PDA

View Full Version : ACAS against unverified SSR


Pierre Argh
13th Dec 2005, 11:44
Yesterday I was providing RAS in Class G (I know, I like to live dangerously!) and called a 7000 squawk indicating slightly below, slowly climbing towards my traffic. The pilot rejected my avoiding action, saying he had TCAS contact and was happy to continue.

I breathed a sigh of relief, but being quiet it gave me time to think. Did the pilot know the Mode C was unverified, and therefore unreliable? I asked, and discovered pilot's do NOT get the Mode A data, just the height readout.

Whilst this would have to be verified in regulated airspace, out here in Injun Country it need not... Seems to me another case, where pilots should be using TCAS/ACAS for collision avoidance (as it is intended) and not using it as a tool to maintain separation? Any thoughts?

threemiles
13th Dec 2005, 11:53
What do you mean by unverified SSR?
TCAS uses Mode C returns for altitude calculation and diversity antennae for bearing, latter being correct to +-10 degrees only.
What would be the use of Mode A? It does only provide the squawk selected on the transponder panel.
Besides this normal separation criteria for uncontrolled airspace apply to all pilots.

Pierre Argh
13th Dec 2005, 12:07
Threemiles...

When I allocate a squawk as a Controller I am responsible for checking that the code has been correctly selected and that the Mode C response (if available) is within tolerance + 200ft. In which case the data can then be used with confidence. You correctly say that TCAS uses altitude comparison of Mode C data... but traffic that departs from a non-radar/ATC environment and selects a conspicuity squawk (7000) has never had the data verified... i.e. the data picked up by another ACAS system is unreliable?... an yes we do see SSR that is out of tolerance!

I did not advocate that pilot's should receive the Mode A.., for although in the case I use the pilot might have recognised conspicuity, it would be unreasonable to expect them to be able to decode every code in-flight and then know which are verified and which aren't... but, I'm suggesting maybe it's not the wisest decision to continue towards a confliction, in IMC, relying on ACAS to prevent collision when the in-cockpit information might be inaccurate?

In addition to the 7000 squawk, many special purpose codes e.g. 0020 - Air Ambulance, 0033 - Parachuting, 0023 - SAR Ops, 0036 - Police, 0047 - Fishery Protection etc etc are considered as unverified data

I hope that makes my original question clearer (and that I don't get shed loads of replies about not providing ATS in unregulated airspace... which is another issue altogether?)

HEATHROW DIRECTOR
13th Dec 2005, 13:04
It's quite worrying when one has a good number of light aircraft (and some larger aircraft) transponders with inaccurate altitude reporting. One morning the north part of the LTMA went into chaos with a squawking aircraft going through the middle at (apparently) FL70. We later learned that it was a small aircraft at around 1500 ft. TCAS is incapable of knowing that an indicated mode C altitude is wrong, with interesting results.

Despite what the book says about "ignoring" unknown aircraft in Class A airspace, it's sometimes very difficult to do so!

bookworm
13th Dec 2005, 13:22
One morning the north part of the LTMA went into chaos with a squawking aircraft going through the middle at (apparently) FL70. We later learned that it was a small aircraft at around 1500 ft.

Interesting to note that the one anecdotal example of an incorrect Mode C return that we have is a discrepancy of more than 5000 ft. Under a RAS, minimum separation of only 3000 ft is provided against unverified Mode C.

Is there any reason to believe that discrepancies of say 300 to 3000 ft are more likely than completely random returns?

Pierre Argh
13th Dec 2005, 13:32
IMHO none whatsoever... I have seen errors of 10,000ft plus, but in fact I guess we are more likely to spot the gross error (which perhaps equally suggests there may be even more errors that slip by un-noticed?).

In one interesting case I was working traffic on the South Coast below 3000ft... when I got a call from Swanwick asking me to confirm the level of my traffic, which their equipment was picking up at FL320. The SSR had been verified and was indicating correctly on my display.

