Vacuum pump failure today

WannFly

Final Approach
Joined
Nov 28, 2016
Messages
6,553
Location
KLZU
Display Name

Display name:
Priyo
Nice day and a lot of people flying around. Went to grab lunch with a “friend” to kbrd, while coming back the AP started acting up, after spending some time to fix it, said to hell with it and started hand flying, a quick glance at the primary AI showed a very slight left bank, verified with G5, straight and level, no increase or decrease in airspeed. Then over 3 min or so the AI tumbled. The vacuum failure light came on.

Complete non event with G5 and el citric backup vacuum, but it real world AI failure experience and what we do in practice failures are 2 diff animals. When u cover up the AI it is completely out of the picture, ur brain is not processing it. In real world it happens slowly, plays with your mind. I can only imagine how dangerous this stuff is if u are in hard IMC.

Obviously didn’t declare, but I was on FF and did let center know that I have a instrument failure. Fargo approach have me a straight in and sent a student out of the downwind to ensure I don’t have to go around, or deal with spacing issues.

Other than that small excitement, awesome day to fly today.

Review those emergency procedures, you will need them. It’s a matter of when, not if


Sent from my iPhone using Tapatalk
 
I thought you quit breaking stuff on your plane for a while?

Sigh.... don’t ask man. While putting her away, put the wire thru the spit window, then closed it as if nothing was going the the window and cracked the pilot side glass, about 2 inch..... overall not a good day for ownership. But my “friend” enjoyed the flight very much, so it was all worth it


Sent from my iPhone using Tapatalk
 
Airwolf wet vacuum pump. Never fail again. Okay, it’s two AMUs and ya gotta install an air/oil separator. It’s sorta a wash on cost vs replacing a dry pump every six hundred hours....
 
Gotta love your new G5! All that vacuum pump is for now is to run your autopilot. Who doesn't like hand flying sometimes anyway?

In real world it happens slowly, plays with your mind. I can only imagine how dangerous this stuff is if u are in hard IMC.
You nailed it with this comment.
 
Airwolf wet vacuum pump. Never fail again. Okay, it’s two AMUs and ya gotta install an air/oil separator. It’s sorta a wash on cost vs replacing a dry pump every six hundred hours....
:(:(:(

i might get rid of the vacuum system some day.. not sure if its worth putting in a 2k pump now....
 
:(:(:(

i might get rid of the vacuum system some day.. not sure if its worth putting in a 2k pump now....
It’s a tough call. No real concerns about a dry pump since 600 hours is several years of operation.
 
My planes got both a dry and a wet pump. They decided they wanted two vacuum pumps so they got the STC for two. I need to make sure both are working... I know at least one is.
 
My planes got both a dry and a wet pump. They decided they wanted two vacuum pumps so they got the STC for two. I need to make sure both are working... I know at least one is.
Your plane sucks?
 
What'll your backup be?

havent thought everything through yet. a lot of people are ditching the vacuum system with G5 (dual G5 have a revisionary mode for the AI), i would still like to keep another AI backup especially for IFR. i just dont know rt now will that backup be vacuum based or another electric based
 
havent thought everything through yet. a lot of people are ditching the vacuum system with G5 (dual G5 have a revisionary mode for the AI), i would still like to keep another AI backup especially for IFR. i just dont know rt now will that backup be vacuum based or another electric based
Make sure you back up power sources, too...all the electric backup gizzies in the world won't help if they lose power.
 
Make sure you back up power sources, too...all the electric backup gizzies in the world won't help if they lose power.
the G5's have 4 hr backup battery in each. not sure if they last 4 hrs or not though
 
Glad it was a non-event. I fly without an AI all the time... and I once thought that was insane.

Me too. Usually the glass is configured for the moving map display with tapes for airspeed/altitude on the sides and instrumentation on a strip beneath. I'm strictly VFR, but two button-pushes gets me the AI if I accidentally wander into the soup.
 
With dual G5’s in my plane my vacuum AI in the TC hole is now my backup. I can tell you with certainty there have been a few occasions that I glanced at that vacuum AI to reassure myself the G5 was telling me the truth.

Before installing the G5’s I remember being aggravated that I had to retain that vacuum AI. In hindsight I love having it.
 
I've had a vacuum failure twice, both in older round dial 172s. Once in day IMC and once in night VFR. The night one was a none issue, but the IMC one was interesting. Those partial panel approaches you learn during instrument training does happen from time to time.
 
