Jump to content
Sign in to follow this  
Ramón Cutanda

I quit: I'm unable to understand the gyropilot

Recommended Posts

Posted (edited)


  • Prepar3D v5 Academic
  • PILOT'S Boeing B314 The Clipper P3Dv4v5 R1.1g, V1.20
  • Active Sky P3D  v7877
  • SPAD.neXt v0.9.11.1 (B314)

Flight Controls

  • Saitek Pro Flight Quadrant
  • Saitek Pro Flight Yoke
  • Thrustmaster T-Rudder
  • Saitek Pro Flight Switch Panel
  • Saitek Pro Flight Radio Panel


CPU:           AMD Ryzen 9 5900X (Vermeer, VMR-B0)
               3700 MHz (37.00x100.0) @ 4516 MHz (45.25x99.8)
Motherboard:   GIGABYTE B550 AORUS ELITE V2
BIOS:          F13, 07/08/2021
Chipset:       AMD B550 (Promontory PROM19 C)
Memory:        32768 MBytes @ 1796 MHz, 18-22-22-42
               - 16384 MB PC28700 DDR4 SDRAM - Team Group TEAMGROUP-UD4-3600
               - 16384 MB PC28700 DDR4 SDRAM - Team Group TEAMGROUP-UD4-3600
Graphics:      GIGABYTE RTX 3080 VISION OC 10G (GV-N3080VISION OC-10GD)
               NVIDIA GeForce RTX 3080, 10239 MB GDDR6X SDRAM
Drive:         Crucial_CT512MX100SSD1, 500.1 GB, Serial ATA 6Gb/s @ 6Gb/s
Drive:         ST8000DM004-2CX188, 7814.0 GB, Serial ATA 6Gb/s @ 6Gb/s
Drive:         Samsung SSD 970 EVO 1TB, 976.8 GB, NVMe
Sound:         NVIDIA GA102 - High Definition Audio Controller
Sound:         AMD Family 17h/19h - HD Audio Controller
Network:       Intel Wireless-AC 9260 160MHz
Network:       RealTek Semiconductor RTL8125 Gaming 2.5GbE Family Ethernet Controller
OS:            Microsoft Windows 10 Professional (x64) Build 19043.1165 (21H1)

My problem

999 out of 1,000 times I am unable to make the gyropilot maintain a fixed altitude. However, it is not impossible. 1 out of 1,000 times it works, so I know IT WORKS; my problem is that I unable to find out WHY it does not always do so. You can see my problem in action in the following video. I apologize for the quality of the audio as today I could not use my main microphone and it seems my English this afternoon was certainly not my best.

My context

1. I have read, re-read, re-re-read, re-re-re-read, re-re-re-re-read (actually, "n"read) both the 2nd part of PILOT'S manual and the Sperry Gyropilot A-3 document.

2. My common sense tells me I must be skipping something or doing something wrong, but after having flown 15,529 nm along 106 hours (I am replicating the 1st Commercial Round-the World Flight of the NC-18602 - https://www.wingnet.org/rtw/RTW003B.HTM), for the life of me, I find myself completely unable to understand the gyropilot. I swear it has a life of its own. I can count with my fingers the number of times it has work as expected. Actually, I can count with my fingers the number of times I can even predict what will happen once engaged. For every flight, I need to spend several hours (literally; sadly I am not exaggerating here) engaging and disengaging the gyropilot over and over and over again before I am lucky enough to get a "lock" in the elevator so that it will keep an altitude. The aileron and rudder adjustments work as expected, but the elevator... At first I tried to keep an specific altitude (usually 8,000 feet)  but now I really don't care; as long as it stays stable I am happy with any altitude. In the rare occasions when it works, most of the times it will stay at an altitude in a rage of 500 feet above, or below, the one I was keeping at the moment of engaging the gyropilot. However, in those in a lifetime situations when it works, it is an incredible solid lock, and is able to maintain a constant altitude even under very strong turbulence. In those situations you can see the logical fluctuations (ups and downs) but the average is solid and eventually, once the turbulent air is gone, it will go back to its original altitude with zero variation in the vertical air speed.

What I know (or I think I know) about the gyropilot

3. The gyropilot is NOT an AUTO-pilot. Once trimmed and in stable flight, the gyropilot's function is to correct deviations from the initial status, so having everything ready BEFORE engaging the gyropilot is a must.

4. In order to activate the system I should engage first the gyropilot (1) and then pressurize the servos (2). I have tried the reverse order too, activating the servos first (2) and then the gyropilot (1). However, I have not seen any difference, either for the better or worse.



5. Once the servos are on (2), it takes the system around 15 seconds or so to get pressurized and "take control". The pressure can be easily monitored with the three available gauges; two of them highlighted in the picture above of these lines (10). Once the pressure is about 30, you can "feel" the gyropilot is starting to get control.

