• This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn more.

My ESCs and Multiwii Stink

#1
Yo forum!

I got a multiwii flip 1.5 back in May. Do not like it that much. I eventually converted my tri to a quad. I got it flying for a little while. One time I tested it, and the thing kept pitching backwards. Could not figure out why after all of the settings I did to it. During that time, it crashed quite a bit, but the quad stayed together. Afterwards, however, The motors would not start all together at the same time. Motor D3 would start, then D9, then D10 and D11. I thought it was an ESC calibration issue. RTF Quads' WIKI points to a code upload strictly for ESC calibration. I uploaded it, then uploaded the original version, and it still does not work. Is this just a case of my ESCs being shot?

Also, I have had nothing but issues with the Multiwii flip 1.5. I can't even calibrate the gyros from the Tx sticks anymore. I have to do everything from the computer. Can anyone offer a better, easy to tune FC under the $70 mark? Thanks.

Mastrmindz
 

cranialrectosis

Faster than a speeding faceplant!
Mentor
#2
Naze 32...

If ever your copter instaflips (flips over on lift off) check your connections from the ESCs to the flight controller first. It is common to connect motor 1 where motor 2 should connect or some such...
 
Last edited:
#3
Right. I will have to double check that. But that should not cause the motors from starting at different times as the throttle is gradually raised, right?
 

jhitesma

Some guy in the desert
Mentor
#5
Have you calibrated the end points on your TX? Not being able to use stick commands to calibrate is generally a sign that your TX endpoints aren't setup correctly.
 
#6
Have you calibrated the end points on your TX? Not being able to use stick commands to calibrate is generally a sign that your TX endpoints aren't setup correctly.
I tried setting everything on my Tx as close as to the Wiki says. There are some things on the Spektrum dx6i cant do what the Wiki says. Like for the DR&E, some of the settings suggested are beyond the Spektrum's settings. Other than that, I have tried to do everything on my Tx by the letter.
 

Craftydan

Hostage Taker of Quads
Staff member
Moderator
Mentor
#7
If your motors are not starting at the same time, but still start up one at a time while at low throttle, and you've already run the ESC calibration sequences, you need to increase the MINTHROTTLE setting in the config.h file and upload. If you switch to the naze, you'll likely still have to adjust this, but you can do it from the CLI interface instead of opening the source code and rebuilding.

This value sets the lowest "ON" signal the board will send an ESC when armed and running, and when it's sufficently high they should all start at the same time, and won't go below this if a downward correction gets too strong.

Now for the instaflip, remove your props and hold the airframe to make it think it's airborne. lower one of the booms and you should see that boom speed up and the one opposite should slow down. Until you've gotten this straight, there's no sense in putting the props back on -- it will continue to flip until the board's corrections are routed to the right motors.

If you go for the Naze, look into the Cleanflight ROM. While it doesn't particularly fly *better* than the baseflight version, it comes with some neat new features, like autotune, which once you've gone through it's procedures can adjust your pitch/roll gains for you. It won't help you troubleshoot setup to make a non-flying craft fly, but it will take a poorly flying craft and tighten up the control quite a bit . . . all with an entertaiing autopiloted wiggle . . .

Also, if you've got an android phone/tablet, consider picking up a bluetooth adaptor for one of these -- it'll work with either the multiwii or naze boards, but it will allow you to change settings wirelessly -- very handy.
 

RichB

Senior Member
#8
Motors starting at different times is probably the number one non-issue I see newbies worrying about on multirotor forums.

There are three reasons this happens:

1. ESCs are not calibrated equally. This does not matter any more than getting your craft's CG perfectly balanced between the motors matters. it's nice to have, but won't really affect your flight unless it is very far off.

2. Craft is in auto-level mode and it's not level when taking off. This does not matter because once the lower arm raises up to level the craft, the higher arms will start spinning.

3. Craft is slightly turned from where the FC wants to be, causing the FC to want to yaw slightly immediately upon powerup. Yaw with anything but a tilt-rotor craft like a tricopter means spinning some props faster than others. This doesn't matter because as soon as the landing gear leaves the ground, the craft will yaw and equalize.


