• Which the release of FS2020 we see an explosition of activity on the forun and of course we are very happy to see this. But having all questions about FS2020 in one forum becomes a bit messy. So therefore we would like to ask you all to use the following guidelines when posting your questions:

    • Tag FS2020 specific questions with the MSFS2020 tag.
    • Questions about making 3D assets can be posted in the 3D asset design forum. Either post them in the subforum of the modelling tool you use or in the general forum if they are general.
    • Questions about aircraft design can be posted in the Aircraft design forum
    • Questions about airport design can be posted in the FS2020 airport design forum. Once airport development tools have been updated for FS2020 you can post tool speciifc questions in the subforums of those tools as well of course.
    • Questions about terrain design can be posted in the FS2020 terrain design forum.
    • Questions about SimConnect can be posted in the SimConnect forum.

    Any other question that is not specific to an aspect of development or tool can be posted in the General chat forum.

    By following these guidelines we make sure that the forums remain easy to read for everybody and also that the right people can find your post to answer it.

Autothrottle failure

Messages
261
Country
ireland
Hi,

I am flying along in my B737, say 14000', say 320kt. I am using the autopilot and autothrottle. I am using VOR/LOC. Suddenly, the speed begins to rise, the OVERSPEED warning goes off and if I don't disengage autopilot and autothrottle rather rapidly and do something about it manually then I break up.

Luckily this is FS2004 or some of the passengers might have good reason to complain. I don't have any official 'Failure' set up. I am using AP_NAV1_HOLD and AP_PANEL_SPEED_HOLD.

It happens rarely but is annoying. My flights would be no longer than 90 mins from loading the sim. The situation can appear anywhere from early on to at the end.

The Autothrottle seems to stay engaged and the MCP reading stays at 320.

The only other gauge affected seems to be a clock that I have set up to record flight-time - it resets to 0.

I wouldn't have been touching any instrument when this happened.

I use ActiveSky 6 but the weather doesn't appear to change radically at the time.

I also have FRAPS and AISmooth running.

I use the 50N B737 with my own (XML) panel.

It just seems to lose control of the Autothrottle - never slows down, only ever speeds up.

My standard way of slowing down had been to gain altitude manually until a reasonable amount of speed had been lost and then recover to the altitude and speed that had been dialled in. It was unwieldy, time-consuming and seemed unprofessional. Now I just go to the map and set the speed there ( even more unprofessional!).

Has anyone here seen this behaviour before. Better still, has anyone got a cure for it (other than FSX!).

Thanks

Walter
 
It happens when the wind changes directions and speed rapidly after a weather update.

Either enable "wind smoothing" in FSUIPC or AS6 and the problem should occur less frequently.

Or turn off "aircraft stress causes damage".
 
You do NOT have to disengage the autopilot. The "speed hold" is a completely separate system from the autopilot.
 
Hi Bill,

The only reason I disengage the Autopilot is to climb rapidly 'right now!!!' with the stick. If I had to go dialing in a new height in the MCP window and pressing the LVL CHG button the speed would have increased so much and so fast I would be travelling at 400 kt or so and there would be some ominous sounds followed by some ominous silence.

Having the Autothrottle on and 320 dialed in on the MCP seems to have no effect - the aircraft just accelerates away.

I have tried disengaging and reengaging the Autothrottle with a lower speed, but I was travelling so fast by the time all that was done it was too late and the aircraft broke up.

I understand that for the aircraft to begin accelerating while in level flight with the Autopilot and the AutoThrottle on would be rather unusual in real life, so I accept that it is most likely a simulator problem (eg AS6 as Heretic says) and that therefore it requires something other than a 'real-life' immediate action - hence straight into Map mode and setting the speed there.

It sort of wrecks the realism a wee bit, (as does pressing the P button while I curse and swear at the inanimate screen!).

I will try the 'wind smoothing' idea and see if my casualty figures decrease in any way.

Thanks...

Walter
 
Last edited:
The stock autothrottle seems to be quite poor at holding airspeed if the wind changes. I also see the same problem as you sometimes (wind smoothing will help but not eliminate it completely). The problem also presents itself during turns when there is a large change in the headwind component - the airspeed rises or falls outside of the target range, but the autothrottle does nothing!
http://www.fsdeveloper.com/forum/showthread.php?t=218515

I ended up writing a couple of XML gauges to disengage the autothrottle and apply extra thrust (or clamp to idle) until the speed is back in the proper range, after which it hands control back to the ATS...

Geoff
 
Thanks for the replies...

I was quite disappointed when I went to AS6 to find that I already had 'wind smoothing' set on. Darn! One possible solution already implemented.

So for something to do, I set the max_throttle_rate = 0.20.

Then went flying from Billund to Stavanger.

I reached cruising speed and height and within 2 or 3 minutes - 'OVERSPEED' - caught me completely by surprise, never got a chance to react and all souls were lost.

I am sure it was the simulator getting back at me for complaining about it!

Absolutely gutted now - I'll try again tomorrow night, maybe.

Walter
 
What I still do not understand is why have AT on a completely different circuit to AP...?


Geoff, that sounds like a pretty good setup. Sharp...
 
What I still do not understand is why have AT on a completely different circuit to AP...?

When the AP fails, you won't lose AT and vice versa.

Also, you won't need AT when you just want to stick to a certain throttle setting at cruise.
 
I wrote myself a little bit of code which I reckon sets out to do something like what Geoff wrote except... mine didn't work!

I wrote it purely to cater for overspeed situations, which are what have been niggling me.