PPRuNe Radar
13th Dec 2005, 13:59
HD

Despite what the book says about "ignoring" unknown aircraft in Class A airspace, it's sometimes very difficult to do so!

The book has changed slightly in recent years. If you believe the aircraft has made an unauthorised penetration of CAS (or have info it has), then you can now take action to avoid it. From MATS Part 1 for airspace classes A-D:

Neither avoiding action nor traffic information shall be passed unless radar derived or other information indicates that an aircraft is lost, has experienced a radio failure, or has made an unauthorised penetration of the airspace.

Pierre

I can see exactly where you are coming from. But to take your argument to its logical conclusion, should pilots no longer react to TCAS RAs since they have no way of knowing whether the intruder aircraft has verified Mode C or not ?? I think there is a global acceptance in the pilot world that on a very small number of occasions the Mode C might not be what it seems, but for the vast majority it will be correct and so TCAS action, when commanded as an RA based on it, is the wisest thing to follow.

5milesbaby
13th Dec 2005, 14:01
Pierre, two points to note:

1: When passing the traffic information did you tell the pilot the height readout was unverified? We are taught to use "indicating FL*** unverified" when passing such information.

2: Any pilot using TCAS as a sole form of separation when under a RAS is just plain stupid. I have seen pilots inside CAS asking about traffic on TCAS and found that the display showed the a/c about 8 miles out of position. I hope that in your case that the pilot was aware of his actions and that you no longer were responsible for any separation against this conflict. I'd have downgraded the service and passed traffic updates as required and seen how he responded to that too.


Edited as PPRuNe Radar posted at the same time!!

Radar - I think Pierre was a little shocked that this pilot used TCAS as the sole source of traffic avoidance, surely we all know that their accuracy isn't perfect as a radar display so I think a compliance to the Controllers avoidance would have been safer.

HEATHROW DIRECTOR
13th Dec 2005, 14:23
PPrune Radar. Yep.. that's what my latest MATS Pt 1 (recently downloaded) says. My wording may not have been "chapter and verse" but it sums up the situation, which is unchanged from when I started radar work back in the 70s.

bookworm
13th Dec 2005, 14:58
I think Pierre was a little shocked that this pilot used TCAS as the sole source of traffic avoidance, surely we all know that their accuracy isn't perfect as a radar display so I think a compliance to the Controllers avoidance would have been safer.

Does Pierre know for sure that the aircraft was in cloud? In cloud, relying on TCAS separation alone seems unwise if improving horizontal separation is a possibility. But in visual consitions, it doesn't seem unreasonable that a crew would expect to be able to see and avoid an aircraft shown on TCAS, even if they hadn't sighted it at the time the avoidance advice was refused.

Pierre Argh
13th Dec 2005, 15:06
PPRuNe Radar, I'm all for TCAS, but IMHO it is not the answer to everything as some seem to think! As 5milesbaby correctly says, I was not saying pilots should ignore TCAS RAs (although with unverified data the situation still applies)... what I'm suggesting is the pilot should IMHO not be using TCAS to maintain separation and should have taken the avoiding action... unless he was visual with the conflictor or had flight conditions that gave him him a good chance of spotting the traffic early... and leave TCAS to resolve the confliction if both he and I screw up!!!

5milesbaby... Tes, I told the pilot it was unverified and in fact repeated it when he said he was happy to continue because his TCAS was showing it below him. Confliction passed we discussed the situation over the radio, and I got the firm impression the pilot didn't fully understand what unverified meant and the implications of that.

Bookworm, Agree and no I cannot be 100% confident the pilot was in cloud, but weather conditions and the level at which the encounter occurred strongly suggest he was either in cloud or in and out of it... Shall we say, unlikely to be conditions in which visual sighting could be acheived? But, my point is the pilot doesn't know/didn't react to the fact the information was unverified... so I humbly suggest the prevailing Flight Conditions are irrelevant.

zzjayca
13th Dec 2005, 16:13
It doesn't only happen in uncontrolled or low level airspace.

