Racerstar 8520 motors controlled by L293D on arduino doesnt take off

Hello everyone, this is my first post; so bear with me for anything that may have been discussed on this space already. I have an 8520 racerstar coreless motor but then i wanted to try if Arduino Nano can power this through an L293D. I’ve also got the MPU 6050 for the required gyro and accelerometer data but then even before i get there i have a problem with the thrust i guess.

Given that am using an Arduino i wish to control speed with the analogWrite() function and then the digitalWrite for the respective pins in play. The whole thing works well - as in functionally i can ramp up from say 50 to 255 in the value range and the motors accelerate as they’re supposed to. Being a small frame i worry that the whole setup isn’t aerodynamic enough and maybe that’s another reason why this is stuck on the ground.

I use the 3.7v 500 mAh batteries - two of them - wondering if 1 wasn’t sufficient. Still no luck. Before i labor with my PID challenges i wanted to know if this will take off at all before i go forward :slight_smile:


Maybe this photo will help understand how this looks. some more detail-

  1. this has the hubsan 65mm propellers
  2. total weight is 90 gms - with the batteries - stuck to the bottom

I read that the thrust per motor is around 30 to 40 gms so it should be able to lift 90 gms for sure; maybe it is the design.

thanks. IMG_20190331_193702046

I think your biggest issue is probably weight, with a side order of motor driver efficiency.

Starting with the motor driver, those motors take more amps than you’d expect. At full tilt it’s 2 or three amps per motor if I recall correctly. The motor driver you’re using is only rated for 1 amp. Additionally it’s a Darlington layout of BJT transistors, which means there’s a voltage drop across them. The datasheet doesn’t say how much. This is an issue as single lipo quads barely have the voltage available to get into the air in a lot of cases, losing some off the top is problematic.
Flight controllers use MOSFETs for this reason. The Arduino should be able to drive MOSFETs directly, so that’s worth looking into.

(Bonus: you need a high current lipo battery, I don’t know what you have but if it’s not built for high current discharge it won’t fly a quad)

Weight wise, given the proper battery and motor drivers those motors should be about to lift off with up to 80 grams or so of total weight. It won’t fly well that heavy, but it should at least get off the ground.
A good goal for a quad powered by those motors is 40-50 grams takeoff weight.
If you can get it lower that’s worth doing, but it will do decently in that weight range.

If the battery voltage is sagging due to load that will drastically lower the amount of weight it can take off with.

Hopefully I haven’t crushed all your hopes and dreams, unfortunately quadcopters on this scale are touchy about weight and power.
What you want to do is very doable, most of the early quads ran Arduino software. Some still do!

1 Like

Thanks much Bobnova for the suggestion. Looked up the data sheet and https://www.arduino.cc/documents/datasheets/H-bridge_motor_driver.PDF says it is a 600 mA output per channel. My battery (for each pair of motors) is currently 3.7v but 500 mAh … maybe i’ll shift to the 1200 mAh alternative this weekend and check if that helps get the desired output.

Regarding the laden weight - i agree; 40 to 50g would’ve been ideal but then i have to get away from the mini bread boards and make a PCB like alternative with soldered connections. That should reduce at least 20g if not a little more. Given my (lack of) soldering skills that is going to be plan C for now :slight_smile:

And thanks for the vote of confidence :slight_smile: really appreciate it.

I totally second @Bobnova. Everything he mentioned is spot on. Listen to him. My first concern was also the target weight of your machine.

Seems you are kind of confusing two things here mAh and mA - the H makes a difference. You want a lipo with a high real “C rating”.
The current draw of 8.5mm motors can be even higher than 2As, a 600mA motor driver just wont cut it. You have two batteries?! You want to fly with a mini breadboard?! 20g weight difference is a massive change in that size. If you can, you really should get rid of those 20g.

Extremely short variant:


You probably want to take your time and reconsider some of your choices. Feel free to ask. There are some people here who can certainly help.

1 Like

Thanks las … i have to find out the C rating for the batteries. With respect to the motor driver choice itself - L293D, i understand it was a better candidate for rolling cars (on the ground); i happened to choose it only coz it was light weight and easier to put into the mini bread board. If you have other obvious motor driver suggestions, kindly advise.

And from what i’ve heard from the two of you - i will now prioritize cutting the weight along with the battery upgrade. Will come back over the weekend once again when i make those changes.

Appreciate the inputs, once again. thanks.

1200mAh will be too heavy. 500-600 is about right, but it’ll have to be a battery built for quadcopter use so that it can put out lots of amps (not to be confused with amp-hours). You can find gnb/tattu packs on eBay for not too much.

Motor driver wise, get some N channel MOSFETs with a low gate voltage and low RDS(on) and use those.

Weight is absolutely key here.

Depending on what your goal is, you may be best off trying to find an old Arduino based flight controller and programming that, or by grabbing something modern, installing the keil ide(free) and learning how to make stm32 microcontrollers do their thing.

Along those lines you could look into Silverware and toy FCs too. It’s an open source flight control software I’m contributing to, and it runs on super cheap flight control boards.

1 Like

Maybe we should restart with the following questions: What do you want to achieve? What’s your plan and goal? Do you want to build a flying quad from scratch and also write the complete flight controller software?

