Spirit System
https://www.spirit-system.com/phpBB3/

Spirit RS governer help needed (nitro)
https://www.spirit-system.com/phpBB3/viewtopic.php?f=20&t=5255		 Page 4 of 5
Author:  kay [ Fri 19. Aug 2022 6:54:38 ]
Post subject:  Re: Spirit RS governer help needed (nitro)
Scott H wrote:
Unless the log shows governor engaged, changing the governor settings won't have any impact.

ok, I see. I wonder what could cause the changing rpm, as it is always the same flight mode with flat trottle curve...
Scott H wrote:
To clarify you added a ferrite ring/band to the RPM sensor (this would include power/ground/signal wire). You mentioned adding it but wasn't clear that is was on the RPM sensor.

It's on the cable of the rpm sensor, so all three wires lead through the ferrit.
Scott H wrote:
I've had, in the past a power and ground bus for my Goblin gasser, but I no longer do, it hasn't and doesn't make any difference to the gov or rpm issue from my perspective.
...was just an Idea to eliminate that possible ripples on the power lines (from the servos maybe) causing the issue.
Scott H wrote:
The log shows events and they don't have to fall on 10 sec intervals. If something happens, it shows.
well, so the time stamp is not to be taken as that serious.
Scott H wrote:
When you connect to the Spirit software and are in the gov tab, where it shows the requested and current rpm. Have you checked to see if it shows reasonable numbers when you spin the motor? For example; if you remove the plug, and spin the motor over with a starter, do you see rpm numbers that make sense, or does that current rpm number go to some weird number?
yes, it works all the time and it seems also to be realistic value, with the former align sensor and one active magnet, with the spirit sensor both, one and two active magnets. The difference between one and two acitve magnets is in the min. speed where the spirit starts showing the rpm within the gov checking tab. Two magnets starts at around 30rpm, one magnet I don't remember exactly but guess around 50rpm.
Scott H wrote:
I'm uploading one of my logs so you can see what it should say.
Take a look at that requested rpm and see what it does.
...Thank you and I like that "Governor was engaged" event ;)

I guess the next step would be to exchange the sensor cable by a shielded one and try again.

best regards, Kay
Author:  Scott H [ Fri 19. Aug 2022 11:46:30 ]
Post subject:  Re: Spirit RS governer help needed (nitro)
Thanks for the update.
I reread the Spirit Wiki and there is a sentence at the very bottom of the gov page: You should see “Governor was Engaged” event in the log after spoolup with enabled governor.
For some unknown reason the gov is still in spooling up and never arrives at engaged. Just odd.

I don't really think the shielded cable is the answer for not engaging, if you had noisy data, yes, but for this, I don't think so.

Too bad you don't have a spare Spirit to try on this to see if it's the Spirit itself.

If I think of anything else, I'll post, but we've reached the end of my suggestions.
Author:  kay [ Fri 19. Aug 2022 12:44:10 ]
Post subject:  Re: Spirit RS governer help needed (nitro)
Scott H wrote:
For some unknown reason the gov is still in spooling up and never arrives at engaged. Just odd.

...well I thought about reasons for this all the time, but without knowing the exact procedures of the software it's more or less trial and error on factors that are influenceable.


I know Tomas is not available, but I hope he will soon recover. Hopefully he will give more details of what the gov is doing from the point being activated until it "thinks" it is engaged. Then we could start thinking of what conditions could interrupt this process and look if such conditions are existing in this case.

best regards, Kay
Author:  Scott H [ Fri 19. Aug 2022 12:53:31 ]
Post subject:  Re: Spirit RS governer help needed (nitro)
Well, I do have an idea, and it's something about how your set up and mine differ. My flight modes are throttle hold, normal, IU1, IU2

Throttle hold is 11% flat curve (constant)
Normal is 3 point 9%, 28%, 32%
IU1 is 81% flat (constant)
IU2 is 90% flat (constant).

Your normal is flat curve (constant) 11% constant.

Aside from the obvious flat vs 3pt curve is the input value of the graph on the right, yours is 50% input and 11% output. Mine is 0% input and 9% output.

The idea, is that maybe this input of 50% input is causing confusion to Spirit.

There are a number of other pilots here, so, please check your systems, if your set up is the same as Kay (that is Constant normal up and input of 50% (and you fly Jeti)) and your gov works fine, then chime in, cause that would automatically mean my "idea" is wrong.

If you are wondering how/when I switch to IU1 it's in the air after I've established a hover. Mine is a behavior that is a holdover from the early days of FBL when helicopters had a tendency to tip over on throttle up. I get that my set up is not what is suggested in the Wiki.

