Limiting Icing Severity in FSUIPC

[jase439]

Pete,

Would it be possible to have an option in FSUIPC which prevents icing levels from exceeding a certain threshold. Alot of people who would like to enable icing either through FSUIPC or thirt-party wx program and are unable to do so because of the FS9 bug which prevents the pitot heat from doing its job. Apparently this is a geneal problem for all icing levels except for "trace".

It would be awesome if FSUIPC could provide a facility which effectively "capped" the threshold for icing.

Is this doable?

J

[Pete Dowson]

Would it be possible to have an option in FSUIPC which prevents icing levels from exceeding a certain threshold. Alot of people who would like to enable icing either through FSUIPC or thirt-party wx program and are unable to do so because of the FS9 bug which prevents the pitot heat from doing its job. Apparently this is a geneal problem for all icing levels except for "trace".

This is news to me. Icing isn't a problem with pitot heat and, possibly also, the anti-icing enabled. At least I've not experienced any problems, and yours is the first report I've seen.

FSUIPC could only reduce or eliminate icing when it is being set be external programs. It cannot do it for weather set by FS's own downloads, nor for weather set from the menus.

Apart from that it would be easy enough, but wouldn't the correct place be in the weather programs? The whole point of providing the extensive weather interface for those programs was to allow them full control, and now you want some taken away.

I don't know what program you use, but have you asked the author about this, as a possibility?

Regards,

Pete

[jase439]

Hi Pete,

There's a number of reports over at AVSIM - 'specially in the ActiveSky, PMDG, and PSS forums on this subject if you'd like to look into the problem further (see posts related to icing and IAS tapes dropping to zero). With the winter weather settling in up north, there are an increasing number of reports of people experiencing pitot-heat failure in FS9. So far, in each of these cases, it has been traced back to FSUIPC's random-icing or third party wx program. Unfortunately, there is no option in ActiveSky to control icing. The conditions, however, can be recreated manually by creating a cloud layer in FS with moderate to severe icing. I've seen this first hand myself where pitot-heat fails.

By the accounts I have read (including my own encounter with this problem) - the problem is most reproducible when enabling pitot-heat on the ground followed subsequently by encountering icing conditions aloft. In some cases, flipping the pitot heat off and then on again (via with SHIFT+H) a few times will sometimes kick the sim out of its "frozen" state, but if you simply leave the pitot heat on, the icing problem persists and the ASI indicator reads zero.

Some have eluded to the notion that FSUIPC is to blame, but a developer for PMDG says the problem is a known bug in FS9; and their recommended practice is to disable all icing in FSUIPC and third party wx programs. This seems like an unfortunate tradeoff.

The ActiveSky folk might implement a workaround for this problem in a coming update (similar to what I describe), but it seems like such a feature would be generally useful to the FSUIPC customer since random icing is something that can be enabled with or without a third party wx program.

Anyway - chances are you haven't heard of this because people rarely raise the issues in your forum. They just ***ch about them elsewhere :). Anyway, Damian Clark or Marc Philibert might have more insight to the underlying problem than I can offer here. I'm not sure what the "correct" fix is or where that fix is best applied. Seems like an FSUIPC'ish sort of thing, though. Some say the pitot heat isn't working at all. Others suggest it's "insufficient" to counter-act the icing effects. It could also have something to do with the state of the pitot heat variable getting silently reset to OFF unbeknownst to the user - so maybe the problem is in the pitot heat logic and not the icing. Not sure what the root cause of the problem is, but it's very reproducible for me by creating icing conditions, turning pitot-heat ON while on the ground, and then flying into the clouds.

Anyway, that's my $0.02 :)

Thanks,

J

[Pete Dowson]

I've seen this first hand myself where pitot-heat fails.

And with anti-icing enabled or disabled?

Some have eluded to the notion that FSUIPC is to blame

FSUIPC gets blamed for all the ills in the world. I get used to it, but it is depressing nonetheless. :(

... their recommended practice is to disable all icing in FSUIPC

There's no icing in FSUIPC unless you elect to enable the random cloud icing. Again, this applies to third party weather inputs only, not to FS "real weather" downloads.

it seems like such a feature would be generally useful to the FSUIPC customer since random icing is something that can be enabled with or without a third party wx program.