6. As I mentioned, I am aware that the gyropilot's function is to keep things "as they are" so BEFORE pressurizing the servos, I do my best to trim and obtain a straight and level flight on my desired heading and I also make sure to use rudder adjustment knob (3) and the directional gyro (6) to select my current heading. I also check that the aileron (4) and elevator (5) adjustment knobs are set to zero.

7. I also know that the recommended setting for the reaction speed knobs (9) is "3" and that the extremes (0 and 5) must be avoided because they create a lock in the controls. I would like to mention here that there is a difference between PILOT's manual and the Operating Bulletin for the Sperry A-3 Gyropilot:

"At full speed (zero on the dials) the control surfaces will go into hydraulic lock." (PILOT's Manual v1.3, page 24)

"Turning the knobs all the way clockwise locks the controls and must be avoided" (Operating Bulletin for the Sperry A-3 Gyropilot, page 9, section 2).

If we turn the knobs "all the way clockwise" we reach 5; and not zero. In any case, I always avoid extremes and set the speed controls to 3 to avoid problems. In that regards, I have tried:

  • Pressurize the gyropilot with the speed set at "zero", and then slowly increasing the value to 3.
  • Make sure the speed knobs are at 3 BEFORE pressurizing the gyropilot.

I am unable to tell the difference of using one method or the other. However, except when desperate, I tend to select the speed of 3 before engaging the gyropilot to avoid the possible lock mentioned.

8. Also BEFORE pressurizing the gyropilot, I make sure to set the directional (6) and bank and climb (7) selectors to UNCAGED. By the way, there is a typo in the VC tip, with an extra "t" at the end of the description:


9. Now comes the "funny" part. It is my believe that after all the previous steps, once the gyropilot is engaged it should keep things as they are. If there is any difference between the current heading and the one selected in the rudder adjustment knob (3), then you can feel the aircraft moving towards searching the selected heading. Most of the times this will happen very smoothly as the difference, if "homework" has been done beforehand, tends to be minimal; if any. For for the elevator, even if the adjustment knob (5) is set to zero, adding pressure in the system usually means "rollercoaster time". I have to admit that MOST of the times it will not be something "wild" but, from the initial steady and level flight, engaging the gyropilot usually means a change in the altitude of dozens (sometimes hundreds) of feet per minute. If I touch nothing, 1 out of 100 times or so the system will finally find a "balance" after about 15 seconds and keep a "more or less" level flight. But most of the times the change in the rate of climb or descend will simply be completely random and unpredictable; my guess is that it simply depends on the meteorological conditions outside. Usually it will keep gaining or losing non-stop until I finally take manual control.

10. Finally, and having said all that, I am aware that, according to the PILOT's Part 2 manual (v1.3 page 26):

On the A-2A gyropilot the Level Flight knob engages the default autopilot if you are having trouble getting the gyropilot to behave as you wish. [...]

When in autopilot mode  
- the rudder knob sets the required heading
- the elevator knob converts degrees of pitch to hundreds of feet per second vertical speed
- setting pitch to zero locks to the barometric altitude at that point

However, that information has little or no use for most of my flights (as described in the video).

My checklists for every flight

Just in case it could be useful to know exactly which actions and steps I take BEFORE engaging the gyropilot, I rigorously follow a custom-made checklist based on PILOT's and Boeing documentation. As it says in the introduction of the checklist, it is my intention to share it with the rest of the forum once it's ready; but before that moment I want to make sure all the steps regarding the autopilot are correct.


Goodby, gyropilot. It was nice having met you

Understanding how the gyropilot works has been, so far, the hardest thing for me during the learning of the B3414. But after dozens of hours of trial and error I simply give up. If I can get any help from you, that would be awesome and thank you so much in advance. However, If I can get no hep I will simply fly manual for the rest of my life... Goodbye, gyropilot. I will always remember you.

Edited by Ramón Cutanda

Ramón Cutanda


Share this post

Link to post
Share on other sites
Posted (edited)

🤣I actually enjoyed reading that - it was genuinely funny. I have hated the gyropilot from the word 'Go' because of the complete lack of information on expected response and the amount of guesswork I had to put into it. So... after the last patch was released and I started on the B314-A, I nuked it. Completely and utterly. I then looked at the Sperry factory maintenance manual (not an operations manual) which of course gives a really good breakdown of the parts of a gyropilot and decided that the only way I was going to beat this thing was to ignore what it is (a gyropilot) and to work out how each individual part will react in isolation and to write code to mimic that. I also talked to an aeronautical lecturer who taught at one of the more well-known universities in England and he gave me his opinion on what he would expect it to behave like under different circumstances. About three weeks later, when I assembled the code blocks, I found I have a gyropilot that has suddenly beome a lot easier to use and whose behaviour makes a lot more sense. Additionally, it does respond exactly as the lecturer thinks it should do. The person I was talking to was involved with the Hawker-Siddeley Trident Autoland system, so he knows a bit about autoflight systems 😄

After  some fine tuning, the B314-A beta group gave it an emphatic 👍 this week, much to my relief. The new version was developed for the B314-A but I have ensured that the systems will automatically back-port the new code to the B-314 and it has recently been tested with the B-314 to check that there are no lurking nasties. It will even do a climbing turn now... 😁

Edited by dragonflightdesign

Share this post

Link to post
Share on other sites

That is SUCH GREAT NEWS! Again, thank you SO MUCH for the update. This is really, really, appreciated. Thank you for all the passion that you show in the development of this wonderful aircraft. I am really (really) looking forwards to the next update.


