Menu Sign In Contact FAQ
Banner
Welcome to our forums

GTX330 FL display showing FL100 or FL515

I would ground D4. If it is an internal fault in the transponder, the problem not go away. If it is a fault in the wiring or in the encoder it will go away.

EGBE - Coventry

Peter,

If your altimeter is bouncing around, it is probably not its fault. Modern TSO altimeters don't have the internal friction that earlier units had and will bounce back and forth if the panel vibrates too much, typically at full power. I would guess that replacing the altimeter will not solve this and you need to address the vibration. Have you dynamically balanced your prop? Does your panel have any vibration isolation?

KUZA

Why? Most modern avionics have their own encoding altimeters and connect to the static system directly. Garrecht, Funkwerk/Filser and Trig transponders, S-TEC autopilots, Avidyne DFC90 autopilot, Aspen PFD etc.

The only place where I use that crappy Gray interface is for the GNS430W. I wonder if it could use ARINC429 from the Aspen...

Integrating an encoder into a transponder would be a mistake IMHO. The encoder fails more often than the transponder and if it is a blind encoder it is a throw away box if it can't be adjusted ($150 for an encoder, $3000 for a transponder). It also means that if a transponder were to be removed for service, a static check would always be required on re-installation, which is a waste of time and effort. Serial blind encoders are available, but many transponders only support the Gray code interface. The Gray code is used so that a momentary change in one of the bits as the aircraft changes altitude doesn't result in a wild change in the altitude because the Gray code only transitions one bit for each 100 feet of altitude change. The transponder transmits a Gray code for this reason.

Aspen might support sending air data on ARINC 429, but last time I checked it did not. They did add a serial output of air data which may be used by the GNS430W. There are at least a half a dozen possible ways of providing the pressure altitude to the GNS430W and it primarily uses it as part of its RAIM algorithm (baro aiding), although with SBAS working, RAIM is not used.

KUZA

I would ground D4. If it is an internal fault in the transponder, the problem not go away. If it is a fault in the wiring or in the encoder it will go away.

The issue is incredibly intermittent.

If your altimeter is bouncing around, it is probably not its fault. Modern TSO altimeters don't have the internal friction that earlier units had and will bounce back and forth if the panel vibrates too much, typically at full power. I would guess that replacing the altimeter will not solve this and you need to address the vibration. Have you dynamically balanced your prop? Does your panel have any vibration isolation?

The issue started maybe a few years ago but was rare. Over the years it got worse, but was never seen in level flight. On the last few flights it has appeared in level flight too, with the big excursions.

The prop was dynamically balanced on several occassions; the most recent being a few years ago when it was overhauled.

The panel has vibration isolation, which was improved about a year ago, with the details being visible in the last pictures in this writeup.

It is vibration related in its magnitude, but it does not feel to me that the entire issue is due to any specific type of panel vibration, because it appears spontaneously and then disappears for a bit, with no change in engine power settings.

The KEA130A is about 12 years old.

The D4 wire could be yet another bodge done by the "145" company which installed the TCAS in which case they will have yet another laugh if this turns out to be a wiring issue and the altimeter was OK all along. But I don't really mind having a spare KEA130A on the shelf because I am unlikely to ever replace it with something better.

Serial blind encoders are available,

Sure, but I need Gray code pressure altitude for the KLN94 also, which definitely doesn't have an ARINC429 option.

Administrator
Shoreham EGKA, United Kingdom

Update:

A new KEA130A altimeter was installed and both the wobble and the D4 behaviour appear to have gone.

Obviously time will tell, and it could just be that the new one responds differently to panel vibration.

I got an overhauled one from Castleberry in the USA, $3200 outright. A new one is over $8k. I no longer use any UK firm for this sort of work; last time I got something done (a KI256 horizon) one "respectable" firm took 6 months and at the end said they could not generate any paperwork because the informal paper-signing deal they had going with some 145 company was no longer available (yeah, really great) so I got a basically worthless job.

Amusingly, the KEA130A is made "for" Honeywell/Bendix-King by United Instruments in the USA (who make loads of "cheap" altimeters), who get somebody to make it for them in .... Japan.

Administrator
Shoreham EGKA, United Kingdom

This issue has popped up again, very intermittently.

Hmmm…. looking at the KEA130A data I have (I have wiring diagrams only) it doesn’t have D4!

It has a DB15 connector and this diagram

accounts for all 15!

So the “overhaul” of the altimeter fixing the above issue was an accident.

The GTX330 does have a D4 input and one wonders where the shop connected it to…

Administrator
Shoreham EGKA, United Kingdom

I am sure the KEA130A has a D4 output since it is certified to 35000ft and above 30700 this D4 is needed. (pin 1 if i remember correctly)
It could be that your (D4) wire is shaving agains ground some where.

another option is to install a blind encoder with both serial and gilham output (eg ACK A-30 mod9 ) disconnect only the D-conn and connect it to the Blind encoder
note: leave the small 3 pin connector connected to the KEA130A (baro correction for AP) add the serial wiring from encoder to TXPDR.
the KFC225 needs gilham altitude, GTX330 can use both and will use the serial input if there is one if this fails the GTX330 will take the gilhamcode

Last Edited by Jetprop at 09 Aug 22:11

@Jetprop – you are right of course. When counting the pins I forgot that 3 of them are on a separate 3-pin connector.

However, on the TB20, D4 is not connected on the altimeter.

It is also not connected to the KFC225.

It is also not connected to the TAS605, according to the wiring diagram in the work pack I have from that installer.

And it doesn’t go to the KLN94:

So D4 should not even be present in the harness at all. I can get that visually checked easily enough…

Also, from what I can see, these signals are wire-ORed active-low, so to activate D4 one would need to ground it. So the suggestion earlier up to ground D4 to suppress this problem would be the wrong way round. The GTX330 has internal pullups – as this dual-GTX330 installation (from the GTX330 IM) shows

IMHO, the best thing would be to verify there is no connection to D4 on the back of the GTX330 and, if there is one, disconnect it.

Can anyone installing a GTX330 confirm whether they ground D4 or leave it unconnected?

This seems to be clear (GTX330 IM)

D4 is pin 1 on the KEA130A – I got this from an EGPWS manual

Administrator
Shoreham EGKA, United Kingdom

The GTX330 has a known bug that occasionally detects D4 as being active. With radar interrogating the transponder once every 5 to 12 seconds, it does not show up very often as the condition lasts for a few seconds at a time, maybe two or three times an hour. ATC computers ignore the one off or two off erroneous altitude spikes. When ADS-B was added to the GTX330ES, the frequency of the squitters with altitude is about once per second and each one shows up on a log file kept by the FAA to evaluate performance of the ADS-B Out unit. I analyzed the data provided by the FAA for one of the errant systems and determined it had to be a spurious D4 and not a big deal. Since then Garmin found a software fix and it will get released sometime this year.

KUZA

The GTX330 has a known bug that occasionally detects D4 as being active

Absolutely astonishing – thank you @NCYankee!

I have spent a bit of today on this and can report that pin 1 on the KEA130A (D4) is a no-connect. This corresponds to the Socata TB MM but one could never 100% rely on that, obviously…

So I suspected it is a duff GTX330.

Could the rate at which the D4=active appears be geographically dependent?

The KEA130A encoding altimeter issue reported elsewhere here was a real duff altimeter, because even the KC225 autopilot complained it was getting an invalid altitude.

Administrator
Shoreham EGKA, United Kingdom
Sign in to add your message

Back to Top