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

Howdy Flite Test Forums!

JimCR120

Got Lobstah?
Site Moderator
#2
Hello sflemdog and welcome to the forum. I see in your signature you already have some equipment and I see in your blog you've got some detailed troubleshooting underway. Unfortunately I am also relatively new to multirotors so the help I can give is pretty limited (almost nonexistent). However, I would encourage you to stay at it and while others peruse the forums you are likely to have someone pop in and offer some things to check.

Until such a time this response will bump your post into some higher visibility. By the way, where in the world do you fly?

Keep us posted.
—Jim
 
Last edited:
#3
I have read your post. I hope that you can get that quad in the air because it looks mean. I am an armature when it comes to quads but I know some about electronics. When you powere your stuff up, feel all your electronics. If you feel something getting hot then you know that is a problem. You have probubly already done this and I dought it could be that easy. You might just have a bad component. I would try the seprate circit like the manufacurer said. Ik you probubly dont want to add more wirers to your rig. Just make sure you use 5 volts for the reciever power. I hope that this helped. If you want to go the extra mile, you could run your 12v from your battery to an arduino board and take the 5v from that for your reciever. Then use the other pins to do cool stuff like power led lights or something. It would be programable on the computer useing c++ and arduino softwhere.
 

PsyBorg

Wake up! Time to fly!
Mentor
#4
Welcome to our humble home mate you have indeed come to the right place for help.

Nice clean build you have started. From what I see your issue is only with lack of knowledge and proper set up procedures. I would be willing to guess that the beeping after power up is the ESC's telling you that the throttle is in too high or to low a condition to safely arm. Why do I say this? Because in your blog you list throttle settings with trims at minimum and trims at maximum. When setting up a multi rotor all that is done in the radio and the FC. Once centered positions are set the radio trims must NEVER be touched like on fixed wing craft.

To save time I would suggest that you delete the model you have been working with and start fresh. If your radio has a script or set up method for multi rotors start with that. If its an older or non "smart" type radio start with a normal model using TAER. Throttle, Aileron, Elevator and Rudder as the channel mapping. This is usually the default for most radios anyway but just make sure. Make sure all trims and sub trims are set to centered in the radio which they should be on a fresh model entry.

Go thru and bind your receiver to the radio again to make sure they are properly bound and talking. Once that is done reconnect the receiver to your flight controller making sure all the channels are connected using TAER for channels 1 thru 4.

Now we can connect the flight controller to the computer. Since you have a NAZE board I will assume you are using one fo the *Flights gui's to connect to it. This procedure is the same no matter which one you use. Make sure for all this your PROPS ARE OFF. You do not need a screaming veggomatic of deaths jumping up at your face when trying to make it work. Once you know the computer is connecting to the flight controller we can start the set up.

Go to the receivers tab in the gui and make sure when you wiggle the sticks that one they move in the gui, and two that each stick is working the correct channel in the gui. If it is not there will be a pull down menu that lets you assign channel maps. Select TAER and check again to be sure things are correct.

The next step now that is all correct is to set the channel centers. You do this thru the sub trim part on your radio. Using those sub trims you want to match each channels center as close to 1500 as you can get it. It may bounce by two points either side of 1500 any thing more you need to change the trim. IF you can not get it that close we can address this later when you reply. I do it for throttle and switches as well although it is not a necessity. More on that later.

Once all four channels are centered we will set end points or travel adjustments depending on your radio's terminology. Each channel needs to have a lowest point of 1000 and an upper point of 2000. You do this as stated in the travel adjustment or end point section of your radio. Keep in mind your upper points should already be at 2000. I drop the radio settings until that goes below the 2000 and then bring it back up for a more exact setting. This allows for the best resolution possible for control.

Now that all of your centers and end points are finished we can go to calibrate your esc's. Again with PROPS OFF... Connect to the gui and go to the motors tab. There will be a check box that confirms you know the quad can eat your face if you do this step with props on. check that to activate the motor control by software. Slide the master slider to full power. Plug the flight battery in and as soon as tones start playing slide it back to zero throttle and the esc's should all reboot. IF you stay at full throttle too long you can potentially put the esc's into programming mode and will have to go back and default each one individually.

After the calibration you can move the main slider up and down a little bit to test the motors and make sure they are all starting at the same time. There may be funky behavior after they sit running for a second or two but don't worry about that. Just make sure they all start at the same time.