A week ago in class A airspace I had an airliner refuse clearance from FL360 to FL370 because of "traffic" displayed on TCAS which, in his opinion, was a conflict. I explained to him that he would pass at least 14nM behind the "perceived traffic". He still would not accept the clearance to FL370 until the aircraft no longer appeared on his TCAS display. The closest the two aircraft came to one another was 15.2nM.

I believe that pilots are not being instructed properly as to the limitations of TCAS, especially with respect to azimuth.

Pierre Argh
14th Dec 2005, 07:45
zzjayca, The UK Guild of Air Traffic Controllers recently published an article warning precisily of this (the dangers of using ACAS/TCAS for azimuth separation)... the pilot is presented with a relative motion picture that makes it very difficult to resolve conflictions. When will we (and by that I mean mainly the pilots) realise that these are great fall-back safety systems but are NOT intended to be used as a primary means of separation?

Scott Voigt
17th Dec 2005, 19:06
The pilot to put it bluntly was an idiot... RADAR is far more accurate in azimuth than TCAS/ACAS. That is one reason why it is expressly forbidden to use turns for TCAS/ACAS RA's. Use the vertical all you want as that is VERY accurate, but azimuth is NOT. We see pilots who do try to second guess what you are doing with TCAS all the time, we have just become accustom to it, and feel, FINE, just do what you are told. (sequencing) It is far more difficult to do the job when a pilot is trying to "help" you with spacing and things of that nature.

regards

Scott

DFC
17th Dec 2005, 19:56
I believe that a pilot using TCAS for separation is like a controller using the STCA for the same purpose. TCAS is a back-up safety tool that hopefully alerts us to an error in time to correct the situation.

There are strict reqirements for the mode C transmitted by transponders to be reguluarly checked. Have to check but I don't think that the UK LAMS requires full altimeter and encoder calibration at all.

The most interesting point about TCAS not being set up to check mode A is that if a controller finds mode C out of limits and the pilot can not turn off mode C, that flight is instructed to set code 0000.

Try explaining to a pilot receiving a RA from a aircraft dead ahead showing the same height that the aircraft is squalking 0000 and is 7000ft below the airway!

Perhaps we will eventually have a requirement for an annual calibration certificate to be given to all transponders and encoders

Regards,

DFC

SirToppamHat
17th Dec 2005, 20:08
I understand completely the issues relating to verification of Mode C, and it is pleasing to see that the general line in this thread is in agreement. It is worrying that pilots are seemingly relying on ACAS as some sort of AI radar.

Back to the Mde C verification issue, it is absolutely the norm for mil ac operating with ASACS units to be verified within the first minute or 2 of arriving on frequency, and this process seems to be well understood by those we work. Just a couple of Qs, though, when an ac takes off from a large airport, who is responsible for Verifying its Mode C? Presumably the first agency capable of reading it whilst in 2-way comms, but these are busy places at a busy stage of flight - does it always get done?

I promise I am not digging here, it's just out of interest.

STH

5milesbaby
17th Dec 2005, 21:00
SirToppam - you are correct that its the first agency capable of doing so verifies a Mode C. The verification can be maintained throughout the next radars/units as long as there is an associated validated Mode A and it has been approved to do so.

As for does it get done at busy airports then yes, always. Departing a/c from say Heathrow that don't say their passing level on first contact will soon realise the stubborn arrogant tone of the controller when they are asked for it and subsequently have futher climb/turns delayed whilst its checked.

Scott Voigt
18th Dec 2005, 06:12
DFC;

In the US we do require that altimetry and Mode C be checked I beleive every two years. If we find that someone has a bad mode c while we are talking with them, we can tell them to turn off the mode C so that it doesn't give false RA's for TCAS. We also are supposed to turn in a report to flight standards so that they can check with the pilot to ensure that the aircraft gets the altimetry or the mode C fixed...


regards

Scott