When the IAS exceeds the speed set on the MCP - the code disarms the AT and cuts the throttles.

Then when the IAS comes down to the speed set on the MCP - the code re-engages the AT and all is more or less well.

Fine, except that when AT is on and the speed is set to 320, for instance, and the aircraft is purring along at that speed, and I then require to reduce speed to 250 and dial it in - the code kicks in, because the IAS at that time is definitely faster than the new speed set.

The same applies if I descend and the speed picks up a bit over the set speed.

I couldn't use my basic XML coding skills to handle these situations, so...

I changed the code to work so that when the IAS exceeds the Barber Pole speed the code takes over, cuts the throttles, waits until the IAS returns to the set speed, then re-engages the AT and all is well.

It appears to work OK under some test conditions (I cannot simulate whatever causes it to happen in the first place), but I presume that cutting the throttles may not always slow me down quickly enough to prevent disaster.

Perhaps, with luck, I won't break up any more, but as a Silver Bullet for the AT problem my code is a poor substitute for a proper AT (or proper XML coding to circumvent the problem - any suggestions welcome).

Thanks

Walter
 
The stock autothrottle seems to be quite poor at holding airspeed if the wind changes. I also see the same problem as you sometimes (wind smoothing will help but not eliminate it completely). The problem also presents itself during turns when there is a large change in the headwind component - the airspeed rises or falls outside of the target range, but the autothrottle does nothing!
http://www.fsdeveloper.com/forum/showthread.php?t=218515

I ended up writing a couple of XML gauges to disengage the autothrottle and apply extra thrust (or clamp to idle) until the speed is back in the proper range, after which it hands control back to the ATS...

Geoff

Vould you share the gauges?
 
Just to update the forum...

I retried the Billund to Stavenger flight today.

I was paying great attention to the gauges, expecting something to happen.

It did - about two thirds of the way there this time.

My copilot shouted "Watch your speed!", the AT disengaged, throttles were cut and the speed, which had been over the 400 kt mark quickly reduced itself to 320, whereupon the AT reengaged and we flew on, undamaged. Whew!.

Because I was glued to the gauges I did notice that the speed was displayed as 0 in the PFD, just before it all hit the fan and the speed jumped to over 400kt.

The code I used was

<Element>
<Select>
<Value>
(A:AUTOPILOT AIRSPEED HOLD, bool)
if{
(A:Airspeed select indicated or true, knot) (A:AIRSPEED BARBER POLE, knots) &gt;
if{ 0 (&gt;K:AUTO_THROTTLE_ARM)
0 (&gt;K:AP_PANEL_SPEED_HOLD)
1 (&gt;L:MySpeedKeeper,bool)
1 (&gt;L:DSD_XML_SOUND_110, number)
(>K:THROTTLE_CUT) }
}

</Value>
</Select>
</Element>

<Element>
<Select>
<Value>
(L:MySpeedKeeper,bool)
if{
(A:Airspeed select indicated or true, knot) (A:AUTOPILOT AIRSPEED HOLD VAR, knots) &lt;
if{ 1 (&gt;K:AUTO_THROTTLE_ARM)
1 (&gt;K:AP_PANEL_SPEED_HOLD)
0 (&gt;L:MySpeedKeeper,bool) }
}

</Value>
</Select>
</Element>

Amateurish, but it worked. (The code is indented in a structured fashion, but that doesn't seem to display properly)

I look forward hopefully to something more comprehensive to replace it.

The flight-time clock also reset - so something of a glitch somewhere.

Thanks

Walter
 
Just to update the forum...

snip-------------

Because I was glued to the gauges I did notice that the speed was displayed as 0 in the PFD, just before it all hit the fan and the speed jumped to over 400kt.


Walter

You never say if the Probe (pitot) heaters are turned on. If you are on AT when the probes ice the airspeed drops to zero and the throttles will move forward (on certain FS planes). Overspeed will occur very quickly.
 
jvile,

On the Overhead Panel of the B737, switches for both of the Pitot tubes are modeled, left and right.

I have each of the two tubes working based on an L Var.

When both L Vars are set to 1 (ON) I set the Pitot Heat on (>K:PITOT_HEAT_ON).

When either of the L Vars is set to 0 (OFF) I set the Pitot Heat off (>K:PITOT_HEAT_OFF).

There is a warning light on the panel for each of the tubes and I set them on when the respective L Var is on.

I just checked the code and the simulator in case I had messed up.

But no, when both left and right tubes are on, as per their lights, the variable (A:PITOT HEAT, bool) shows 1. When either tube is off the variable goes to 0.

I never fly without Pitot Heat on.

I never even questioned that I hadn't coded the procedure correctly!

You had me worried there for a while, knowing my messy XML code standards.

Thanks for the heads-up, but at least now I know that I coded it correctly.

I am going to set up a (temporary) alarm to alert me to the situation where either of the L Vars goes to 0 in flight, just in case the glitch that affects the flight-time clock affects those as well. (edit: I don't have to do this after all, because the Right Master Caution panel shows a warning if the pitot heat is off)

Thanks

Walter
 
Last edited:
Amateurish, but it worked. (The code is indented in a structured fashion, but that doesn't seem to display properly)
See the # icon in the editor's toolbar? That is used to wrap code or XML scripts in a
Code:
 tag pair. This is to allow indenting to be preserved, and switches automatically to a monospaced font...

Simply highlight all of your code or script with the mouse and click on the [B]#[/B] icon.
[code]
This is
   an example
       of
   indented
formatting.
 
Back
Top