Phidgets, AT, and FS9

[Rob Archer]

I have my throttles going thru a phidget 8/8/8 which is going thru
FS2Phidget on a remote computer.
I use the entire Project Magenta suite as well. :)
Now here is my problem. I have servo's hooked to my throttles. I engage
the AT and the servo's move the throttle arms. PM MCP, commands say 98%
then 96%, The throttle movement lags behind slightly do to servo and signal
play. This is measured in milliseconds, but, this lag causes the throttle
position to "fight" against PM MCP and phidgets servo and 8/8/8. Recap,
after AP and AT engaged, if i move the throttle our the servo's move the
throttle, the PM MCP sends the throttle into convulsions trying to get it
right. If i eliminate the PM MCP and have all the rest of the software
running, and default FS9 AP running, the throttle move smoothly and don't
input or "fight" with FS9 on their position. Hope this makes sense?
Any ideas?
Rob

--
Rob Archer

Hoosier Performance Specialist
"No Horse Play Just Horse Power"

[Enrico Schiratti]

Hi,

depending on the commanded thrust, the value you will read
for the throttles will jump back and forth. Is there any way
you can dampen the value you send out to your servos?

Ciao

Enrico


"Rob Archer" wrote in message
news:379663.76204@wb.onvix.com...
>I have my throttles going thru a phidget 8/8/8 which is
>going thru
> FS2Phidget on a remote computer.
> I use the entire Project Magenta suite as well. :)
> Now here is my problem. I have servo's hooked to my
> throttles. I engage
> the AT and the servo's move the throttle arms. PM MCP,
> commands say 98%
> then 96%, The throttle movement lags behind slightly do to
> servo and signal
> play. This is measured in milliseconds, but, this lag
> causes the throttle
> position to "fight" against PM MCP and phidgets servo and
> 8/8/8. Recap,
> after AP and AT engaged, if i move the throttle our the
> servo's move the
> throttle, the PM MCP sends the throttle into convulsions
> trying to get it
> right. If i eliminate the PM MCP and have all the rest of
> the software
> running, and default FS9 AP running, the throttle move
> smoothly and don't
> input or "fight" with FS9 on their position. Hope this
> makes sense?
> Any ideas?
> Rob
>
> --
> Rob Archer
>
> Hoosier Performance Specialist
> "No Horse Play Just Horse Power"
>

[Rob Archer]

Hi Enrico,
I think what is happening is that PM MCP is not enabling offset 310A so
that
the throttle input is ignored. Or is that possible. Somehow, in default AP
of FS9, it is ignored until AT is disengaged. I f i could get the 310A
offset triggered when the PM MCP AT is active, ie; N1, SPD, or VNAV then
all
would work fine.
Rob
"Enrico Schiratti" wrote in message
news:379894.76204@wb.onvix.com...
> Hi,
>
> depending on the commanded thrust, the value you will read
> for the throttles will jump back and forth. Is there any way
> you can dampen the value you send out to your servos?
>
> Ciao
>
> Enrico
>
>
> "Rob Archer" wrote in message
> news:379663.76204@wb.onvix.com...
> >I have my throttles going thru a phidget 8/8/8 which is
> >going thru
> > FS2Phidget on a remote computer.
> > I use the entire Project Magenta suite as well. :)
> > Now here is my problem. I have servo's hooked to my
> > throttles. I engage
> > the AT and the servo's move the throttle arms. PM MCP,
> > commands say 98%
> > then 96%, The throttle movement lags behind slightly do to
> > servo and signal
> > play. This is measured in milliseconds, but, this lag
> > causes the throttle
> > position to "fight" against PM MCP and phidgets servo and
> > 8/8/8. Recap,
> > after AP and AT engaged, if i move the throttle our the
> > servo's move the
> > throttle, the PM MCP sends the throttle into convulsions
> > trying to get it
> > right. If i eliminate the PM MCP and have all the rest of
> > the software
> > running, and default FS9 AP running, the throttle move
> > smoothly and don't
> > input or "fight" with FS9 on their position. Hope this
> > makes sense?
> > Any ideas?
> > Rob
> >
> > --
> > Rob Archer
> >
> > Hoosier Performance Specialist
> > "No Horse Play Just Horse Power"
> >
>

[Enrico Schiratti]

> I think what is happening is that PM MCP is not enabling
> offset 310A so
> that the throttle input is ignored. Or is that possible.
> Somehow, in default
> AP of FS9, it is ignored until AT is disengaged. I f i
> could get the 310A
> offset triggered when the PM MCP AT is active, ie; N1,
> SPD, or VNAV then
> all > would work fine.

But then the joystick position would interfere with the
commanded thrust. I do not follow what you want to do to
solve this.