Now we have to find the minimum throttle setting. To do this we will high light each channel one at a time and use the up arrow to more accurately move the slider up until we fond the spot where each motor turns steadily with no twitching or jittering. Write that number down for each motor. Now that you have all four motors lowest number you will take the largest of the four and add 15 - 20 points to that since you are using larger props to account for the weight. That number will go into the configuration tabs minimum throttle setting. The last thing to do is make sure the motors are spinning in the correct direction. Use the picture in the gui that shows correct rotation. This usually means front left spins clock wise, front right spins counter clock wise, rear left spins counter clockwise, and rear right spins clockwise.

Now all that is left is to make sure everything works as it should. Unplug from the Gui and reboot your radio. Now power the quad STILL no props please. After the flight controller and esc's boot up arm the quad and test the throttle spins all four motors at the same time. Make sure there is a response when pitch roll and aileron is made with the sticks. throttle up a little bit again and pick up the quad. Tilt one motor corner lower then the rest. That motor should speed up. Do this for all four motors. If they respond as they should set up is now complete. Power down and dis arm the quad.

The very last thing you need to do is set the failsafe up for your radio and receiver. Personally I fly race quads and set mine to stop all motors and fall from the sky as I do not want a low flying missile going rogue. It is better to lose gear then hurt someone or damage property. That choice will be yours but make sure you have something set up and then test it buy powering the quad then turning off your radio and make sure the expected behaviors happen correctly.

This completes your set up with the radio and flight controller minus any switch settings you use or have available. Once you go thru all that you can put the props on and do a final test OUTSIDE.

For this I stand back from the quad a bit after powering it up. I arm the quad and make sure the motors spin smoothly at idle with the props on. If they do not I make a note and will go back and raise min throttle a touch more later. Keep in mind when you arm the quad already thinks it is flying so some motors may spool up over time or slow down and stop if it sits too long with now inputs. This is normal behavior. Anyway at this point you want to SLOWLY raise the throttle until you see the quad start to get light but NOT lift off. then I will gently ease the pitch stick forward to make sure it starts to push forward the same for backwards. Then I repeat this for roll. Then once again to see that the quad tries to twist in the correct direction when yaw is applied. If all responds as it should you can now pop it into the air for your first test hover.

If at any point something is not working or you have troubles feel free to post here again and ask for more help. good luck with your quad and make sure to get us more pictures and possibly video of the maiden flight.
 

Snarls

Gravity Tester
Mentor
#5
Yay an old school quad build with DIY frame! How rare you are these days :( After reading your blog post here are some quick tips I can offer. Sorry if you already wrote some of this Psyborg, I didn't want to cross check with the wealth of knowledge you have already given.

From your pictures and descriptions it looks like your wiring is fine. I think your ESCs are not getting a signal, or not getting the signal they want. When not armed the flight controller sends a constant signal value to the ESCs. By default the value is like 1000us. For most ESCs this is fine and they will boot up ready to go. But for some ESCs they expect a higher value in order to become 'ready'. This may be what is happening with you. Especially if you are using an airplane designed ESC and radio which typically operate closer to 1100 and 1800 as you saw.

To change the value you want to look at 'mincommand' (or something with that sort of name). I'm not sure if you can change it in the GUI or if you have to go to the CLI. Increase it slowly until your ESCs are satisfied on startup. Be careful, increase it by small increments, and keep the props off! You may also have to bump up min throttle if you start approaching it with min command.

Good luck, I want to see it fly!
 
#6
Thanks all

JimCR120 - Thanks. I am in California, in the Silicon Valley. I will change that on my profile later.

Gatormurica - My ESCs are BEC (linear) and do get hot when using the BEC circuit to power the flight controller. I haven't decided it's necessary to undo the power lead because they are not switching BECs. Still might change it for troubleshooting purposes temperarily, so as to make progress in some way. Thanks for the kudos on the frame, I have designed and machined other things, so for me it was extra fun. I actually got a new power distribution board with a 5V output and plan to play with that instead of using the built in BEC on the ESCs to change the setup for troubleshooting, as I said before.

PsyBorg - Thanks for the basic setup advice, as they are a good set of instructions...until something goes wrong like mine, doh (palm to forehead). I've gone down my diagnosis path to try to rule problems out and I feel I am getting close to the true issue.