I can certainly add a limit option, but you are under a misunderstanding, as the random icing cannot be applied to FS's downloaded localised weather. It can be applied to the global weather but that is only used by FS where there are no nearby WX stations, and even then an additional option in the Technical page has to be enabled in FSUIPC (with it on you cannot retain Themes).

Anyway - chances are you haven't heard of this

Well, it hasn't been brought up in the Microsoft Beta newsgrups either. Generally identified FS errors are mentioned there, else it is unlikely they'll be fixed in the next version either. And why don't the problems occur with FS downloaded weather? Don't the Jeppesen/Microsoft weather reports ever have icing in clouds?

Not sure what the root cause of the problem is, but it's very reproducible for me by creating icing conditions, turning pitot-heat ON while on the ground, and then flying into the clouds.

In any aircraft, or only some? If the aircraft is equipped with anti-icing, does this make any difference? Possibly it occurs on aircraft where the pitot tube is on the wing or in other places more liable to severe icing? Is the aircraft realism turned up to 100%? What if that is lowered? Are folks really actually complaining about some additional realism introduced into FS2004 which wasn't there before?

Anyway, that's my $0.02 :)

Well, I've added the item to my list, and it is reasonably easy to do (for third party inputs, and the random icing option) so it'll probably get done in the next version.

I'd like to know what icing level you need to set to force this failure -- is it only the top level, or anything other than 'slight' (1)?

I'd also still like to know whether it is really an FS bug or not. Maybe MS just changed the icing scale in FS2004 compared with previous releases, and the values used in the Weather programs need scaling down for the same effects as before. Or maybe the new effects are more realistic. There's a difference.

Regards,

Pete

[jase439]

And with anti-icing enabled or disabled?

That would be with anti-icing enabled.

There's no icing in FSUIPC unless you elect to enable the random cloud icing. Again, this applies to third party weather inputs only, not to FS "real weather" downloads.

Right. From what I have observed, I *think* ActiveSky 2004 will randomly add icing regardless of whether the user specified it in FSUIPC. It does not seem to care whether "random icing" is checked or unchecked. FSMeteo, on the other hand, doesn't explicitly introduce icing; if you enable it in FSUIPC you get it from time to time...otherwise you don't get it at all. I am new to AS2K4, so am not quite as familiar with it as FSMeteo.

And why don't the problems occur with FS downloaded weather? Don't the Jeppesen/Microsoft weather reports ever have icing in clouds?

Good question. I don't know the answer to this. I have never seen it personally. The only times I've been able to get icing conditions is if I download my weather through a 3rd party app (like AS2K4 or FSMeteo) or I explicitly create the icing conditions by hand.

I'd like to know what icing level you need to set to force this failure -- is it only the top level, or anything other than 'slight' (1)?

Well, "trace" seems to work fine - all the time for me. From what I've read from others is that the problem is isolated to "severe" icing (which has led to the theory that the pitot-heat cannot cope with severe icing). Let me create some conditions and experiment with different aircraft and see if I can't pull together a more controlled test case for you.

More to follow.

Thanks,

J

[Mark McG]

Pete,

this problem first occurred when we were beta testing AS2004. In the beta versions, Damian was able to successfully isolate it and prevent it from happening. Some how it slipped back into the release version.

Numerous theories abound as to what causes it, but as Jason has said it is limited to when severe to extreme icing conditions are present. Using Wing and Engine Anti Ice along with the Pitot Heat seams to cure/prevent the problem. Hopefully within the next few days, testing of the AS2004 update will begin, and I will be able to see if this problem has been fixed.

I definitely do not think this is an FSUIPC issue. As Pete has mentioned, it seems to be a slightly different way that Icing Strength is coded in the new FS9 weather engine that is causing this - Icing being imported to FS9 from the weather programme using FS2002 values and being too powerful/strong for FS9's weather engine.

A good Weather SDK from MS would go a long way to solve many of the reported issues with weather in FS9. Unfortunately I do not see something with the depth required by Damian or Marc being released - more likely a watered down version.

Jason, could you write down what tests you are carrying out and email them to me mark@ntlworld.com

I will then test them with MR1 of AS2004 and post my conclusions.

