Thanks very much for this extra info, Duncan.
If you can find that spare stop that would be great. I had seen a picture s
imilar to the one you sent of the stops in another aircraft application.
Following your comments about the shear pin, I think this is the small bras
s screw located just above the main central cross-head screw that holds the
control arm onto the servo in the photo I=99ve attached here? I looked at mi
ne more closely and saw that the head of that brass screw appeared to be sli
ghtly loose and with some tapping I could get the screw head to drop out of
the arm. If that is the shear pin you=99re referring to then it appears mine
is already sheared! However, the arm doesn=99t show any tendency to rotate
on the hub? I guess the cross-head screw is holding it onto the hub pretty f
irmly, but I can see that the arm could potentially rotate against the hub w
ithout the pin being there (since there is no positive locking mechanism aga
inst rotation), but the cross-head screw would need to be looser than it cur
rently is for rotation of the arm to be possible. Should the arm be free to
rotate against the hub without the pin there? Question also now is how to ge
t the threaded bit out and the pin replaced?!
Best regards
David
I'd not seen before the clip on the servo arm; there was an earlier Factory
'fix' of Loctited replacement screws. Later servos have a stud and pinned l
ocknut instead (I guess the stud is held in only by Loctite!).
The control column stops should take priority, as the servo stops are proba
bly too light to limit enthusiastic control movements by the pilot, but shou
ld at least contain a runaway servo or some other breakage in the servo. I m
ight have some spare stops (if I can find them). The stops are half-visible
and annotated 'F' in the photo at: TruTrak-RV10-Roll-Installation-Guide.pdf
(bendixking.com)
The shear pin is merely a 6-32 screw with a short part of its length (where
this crosses the interface between the mating arm and hub) necked and polis
hed to the roots of the screw thread.
My own servo is mounted towards the back of the Europa 'tunnel' and connect
s to a horn bolted to the rear end of the roll torque tube. This area used t
o get hot under certain situations via heat from the stainless firewall and
no tunnel ventilation. Subsequently a 'fire blanket' (two thin layers of wov
en Kevlar with thin fibreglass loft insulation sandwiched between, rather li
ke Orkotek) placed on the back of the firewall enabled cooler temperatures.
I've not heard of the standard under-seat servo location getting too hot.
Duncan McF.
On 03 January 2023 at 22:30 David Cripps <dpc@knightonweb.com> wrote:
Thank you for the comprehensive reply, Duncan, and also for the replies fro
m Alan and Peter. I really appreciate the feedback.
It seems that the Skymap III should work OK for us, despite its 0.5Hz data
refresh rate. A bit of =98hunting=99 shouldn=99t really be a problem. We had
n=99t planned to have it fly a full route and navigate itself around turning
points. I had thought we=99d be manually changing the selected track at eac
h turning point, and then lock the tracking to the new direction required. H
owever, the Skymap III does have a =98Turn Anticipation=99 feature if we sho
uld in the future want it to follow a route, and not overshoot a waypoint. D
o most people use it to follow a full track, and make all the waypoint turns
?
Good to know about the shear pin in the arm as a last resort! Do you have a
ny drawing of the necked stainless pin that you got the LAA to agree to?
The servo that I have came out of an old Europa and didn=99t seem to have a
ny physical stops on the servo arm. It did, however, have the extra alloy
=98clip=99 to secure the screw that attaches the arm to the servo. I=99ve seen
that that clip was an LAA requirement too (see photo). For the physical sto
ps, I did see on another forum something that looked like it would do the jo
b on this servo (may even have been a Trutrak part). Does the LAA specify wh
at these stops should look like or be made of? Should they be set so that th
e servo arm touches them at the same time as the control column reaches its
full deflection (if not, then which stop should =98take priority=99 and be h
it first)? Does anyone have a photo of what the stops look like in their set
up? I imagine them being mounted on the bolts that attach the servo motor to
its mounting frame.
My servo would be mounted under the pax seat per Mod 75, so there is not re
ally any ventilation there. Is it worth adding a little cooling fan to the s
eat locker to keep the air circulating around the servo?
Many thanks for all the input!
David
i). The Trutrak will work on a Europa at 0.5Hz update frequency (i.e. every
2 seconds). Sometimes the AP will hunt gently (yesterday was a case in poin
t, but there were sites yesterday that were GPS-jamming as well, so that mig
ht have been a cause' or possibly I have a too low torque setting than optim
al in order to limit servo overheating). If too many options of output data
are requested in the GPS NMEA output setup then there won't be enough time t
o output all that data within a faster update period; the GPS set will norma
lly tell you if this is going to be the case.
iii). The servo does not care about torque feedback. The servo will provide
a force (torque) only up to the maximum value that has been preselected.
iv). The Europa needs quite a high torque setting (10 or 12 from memory, wh
ich can on hot days plus a high 'activity' setting result in servo overheati
ng and shutdown if you have a trigear with 'heated'/unventilated fuselage tu
nnel and depending where the servo is mounted).
The pilot stick force required to overcome maximum torque will be applied w
ithout even thinking about it and in any case there is a necked brass shear
pin (comprising a modified 6-32 brass instrument mounting screw) in the serv
o arm that can be easily sheared (even without knowing it, and then you wond
er why the AP is not tracking!). I agreed with LAA to replace this with a (n
ecked) stainless steel pin in order to provide a little more strength and ha
ven't sheared it since.