Snarls - Thanks for this information ("ESC...operate closer to 1100 and 1800 as you saw"). I am learning that some ESCs, especially in the era that I got mine, are different than the mainstream and inconsistant to most genereal signals for "arming" or throttle calibration. I find your suggestion very helpful about the min command changes and will try this later...as I know if you start to change too many settings at once you end up with a mess.

I just wanted to say too, I have really basic 4 channel Rx with manual trim settings...nothing fancy like the ones people have with screens and like 8 different auxiliary switches. I am hoping I can get through this without feeling the need to go buy a more advanced radio setup before I even get off the ground.

Sometimes I do feel discouraged, but I have solved many odd problems like this before and don't believe in giving up. I have hope still and plan to shove forward. I'll keep you updated.

One of the oddest things I am currently running across is that my ESCs aren't talking to my naze32_rev6 yet AND that power is routed oddly. From my blog, you will see I am actually getting quiet beeps from the ESCs when connecting the naze32 to USB...I found out later that this naze board has no diode from the board's main power (via USB at this point) to the ESC rail, so they are getting power without the battery plugged in. Normally, with no USB power, the ESCs would be powering the flight controller (naze32_rev6), but from the beeps you can tell it seems it getting power through them (bad because it's reverse). Maybe I'll upload a video of this odd occurrence.

Going to check some replies to my blog now. Thanks all.
 
#8
(posted in blog as well): USB voltage was waking up the ESCs early because it was getting pushed through the board to the power distribution board (PDB). I was using a servo extension to connect the naze32 ESC rail to power from the 5V regulator on the PDB. This naze32 rev6 apparently was one of the early ones without a diode connected to the power end, so I added one in-line to the servo extension wire I talked about before. This keeps the ESCs from waking up during USB connection, but still can't get the ESCs to beep when I try to do the Cleanflight throttle calibration procedure. Hmm.
 

Snarls

Gravity Tester
Mentor
#9
I was using a servo extension to connect the naze32 ESC rail to power from the 5V regulator on the PDB.
You should not have to do that. Your ESCs have BECs that should power the naze on their own. That 5V regulator is if you had opto escs or have an accessory that requires 5V, such as an FPV system. When calibrating the escs they should have no power when you have just the usb plugged in. Then in the calibration process you will power the entire quad (escs) with the battery while it is still plugged in to usb.
 
#10
You should not have to do that. Your ESCs have BECs that should power the naze on their own. That 5V regulator is if you had opto escs or have an accessory that requires 5V, such as an FPV system. When calibrating the escs they should have no power when you have just the usb plugged in. Then in the calibration process you will power the entire quad (escs) with the battery while it is still plugged in to usb.
I understand I could use the BEC on my ESCs, but after finding out there wasn't polarity protection I'd really rather have power going through the PDB than my ESC (also, sitting still ESC was getting hot). PDB is cheaper to replace if need be; don't want to blow an ESC accidentally. Also, I could solder to a servo extension instead of making a permanent diode addition for now. Should still work. Thanks.

See my wiring diagram here
 
Last edited:
#11
Still can't get my ESCs to calibrate...even tried this(found somewhere on fpvlabs forums, not sure where now):
From the CLI prompt: type "set min_command = 2000" and enter.
type "save" and enter.
Wait for re-boot and re-connect to the CLI. type "set min_command = 1000" DO NOT hit Enter yet.
Connect LiPo to power the ESCs (Note: a current limiter is not a bad idea) Once the ESCs sing their song hit enter.
Now the ESC will be getting the 1000usec pulses and sing again. Once they
stop type "save" and enter. Disconnect LiPo.​

But this did not work. No response from the ESCs. They are Hyperion Altas ESCs and I have been told they just use a basic PWM protocol. Is there a possibility that there is another setting for the timing for communication that is not set properly yet? Has anyone used these ESCs?
 
