P even though the pitch servo motor is new, the sensor is new and=20= my=20 computer was checked by S-Tec within the last 30 days.
 
>>>>>>>>>>>>>>>>> (t= he=20 originating message is at the end of this missive)
 
More importantly, the sucking and blowing produced voltages of the same= =20 sign at the servo.  So.........................
 
With agreement from S-Tec, I sent my AP computer in once again.
 
After 10 days and two calls later, I was informed that S-Tec tests show= ed=20 the computer to be operating properly.  In a state of exasperation, voi= ce=20 moving up an octave and several Db in volume, I went thru all of the problem= s=20 over the last few years, including the prior $200 S-Tec ck that said it was=20= OK=20 and our recent tests while the tech was on the phone with me (it is=20 important to stay with the same person).  I said, "don't bother to send= it=20 back without it being fixed," as I reminded him of his participation in the=20 tests with me.  After having his memory jogged, he said that he would h= ave=20 the pitch control boards replaced and would do so under warranty.
 
Yesterday I had a chance to install and test fly the AP with the n= ew=20 pitch boards.  After ensuring I had fine tuned the roll computer, I was= off=20 into the mild blue yonder.  All navigation functions were first=20 checked and then Alt Hold was engaged.  Remember that elevator neutral=20= now=20 had the servo set with the push-rod connector at 12 o'clock - max control=20 sensitivity. One minor oscillation and the altitude was held within 50=20 feet.  The stick was bumped, pushed/pulled and the proper trim lights c= ame=20 on, afterwards always returning to the "held" altitude.  I even mis-tri= mmed=20 the elevator and engaged hold resulting in the AP settling down after a few=20 minor oscillations to the desired altitude.  This is the best=20 that altitude hold has ever worked.  I will need no adjustments to= the=20 "sensitivity" unless the AP gets weird in turbulence, a function yet to be=20 checked.
 
Conclusions:
 
1.  There are serious problems with S-Tec's ability to check the=20 proper performance of their autopilots.  They have been informed of thi= s=20 and whether anything comes of it remains to be seen.  In this case, the= y=20 were unable to detect or note the sign (controls motor direction) of the vol= tage=20 sent to the pitch servo!
 
2. I believe that the pitch function was deteriorating over time -=20 eventually destroying the original servo motor ($600 to replace).  This= was=20 probably caused by randomly signed short burst low voltages being sent to th= e=20 motor even though "hold" was not engaged.  Note that the motor can= be=20 operated independently of hold having been selected.  Heaven knows= =20 what the computer was sending during the periods when hold=20 was activated but controlled the airplane poorly.  Hmmmmm, now tha= t I=20 think about it, maybe the motor wasn't destroyed -- If only the boards had b= een=20 fixed before I replaced the motor.  Of course, I had to disassemble the= =20 motor and gears - a curious mind must sometimes de-construct to learn.
 
3. The old board (circa 1992) may have had inherent pitch sensitivity=20 problems.  S-Tec noted that there have been several re-designs of that=20 function.
 
4. Persistence pays off (to whom I don't yet know). $1600, 3 months,=20 numerous test flights, many "service" hours in the hangar, numerous phone ca= lls,=20 special test equipment and an understanding tech finally got this thing=20 operational.  Or, at least ready for a sale so I can get a Tru-Trak.
 
5. Note that I have a amateur-built break-out box suitable for tes= ting=20 50-pin D-sub connections (9, 15 and 25 too).  Fly-in sometime and we ca= n=20 check your broken electrical stuff.=20
 
Scott Krueger=20 AKA Grayhawk
N92EX IO320 Aurora, IL (KARR)

Some Assembly Required=20
Using Common Hand Tools
UPS and a Telephone.
 
<<<<<<<<<<<<<<Original messag= e=20 10-19-04
 
I have an S-Tec 50 (circa 1992) that has been updated over time and=20 utilizes GPSS via a G430.  I have been having altitude hold problems al= l=20 summer and so far this fall.=20
 
But first, the main connector to the computer is locked together with a= =20 spring-loaded latching mechanism.  I have had connectors apart many tim= e=20 and have installed my breakout box between the cable and the computer and I=20= have=20 not seen a problem on the ground.  Colyn and Gary, I hope that the=20 connectors are not the problem as other A/P functions are fine.
 
