Pneumatics Question

On 25 May, 02:32, jsbrake wrote:
Hi All,
Whenever I accelerate, my varios bump up and tell me I'm climbing,
then settle back to reality. *Any idea of what I need to look for to
fix this?
My setup:
Kestrel 19, two static ports in the tail boom, pitot and TE in the
fin. *The TE line is split under the seat using a Y connector and then
travels to two varios (a Winter mechanical and a 1990-era audio/final-
glide called a VariCalc). *Pitot and static lines are split behind the
panel using T connectors.
I replaced all the non-permanent tubing last year and they all seemed
to be leak-free.
Thanks!
-John


A TE system in which the negative dynamic head compensation is exactly
'right' will do just this. It does not not necessarily indicate a
fault in installation. It is generally manifest during vigorous
changes in load factor. A TE vario reflects rate of change in energy
height not actual height so if you pull/push g the change in induced
drag as a result of load change represents a change in energy. e.g.
Pull up and increase load and therefore induced drag and your total
energy reduces which is manifest as sink on your total energy vario.
And vice versa.

Some commercial probes compensate for this by being deliberately not
quite exactly 'right'. Many probes are not spot on because the maker
can't manage it which is almost invariably true of any home made
device or one of poor design such as the Nicks and Braunschweig tubes.
Thus your observation is relatively uncommon.

If your system is truly reflecting your energy changes then try flying
a little more smoothly for you are throwing energy away.

There is no reason why you should not couple two varios to one TE
source but put the Y connector as far away from the panel as possible
to prevent one vario puffing into the other.

To answer some of the questions posed regarding my setup:
- I replaced the non-permanent lines last year because they were quite
old and looking dry. I was also having TE troubles. The replacement
lines tested leak-free (I hooked up a hyperdermic and an ASI, the
speed held when I pushed the plunger). I left the hard plastic
original lines in place, figuring they weren't part of the problem.
- I split the TE line under the seat when I replaced the other
tubing. It was originally split with a "T" at the panel and I read
that this was a no-no (article by Mike Borgelt)
- I tested all the hard plastic lines with the hyperdermic an ASI,
they all held the pressure.
- then traced my original TE troubles to a leak in the 4-way connector
at the panel. Replaced the connector this year.
- I don't recall the varios doing a quick jump when I accelerate.
It's rather reminscent of a "stick thermal" on non-TE varios, lasts
only a couple of seconds
- I'm not doing hard accelerations, at least to my mind. Not feeling
much force on my body
- Winter mechanical has a capacity flask mounted under the instrument
hood. VariCalc has no flask.
- I'm not sure if there were vario changes when I opened or closed the
vents as I wasn't paying attention to that aspect.
- the VariCalc has connections for all three lines; the Winter
mechanical connects to TE and capacity. A problem in static shouldn't
affect both instruments.
- the TE probe is a simple straight out and then bend down, with 2
slots. No idea who made it, but I suspect it came standard from the
Slingsby factory in 1972.

According to Mike Borgelt's article, it's okay (but not best) to
connect two varios to TE, but they should be split as far away from
the instruments as possible. The farthest back I could get was under
the seat, since I'm too big to fit into the boom and all those pesky
control connections were in the way.

I'll try blocking off one vario at a time to see if that changes
things.

The VariCalc has an option for electronic compensation, but I read
somewhere that it's not as good as using a TE line.

. *Replaced the connector this year.

- the VariCalc has connections for all three lines; the Winter
mechanical connects to TE and capacity. *A problem in static shouldn't
affect both instruments.

Faulty multi-connectors can also cause your described symptom as well.
Check the O-rings well, and silicone grease is your friend.

Hmmm, this puts possible tailboom static issues back on the map and
your VariCalc could be cross-talking that or an internal pitot leak
error to your Winter. Run it once with cockpit static before getting
too fancy with time consuming ground testing. It's real easy and you
get to fly... Your vario isolation test will also be useful here, I
suggest unhooking (plug loose lines!) the VariCalc first.

-Paul

Get rid of the 4-way connector. They are unreliable, sometimes
causing cross-talk between paths. For peace of mind, use separate
connectors.

Bob


- then traced my original TE troubles to a leak in the 4-way connector
at the panel. *Replaced the connector this year.

Get rid of the 4-way connector. *They are unreliable, sometimes
causing cross-talk between paths. *For peace of mind, use separate
connectors.


He mentioned it's a 5 way (although only 4 ports are currently being
used) in another post, which I mean to take is a Winter unit. I've
personally had good luck with those but I still prefer the positive
locking threaded compression connection of my 4 way PZL version. I've
had no leaks with either though, nor any crosstalk (ahem- well just be
sure you didn't x-thread the PZL unit back together on after you show
it off just before takeoff... heh.) Only problems I've heard of with
the quick connects are leaks when the stock O-rings are old/crusty,
which is an easy inspection/replacement (spend the extra $.12 on
Viton!). While crosstalk is possible on the PZL 4-way with it's
compression loaded seals, on the Winter 5-way there is no *real chance
of it as the o-rings form a male seal at least 1/4" into the female
sockets, effectively isolating the individual lines from each other in
case of seal failure, even with dual seal failure. Rubber is not
ageless... and again, silicone (spray, applied via cotton swab) is
your friend, same for the probe/socket...

It's also unlikely a probe connection problem since it is not a double-
probe and I feel safe to assume the probe gets taped at each assembly
of course. Borrowing a probe from another ship will help isolate any
problems with the probe itself it It winds up being suspect.

I may just have tunnelvision but really I think an old vario (with
aged internal seals...) that has pitot and TE seems more suspect of
crosstalk issues than a new Winter quick connect or probe performance
issues...

Slightly OT, I like to use these to seal pneumatic instrument hoses on
barbs:
http://www.jollygerman.com/products/goats/rings.shtml cheap, last
bout 5yrs and gives a uniform 360deg seal (UNlike zipties...) that is
easily future-accessed with no damage to hose/barb. Got idea from
Borgelt but it took me a while to figure out what these little green
rubber Cheerio things were *really* used for... OUCHY!!!

-Paul

It's also unlikely a probe connection problem since it is not a double-
probe and I feel safe to assume the probe gets taped at each assembly
of course. Borrowing a probe from another ship will help isolate any
problems with the probe itself it It winds up being suspect.

The Brunswick socket won't have any internal seals. Just one tube
inside another. Tape is unreliable. A short length of silicon tube
tightly fitted over the joint is better. If the tube can stay in situ
in the trailer then wire it or castrate it with a couple of Cheerios
and leave it well alone.

I may just have tunnelvision but really I think an old vario (with
aged internal seals...) that has pitot and TE seems more suspect of
crosstalk issues than a new Winter quick connect or probe performance
issues...

Seems the first port of call for an elimination exercise. I await some
more real data with fascination.

Whee ! Looks like I'm in for some fun this coming weekend, trying to
figure out what's leaking. I'll probably fly it in the daytime and
try to test it during the evening, or bring the panel home to test the
varios.

I think I was happier before I asked the question... oh well, test it
out and get it working nicely.

Thanks to all for the information and advice; I'll report back when I
figure it out.

-John