Airspeed drops to zero approaching cloud layer

[airforce2]

Pete;

Using ActiveSky wxRE 1.91 beta 6, I am seeing a recurring problem where the indicated airspeed reported by FSUIPC via WideFS drops to zero for 20-30 sec as I approach a cloud layer in the descent. The latest episode occurred descending through ~14000 ft MSL close to the top of a scattered cloud layer with bases at perhaps 12000 ft. It appears that only airspeed is affected...other params appear normal.

This behavior began with installation of both FSUIPC 3.06 and the Beta 6 of wxRE. I also reported this on the ActiveSky forum, as has at least one other person.

Regards

[Pete Dowson]

Using ActiveSky wxRE 1.91 beta 6, I am seeing a recurring problem where the indicated airspeed reported by FSUIPC via WideFS drops to zero for 20-30 sec as I approach a cloud layer in the descent. The latest episode occurred descending through ~14000 ft MSL close to the top of a scattered cloud layer with bases at perhaps 12000 ft. It appears that only airspeed is affected...other params appear normal.

Sounds exactly like icing. Switch on your pitot heat! I never leave the ground without it, neither should you!

Regards,

Pete

[airforce2]

Pete;

I wish it were that simple...probe/pitot heat's on and verified (in both the panel and using FSLook to make sure the sim thinks so) in all three occurrences. PM also reports Pitot Heat on at the time it happens.

The A/S drops abruptly to 0...then without any other action on my part, it returns 20-30 sec later. In the meantime, the Project Magenta MCP has poured the coals to it, sensing that we're seriously slow (can't get much slower!), so when A/S indication returns, I'm at Vmo+20 or so.

Cheers

[Pete Dowson]

I wish it were that simple...probe/pitot heat's on and verified (in both the panel and using FSLook to make sure the sim thinks so) in all three occurrences. PM also reports Pitot Heat on at the time it happens.

Sorry, then. I've got no other theories. Maybe there's a bug in FS2004 where is doesn't realise the pitot heat is on? Have you checked whether there is icing set for the clouds? Can you tell me how "intense" it is? (I.e. what's the "icing" value shown by WeatherSet2?).

I still think it must be something to do with that. I know of no other mechanism for a wrong airspeed reading, and since it seems to be associated with clouds it must have somerthing to do with icing, surely?

Maybe FS2004 likes the Anti-Ice to be turned on too. did you try that?

Regards,

Pete

[airforce2]

Pete;

The icing level reported by WeatherSet2 during another zero-airspeed incident was 4...I think the problem may be that even pitot heat is not effective in extreme icing (as would be the case in reality). But...to see that level of unforecast icing routinely isn't realistic...perhaps the probability programmed into the randomizer is set too high.

Also, significant structural icing at temps below ~-20 deg C is very rare, as ice tends to sublimate off the wings. Hence icing at 35000 or 52000 ft (as I've observed a few times in the last couple hours) isn't very realistic...not sure if that's the AWI or ActiveSky doing that.

Anyway, I'm turning random icing in clouds off in FSUIPC...if I see another zero-airspeed hit with it off I'll readdress here.

Cheers

[Pete Dowson]

The icing level reported by WeatherSet2 during another zero-airspeed incident was 4...I think the problem may be that even pitot heat is not effective in extreme icing (as would be the case in reality). But...to see that level of unforecast icing routinely isn't realistic...perhaps the probability programmed into the randomizer is set too high.

Do you mean the randomizer in FSUIPC? You had it on? Or is this something in ActiveWeather?

The probability system for the random events in FSUIPC hasn't changed since FS2000 days. I've rarely if ever seen icing being set to more than 2, and it is usually 1 or 0. And no one has mentioned any such problem in the three or four yers it's been there.

Certainly, it wouldn't be much of a randomiser if you didn't rarely get a 4. But you are implying you get it quite often.

:?

Regards,

Pete

[Pete Dowson]

I've reproduced the symptom of the airspeed dropping to zero.

