Jump to content

GFC500 Update


81X

Recommended Posts

For Long Body Mooney owner who is not in immediate need of an autopilot and insists on Garmin I'd wait for the GFC600 - much better autopilot. I bet that ends up on the list down the road. If a Bonanza gets it why not eventually a Bravo or Ovation?

http://www.aviationconsumer.com/issues/49_9/avionicsreport/Garmins-New-Autopilots-Flawless-Performers_7111-1.html

 

It’s also over 25k installed I’d think and that’s before addressing the AI... The king is a 10k swap out and new warranty on the servos.

 

 

Sent from my iPhone using Tapatalk

  • Like 2
Link to comment
Share on other sites

Reasoning of Garmin on requiring GPS with the GFC 500.  Still, I don't like the fact that it doesn't have CWS..
1447285294_ScreenShot2018-12-21at12_22_22PM.thumb.png.313388f0a4a990e360aefe594835d694.png


Maybe I’m confused what they mean by coupled. No roll steering but will it “fly the needles”?


Sent from my iPhone using Tapatalk
Link to comment
Share on other sites

Twenty five years ago in 1993 a KFC150 was 21,000+ installed in a Mooney
357138539_ScreenShot2018-12-21at2_30_08PM.thumb.png.820b8783cf316ff3565d9b1b208d913c.png
 


But the king is priced at under 10k and is a slide in replacement for a kfc150. It’s pretty tempting. Let’s see when it’s out.


Sent from my iPhone using Tapatalk
  • Like 1
Link to comment
Share on other sites

24 minutes ago, gsengle said:

 


Maybe I’m confused what they mean by coupled. No roll steering but will it “fly the needles”?


Sent from my iPhone using Tapatalk

 

Sounds like the answer is yes, just not in Approach mode, until gps signal is reacquired. 

Link to comment
Share on other sites

I agree completely - that seems like the way to go if you have a KFC-150.


If they are going to start to win back GA business it will be with a good G5 competitor paired with a good competitive autopilot. Will have to be cheaper/better.


Sent from my iPhone using Tapatalk
Link to comment
Share on other sites

Sounds like the answer is yes, just not in Approach mode, until gps signal is reacquired. 


Without approach mode it will only fly a lateral not a vertical needle, so it won’t fly the needles... just the needle.


Sent from my iPhone using Tapatalk
  • Like 1
Link to comment
Share on other sites

2 minutes ago, gsengle said:


Without approach mode it will only fly a lateral not a vertical needle, so it won’t fly the needles... just the needle.


Sent from my iPhone using Tapatalk

 

My reading is it won't fly any needles without GPS. It will display pretty green needles to fly either by manual commands(hdg bug, etc) or by hand.

33 minutes ago, gsengle said:

But the king is priced at under 10k and is a slide in replacement for a kfc150. It’s pretty tempting. Let’s see when it’s out.
 

 

From what I've seen that's only if you have an existing digital attitude source(KI-300, G500, Aspen) otherwise you'll have to add one of those.  Similarly it doesn't include an HSI, but it can use the existing one. I thought I saw a vapor announcement for King doing a digital HSI, but I can't find it now.

 

  • Like 1
Link to comment
Share on other sites

2 hours ago, aviatoreb said:

One can model attitude (meaning an equation best guess as to likely attitude) based on time varying observations of ground track and altitude measured from GPS, under assumptions such as coordinated flight.  Such computed attitude is inferior to a good directly measured attitude but fuse the two together intelligently and you get a new observation that is better than either separately.

In fact, in principle one can data fuse a bunch of crappy measurements to infer an over all high quality measurement.  The magic of Bayesian statistics based data fusion.

Right, and as you probably know, the output of a good primary sensor can be muted by inclusion of higher-order sensors if the combining is done poorly.   This can lead to things like Aspens red-x-ing most of the display on pitot failure, or probably the sort of thing here where Garmin turns off ILS/LOC/VOR autopilot guidance if they can't fuse it with gps.

Sensor fusion is a good thing when it works, but it seems like in many systems it can be very difficult to get it to work well and the failure handling (i.e., how to detect and what to do when various inputs fail) can get very complicated.   Hence the sort of apparent punting we're seeing with Aspen and Garmin.   As many are saying, the benefit isn't clear when you lose capabilities that were there with older, simpler technology.   Babies and bathwater and all that, I think.