> Rob
> "Enrico Schiratti" wrote in message
> news:379894.76204@wb.onvix.com...
>> Hi,
>>
>> depending on the commanded thrust, the value you will
>> read
>> for the throttles will jump back and forth. Is there any
>> way
>> you can dampen the value you send out to your servos?
>>
>> Ciao
>>
>> Enrico
>>
>>
>> "Rob Archer" wrote in message
>> news:379663.76204@wb.onvix.com...
>> >I have my throttles going thru a phidget 8/8/8 which is
>> >going thru
>> > FS2Phidget on a remote computer.
>> > I use the entire Project Magenta suite as well. :)
>> > Now here is my problem. I have servo's hooked to my
>> > throttles. I engage
>> > the AT and the servo's move the throttle arms. PM MCP,
>> > commands say 98%
>> > then 96%, The throttle movement lags behind slightly do
>> > to
>> > servo and signal
>> > play. This is measured in milliseconds, but, this lag
>> > causes the throttle
>> > position to "fight" against PM MCP and phidgets servo
>> > and
>> > 8/8/8. Recap,
>> > after AP and AT engaged, if i move the throttle our the
>> > servo's move the
>> > throttle, the PM MCP sends the throttle into
>> > convulsions
>> > trying to get it
>> > right. If i eliminate the PM MCP and have all the rest
>> > of
>> > the software
>> > running, and default FS9 AP running, the throttle move
>> > smoothly and don't
>> > input or "fight" with FS9 on their position. Hope this
>> > makes sense?
>> > Any ideas?
>> > Rob
>> >
>> > --
>> > Rob Archer
>> >
>> > Hoosier Performance Specialist
>> > "No Horse Play Just Horse Power"
>> >
>>
>

[Rob Archer]

"But then the joystick position would interfere with the
commanded thrust. I do not follow what you want to do to
solve this."

That's exactly the problem your mcp is causing on my sim. :) When project
magenta mcp is controlling the flight, if i move the throttles in my sim,
the N1 tries to change to where the throttle position I manually set is but
a split second goes by and the MCP takes back over.

Example, we are flying along at 75%N1, your MCP commands 90%, the software
"sees" the throttle position change in the sim so the servo's move the
physical throttle levers in response. Now here is where the trouble begins,
since there is an inherent lag in system response, the MCP sees the
throttle
input lower than what it commanded just a moment before because as soon as
the physical levers moved, THEIR input took over. So now the AP
"re-commands" a higher N1 to compensate. This is how they the system fights
itself for position. We are talking
milliseconds lag though. Again, if I fly with the default FS9 AP and AT,
the levers move along with the sims throttle with no feedback issues. You
can move them physically up and down and the sim doesn't care. It stays
where commanded. So that is why it doesn't fight for position in default
like it does with PM MCP. Does that make anymore sense?
Here is another example from a "professional" sim make on this issue,
"Hi Folks,
> >
> >I wonder if anyone would be willing to take the time to detail how
> they have
> >
> >gone about getting their autothrottles functioning with Boeing type
PM
> >software?
> >
> >I am interested in what offsets your read from and write to, and how
you
> >find it is behaving.
> >
> >I am VERY interested in whether you can override your autothrottle
> operation
> >
> >by grabbing the throttle and moving it manually.
> >
> >In our setup our throttles are connected to optical encoders and
> in parallel
> >
> >are connected to worm drives through slipping clutches.
> >
> >Our 'Thrust Management Computer' software reads the encoder pulses
and keeps
> >
> >track of the absolute position of each lever. It writes the
> current position
> >
> >to FS using offset 0x88C and 0x924.
> >
> >Our 'Autothrottle' software reads the desired position of the
throttle, and
> >
> >if the desired position of the throttle does not match the position
being
> >published by the Thrust Management Computer, the authrottle set the
> >direction and speed of the worm drive and then engages it. We read
our
> >desired position from a new offset that Enrico kindly added for us,
0x553A.
> >
> >Our software also writes to and uses bit 8 of 0x5528 to detach the
A/T from
> >
> >FS.
> >
> >How do other people do it?"

My understanding was that when the MCP AP was in control, it ignored input
from axis unless set in the ini file?

