Monday, August 25, 2014

T28 Gear Door Follow-up

If any of you have been following my blog, you know that I had issues with my Airfield/FMS 1400mm Trojan T28 electrical system.  In blog post T28 update and putting test equipment to good use the result was a smoked front landing gear door servo.  In this post, I'll review what I did to fix this as it did not turn out to be as easy as it seemed.

On the outset, it looked like all I needed to do was to replace one servo and be done with it.  Such was not the case.  Not only did I need to source a replacement servo, but I still needed to make sure that what blew the first servo would not destroy the replacement.

Sourcing a replacement servo did not initially seem like a big deal as when I removed the original, it looked to be the same size as the cheep HK servos I already had.  The only difference that I could see was the the mounting flanges were slightly off, with the original black plastic servo sitting a little lower (one, maybe two, mm).

This did not turn out to be that big of a deal and pretty soon I had the replacement in.  I had already centered the servo and so I just screwed the servo arm in place at a 90 deg. angle, facing inboard, as the original was.  Then I hooked up the linkage, in what seemed like a neutral setting.  I then plugged it into the sequencer.  Then I plugged the sequencer into into my servo tester.  And finally, a battery pack into the servo tester to power everything.

Suddenly everything is going wonky!  The doors are closing and the landing gear is cycling down!  I reverse the servo tester and then the landing gear is retracting before the doors are opening.  Dog gone it!  I pulled the power.

What the heck is going on!?!

After some research on the Internet, I find out that the sequencer and my wiring are not messed up, but the front landing gear door servo needs to be reversed!

Out it comes again.

I don't want to wait and order just one reversed servo.  Especially as I'm not even ready for another parts order.  So, online I go again for more research and find a couple of good videos on how to reverse a servo.  Since I have a background in electronics and am proficient at soldering, I decide to tackle it.

An hour or so later (I took it slow to make sure I had it right), I had a reverse servo ready to go (sorry, but I don't have any photos of this).

Back in it goes and I'm ready to test.  Servo to sequencer, sequencer to servo tester, servo tester to battery pack.....and...phew!  All is working fine and the doors and landing gear are sequencing as they should!

A little adjusting of the linkage and the doors are closing flush and appear to be opening fine.  I can hear a little servo noise with everything sequenced open, but don't think too much as it seems like that is the norm with servos today.

Now, knowing that this is not the end, I add my servo and system power meter into the mix (check out this post on how I made this meter: Can't get enough test equipment).  Low and behold, I'm pulling similar amp readings as when the original servo blew (over 1 amp!)  I notice that this is when everything is sequenced open and when the servo is making noise.  This is not just some stray noise.  This is a jammed servo.  No wonder the original blew out.

This is the original servo and arm position when closed.
I studied and studied the linkage and servo movement.  The door linkage did not need to move as far as the servo was trying to make it go and could not move that far.  I tried moving the pin in the servo arm in one hole, but then it did not have enough throw and the doors would not open far enough.  What to do?

I thought about using my programmable radio to just reduce the throw, or end limits, of that servo, but with the sequencer in the mix, that would not work.  No way was I going to try to set this up without using the sequencer.

The more I looked at it, the more frustrated I was getting.  There had to be something to do.  Then I looked again at the linkage and how the main push rod from the servo was slender and maid of spring steel, and how it had some flex to it.  Then I remembered mechanical equipment that I had worked on in the past with levers connected to push rods, and how they would lock into place by essentially going past top dead center.  Now any force on the lever (or servo arm in this case) is back toward the shaft and not back on the gears.  That is what I would do.  I set the arm position a few notches forward on the servo, adjusted the linkage so the doors would be closed in that position.  With the doors unhooked, I watched as I sequenced the door servo.  Sure enough, it went up to dead center on its rotation, compared to the link, flexing the push rod as it did.  It came to rest with a little push rod flex but no strain noise from the servo!  I tested it several more times and then connected the doors again.  All was working well.

It felt good to resolve a problem and find a solution that I was happy with.

To help out with visualizing what I did, I added some pictures and a video below.

End position
Start position
The left photo shows the starting point and the right photo shows the end point of the arm travel.  With the arm in this position, any pressure on the arm is straight back and not on the gears.

Please excuse the condition of the plane in these pictures and video as they were not taken after the repair, but after a subsequent crash that bent the front landing gear (that repair might be in a future blog entry).

While it was apart, I used this opportunity to take the pictures and video that I had not done before.

Some lessons learned:
If servos are making noise, check it out and make sure its not because they are jammed or stressed.
If you are replacing a blown servo, find out why.
If you run into a problem, relax and keep at it, and make use of the wisdom out there on the Internet.
And most of all, remember that this is supposed to be fun, enjoyable, and relaxing.

Thanks for stopping by my blog.  Please feel free to post comments, good or bad, and be sure to come back and check for future posts.