Link to comment
Share on other sites

The GPS requirement for a coupled approach is disappointing.  Hopefully that's a limitation that goes away with further development of the system.

1 hour ago, gsengle said:

But the king is priced at under 10k and is a slide in replacement for a kfc150. It’s pretty tempting. Let’s see when it’s out.

 

Plus the cost of a KI300 and a KA310 converter which runs an additional $5K plus the cost of install.  Still cheaper than a GFC 600.  Total cost is probably a little higher than a Garmin autopilot, but it doesn't have the limitation requiring GPS which is an advantage.

58 minutes ago, gsengle said:

If they are going to start to win back GA business it will be with a good G5 competitor paired with a good competitive autopilot. Will have to be cheaper/better.

 

The KI300 is substantially more expensive than the G5 for attitude, and the best I can tell Bendix King doesn't have an HSI alternative- so you're likely to have one G5 in the system anyway.  Add to that, you still have original servos in the airplane which are very expensive to replace, although it makes me happy as I have a set of KFC200 servos I need to sell. 

I don't believe a KI300 and a KFC230 will give many people any confidence that Bendix King can be relied on as a vendor.  It certainly doesn't give me any confidence.  How many people do you see installing a KSN 770?  Who even knows what a KSN 770 is without google?  Yes- I had to google the model number.  What about a Bendix King glass panel?   I don't have any confidence in spending money on Bendix King avionics.  I bought a GFC 600.  The fact that the KFC 230 wasn't available to me is irrelevant; I wouldn't have changed my purchase decision.  I'll bet that the KFC230 will be a good autopilot, but I don't have any confidence in the long term support of the vendor.

The first Mooney to be approved is the M20L, which doesn't make much sense.  The rest of the long bodies are on the list for 2Q 2019.  We all know Bendix King has done a poor job hitting their deadlines.  Hopefully KFC230 is eventually an option for mid body Mooneys. but who knows when (or IF) that will happen.  

 

 

Link to comment
Share on other sites

1 hour ago, EricJ said:

Right, and as you probably know, the output of a good primary sensor can be muted by inclusion of higher-order sensors if the combining is done poorly.  

There is no reason to do it poorly.  If done appropriately a professional engineer should not be doing bad math - combining two sensors poorly just should not be happening.  The math behind the concept it is quite plausible that combining a low quality observation to a high quality observation to make an even better observation.

I do not know anything about the specifics of their engineering but they seem to have a good reason to reject VNAV approaches if the GPS source is also absent.  

Now the old simple stuff also has nasty failure modes - just not that one.  For example, my steam gauge KFC200 has a bad habit of going entirely inop when the vacuum goes inop and thus the attitude.  I would prefer the GFC500 failure mode.

Sure then you can say that ok, I could get a Aspen for my KFC200 and then I would not worry about vacuum.  Sure but then the Aspen has a similar failure mode due to loss of GPS as the G5.

I have a vacuum and a backup electric vacuum.

I have been holding off on glass until this moment - digital autopilot.  Nothing I have heard so far has convinced me I don't want the GFC500 - yet.

I am unexcited by the aerocruz 230 since it does not seem much cheaper than the GFC500 despite the drop in replacement concept.  10k.  And I must buy the vaporware KI300 and I still need some kind of HSI - or I could get an aspen (which is nice) but all, work included in I think installed prices are close.  And then I still have my old servos versus new and modern brushless motors.  And I don't have the options I want most - ESP, and straight and level.

Plus I am very turned off by investing in anything BK ever since about a year ago when BK temporarily announced they would put all the overhaul houses (autopilot central, etc) out of business and require all work to be done by them.  And 3 years ago I had a trim switch micro switch ( the switch on the yoke) go bad - and they wanted something like 2k for a micro switch.  After much calling I found one on someone's shelf new-old stock for $950.  Something that just a few years before was $400.  And a micro switch.  I don't think keeping BK is anything more than throwing good money after bad....