#12
I wanted to post this here too:

Update: I went to a local hobby shop and had one of their experience quad builders / flyers look into my setup. We did some troubleshooting, most of what I have done in the past, but to no avail. He did recommend I just get different ESCs because the headache is not worth it anymore. He might be right. It's really too bad that after even contacting Hyperion themselves and a local expert that there's no solution.
I would NOT recommend Hyperion products, for beginners at least.
The above was not a recommendation stemming from anger or quick decision...I have spent years trying to get there products to work, maybe they work really well on just straight radio to ESC setups like in airplanes, but they do not seem to agree with the signals coming from flight control boards. Maybe they have their ESCs flashed with firmware that doesn't accept a signal that is not constant from boot and they always failsafe. If you've heard of this issue / have some insight, please comment here about it...would love to hear. Thanks.
this is from my blog post, but same message: probably just going to get new ones (different brand).
http://forum.flitetest.com/entry.php?2638-Naze32-(rev6)-via-Cleanflight-to-Hyperion-Atlas-ESC-communication-issue

p.s. ... I will be keeping the motors, I think I read somewhere that they were German, so fingers crossed for performance when actually in the air, some day.
 
Last edited:

Snarls

Gravity Tester
Mentor
#13
Sorry it's not working out. Maybe your right and it's better to get some ESCs that you know work with multirotors. You should be able to find some 30a ESCs for fairly cheap these days.
 

Snarls

Gravity Tester
Mentor
#15
I'm not sure what you'll find at hobby shops these days, but any 25a - 30a ESC with blheli should do the job just fine. It's really not about the manufacturers these days, but what the ESC is capable of. Most miniquad fliers are looking for blheli_s (special form of blheli) and dshot 600 right now. You're flying a larger quad so you might not want to worry about those specifics, except that most multirotor escs run blheli. Of course if you find some with those characteristics for a good price, go for it. Also make sure to get at least one that has a BEC, or you can get four opto ESCs and use the BEC on your power distribution.
 
#16
I'm not sure what you'll find at hobby shops these days, but any 25a - 30a ESC with blheli should do the job just fine. ...
Snarls, thank you for voicing your opinion it helps, but I respectfully disagree. Maybe the chips are all made in the same place; have the same quality and I am not sure what BLHeli has to do with it because it's a firmware...it's all virtual as far as I'm concerned, but what I mean is QUALITY for reliability as in more of the physical side of things. (Side note: saw that BLHeli is run primarily on SILABS MCU, can be on ATMEL...long story short ATMEL is old, more reliable, whereas SILABS supposed to be faster, but newer so not as much reliability data). Like, maybe these things are being shoved out in mass production, but what's the soldering like? Do they use higher quality surface mounted components? Things like this...to be honest if I just go with what is sold most at local hobby shops it is bound to work, that's why the hobby shops carry it. Harder to decide these days when you can get just about anything through the internet. If I go to a local hobby shop here (Silicon Valley, CA, USA) and want ESCs for brushless motors I find these brands on the shelf: E-flight, Castle Creations, Jeti...each brand with a various line/type depending on, presumably, your experience/demand for the electronics to operate well. I might have this started in a new thread or post my 2 cents in a current one when I find it, but I might just stick with something from E-flight. They are very mainstream over here from what I can tell and sometimes you get better luck deal with what is the most mainstream product. Any real insight into ESC quality? Let me know, thank you.

p.s. this may be overkill discussion for such a beginner, but if everyone just bought what was cheapest then that's what will be everywhere: the cheapest junk. I say junk because unfortunately that usually what "cheapest" equates to on the back-end is the use of cheaper components / cutting corners on process quality to give you a lower-priced product. I believe in "investing" in the hobbys I go for...unforutanetly I tried this in the past with Hyperion and didn't do enough research. Ultimately, even the tech rep from Hyperion admitted to not "waste my time on" getting those to work on a quad build and just buy one of their newer products...unfortunately again, I have already bought them and tried to stubbornly just make them work.
 
#17
So, I decided to order a Emax Bullet 30A OPTO ESC. At least in the descriptions, they claim to have used some nice quality components and have tried to improve on older issues with heat. Also other reasons for decision are: 1) Lots of people seem to like these and there are people out there talking about them and how to use them. 2) Seem to be a newer release, as I seemed to have order old ESCs for the time when I got my last ones. 3) Not super expensive, but still need to buy/solder the motor wires and connectors when it arrives.

I am going to check out the basic operation of this ESC then go for getting 3 more, hopefully.

Thanks all, for the help.

I was given a little ProtoX to keep me from going too far off the deep end with frustrations on this build, will suffice for now ;) Haha

and bottom line: Do not use Hyperion Atlas ESCs with flight control boards. Even local hobby shop experts were confused as to why the setup wouldn't work because they should take basic PWM signals. Maybe you can use them on airplane setups just fine, but not multi-rotors.
 
Last edited: