- Posts: 18
Trimming and curves...
- silverAG
- Topic Author
- Offline
So, here is my problem: I have Devo 7e with additional modules and Deviation firmware 5.0.0) and I have question regarding trims. One of my quads (original Bayangtoys X9) started to drift a bit backward and I need to trim it. When I do this using trim on stock controller and uses stock controler to fly, it flys fine in all 3 rates . But when I use trim in Devo, trim values used on ELE chanel affect quad different and depends on selected flying rate - the reason is because model I use (can't remember where I downloaded it but it has also 3 rates and trim available on channels) uses different curves to "simulate" rates so trim values that work fine for one curve does not work for another (I think that's the reason).
Is there any solution which will cause trim on Devo to work the same way on all curves like it works on stock controller on all rates so that Devo curves does not affect trimming (somehow force Devo to trim before applying curve - that will maybe help)? Or maybe there is other way to correct drift in Devo?
Thanks for help in advance.
Please Log in or Create an account to join the conversation.
- silverAG
- Topic Author
- Offline
- Posts: 18
Please Log in or Create an account to join the conversation.
- FDR
- Offline
This is because the trim shouldn't rely on them, it is only a fix correction to achieve stable hovering.
Deviation usually doesn't use the built in switchable rates of the different flight controllers, but operate them always on their highest rate, and handle the rate calculations in the TX.
Of course there might be exclusions, you can ask about in the protocol specific threads...
It is hard to suggest anything until we see the actual model.ini file, so please attach it.
It is because you can turn on or off the trims for each mixer, at least using the complex mixer type. It is a common mistake to apply the trims more then once in the mixers.
On the other hand trimming is not the recommended method for achieving a stable hover in modern flight controllers.
You had better calibrate the accelerometers of the fligh controller instead...
Please Log in or Create an account to join the conversation.
- silverAG
- Topic Author
- Offline
- Posts: 18
That's why I thought that Devo trims after applying every other settings and you conmfirmed that.So, when trim values are applied, they will cause different result depending on curves in mixers.
Also, I am aware that trimming is not solution for drifting - but my X9 drifts no matter how much I calibrated it - and I did it several times on 100% flat surface + I usually calibrate and avoid trim buttons but it keep drift to back and only trimming helps. Maybe motors are starting to go bad etc.but trimming is only way I can stabilize it for now.
I attached INI - ignore trim values (I experimented with all trim buttons because I was not sure it works so values are now messed up for all directions), and also ignore Virt1 channel I created for deadband values (I have problem with centering elevator stick which I soved with deadband and I also removed that to be sure that it does not cause trim to behave diferently but also does not help).
So, the point is - X9 drifts completely the same with Devo and with stock controller. Trimming with stock controller solve problem on all rates while trim on Devo solve problem only for currently active curve - by switching to another one, trim values force quad to behave differenlty. And if Devo apply trims on final values, looks like there is no solution except if there is a way I can have different trim values for different curves somehow...
Please Log in or Create an account to join the conversation.
- FDR
- Offline
No. That's the point: the same trim value is applied no matter of the stick position or curves (which is only a multiplication on the stick position), because it should correct the same amount of drift.silverAG wrote: That's why I thought that Devo trims after applying every other settings and you conmfirmed that.So, when trim values are applied, they will cause different result depending on curves in mixers.
However if you need different trims for the different rates, set your rate switch (in this case SW B0) as the switch on the trim setup pages, and you will have independent trim values for all the switch states, i.e for all your different rates.
BTW I don't see anything wrong in your ini, only the Virt1 source was strange. While the mux type is replace on all pages of the complex mixer, the trims shouldn't affect other pages...
Please Log in or Create an account to join the conversation.
- silverAG
- Topic Author
- Offline
- Posts: 18
FDR wrote:
No. That's the point: the same trim value is applied no matter of the stick position or curves (which is only a multiplication on the stick position), because it should correct the same amount of drift.silverAG wrote: That's why I thought that Devo trims after applying every other settings and you conmfirmed that.So, when trim values are applied, they will cause different result depending on curves in mixers.
Hm. I am obviously thinking wrong. OK - if trim is at end of "curve chain" (sort to speak ), it will affect behaviour always in the same way no matter what curve is set before it?
FDR wrote: However if you need different trims for the different rates, set your rate switch (in this case SW B0) as the switch on the trim setup pages, and you will have independent trim values for all the switch states, i.e for all your different rates.
Need to try that. Thanks for tip!
FDR wrote: BTW I don't see anything wrong in your ini, only the Virt1 source was strange. While the mux type is replace on all pages of the complex mixer, the trims shouldn't affect other pages...
As I said, Virt1 is used for setting deadband before other curves but I turned deadband off to test does it affect trim problems (and it does not).
I definitelly have different behaviour of the same trim value when I switch to different curve with this INI. But if the same trim value should work the fine no matter what curve I used, I'll experiment more as soon as I find time.Thanks for reply!
Please Log in or Create an account to join the conversation.
- FDR
- Offline
Exactly! That's why it's weird, it doesn't work for you the same.silverAG wrote: Hm. I am obviously thinking wrong. OK - if trim is at end of "curve chain" (sort to speak ), it will affect behaviour always in the same way no matter what curve is set before it?
However I've got an other idea.
Since the curves only affect the stick value, it is possible, that your stick value isn't zero, so the different rates (multiplications) give you different output with the same trim value applied.
I.e. calibrate your sticks first!
You can verify it on the input monitor if you have zero values at mid-stick...
Please Log in or Create an account to join the conversation.
- silverAG
- Topic Author
- Offline
- Posts: 18
I just tested trims on Eachine H8 by putting some weight to have drift to right side to exclude elevator inprecision as a problem (using the same INI - it has the same protocol). I have no problem with AIL (when it is centered, it is 0 and no deadband need) but trimming to the left reacts completely the same as for trimming ELE on X9 - with left/right trim button I can stabilize H8 but when I change curve (rate) it start driftting to the right until I put back curve where I set trim. Stock TX works OK - when I trim it stays trimmed in all rates. And, to emphasize, I have no problem with aileron stick poitions (0 as I said - no deadband need and perfectly centered).
Strange. But if understandt how to assign trim values for various mixer setings as you suggested in few previous posts. That might solve it. Need to try that...
Please Log in or Create an account to join the conversation.
- Home
- Forum
- General
- General Discussions
- Trimming and curves...