Thanks for your help,
Rob

"Enrico Schiratti" wrote in message
news:380235.76204@wb.onvix.com...
> > I think what is happening is that PM MCP is not enabling
> > offset 310A so
> > that the throttle input is ignored. Or is that possible.
> > Somehow, in default
> > AP of FS9, it is ignored until AT is disengaged. I f i
> > could get the 310A
> > offset triggered when the PM MCP AT is active, ie; N1,
> > SPD, or VNAV then
> > all > would work fine.
>
> But then the joystick position would interfere with the
> commanded thrust. I do not follow what you want to do to
> solve this.
>
>
> > Rob
> > "Enrico Schiratti" wrote in message
> > news:379894.76204@wb.onvix.com...
> >> Hi,
> >>
> >> depending on the commanded thrust, the value you will
> >> read
> >> for the throttles will jump back and forth. Is there any
> >> way
> >> you can dampen the value you send out to your servos?
> >>
> >> Ciao
> >>
> >> Enrico
> >>
> >>
> >> "Rob Archer" wrote in message
> >> news:379663.76204@wb.onvix.com...
> >> >I have my throttles going thru a phidget 8/8/8 which is
> >> >going thru
> >> > FS2Phidget on a remote computer.
> >> > I use the entire Project Magenta suite as well. :)
> >> > Now here is my problem. I have servo's hooked to my
> >> > throttles. I engage
> >> > the AT and the servo's move the throttle arms. PM MCP,
> >> > commands say 98%
> >> > then 96%, The throttle movement lags behind slightly do
> >> > to
> >> > servo and signal
> >> > play. This is measured in milliseconds, but, this lag
> >> > causes the throttle
> >> > position to "fight" against PM MCP and phidgets servo
> >> > and
> >> > 8/8/8. Recap,
> >> > after AP and AT engaged, if i move the throttle our the
> >> > servo's move the
> >> > throttle, the PM MCP sends the throttle into
> >> > convulsions
> >> > trying to get it
> >> > right. If i eliminate the PM MCP and have all the rest
> >> > of
> >> > the software
> >> > running, and default FS9 AP running, the throttle move
> >> > smoothly and don't
> >> > input or "fight" with FS9 on their position. Hope this
> >> > makes sense?
> >> > Any ideas?
> >> > Rob
> >> >
> >> > --
> >> > Rob Archer
> >> >
> >> > Hoosier Performance Specialist
> >> > "No Horse Play Just Horse Power"
> >> >
> >>
> >
>

[Stuart Whelan]

Hi Rob, our sim, the one you quoted below, has exactly the same problem.

Our throttle seek backwards and forwards constantly, and during TOGO they
back off to almost idle.

We dont use AT much at the at the moment for this reason. In the end I ran
out of time and patience, and just let it lay.

Maybe I am missing somethig, but I just can't get a grasp how all the parts

of the PM autothrottle were intendend to function.

Enrico would not explain how it was intended to work, so in the end I just
asked for the mcp to output its desired throttle positon to an offset (other

than 0x88c).

Hence my request for someone to do a howto...

[Rob Archer]

I'm with you there Stuart. It's a shame to since it functions so well thru
the default FS AP. I'm sitting in the cockpit now, TOGA works great! NO
jittering, nice smooth, although a little fast, movement. :) Right now, i'm
flying with everything but PM MCP as i like the working AT's

Did i explain correctly in my post to Enrico? Just checking for a 3rd party
opinion.
Will advise if i find a work around.
Thanks,
Rob

"Stuart Whelan" wrote in message
news:380404.76204@wb.onvix.com...
> Hi Rob, our sim, the one you quoted below, has exactly the same problem.
>
> Our throttle seek backwards and forwards constantly, and during TOGO they
> back off to almost idle.
>
> We dont use AT much at the at the moment for this reason. In the end I
ran
> out of time and patience, and just let it lay.
>
> Maybe I am missing somethig, but I just can't get a grasp how all the
parts
>
> of the PM autothrottle were intendend to function.
>
> Enrico would not explain how it was intended to work, so in the end I
just
> asked for the mcp to output its desired throttle positon to an offset
(other
>
> than 0x88c).
>
> Hence my request for someone to do a howto...
>

[Andras Kozma]

Hi Rob,

> magenta mcp is controlling the flight, if i move the throttles in my sim,
> the N1 tries to change to where the throttle position I manually set is
> but a split second goes by and the MCP takes back over.

I think you have the same problem I used to have almost a year ago, until
Pete Dowson came to my help. He had introduced two new variables in FSUIPC
where the actual throttle positions are copied to, 089A and 0932 and
reading
those instead of the 'stock' throttle variables solved my problem at least.
It was added in version 3.482 if I'm not mistaken for this purpose only.

regards
Andras

[Hans Krohn]

Hi Rob,

how about this solution: Just install an additional pot in the throttle that

is responsible ONLY for reading the actual, physical lever position.

Additionally, you could use a relay to disconnect the original throttle pot

during AT op's. Or maybe redirect it to another joystick axis (even though
that would probabely make sense only if you use EPIC hardware.

Regards,
Hans



"Andras Kozma" wrote in message
news:380942.76204@wb.onvix.com...
> Hi Rob,
>
>> magenta mcp is controlling the flight, if i move the throttles in my
sim,
>> the N1 tries to change to where the throttle position I manually set is
>> but a split second goes by and the MCP takes back over.
>
> I think you have the same problem I used to have almost a year ago, until
> Pete Dowson came to my help. He had introduced two new variables in
FSUIPC
> where the actual throttle positions are copied to, 089A and 0932 and
> reading
> those instead of the 'stock' throttle variables solved my problem at
> least.
> It was added in version 3.482 if I'm not mistaken for this purpose only.
>
> regards
> Andras
>