Tello brushless conversion

[Mud]

Hi there,

I would like to convert to brushless motors on my Tello and give it a carbon fiber frame. I have all the parts including four brushed to brushless converters from back in the days of the Blade MCPX and other micro helicopters which were only available with brushed main motors which a lot of people wanted to upgrade to brushless. Then there was "TheSteve", a tiny converter which outputs a PWM signal readable for standard, also tiny high speed brushless ESCs - birth of BlHeli I guess the converter must be soldered to the pin of an IC on the circuitboard which puts out the PPM signal to the onboard brushed DC drivers. See this diagram as an installation example on the Walkera 4#3 micro helicopter: HighSpeed Brushless Converter's by TheSteve Since I am not an electronics expert, nor do I own an oscilloscope to find out where there is a typical PWM signal used in RC. But I might be wrong about how the signal should look like.
So I thought I post this here since there is most likely some expert in this Forum who might be able to tell me where on the board to get the signal, or maybe someone even did that before...
?

Cheers

 

[Mud]

I found this video about a brushed to brushless conversion:

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies

He just uses 1k Ohm SMD resistors between the contacts where the wires for the DC motors were attached to and pulls the signals from each of the negatives which are then directly connected to the PWM inputs of the brushless drivers. His explanation makes sense to me and it is super simple compared to the other method. But I wonder why I never found someone in the web doing it this way before. Maybe this is not the optimal way to do it, I wonder how much energy just gets turned into heat in these 1k resistors. Or it is because the inverse polarity of the signal which is not a problem for BlHeli escs but has been an issue before. Any thoughts or advice?

 

[doug01n]

I don't think this resistors will heat too much. The 1kOhm will decrease the current consumed and it will help to filter the PWM signal without having a short circuit on the end. It seams pretty solid, but it requires a good iron solder skills. It would be great to have a spare Tello frame so you can cut the arms and use the original housing if you want to keep the camera position, sensors, as well as the stock battery.

Looking forward to see your progresses on that.

 

[Mud]

Thanks, yes I am going to try it this way. I just ordered the resistors on ebay, will take a few days. I bought a proper 90mm carbon race quad frame which weighs 16g and 1s DYS 1103 10KV motors around 3.5g each and matching the mounting holes. The brushless esc is a 4in1 sunrise 6A, also just a few grams. I will just remove the motors and keep the frame for now as is since it is a good housing with battery holder and nice landing gear and all. Also in the rear legs the two antennas for the wifi are positioned nicely straight and away from the carbon. Maybe later if I want it to lose weight I strip it completly and coat the board with silicone conformal stuff or wet protect. Not so sure how to mount it below the carbon frame though while needing the least amount of additional material. The top of the tello is neither straight nor flat. Also it is very thin and weak so screws disqualify. For starters I might just use very strong double sided tape and stack it a little in the front so i can level it out. Another idea I have is to use silicone glue. Both methods would help even reducing vibration, while the silicone glue would stick better I guess. CA with nano glass balls is also light and very strong, but thats harder to shape. Or good old epoxi...

 

[doug01n]

This will take some good investment to change it.
With the total money (Tello + new parts), wont you be able to get a BL drone with similar characteristics?

 

[Mud]

True, but there is no drone on the market which totally meets my liking so I need to tinker with the ones there are. The Tello could be it if it had brushless motors, a gimbal and no restrictions. In the DIY market there are no optical flow flight controllers at this size and weight.
I like them very small, but still want quite long flight times. I have a DJI Spark but it is just too noisy and heavy, there are more places you can fly with a drone the size of the Tello and smaller. Sizes for props around 2.5" seem to be the sweet spot atm for beeing as small as possible while still being powerful and not too heavy with the downsized electronic parts of today. There are two other, even smaller wifi selfie- or smart drones drones on the market I know of. But I believe the Tello has far more capabilities because of the powerful space tech computer that's under the hood and has way longer flight times.
But the biggest plus atm is that it's hackable and this community and people who develop Apps like aTello and another one that does edge detection. I also thought the little thingy might motivate me getting into that sort of stuff.

 

[doug01n]

You have the same spirit I have. Why not taking a great drone and make it even better!
Good luck, I'm looking forward to clap on your progress and help on anything I can.

 

[Mud]

Thanks! Yes I am looking for help here... I posted some other threads, too. 1. Would like to remove lower and upper height limits. 2. Possibility to turn off optical flow stabilization if desired. 3. Interested to know whether it is possible to use the USB on the drone via OTG cable to connect other things. I would be happy about some hints on where too look for or what to get into. I've looked up some former projects. With Parrot's AR drone 2, people managed to run their own software, even hooked up another wifi dongle via USB and an arduino somewhere on the mainboard. Funny stuff, I wonder if the Tello is also that hackable, I fear not, though. But if anyone finds some interesting project with the Tello or even another tiny drone, please let me know. I am aware of the great GoBot project, though! That's already awesome!