I've had a vacuum failure twice, both in older round dial 172s. Once in day IMC and once in night VFR. The night one was a none issue, but the IMC one was interesting. Those partial panel approaches you learn during instrument training does happen from time to time.
Just hope if you have to make an approach without your DG that you can do one with south bound headings!
 
I don't mind having a vacuum backup either. At some point the redundancy can get a bit ridiculous, but hey, it's attitude information... the most critical display in the airplane for IFR flying.

Right now I'm running a G5 ADI, with a G5 HSI to be installed soon. That'll give me a primary ADI at all times backed up with the HSI which will revert to primary if the ADI fails. Backing that up is a vacuum powered ADI which I need to keep for my autopilot. Backing THAT up is a second vacuum ADI which I installed many years ago. And last but not least -- okay, maybe it is least -- I have AHRS from my GTX345 available on the iPad.

I think we have to consider a new non-redundant failure mode which is shared software across instruments. With two G5s both running a new firmware revision, both are susceptible to any programming bugs or software issues. That being the case the backup battery powering both can't be considered redundant. It would take something non-G5 to back up the attitude source. Vacuum makes sense as the presence of electrical power makes no difference.

I doubt by the time I'm ready to install a GFC500 (here's hoping the PA-30 ends up on the AML) I'll still have this concern. I'd ditch vacuum in a heartbeat as soon as I have an autopilot that doesn't require looking at a legacy vacuum ADI.
 
I thought the AP went off the turn and bank, which is electric.

Some autopilots are rate based (T&B), others are attitude based (ADI). Mine is attitude based. So I have to keep my old attitude indicator, which in turn requires that I keep my vacuum system.
 
...........Complete non event with G5 and el citric backup vacuum, but it real world AI failure experience and what we do in practice failures are 2 diff animals. When u cover up the AI it is completely out of the picture, ur brain is not processing it. In real world it happens slowly, plays with your mind. I can only imagine how dangerous this stuff is if u are in hard IMC................

Yeah. After reading about an accident where this could have been an issue I got a CFII and a simulator and did some loss of Instrument training. I had done of lot of partial panel but that was always 'instant' simulated failures. Never 'recognition' of failure. It's a whole different thing when instruments start lying to you than when they just get covered up.
 
I thought the AP went off the turn and bank, which is electric.

Most of the Piper-branded Century autopilots, including the one in my Aztec, require the vacuum AI.
 
With dual G5’s in my plane my vacuum AI in the TC hole is now my backup. I can tell you with certainty there have been a few occasions that I glanced at that vacuum AI to reassure myself the G5 was telling me the truth.

Before installing the G5’s I remember being aggravated that I had to retain that vacuum AI. In hindsight I love having it.

I am thinking about doing exactly what you have for the same reason. Piper autopilot. I'm not fussed about the vacuum system as I need it for the boots, and I have 100% redundant vacuum pumps.

But I get concerned about having two AIs and no TC. As long as they both give you the same information everything is fine. But if they start deviating from one another, one is faced with resolving which one is correct. In solid clag that could be a wee bit stressful imo. I'm referring to the insidious gradual failure of one.

Any opinions on using the reversion of the G5 HSI to AI to resolve such a situation?
 
I am thinking about doing exactly what you have for the same reason. Piper autopilot. I'm not fussed about the vacuum system as I need it for the boots, and I have 100% redundant vacuum pumps.

But I get concerned about having two AIs and no TC. As long as they both give you the same information everything is fine. But if they start deviating from one another, one is faced with resolving which one is correct. In solid clag that could be a wee bit stressful imo. I'm referring to the insidious gradual failure of one.

Any opinions on using the reversion of the G5 HSI to AI to resolve such a situation?

I can hit the knob and choose PDF, but I will also verify with other mechanical instruments like ASI and VSI


Sent from my iPhone using Tapatalk
 
I have AHRS from my GTX345 available on the iPad.
I've only used the AI on my Foreflight twice with yesterday being the second time. (Same GTX345 setup.) The first time it worked well. Yesterday I couldn't get the attitude information to display correctly. I tried re-syncing the straight and level but it didn't hold correctly. I googled it a bit and it seems that the installer is suppose to program/sync the AHRS on the 345 correctly for it to work. Have you ever noticed it act up? I'm guessing at this point that the installer forgot this step? Thoughts?
 
I am thinking about doing exactly what you have for the same reason. Piper autopilot. I'm not fussed about the vacuum system as I need it for the boots, and I have 100% redundant vacuum pumps.