We have competent personal here who will very likely help you with whatever goal (w.r.t. to micro quadcopters) you want to achieve in whatever way. But the initial approach to whatever you are trying to achieve (which isn’t exactly clear to me, thus the questions) seems to be burning a bit too much cash for a too little fun gain. We can certainly help you with burning your money in way more entertaining ways. :smiley:


thanks once again, las and bobnova. Here is what i figured …

  1. I dont have a C rating on the lipo i purchased but am reading (on websites that sell similar stuff- this one i got from amazon) they range between 25C and 40C for the discharge rate. For now i guess i will try with 2 of them as before; just that i’ll work on removing the bread board and soldering a PCB myself.
  2. Secondly - the motor driver choice - was simply convenience; the l293d was light; breadboard ready; and could power 2 motors; but if the output Amps is not good enough maybe i have to get to the MOSFET alternative.
  3. To las’s questions - well - it was a simple intent to get something to fly- thats all :slight_smile: am more a coder than an electronics person so am definitely keen i write the flight controller myself (i hardly understand PID to start with for now :)). I started with the Alan Brokking site that gave me details on the larger Quad (quite wonderfully captured detail on his site) but then testing the larger brushless motors and the larger weight was actually challenging for me. So i moved on to the smaller version - just to get something light-weight, to fly. I chose an Arduino Nano coz it fit the adjoining breadboard as well very well; so i have 2 small bread boards inter-locked next to each other with a maze of cables and then 2 batteries stuck to the bottom.
    Thanks once again. I will try the lighter alternative this weekend and update. Really appreciate it.

If coding is your game check out silverware and the E011C(which I believe is still silverware compatible).
The whole quad is $17 or so and flies decently, and you can flash your own firmware on it. That’s where silverware comes in, it was initially written by silverxxx for reasons I’m not sure of, and has recently been forked and developed by @notfastenuf and a variety of others.
It’s something you can use as a framework and write your own PID loop for. That way you don’t have to develop an entire firmware in one go.
Unless you want to, in which case it’s still a nice starting spot.
You can do the same thing with betaflight controllers, it’s also open source, but there’s a lot more code to digest.

1 Like

Or there is the old “alien classic narrow” available from Micro Motor Warehouse if you want to stick with the Arduino platform as opposed to go with c language. That is an all in one arduino PCB solution for you that was hand assembled lovingly by one of our own members here. It has your atmega mcu, proper n channel mosfets, your gyro, a USB plug, and proper motor plugs all ready to go on a convienent and lightweight board. Just add your Rx.

If it were me… I’d keep the breadboard on the desk next to the pc. I have 5 or 6 now next to my PC with junperwires and other unflyable heavy nonsense going all over the place. I write my code… Flash my dev boards… Test and debug.
BUT THEN… I flash what I’m happy with over to a properly lightweight FC and fly till my heart is content where I can enjoy the fruits of all my labor in something high performance!


Just wanted to point out that the top left prop in the pic is the wrong orientation (i think). Whatever… double check prop and motor directions.

Acknowledged, Bobnova and Notfastenough… Will go with the Arduino based version to start with. Will get hold of the mosfets this weekend and give it a go. Also I had planned to control via hc05 the Bluetooth module. Right now, it is rooted to the laptop with the cable where I send serial input to test the speed increase and thrust. Once this works I will then remove the cable and send commands over BLE.
@DsAds, I will check again. Unlike the altura hold commercial version s blades these hubsan ones almost look the same whichever way I hold them. Thank you.

1 Like

Just an update … bought the IRFZ 520N and the thrust is definitely better. I have to redo the MPU connection as i’ve got a fresh setup from scratch going. Will update everyone once i get the MPU and the BLE going in the coming week. One challenge i may have is - the weight still hovers around 87g. It looks like this will take off but after some attempt with my own soldering on PCB I think i may have to change the type of wires I use (i use the single strand cables which i think are adding some 7 to 10g here).

If anyone has a view to the IRFZ520N do let me know; i couldn’t get hold of SQ2310ES N channel, which was my original chase (per some link i saw on the web that uses mosfets).

thanks once again.

They’re decent enough.
Something built to be driven by 5v gate voltage would be better, little sot23 things reduce weight a lot which is nice.

They should be able to get you into the air though.

1 Like

This took a while, everyone… But I got the basics fixed I hope. I have a PCB done, N channel mosfets in place, weight reduced to 87 gms even though I have only 3.7v batteries powering the Nano and the motors. I even moved away from racerstar to chaoli 8520 coreless just to see if they made a difference but after all this my quad is still grounded :-(… I even validated the thrust with a long winded formula but nothing that suggests that 87 gms can’t be lifted off. Any thoughts?


That should be able to get into the air.
I’d check the voltage across the MOSFET at full throttle.
(Make an Arduino program with no PID loop, just set one motor to 100% for a couple seconds, then check the voltage between motor - and battery - when the motor is going full tilt. It should be very low).

Probably also worth checking the voltage between motor+ and battery- to make sure the full battery voltage is making it to the motor.

Those motors and props should be able to lift 83g(that’s including the battery right?). It’ll be a short, high throttle, flight, but it should at least be able to hover on ground effects.

Yes … they’re 87 including all 3 3.7v batteries as well. Will try measuring the voltage on this and come back. I even got this little gadget to measure up for my remote car but haven’t tested it across the mosfet. Thanks @Bobnova.
The other concern I had was on center of gravity… but then the actual thrust numbers come up to as high as 280 gms for this design and I was hoping that in reality I should be able to pull atleast 90.


The voltage was a little easier for me to measure between motor + and motor - … Came to around 2v when the throttle is like 190 out of 255. Am using simple analogwrite function on PWM pins and getting this reading. Will do some more plumbing to get the voltage across battery - and motor -.

1 Like

@Bobnova, thanks for the suggestion … i figured it was the C rating on my battery that’s not good enough to provide sufficient AMPS. I used the Volt/ Amp meter given above to validate how much each of these motors need - and roughly (if am not wrong) they seem to need ~2 Amps at least to get to their peak rev. My current batteries i guess are not 25C as i suspected; they end up showing as low as 0.1 amps while when i use a more liberal 11.1v 3S Lipo, i see that a single motor can burn up to 3.2 amps. Have ordered for better rated 3.7v batteries and hopefully that will work.