Good luck
Author:  kay [ Fri 19. Aug 2022 16:08:40 ]
Post subject:  Re: Spirit RS governer help needed (nitro)
Hi Scott,

well, I'm sorry probably I didn't catch your Idea correctly... You mean the 50% that is shown at the curve setup in the DS12? If so, from my point of view this value never is transfered to the spirit, it only sees the corresponding absolut servo position of each point of the curve.
If I understand correctly, you are using the trottle hold for starting the engine, and after the engine is running you use the 3point curve for "manual spoolup", take off till hovering. After that you switch to IU1 or IU2. In that case the difference between your setup and mine is the difference of the current servoposition to the requested one.
I have also a 11% idle (flat) curve for engine start. The direct servo output sent to the spirit is than a value of -75% (channel 1 output of the transmitter). As soon as the engine is running, I switch to 80% flat curve what should activate the gov and tell the gov that I want to have ~1630 rpm. The corresponding servo output on channel 1 is now 65%.
In your case as it seems to me that you activate the gov at hoovering, means the trottle current position is much closer to the value the spirit would have to set it. This is valid for both the channel 1 output at this moment and also for the real servo position. Your gov would probably need less time for adjusting the requested rpm. This is interessting, I will check, first programming a much slower rpm, and seeing if the gov is struggeling if it takes to long for complete spoolup from idling. I also could try a 3point curve also for hovering and activating the gov at this point.
thanks Scott.

best regards, Kay
Author:  Scott H [ Fri 19. Aug 2022 16:53:36 ]
Post subject:  Re: Spirit RS governer help needed (nitro)
Attachment:
85715721-1392-45D1-A3D3-012AB9F248D9.jpeg
85715721-1392-45D1-A3D3-012AB9F248D9.jpeg [ 3.41 MiB | Viewed 229 times ]
Here are two screen captures from my model and yours.

The input of 50% is what is was seeing. But I don’t know what it means or if it impacts anything since it is t an output.

Attachments:
4B385E6D-73E2-467E-A97A-85B2A470A95E.jpeg
4B385E6D-73E2-467E-A97A-85B2A470A95E.jpeg [ 3.39 MiB | Viewed 229 times ]
Author:  Scott H [ Fri 19. Aug 2022 16:54:36 ]
Post subject:  Re: Spirit RS governer help needed (nitro)
That post looks awful, sorry it was from my phone, hope you can see the images.
Author:  kay [ Wed 24. Aug 2022 20:47:13 ]
Post subject:  Re: Spirit RS governer help needed (nitro)
Hi Scott,
unfortunately we are the only ones talking...
Well, new test today, I only lowered the programmed flat curve from 80% to 55%. I got rather stable head speed and the message "governor engaged" at the log. Next check was to increase the headspeed limit from 2000 rpm to 2500 rpm and same flat curve (55%), no sucess, no governor message at the log, but some "vibration level high" messages. Next I tried a 3 point curve for manually control the trottle near the requested head speed. But also no success. At manually control of the trottle the engine runs very rich at that trottle position and it could be seen that the heck is twitching.
Could it be that the vibration level prevents the gov for beeing engaged? This seems the only difference between acitvated gov at low rpm and no gov at any level above.

best regards, Kay
Author:  Scott H [ Wed 24. Aug 2022 21:07:49 ]
Post subject:  Re: Spirit RS governer help needed (nitro)
Congratulations for some good news.

This is a TREX700 nitro, right, isn't your max headspeed 2200 or so?

With flat curve of 55% what headspeed was that?

I would go back to the headspeed limit of 2000 and 55% flat curve get the gov to engage in the log. I might inch that up to 60, and then 65% to see if/when it stops engaging. The governor performance will be awful without tuning, so ignore that deficiency for the moment.

With your equipment do you capture vibration as one of the telemetry fields? (I didn't know I had this until I looked). I don't know if high vibration can prevent gov engaging, I don't think so, as I've always had gov engaged even though I have had plenty of high vibrations.

I use an automotive sniffer to listen to different parts of the heli as I manually rotate the parts. This has been a real time saver to determine what bearings are bad.

I intend to fly in an hour or so and I'll raise my headspeed limit and see if that prevents the gov from engaging. See if I can duplicate your condition.
Author:  ZeXx86 [ Thu 25. Aug 2022 7:20:16 ]
Post subject:  Re: Spirit RS governer help needed (nitro)
Hello,

if Governor gains are too high and RPM is too unstable the Governor will not engage.
If your tail is vibrating, it could cause additional oscillations of the RPM. So this could be the issue.

It could be enough to lift off with the lower headspeed or to decrease Governor Response.
Page 4 of 5 All times are UTC + 1 hour
Powered by phpBB® Forum Software © phpBB Group
http://www.phpbb.com/