[jase439]

To rule out any potential funny business involving third party aircraft, I decided to find a test case using a stock MS aircraft and a simple global weather setting consisting of a single cloud and temperature layer favorable for icing:

Here are the parameters I used:

I started with the default KSEA situation, using the Cessna 182S. I modified the default fair weather theme, and selected global weather, then using the advanced weather editor I changed the following:

* Changed the lower cloud deck from cumulus to stratus. Changed the coverage to overcast. Set tops at 5000, bottoms at 1000. Icing to Severe (initially).

* Changed the surface temperature to 5C and the dewpoint to 5C.

* Visibility to 1 mile (just for giggles - should have no bearing on this, but mentioned here for completeness :)

With each icing level I ran two tests. The first was with the pitot heat off. This was to verify that icing conditions, did in fact exist. I verified this by observing the airspeed indicator drop to zero shortly after passing 1000' and entering the cloud layer. In ALL cases, I was able to verify that valid icing conditions existed.

In the second round of tests, I ENABLED pitot heat on the ground while motionless on the runway. I departed KSEA and allowed the aircraft to climb into the clouds. With SEVERE icing enabled and pitot-heat ENABLED, my airspeed indicator dropped to zero passing 1200' MSL. The gauge remained frozen for over a minute, at which time I began to cycle the pitot-heat off and on again using SHIFT+H. Within 10 seconds, my airspeed indicator returned and I had no further icing conditions.

The same behavior was observed with MODERATE icing save the fact that it took a bit longer before the AS indicator froze. Again, cycling the pitot-heat using SHIFT+H after a minute or longer with the AS gauge frozen eventually brought the AS indicator back.

LIGHT and TRACE icing DID NOT yield these results. They behaved as they did in FS2002. If the pitot-heat was enabled on the ground (or at anytime in flight), I did not observe any freezing of the AS indicator. If I turned it off while in the air, it froze up very shortly, and would "unfreeze" shortly after turning it on again - as one would expect.

I repeated the SEVERE and MODERATE tests using the PMDG 737-600 SU1. I observed similar behavior as that observed in the Cessna 182 - although the freeze ups were somewhat intermittent and much less predictable. I did not observe any definitive benefit one way or another with engine or wing anti ice. These latter tests with TAI were inconclusive at best. The behavior of the pitot heat (or probe heat as it would be in the 737), however, was more or less the same. In all cases, LIGHT or TRACE icing caused no problems at all.

As Mark states, while this might not be a bug per se, the way icing strength is modelled in FS9 is different and can render third party panels designed for FS2002 helpless in icing conditions. I certainly believe it would be helpful if apps like AS and FSMeteo could deal with this at their level - but the problem is still present even for those who would set their weather globally - so i believe an icing threshold is a good candidate for a future FSUIPC release.

Hope this is helpful.

Cheers,

J

[Mark McG]

In the second round of tests, I ENABLED pitot heat on the ground while motionless on the runway. I departed KSEA and allowed the aircraft to climb into the clouds. With SEVERE icing enabled and pitot-heat ENABLED, my airspeed indicator dropped to zero passing 1200' MSL. The gauge remained frozen for over a minute, at which time I began to cycle the pitot-heat off and on again using SHIFT+H. Within 10 seconds, my airspeed indicator returned and I had no further icing conditions.

Is there some way of verifying that Pitot Heat is in fact on if enabled on the ground. It could be something as simple as a line of code in the pitot heat logic that prevents it from being enabled on the ground.

Mark

[Pete Dowson]

Is there some way of verifying that Pitot Heat is in fact on if enabled on the ground. It could be something as simple as a line of code in the pitot heat logic that prevents it from being enabled on the ground.

The only "pitot heat" indicator I know of is the switch itself, plus an electrical circuit flag ("PITOT_HEAT_CIRCUIT_ON"). You could check the latter. It is in the list of Gauge values which FSLook provides, and as far as I can see the circuit normally stays on whilst there's an active electrical system whether ot not the switch is operated. Possibly in extreme icing the circuit blows a fuse?

There's also an FS control called TOGGLE PITOT BLOCKAGE which you could try using -- assign it in FSUIPC or in the FS Options-Controls-Assignments (where it's called "Pitot tube blockage on/off").

Regards,

Pete