A flight controller's job is NOT to apply even throttle to each motor to keep things stable. If equal throttle to each motor keeps a craft stable, we would not need flight controllers. A flight controller's job is to monitor the attitude of the craft and adjust throttle on each motor minutely hundreds of times per second. If one ESC's calibration doesn't match up with the other three, it will be compensated for faster than you can blink.

Props spinning up at different times as you throttle up means absolutely nothing when searching for a reason your craft will not fly right. It is the easiest scapegoat to identify for people who aren't sure how flight controllers work, and the tendency is for people to think that if one motor is spinning too fast, then that must be the culprit for the flip-over, lean, uncontrolled yaw, or whatever issue they are having. It's a highly-visible and alluring red herring that causes so many people to focus on the wrong thing and waste lots of time.

If your props spin up at different times, but you are NOT doing an insta-flip, then don't worry about it, everything is OK.

If your craft is pitching backwards: then describe the pitch. Is it a constant backward lean that causes drift, or is it a constant rotation that will eventually flip? What flight mode are you in? Angle or Acro? Dis you adjust trims so that your flight controller is seeing 1500 in the pitch channel when you aren't touching the sticks?

If it's drifting and not a trim issue, and you are in angle or horizon mode, try doing the angle mode trim procedure.

If you can't do that because your sticks won't trigger it (which might be likely given that you are having an issue arming), then you need to adjust your endpoints.
 

jhitesma

Some guy in the desert
Mentor
#9
One thought on the TX setup since you said you're using a dx61 - some people using the Spektrum gear have run into issues setting it up as apparently setting the end points for each direction is done separately which isn't initially obvious. A couple of people have had the same issue with not being able to use stick commands only to find that the problem was they only extended their range in one direction on each axis due to that.
 

joshuabardwell

Senior Member
Mentor
#10
If your motors are not starting at the same time, but still start up one at a time while at low throttle, and you've already run the ESC calibration sequences, you need to increase the MINTHROTTLE setting in the config.h file and upload. If you switch to the naze, you'll likely still have to adjust this, but you can do it from the CLI interface instead of opening the source code and rebuilding.
FYI, changing this value in config.h does nothing on my MW Flip 1.5 with MW 2.3. In MultiWiiConf, in the upper left, if you click "Settings", you can change MINTHROTTLE on that screen, and it appears to work.
 

Craftydan

Hostage Taker of Quads
Staff member
Moderator
Mentor
#11
FYI, changing this value in config.h does nothing on my MW Flip 1.5 with MW 2.3. In MultiWiiConf, in the upper left, if you click "Settings", you can change MINTHROTTLE on that screen, and it appears to work.
You see, this is one of the things I *DETEST* about multiwii . . . go between different versions and things simply stop working, but they don't go away.

There are far too many depricated items in the multiwii source.

I liked the idea of moving a few parameters into the settigns tables instead of global constants in the source code, but if you're not going to use the global constant any longer it should be removed from the code. Makes lending a hand to someone unnessicarily difficult becasue you need to know which version they use when they might barely be aware that source code even exists. In this respect, I think Multiwii is oversold as a refined, plug-and-play project.
 

Craftydan

Hostage Taker of Quads
Staff member
Moderator
Mentor
#12
Motors starting at different times is probably the number one non-issue I see newbies worrying about on multirotor forums.
Rich, I agree and disagree with this.

I agree there are far too many beginners who obsess about this, for the wrong reasons.

I disagree this is a non-issue.

The beginner looks at the motors spinning up in a random way and think the board isn't talking to the ESC's properly . . . basically for the reasons you've discribed . . . but . . .

If all four of your motors do not come on at the same time, that means the smallest "ON" throttle level the control board can send to the ESCs can be below one or more of the ESC's "OFF" state -- effectively the board can be in flight and maintaining control and inadvertantly shut off one of the late-starting ESCs, becasue the correction values went too low while on a low throttle setting.

This crashes airframes.