 

[doug01n]

I'm curious about how to add a gimbal to it... But I don't know the available gimbal boards available...WIth that and some tiny motors, I think it would be very interesting!
About the other questions, I have no clue. I'm not on that level!

 

[Mud]

I don't need a real gimbal for stabilization, but just one servo with less than 2 grams swieveling up and down would be great. As for the other stuff, yeah that's why I am writing here... Maybe someone has a clue and can point me in some direction.

 

[doug01n]

The stabilization already available on Tello is really good. I was thinking about a way to add a gimbal and other controls as a separated control, with a second receiver doing that, but that would also add another joystick on the game.

 

[Krag]

Thanks! Yes I am looking for help here... I posted some other threads, too. 1. Would like to remove lower and upper height limits. 2. Possibility to turn off optical flow stabilization if desired. 3. Interested to know whether it is possible to use the USB on the drone via OTG cable to connect other things. I would be happy about some hints on where too look for or what to get into. I've looked up some former projects. With Parrot's AR drone 2, people managed to run their own software, even hooked up another wifi dongle via USB and an arduino somewhere on the mainboard. Funny stuff, I wonder if the Tello is also that hackable, I fear not, though. But if anyone finds some interesting project with the Tello or even another tiny drone, please let me know. I am aware of the great GoBot project, though! That's already awesome!

Upper height limit can be raised to 30m. No way to remove lower yet. Nor to disable optical flow (maybe put a piece of tape over the camera?). I have tried to use the usb port as an OTG and it doesn't work. No current coming out. I have seen this hack for getting power out but haven't tried it yet.

https://tellopilots.com/attachments/camerawiring-jpg.583/

And a lot of hacking is going on in the development section:
Tello Development

I hope this project works. I would do it in a second.

 

[Mud]

The stabilization already available on Tello is really good. I was thinking about a way to add a gimbal and other controls as a separated control, with a second receiver doing that, but that would also add another joystick on the game.

Same thoughts, brainsnailing it with RC receiver or a raspberry pi zero w. I have a little FrSky RC controller from DYS which can hold a phone or a small fpv screen on top. Then there are OTG capable game controllers wich have the sticks to the side and attach the screen in between. So I will try to have a unit with 8 control sticks and one screen by combining these two types of controllers. Also steering via gravity, that is tilting of the phone could be an option, too. The Eachine e014 does that.
I worry about the tiny and short ribbon cable of the Tello's camera, wiggling it up and down repeatedly might break it after time.
I am going to build my brushless setup modular, so it is easier to try out different setups.

 

[Mud]

Upper height limit can be raised to 30m. No way to remove lower yet. Nor to disable optical flow (maybe put a piece of tape over the camera?). I have tried to use the usb port as an OTG and it doesn't work. No current coming out. I have seen this hack for getting power out but haven't tried it yet.

https://tellopilots.com/attachments/camerawiring-jpg.583/

And a lot of hacking is going on in the development section:
Tello Development

I hope this project works. I would do it in a second.

Thanks, just waiting for the SMD resistors to arrive in the mail. I already have everything else that is needed. I hope I bought motors and ESCs that match more or less. Yes a 5V source from the USB port would be great, but a 5V step up converter doesn't weigh and cost much. Nice that you have already tried out OTG functionality. Too bad it doesn't work. We'll see what people find out about the Tello's OS etc, I have good hope because the little thingy is so capable and has high end electronics imo which makes the effort worthwhile and therefore many people will try. Also DJI products don't have such a short shelf life as many products the competition. Even though that pushes forward development, it is not very sustainable at all. I have seen a project of someone successfully hacking a little chinese drone but then two months later, a slightly better version came out with different protocols etc. Two months later, same story again and the guy lost interest. They were too fast with putting out new versions and he couldn't keep up hacking. By the time his hacks became more elaborate, the product wasn't available anymore. Btw I don't understand why probaly hundreds of companies try to sell just about the same thing at similar price points trying to beat the competition. Why aren't there any companies making a thing that costs more but does all that we want? But that's another topic we don't have much influence in... Also, there would be no fun tinkering
Keep up your great work and contribution, Krag - I appreciate a lot

 

[Mud]

Hi there,

I am back again now with the Tello converted to brushless. Unfortunately, there is an issue of which I hope can be solved by changes to the control app. What I did is solder 1 kOhm resistors between the contacts for each of the original motors and connected each negative pole to the corresponding 4in1 brushless esc input. The negative poles are taken because the signal is expected to be cleaner, but inverted which doesn't matter since the BL-heli brushless escs can handle it ( info taken from above linked video on youtube ). From the empty cans of the brushed motors and plectres I made motor mounts for 5g AP05 brushless motors. Instead of transferring to another frame, I decided to aim for using the original frame and trying to add the least amount of mass possible. The Tello's frame is pretty stiff, but not against twisting the arms so if it runs I will add anti twist tubes between them.