I did this with no FSUIPC, no add-ins at all, on a completely virgin FS2004 installation I keep for such elimination tests.

I set a thick cloud layer from 10000 to 15000 feet with severe icing set. I slewed up to 17000, then, with Engines idling, I started a slow descent.

Unlike you I had to actually fly within the cloud for a while before it occurred -- but the difference there might be the cloud top variations. I expect FS considers you in the cloud layer if you are between the very top and the very bottom.

I had Pitot Heat switched on, but I deliberately left Anti-Ice off. When I switched Anti-Ice on, the airspeed indication returned very quickly.

I tried exactly the same test again but with Anti-Ice on initially, and couldn't reproduce it.

So, my conclusion is that it is wing ice blocking the pitot entrances. Pitot heat alone is not enough for these extremes.

Incidentally, I noticed a control called "toggle pitot blockage", which I thought might be interesting to try -- it appears in the drop down lists in FSUIPC's Keys and Buttons pages. However, I tried it and it appears to do nothing I'm afraid.

Regards,

Pete

[airforce2]

Pete;

I tried same experiment...saw the same. In severe icing, I'd expect massive accumulation on the wings, even with anti-ice on...but I can't conceive of severe icing that'd accumulate on the probes themselves when they're heated. Those things will burn the fingerprints right off your fingers if you make the mistake of leaving 'em on while on the ground. Been there, done that.

I actually got one of these dropouts in a clear spot between the clouds...don't think FS differentiates between clouds and clear spots...it just looks at layers.

Is the "random" icing level in clouds written by FSUIPC or an FS feature FSUIPC activates? If an FSUIPC feature, would it be possible to limit random icing in clouds to level 3 or below?

Cheers

[Pete Dowson]

Is the "random" icing level in clouds written by FSUIPC or an FS feature FSUIPC activates? If an FSUIPC feature, would it be possible to limit random icing in clouds to level 3 or below?

Well, yes, I could do, but you shouldn't be seeing it generate a 4 more than once in a blue moon. The probability of a 4 is kept very low. Are you saying you are seeing FSUIPC generate it more often? I don't think I've ever seen it in the three or four years it's been implemented.

I thought you and others were only noticing it with some new ActiveWeather release, so couldn't it be generated by that?

Why not simply turn Anti-Icing on? I don't understand the problem. If you don't want FSUIPC generating ice at all, you can turn that off in any case.

Pete

[airforce2]

Pete;

Well, suppose it's possible that ActiveSky is doing it...I need to re-engage with them on that. I did see ice values of 4 several times watching it with WeatherSet2 over the course of 15-20 min.

The issue is realism...one doesn't normally fly with anti-ice on unless either you expect to climb/descend through forecast icing, or in-cloud with temp < 10 deg C (or as specified by acft ops manul). Problem is, you can be in the cloud layer some time before actually visually entering the clouds themselves.

I sent a note to you and Enrico re: possibility of adding an FSUIPC offset for current icing condx, and adding an amber "ICE DET" EICAS message in PM to announce icing is present. With an icing indication, we can reach up and turn on engine A/I and solve the problem.

Cheers

[Pete Dowson]

I sent a note to you and Enrico re: possibility of adding an FSUIPC offset for current icing condx, and adding an amber "ICE DET" EICAS message in PM to announce icing is present. With an icing indication, we can reach up and turn on engine A/I and solve the problem.

I replied to it. FSUIPC already provides full weather information including the icing data for each cloud layer. I don't believe FSUIPC should have to work out which cloud you are in just so that it can copy the current icing value to another location. That's really an application activity -- having FSUIPC doing such computations continuously just in case something might need it is not a good idea, performance wise.

Regards,

Pete

[airforce2]

Pete;

OK...fair enough. I think I'm going to try writing a CAS alert program that will overlay a black text box over PM's EICAS to display CAS messages. I can code the ice detection in there.

Cheers