But I get concerned about having two AIs and no TC. As long as they both give you the same information everything is fine. But if they start deviating from one another, one is faced with resolving which one is correct. In solid clag that could be a wee bit stressful imo. I'm referring to the insidious gradual failure of one.

Any opinions on using the reversion of the G5 HSI to AI to resolve such a situation?
Keep the partial panel skills polished. An AI is just one vote in the system. Any bank should have a heading change which verifies it so conflict between two AIs should be resolved quickly with no stress.

This is one of the things that bothers me about instrument training which focuses on the AI. I was re-reading Machado a while back and put it down quickly once I hit the section about focusing on the AI. Yes, the Air Force trains that way and I'm sure it's great when ya have jets with government budgets and enlisted maintenance staff. For those of us in spam cans I think we are better off keeping partial panel skills sharp and remembering that the AI is just one vote in the system and must be supported by the other instruments. Keep the scan going and take action when certain.
 
Keep the partial panel skills polished. An AI is just one vote in the system. Any bank should have a heading change which verifies it so conflict between two AIs should be resolved quickly with no stress.

This is one of the things that bothers me about instrument training which focuses on the AI. I was re-reading Machado a while back and put it down quickly once I hit the section about focusing on the AI. Yes, the Air Force trains that way and I'm sure it's great when ya have jets with government budgets and enlisted maintenance staff. For those of us in spam cans I think we are better off keeping partial panel skills sharp and remembering that the AI is just one vote in the system and must be supported by the other instruments. Keep the scan going and take action when certain.

Yup. Ya gotta use other instruments to verify the AI isn't going bad on you. There are simulators that mimic that slow 'wind down' where it starts lying to you before it finally just goes TU.
 
Keep the partial panel skills polished. An AI is just one vote in the system. Any bank should have a heading change which verifies it so conflict between two AIs should be resolved quickly with no stress.

This is one of the things that bothers me about instrument training which focuses on the AI. I was re-reading Machado a while back and put it down quickly once I hit the section about focusing on the AI. Yes, the Air Force trains that way and I'm sure it's great when ya have jets with government budgets and enlisted maintenance staff. For those of us in spam cans I think we are better off keeping partial panel skills sharp and remembering that the AI is just one vote in the system and must be supported by the other instruments. Keep the scan going and take action when certain.

:yeahthat:
 
Keep the partial panel skills polished. An AI is just one vote in the system. Any bank should have a heading change which verifies it so conflict between two AIs should be resolved quickly with no stress.

This is one of the things that bothers me about instrument training which focuses on the AI. I was re-reading Machado a while back and put it down quickly once I hit the section about focusing on the AI. Yes, the Air Force trains that way and I'm sure it's great when ya have jets with government budgets and enlisted maintenance staff. For those of us in spam cans I think we are better off keeping partial panel skills sharp and remembering that the AI is just one vote in the system and must be supported by the other instruments. Keep the scan going and take action when certain.
I wholeheartedly agree with keeping sharp on your backup, whatever that may be. When it comes to jets, the backup is generally a third AI. Unfortunately many pilots apparently don't learn to use the AI properly when they focus their scan on it.
 
I've only used the AI on my Foreflight twice with yesterday being the second time. (Same GTX345 setup.) The first time it worked well. Yesterday I couldn't get the attitude information to display correctly. I tried re-syncing the straight and level but it didn't hold correctly. I googled it a bit and it seems that the installer is suppose to program/sync the AHRS on the 345 correctly for it to work. Have you ever noticed it act up? I'm guessing at this point that the installer forgot this step? Thoughts?

The steps are basically to enter the angles the 345 is mounted in the aircraft into the setup pages of the 345 so it knows which was is up, and whether it’s angled toward the pilot in the panel (which would throw off the accelerometers since the default is that it’s square with the airplane’s direction of flight), etc.
 
With two G5’s, a vacuum AI, and an IFD 550 I have 4 independent attitude sources. Changing the G5 HSI to pfd takes about 3 seconds. My CFII and I were talking today that “failing my instruments would be a challenge for the check ride. The independence of the G5’s makes them as close to bullet proof as you can get. The ifd 550 (synthetic vision) and the vacuum AI are a bonus.
 
Keep the partial panel skills polished. An AI is just one vote in the system. Any bank should have a heading change which verifies it so conflict between two AIs should be resolved quickly with no stress.

we are better off keeping partial panel skills sharp and remembering that the AI is just one vote in the system and must be supported by the other instruments. Keep the scan going and take action when certain.


:yeahthat: x two!
 
Back
Top