Link to comment
Share on other sites

There is no reason to do it poorly.  If done appropriately a professional engineer should not be doing bad math - combining two sensors poorly just should not be happening.  The math behind the concept it is quite plausible that combining a low quality observation to a high quality observation to make an even better observation.
I do not know anything about the specifics of their engineering but they seem to have a good reason to reject VNAV approaches if the GPS source is also absent.  
Now the old simple stuff also has nasty failure modes - just not that one.  For example, my steam gauge KFC200 has a bad habit of going entirely inop when the vacuum goes inop and thus the attitude.  I would prefer the GFC500 failure mode.
Sure then you can say that ok, I could get a Aspen for my KFC200 and then I would not worry about vacuum.  Sure but then the Aspen has a similar failure mode due to loss of GPS as the G5.
I have a vacuum and a backup electric vacuum.
I have been holding off on glass until this moment - digital autopilot.  Nothing I have heard so far has convinced me I don't want the GFC500 - yet.
I am unexcited by the aerocruz 230 since it does not seem much cheaper than the GFC500 despite the drop in replacement concept.  10k.  And I must buy the vaporware KI300 and I still need some kind of HSI - or I could get an aspen (which is nice) but all, work included in I think installed prices are close.  And then I still have my old servos versus new and modern brushless motors.  And I don't have the options I want most - ESP, and straight and level.
Plus I am very turned off by investing in anything BK ever since about a year ago when BK temporarily announced they would put all the overhaul houses (autopilot central, etc) out of business and require all work to be done by them.  And 3 years ago I had a trim switch micro switch ( the switch on the yoke) go bad - and they wanted something like 2k for a micro switch.  After much calling I found one on someone's shelf new-old stock for $950.  Something that just a few years before was $400.  And a micro switch.  I don't think keeping BK is anything more than throwing good money after bad....


With 25 year old gear I fly ILS approaches every day with zero reliance on a GPS signal.


Sent from my iPhone using Tapatalk
  • Like 1
Link to comment
Share on other sites

5 minutes ago, gsengle said:

 


With 25 year old gear I fly ILS approaches every day with zero reliance on a GPS signal.


Sent from my iPhone using Tapatalk

 

That is a different topic.  Like a change of topic.  I was discussing if a system is capable of performing a task.  I was asserting that the mathematical technology exists to fuse to signals to produce a higher quality signal.

Then you say so what you fly the ILS by hand.  That's great.  I practice hand flying my ILS's too. I did that on Saturday.  But still data fusion is a useful concept across many fields of sensors engineering.

Edited by aviatoreb
Link to comment
Share on other sites

That is a different topic.  Like a change of topic.  I was discussing if a system is capable of performing a task.  I was asserting that the mathematical technology exists to fuse to signals to produce a higher quality signal.
Then you say so what you fly the ILS by hand.  That's great.  But still data fusion is a useful concept across many fields of sensors engineering.


No, the original topic is that we lose the ability to fly a coupled ILS with this new gear even when we aren’t relying on GPS *at all* for navigation. If King can do that and Garmin can’t, that’s a compelling consideration.


Sent from my iPhone using Tapatalk
  • Like 1
  • Thanks 1
Link to comment
Share on other sites

2 minutes ago, aviatoreb said:

There is no reason to do it poorly.  If done appropriately a professional engineer should not be doing bad math - combining two sensors poorly just should not be happening.  The math behind the concept it is quite plausible that combining a low quality observation to a high quality observation to make an even better observation.

I spent a career doing the math combining comm signals for improved performance.   There's *always* a way to do it poorly, and people come up with very creative ways to do it poorly.   I spent the last part of my career consulting to fix some of those, i.e., clean up after the "no reason this should happen" guys.  ;)

Spent a little while doing airliner avionics, a little while doing rockets, and long time doing other complicated things, and the ways by which things can be screwed up are pretty much unlimited, whereas the ways that they will work properly are limited.   Take the ratio.  ;) 

  • Thanks 1
Link to comment
Share on other sites

4 minutes ago, gsengle said:

Where do you see that the KI300 is substantially more expensive? I expect M20L is a typo.

 

https://sarasotaavionics.com/avionics/ki300

 

 

Sent from my iPhone using Tapatalk

 

KI300 plus a KA310 is $4700.  G5 Certified is $2150.  Cost of the AI is double.  https://sarasotaavionics.com/avionics/g5-certified

On paper the GFC 500 may not need a G5 HSI, but I can't imagine doing a GFC500 without it- regardless the total cost of the KFC230 is a little higher.  In addition, there is core value in the King servos which would further reduce the cost of the Garmin install.  There's greater capability to the King, but it's a more expensive solution.  Nothing wrong with that.

There is a Bendix King rep on Beechtalk.  Our own @KSMooniac challenged the rep and he confirmed that the M20L was with the initial STC release, the rest of the long bodies will be 3 months later.  I have to correct myself- looks like the J and the K are coming in the first round of releases which is fantastic.  https://www.beechtalk.com/forums/viewtopic.php?p=2262791#p2262791

Link to comment
Share on other sites

20 minutes ago, gsengle said:

 


No, the original topic is that we lose the ability to fly a coupled ILS with this new gear even when we aren’t relying on GPS *at all* for navigation. If King can do that and Garmin can’t, that’s a compelling consideration.


Sent from my iPhone using Tapatalk

 

Ok, but I don't see how that is a response to what I said and you quoted me.

If I talk about the possibilities of engineering a system and discuss its various failure modes I am not a vote for or against hand flying.  I actually like engineering as a topic unto itself.  I asset that every system, King, Garmin, vacuum, digital, whatever has failure modes.  yes king has some very nasty failure modes but they are not the same failure modes.  And one should understand the failure modes.  

Link to comment
Share on other sites

23 minutes ago, smccray said:

KI300 plus a KA310 is $4700.  G5 Certified is $2150.  Cost of the AI is double.  https://sarasotaavionics.com/avionics/g5-certified

On paper the GFC 500 may not need a G5 HSI, but I can't imagine doing a GFC500 without it- regardless the total cost of the KFC230 is a little higher.  In addition, there is core value in the King servos which would further reduce the cost of the Garmin install.  There's greater capability to the King, but it's a more expensive solution.  Nothing wrong with that.

There is a Bendix King rep on Beechtalk.  Our own @KSMooniac challenged the rep and he confirmed that the M20L was with the initial STC release, the rest of the long bodies will be 3 months later.  I have to correct myself- looks like the J and the K are coming in the first round of releases which is fantastic.  https://www.beechtalk.com/forums/viewtopic.php?p=2262791#p2262791

I'm not seeing where the king aero cruz is promising more capability than the garmin.  I don't mean that as an argument.  I'm just asking.

I mean separate from the different failure modes that I was already discussing above, does the aero cruz offer more features?

Link to comment
Share on other sites

33 minutes ago, EricJ said:

I spent a career doing the math combining comm signals for improved performance.   There's *always* a way to do it poorly, and people come up with very creative ways to do it poorly.   I spent the last part of my career consulting to fix some of those, i.e., clean up after the "no reason this should happen" guys.  ;)

Spent a little while doing airliner avionics, a little while doing rockets, and long time doing other complicated things, and the ways by which things can be screwed up are pretty much unlimited, whereas the ways that they will work properly are limited.   Take the ratio.  ;) 

I agree completely.  One can always do a bad job. And sometimes we do!  I guess when I say one should never do a bad job engineering the system - I might as well say one should never design a bad bridge - one that will fall apart. And yes sometimes people do exactly that.  And sometimes bridges collapse, buildings collapse, etc. But my assertion that one should not do bad engineering I am trying to say that the concept of how to do it well exists even if in practice the people doing the practice may make bad designs.  I'm standing up for the math vs standing up for the people doing the math.  I have not spent a career doing the math combining comm signals specifically as you have, but I am spending a career doing math generally, especially dynamical systems and data science (data enabled science is a trending phrase).  I am a math/ece prof.

Edited by aviatoreb
Link to comment
Share on other sites

10 minutes ago, aviatoreb said:

I'm not seeing where the king aero cruz is promising more capability than the garmin.  I don't mean that as an argument.  I'm just asking.

I mean separate from the different failure modes that I was already discussing above, does the aero cruz offer more features?

I suspect the king system offers control wheel steering. I believe the GFC500 includes envelope protection and the level button. 

There’s not a perfect system out there- pick your poison :) 

  • Like 1
Link to comment
Share on other sites

2 minutes ago, smccray said:

I suspect the king system offers control wheel steering. I believe the GFC500 includes envelope protection and the level button. 

There’s not a perfect system out there- pick your poison :) 

What is control wheel steering?

I really like the envelope protection (ESP) and level button - these are the killer aps that make me want to get that A/P above all.

Link to comment
Share on other sites

34 minutes ago, smccray said:

KI300 plus a KA310 is $4700.  G5 Certified is $2150.  Cost of the AI is double.  https://sarasotaavionics.com/avionics/g5-certified

I haven't seen anything that says the KA310 is required if you use a KFC-230, only for legacy(analog input) autopilots. I'd assume they were smart enough to put the ARINC output for the KFC on the KI-300 itself... Then again, it is BK.

  • Like 1
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.