It's not theoretical, it happens. There is a small lag restarting the ESCs, getting the motor back in sync, and if the control board simply wanted to idle the prop to have minimum thrust, turning it "OFF" won't allow it to rapidly come back up to speed.

The control board should not have the ability to spin down any motor by accident -- it must be commanded by the pilot. The Minimum Throttle setting is specifically to prevent this, and should be set properly for every airframe. No matter how much correction or reduced throttle, so long as the pilot's throttle stick doesn't pass the magicall "off" level or the arm switch isn't flipped off, every motor spins *at least* at idle. Pass the "off" point, they all go off.
 

joshuabardwell

Senior Member
Mentor
#13
You see, this is one of the things I *DETEST* about multiwii . . . go between different versions and things simply stop working, but they don't go away. There are far too many depricated items in the multiwii source.
LOL. It's even worse than you might think. I commented out the #define line and tried to compile the sketch, to see if it had been deprecated without being removed, and the sketch won't compile without it. So it's still referenced somewhere, but it doesn't actually do anything. Ugh.
 
#14
One thought on the TX setup since you said you're using a dx61 - some people using the Spektrum gear have run into issues setting it up as apparently setting the end points for each direction is done separately which isn't initially obvious. A couple of people have had the same issue with not being able to use stick commands only to find that the problem was they only extended their range in one direction on each axis due to that.
Yeah. I remember trying to figure that out. You have to turn the sticks in the other direction in order to adjust those values.

So far, I have had no luck. I ordered a new set of Multistar 30A ESCs to see if they were the problem, plus they were getting pretty old and beat up. I noticed that the motor that started last had a bad bullet connector, so I soldered a new one. With the new ESCs connected, the same motors were having the same issues, so it has nothing to do with the ESC calibration.

That leaves FC and the motors to being the culprit to my problems. I am inclined to think that the problem lies with the motors, which would be a bummer. I use the suppo motors from Altitude Hobbies. I got a brand new one and it seems to be the one that starts up last out of all of them.

However, the motors seem to spin freely and do not feel like anything is restricting them from moving. I hope that the problem is with the FC. I am going to switch to the Naze32 eventually. It is an expensive trying to figure out what my problem is.
 
#15
The beginner looks at the motors spinning up in a random way and think the board isn't talking to the ESC's properly . . . basically for the reasons you've discribed . . . but . . .

If all four of your motors do not come on at the same time, that means the smallest "ON" throttle level the control board can send to the ESCs can be below one or more of the ESC's "OFF" state -- effectively the board can be in flight and maintaining control and inadvertantly shut off one of the late-starting ESCs, becasue the correction values went too low while on a low throttle setting.

This crashes airframes.

It's not theoretical, it happens. There is a small lag restarting the ESCs, getting the motor back in sync, and if the control board simply wanted to idle the prop to have minimum thrust, turning it "OFF" won't allow it to rapidly come back up to speed.

The control board should not have the ability to spin down any motor by accident -- it must be commanded by the pilot. The Minimum Throttle setting is specifically to prevent this, and should be set properly for every airframe. No matter how much correction or reduced throttle, so long as the pilot's throttle stick doesn't pass the magicall "off" level or the arm switch isn't flipped off, every motor spins *at least* at idle. Pass the "off" point, they all go off.
Thanks for all of your tips CraftyDan. I am not necessarily saying that it is the Multiwii's fault that the motors do not start properly. I was inclined to think that it was with the ESCs. But, as I mentioned in my previous post in this thread, I was wrong. I have narrowed down the problem to the motors and the FC. The main reason I mentioned the Flip 1.5 is because I have had issues with it before which I have mentioned in this forum. It is not as plug and fly as I thought it was, especially for tricopters. Right now, I am pretty fed up with it. It might just be me, but regardless, it is not for me.

Also, the quad does not necessarily instaflip.....kinda but not really. More like its goes toward one direction thinking left is up or something, and eventually crashing.
 

RichB

Senior Member
#16
Rich, I agree and disagree with this.

I agree there are far too many beginners who obsess about this, for the wrong reasons.

I disagree this is a non-issue.