I also think that when the servo gets very hot (as above) it's internal vol
tage regulator starts to shut down (as many are designed to do). Again, ther
e is no way of knowing when this happens until realisation that the AP is no
t tracking!
v). The LAA will insist upon the stops on the servo, although these stops a
re not particularly robust and the set up would be better adjusted so that t
he stops never have to resist the pilot's stick inputs.
Otherwise, the Trutrak system and its dynamics work extremely well and can
do a much better job than the pilot, especially in turbulent conditions.
Have you found a setting on your Skymap that allows adjustment of the pre-e
mptive steering sensitivity (or "turn anticipation")?
Duncan McF.
On 03 January 2023 at 13:10 David Cripps <dpc@knightonweb.com> wrote:
So, I've slowly been making some progress on this, having mocked up the who
le system out of the aircraft. I have now been able to test it in the worksh
op and yesterday in a car (it needed to be moving to get a track signal). Se
e attached photo. However, I now have a couple more questions/observations!
i) It seems that the Skymap III that I used in these tests works in terms o
f making the servo move and display a track on the Trutrak, so that's progre
ss! However, I note from the Skymap manual that it only gives its GPS info e
very 2 seconds, rather than the once every second that the Trutrak recommend
s to avoid 'wandering'. Peter, in your post you mentioned that you were usin
g a Skymap III - have you ever had a problem with this (assuming your A/P is
the Trutrak Digitrak too)? Alan, is the Garmin 196 that you've mentioned as
suitable, able to be set to give data at the 'once per second' rates, as fr
om the manual it seems it may too default to once every 2 seconds?
ii) I noticed when testing in the car, that when the track that was display
ed in the Trutrak was close to what we were actually tracking (as read from
the Skymap), that the servo was more or less static, and moved in the correc
t direction when there was a small deviation between what was set on the Tru
trak and what we were tracking. However, because our roads are not straight,
there were frequently occasions when the actual track from the Skymap was m
iles away from that set on the Trutrak. In that situation the servo went a b
it crazy and rotated its arm by many more degrees than it could ever do in t
he plane. Is this to be expected in this rather 'offline' mode? I can imagin
e that when the Trutrak senses that its control input is not creating the so
rt of aircraft track change that it had expected, that it just moves the ser
vo even further in order to try to get aircraft to do something?
iii) Clearly my mock-up does not provide any sort of 'force feedback' to th
e servo control arm (ie the arm is completely free to rotate as it likes). I
n an aircraft, would this be taken care of by aerodynamic feedback via the a
ileron torque tube? What happens to the servo when it reaches the full movem
ent that the aileron would allow (the point when the control column hits its
roll stops)?
iv) I can imagine that I wouldn't want the servo ever applying full aileron
to create a turn as that would make for a very violent manoeuvre! I see tha
t there are settings that can be adjusted in the Trutrak set-up menus that c
ontrol activity level and also control maximum torque. What values have peop
le set in theirs? I note that when I have set the recommended value of 12 fo
r the torque, it is almost impossible to override the servo arm when the A/P
is engaged (though I appreciate that with no control column attached the le
verage I can apply to the servo arm with my fingers is very low. I also appr
eciate that normally if one wanted to override the A/P you would disengage i
t first).
v) Lastly (for now!) has anyone fitted physical 'stops' to the servo body t
hat the servo arm would hit to ensure that the servo can never try to move t
he pushrod beyond the point where full aileron is applied?
Apologies for all the questions but I want to make sure it is really workin
g correctly before I contort myself to fit it into the aircraft itself!
Best regards
David
On 19/11/2022, 16:26, "Alan Burrill" <owner-europa-list-server@matronics.co
m <mailto:owner-europa-list-server@matronics.com> on behalf of alanb@dpy01.c
o.uk <mailto:alanb@dpy01.co.uk>> wrote:
alanb@dpy01.co.uk>>
Ok I=99ve had a. Ha ce to fly mine with the GPS feed off.
As Duncan says below the display drops to show -|=94|- but maintains the tr
ack last set or you are holding when you switch it on.
The are no heading digits on the display so you will need use your compass/
DI to show that.
If you press the left or right =A4=B5=EF=B8=8F buttons then you can alter the
track
the AP is following either left or right and a number appears which if you p
ress the button you can increase or decrease after the AP locks on the numbe
rs disappear and you have the -|=94|- on the display.
Useful if you have a GPS failure but wouldn=99t want that as normal mode of
operation so my suggestion is you need a GPS feed with the right NEMA messa
ge set to get the most out of the AP and the ability for it to follow a trac
k you have programmed in is a bonus.
One word of caution, I have come across some GPS that don=99t put out any m
essages, even the position message for driving a Transponder ADS-B output, u
nless there is a track in the GPS. That was how some of the older version wo
rked, GARMIN 430 and possible the early AVMAP were guilty of that.
Hope that helps.
Alan
Sent from my iPad
On 10 Nov 2022, at 10:52, D McFadyean <ami-mcfadyean@talktalk.net <mailto:a
mi-mcfadyean@talktalk.net>> wrote:
<mailto:ami-mcfadyean@talktalk.net>>
Digitrak only needs basic positional information from the GPS on a regular
basis. It will use this information to fly the track that has been set on th
e Digitrak.
If a route is programmed in to the GPS, it will fly that too (with differen
t commands being set at the Digitrak control head to enable this), so long a
s the GPS is also putting out the crosstrack error signal; not all of them d
o.
Duncan McF.
s.com/Navigator?Europa-List>
ion>
|