Here's where I am at:
 
1.  The first thing I did was to clean the brushes and armature on= the=20 pitch servo motor.  It ran fine on the ground and in place although it=20 sounded different when run in opposite directions.  It could be heard=20 running from about 3 to 36 volts.  Higher voltage (up to 40 volts) is u= sed=20 by the local avionics shop (JA) to clean the buildup of carbon and I have be= en=20 told this is good for about a year.  Physically cleaning is a longer te= rm=20 solution although I was recently told by S-Tec that the FAA does not allow S= -Tec=20 to "overhaul" or service these motors - they must replace them with new beca= use=20 they have been known to jam from a brush blocking the armature or a short ac= ross=20 armature segments.  The motor should turn slowly at very low=20 voltages.  The tech also told me that the brushes can be "hardened" aga= inst=20 chipping by heating them with a hair dryer until they smoke a bit, thus gett= ing=20 all the moisture out. (I have not done this because ????).  Anyway, the= =20 next flight test provided no control of altitude - that is, hand trimming th= e=20 elevator was the only way to obtain relatively level flight.
 
2.  There is no way to fully check out the system on the ground=20 because the accelerometer affects elevator management also.  Thus, I=20 suspected the pressure sensor.  Previously I had removed the pressure=20 sensor from the static system thinking that might be a problem.  Using=20 cabin pressure made no difference.  Note that I have carried a spare=20 altimeter aloft using the cabin air pressure which is generally .2  les= s=20 (200 feet higher than the external static source.  OK, I bought the new= ,=20 improved sensor for about $800 and temporarily mounted it in the horizontal=20 orientation in a relatively vibration damped area using cabin pressure. = ;=20 Drat, no difference!
 
3.  S-Tec tech suggested there might be a problem with the compute= r -=20 so I sent it in.  Returned with a clean bill of health for a mere=20 $215!
 
4.  I cleaned and tested the motor again - this time after complet= ely=20 removing the servo located in the battery compartment behind the co-pilot se= at=20 and requiring that the battery be removed also.  BTW, I am now utilizin= g=20 more nutplates on the battery hold-down and certain A/P servo mounting=20 brackets.  After another failed test flight, I found that low voltage w= ould=20 not turn the motor.  So.... I ordered a new one (note that S-Tec will o= nly=20 sell thru a distributor) for about a mere $600!  The new motor will tur= n=20 even though I used a single AA battery.
 
5. Let's see, I have now invested about $1600.  Why don't I just=20 switch to a Tru-Trak?  I said to myself, "Self, you have a beautiful=20 fiberglass panel with the three large pilot side instrument holes (TC, DG an= d=20 AP) that would be replaced by a little airplane and a knob.  What would= I=20 do with those other big holes?"
 
6.  I reconnected the original pressure sensor and went for a=20 flight.  The first engagement of the ALt Hold function caused a rather=20 large climb and dive.  The second engagement got me in a climb angle I=20 didn't like.  OK, OK I did change something else too. I noted that= at=20 elevator neutral, the connecting rod end was positioned at about 10 o'clock=20= -=20 towards the nose.  Thus, "large" movements of the servo would result in= =20 small fwd-aft movement of the elevator pushrod.  Hmmmm..... pretty much= =20 de-sensitized.  Thus, I repositioned the pushrod so that it was at the=20= 12=20 o'clock at elevator neutral (having forgotten why I had set it up the way it= =20 was).
 
7.  My next two flights will be such that first the new expensive=20 (Gary?) pressure sensor will be checked and second (if necessary) the pushro= d=20 will be returned to its' original postion.  I will report the results a= nd=20 any further conversations with S-Tec.
 
>>>>>>>>>>>>>>>>>>= ;>>>>>>>
 
 -------------------------------1101234687--