The beginner looks at the motors spinning up in a random way and think the board isn't talking to the ESC's properly . . . basically for the reasons you've discribed . . . but . . .

If all four of your motors do not come on at the same time, that means the smallest "ON" throttle level the control board can send to the ESCs can be below one or more of the ESC's "OFF" state -- effectively the board can be in flight and maintaining control and inadvertantly shut off one of the late-starting ESCs, becasue the correction values went too low while on a low throttle setting.

This crashes airframes.

It's not theoretical, it happens. There is a small lag restarting the ESCs, getting the motor back in sync, and if the control board simply wanted to idle the prop to have minimum thrust, turning it "OFF" won't allow it to rapidly come back up to speed.

The control board should not have the ability to spin down any motor by accident -- it must be commanded by the pilot. The Minimum Throttle setting is specifically to prevent this, and should be set properly for every airframe. No matter how much correction or reduced throttle, so long as the pilot's throttle stick doesn't pass the magicall "off" level or the arm switch isn't flipped off, every motor spins *at least* at idle. Pass the "off" point, they all go off.

You make a very good point about getting down to shut-off throttle. I'm not sure where in flight one would encounter this, unless maybe you went zero-throttle to try to drop as fast as possible and catch just before hitting the ground. Perhaps over-powered setups that can hover at like 20% could run into this during normal maneuvers. I could also see low-throttle acrobatics hitting that limit.

But it's really two-issues: One, calibration of the ESCs being unequal, which I maintain is a non-issue for flight so long as you are not operating at the extreme edges of power-to-weight ratio, and have minthrottle set. Two, your point, which sways me, minthrottle is necessary to avoid accidental shutoff. Very good, fine sir.
 
#17
Ok. I do not why I did not do this before.

As stated before, I have narrowed down the problem to the motors and the FC. Turns out to be the FC. I determined this by simply switching two of the motors via their ESC connections to the FC. I switched the D3 and D11 positions. The theory is that if it is the motors that are causing the issue, then the same ones will start at the same times. If it is an FC issue, then the timing of the motors will switch between the two. Sure enough, it was the latter. I do not know if there is a fix to this problem. Let me know if there is one for the Flip 1.5. But regardless I think it is time to switch to a Naze32.
 

jhitesma

Some guy in the desert
Mentor
#18
30A multistar ESC's? Oh no. Multistar ESC's in general are horrible and the 30a's are the worst of the worst :(

I can't think of anything in the FC that could cause what you're describing at startup.

But those 30A multistars are horrible ESC's that should be taken off the market and whoever was responsible for billing them as appropriate for multi rotors taken out back and made to cut their own switch. The firmware on them is anything but optimized for multi use and the hardware on them has gone through several revisions (to save money not to improve anything) but none of the revisions have been very good.

The multistar ESC's can't be calibrated so doing a calibration on them is a waste of time. But they are also notorious for stalling out and not starting motors reliably.

Even brand new ones I'd still be seriously questioning the ESC's.
 
#19
30A multistar ESC's? Oh no. Multistar ESC's in general are horrible and the 30a's are the worst of the worst :(

I can't think of anything in the FC that could cause what you're describing at startup.

But those 30A multistars are horrible ESC's that should be taken off the market and whoever was responsible for billing them as appropriate for multi rotors taken out back and made to cut their own switch. The firmware on them is anything but optimized for multi use and the hardware on them has gone through several revisions (to save money not to improve anything) but none of the revisions have been very good.

The multistar ESC's can't be calibrated so doing a calibration on them is a waste of time. But they are also notorious for stalling out and not starting motors reliably.

Even brand new ones I'd still be seriously questioning the ESC's.
Oops. Should have thought of that. But still, the fact is that when I switched spots on the FC, the timing of the motors changed. That means it has to be something in the FC.
 

RichB

Senior Member
#20
Oops. Should have thought of that. But still, the fact is that when I switched spots on the FC, the timing of the motors changed. That means it has to be something in the FC.
Well then let's narrow it down.

First, hook your wires back up properly.

Next, place a book or something under one arm so that the motor which always starts first sits up higher. See if that changes anything