Here is what it looks like so far:
Pluging in the battery, I can hear the escs 3 tone sequence but unfortunately the motors do not start. I assume the reason is the escs do not arm since they are not getting the expected throttle input signal. Moving left and right sticks inward and down ( startup sequence ) triggers repeated beeps from the motors seeming to be some programming mode. The same happens when I want it to auto take off.I am thinking that because it probably has to do with that it is an altitude mode type of throttle input instead of the usual direct throttle control, the esc's won't arm. I wonder if it would be possible to create the required throttle sequence with GO?

Any ideas?

Cheers

 

[Mud]

I am wondering if there is a suitable brushless esc as a spare part or salvaged from some other altitude-hold wifi drone that could be used here instead of what I installed. This is a sunrise cicada 4in1 5A 1s BL-Heli brushless esc.

 

[Storyline]

I am unable to contribute anything technical to this thread but just want to say that I feel that Mud is absolutely on the right track here.

Imo the Tello is the best UAV by far at the moment for such development work due to it's diminutive size, sound and 'presence' in the air. When I got my first gen Phantom I took it to my local park which is historically quite important. There are quite a few monuments and statues that I used to practice flying around - there were no flight modes to do this - there wasn't even a gimbal ! Anyway, drones were new then and I would often get quite a positive little crowd watching and wanting to know more - letting kids move the sticks whilst I held the controller etc. I was careful to keep the aircraft at a safe distance and well away from anyone but it was all good fun.

Now I have just got a Mavic Pro but would not dare take it down to the same park (even though it could produce silky smooth footage going round the same statues by itself) because of the way the perception of drones have changed. I am sure a majority would still be positive but also sure there would be some who would think it was dangerous, noisy and maybe even some who would accuse me of invading their space. You just would not get this with the Tello. It is almost toy like and non threatening. We know though that although the footage is not Mavic Pro quality it is probably as good if not better than my original Phantom.

So good luck Mud . I am reasonably proficient with a soldering iron and providing any conversion to brushless solution has intelligible instructions I would definitely be up for giving it a go !

 

[doug01n]

I'm interested on that too. Changing motors after some time of flying would be really annoying to me. Brushless is also more efficient, so it would have more power and longer flight times.
I wouldn't care paying 50% more only for adding brushless motors and tilt camera.

 

[Mud]

Thanks for motivating!
Same here, I would be paying 50-100% more for a brushless version with tilting camera, too. And even more if it was open source. I guess only most people just mainly look at the price and the manufacturers aim at maximum sales volume so that's what we get. Tinkering is fun though but often more expensive and you can get stuck with some unforseen problem like the one I'm having now. But I am not giving up yet and I am not alone with this desire, so maybe someone in this forum has a solution or some idea? Also, you are right doug, if the Tello is flown often, the brushed motors will eventually have to be swapped anyway and that's gonna be way more costly than having brushless motors from the beginning on. Actually having taken apart the whole thing made me realize that it is not made to be used for a long time. The original motors are way too expensive and too hard to swap for an average person and it is horrendously expensive to let it get fixed by DJI, respectively RYZE. As for tilting the camera, I must say that I doubt the ribbon cable can stand being wiggled up and down many times. The ribbon cable is very stiff and short, it will most likely break. So we would need to find a longer and flexible ribbon cable. Maybe with some more hacking another channel for controlling the tilt servo could be unleashed. Idk if it would be possible, I could imagine that there could be done something with the USB port, but I am no expert in that, although willing to learn if I knew how. As mentioned one could add a raspberry pi zero w or regular rc receiver to add more channels. But it would be much nicer not needing additional control hardware. But if using an extra receiver or raspi I would keep the front facing camera at its angle for the sake of working follow mode and such and add an extra fpv camera with tilt or more altogether. Way more fun flying with almost lag-free video transmission as well.

 

[Mud]

Bounty hunt:

If anyone can make a change to the atello app for example, so that the escs would arm - the thingy would fly again and I would give it away as a reward XD