Ramón Cutanda


Share this post

Link to post
Share on other sites
On 8/28/2021 at 11:56 AM, dragonflightdesign said:

After  some fine tuning, the B314-A beta group gave it an emphatic 👍 this week, much to my relief. The new version was developed for the B314-A but I have ensured that the systems will automatically back-port the new code to the B-314 and it has recently been tested with the B-314 to check that there are no lurking nasties. It will even do a climbing turn now... 😁

Awesome to hear, thank you! I am eyeing the clipper since I heard it's finally released and was following the Gyropilot saga. Learning it's improved makes me think it's finally time pull the trigger. 

Is there any timeline on the release of that patch? Also, seems like 314A is also being developed, might I ask if that's going to be an update/expansion or a seperate product? It's surely the variant I would be much more interested in.

Apologies if I asked about things you can't, or would rather not, talk about. 

All the best. 

Share this post

Link to post
Share on other sites

So much for a $92.00 addon.  Disaster and cash grab from start to finish.


Share this post

Link to post
Share on other sites

You are absolutely entitled to your own opinion and I am not going to argue. I see that you claimed a refund on the 4th October.

For anyone else: I am still working on the B314-A but as stated in a previous post, I am one person doing coding, textures, fx and everything else except the mdl. If you have ever been involved in a development cycle you will be very aware of how long it can take.

Edited by dragonflightdesign
  • Like 1

Share this post

Link to post
Share on other sites

Hello Ian

I can give you some good news ahead of time: I got so fed up with the gyropilot that I nuked the lot and started again. This time I used the A-3 maintenance manual and replicated each major block in code without worrying about what the end product was 'supposed to do'. When I stitched everything together, my very picky and excellent beta tester tells me that we've finally beaten it. It now matches almost precisely what was predicted by an autoflight systems engineer that I contacted. If you recall, the biggest problem was a lack of information on how the A-2 gyropilot works, so it was all (hopefully educated) guesswork. Three days ago I was reading a technical book on the Short Brothers 'C'-class Empire flying boats which also uses the A-2 gyropilot, albeit with some minor differences to the B-314. Almost in the middle of the book was a full-page description of the setup and operation of the Empire's A-2 gyropilot. What I have built and what is written, match, with the exception of those minor differences. The minor differences are:

- the Empire runs the gyropilot servo system at 300psi vs the B-314's 100psi

- the Empire has no 'Level Flight' control

- the Empire can change heading using the Bank control whereas Elmer says 'don't do that' in the B-314.

In the case of the latter, I believe that the Empire book may have missed some operational information. It isn't possible to change direction on the Bank control unless you cage the Directional Gyro first because if you try it, the heading gyro and the bank gyro will end up fighting each other. If you cage the DG, change heading on the Bank control, realign the heading request with the new heading on the DG and then uncage it, there should be no unexpected movement of the aircraft. However, the Empire weighs almost 40% less than the B-314 so this may have something to do with the difference in A-2 usage (I'm guessing here).

If the Heading gyro fails on the B-314 then you are able to change direction using the Bank gyro.

The author of the book does some comparisons with BOAC's B314-As. Apart from finding that in a couple of cases we have used the same primary sources for research, what did make me almost laugh out loud was finding that for all of Boeing's claims about a noise-insulated cabin, it didn't apply to the the flight deck. This explains a crew complaint about the noise of the engines when trying to get some off-duty rest. Apparently BOAC soundproofed their crew decks during the first major overhaul.

Edited by dragonflightdesign
  • Like 2

Share this post

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

  • Tom Allensworth,
    Founder of AVSIM Online

  • Flight Simulation's Premier Resource!

    AVSIM is a free service to the flight simulation community. AVSIM is staffed completely by volunteers and all funds donated to AVSIM go directly back to supporting the community. Your donation here helps to pay our bandwidth costs, emergency funding, and other general costs that crop up from time to time. Thank you for your support!

    Click here for more information and to see all donations year to date.
  • Donation Goals

    AVSIM's 2020 Fundraising Goal

    Donate to our annual general fundraising goal. This donation keeps our doors open and providing you service 24 x 7 x 365. Your donation here helps to pay our bandwidth costs, emergency funding, and other general costs that crop up from time to time. We reset this goal every new year for the following year's goal.

    $13,405.00 of $25,000.00 Donate Now
  • Create New...