ModalAI Forum

@Alex-Kushleyev @Eric-Katzfey Managed to get everything up and running with your advice - One thing I needed to do first was flash V1.0.0 as the system was running V1.0.0beta and had an older version of ESC tools V1.08. Once V1.0.0 was flashed the ESC tools was update to 1.20 and allowed me to update to the ESC firmware version 38. After everything flashed and updated I now have the battery information in QGroundControl! Many thanks again for your assistance!

@Aidan-Dempster , Good question. As you may already know, the 4 ESCs normally only send their feedback status packet when the host (flight controller) queries them. The feedback query bit is set for one ESC at a time when sending out the control packet to the 4 ESCs. The Flight controller requests the feedback from one ESC at a time in round robin fashion in order to avoid collisions on UART line. This is true for all ModalAI ESCs. Now, the mini ESC M0129 has a special setup in terms of current sensing. IDs 0-2 do not have any current measurement, but ID3 measures the total current. In order to avoid confusion with individual current measurement, we have added a new Power Status message, which ID3 sends out in addition to the standard feedback. So when the Flight Controller asks ID3 to send feedback, the ID3 will send back the normal feedback packet and the power status packet. Here is a link in PX4 for your reference: parsing of standard packet: https://github.com/modalai/px4-firmware/blob/voxl-dev/src/drivers/actuators/modal_io/modal_io.cpp#L303 parsing of power status and sending out battery status to PX4: https://github.com/modalai/px4-firmware/blob/voxl-dev/src/drivers/actuators/modal_io/modal_io.cpp#L393 It looks like the power status feature is only in voxl-dev branch of PX4 code at the moment. Hopefully that helps! Let me know if you have any other questions. Alex

Hello @andrespalomino , It is possible that during crash the motors have sustained some critical damage. Since you indicated the motors "burned" that usually means the winding insulation has melted and there is probably internal short inside the motor. when you say motors spin correctly when they are separately - are all 4 motors spinning at the desired 4500 rpm with very little current, if tested individually? Hopefully the ESC is still working fine (please check it for any indication of burned components). You can do a quick test in open loop mode before replacing the motors: python voxl-esc-spin.py --id 255 --power 10 --timeout 1 This will apply a fixed power of 10% and it is a very "gentle" test mode to make sure the new motors and ESC are working properly. You can test with propellers or without (maybe better without propellers at first). All 4 ESCs should spin around the same RPM and current should be very low. You can increase the test duration from 1 second if all goes well. As next step, I would suggest replacing the two motors that are not spinning correctly. Burned motors typically means they need to be replaced. After motor replacement, you should repeat the spin test in open loop mode (in previous step). Also, are you able to share the motor and propeller used on your vehicle? Alex

Hi @Jon305Johnson We will not be offering 60A support for some months yet, but we are now testing our standard 4-in-1 ESC at 6S support. We built the first few proto units and expect to have some results in a matter of weeks. All forward progress. Thanks! Vinny

@ErezI When it comes to comparing the quadratic curve fits, it is difficult to look at just the coefficients and compare them. If you have multiple tests, you should plot the quadratic plots on the same figure and see how repeatable they are. What may seem like a noticeable variation in the coefficients, could still result in curves that are very similar. By the way, you should not just average values of a0, a1, and a2 to get a0_average, a1_average, and a2_average, as this will result in wrong curve. If you would like to find an average of all 4 runs, then you can just create a quadratic fit of all the points from the 4 data sets - that is the proper way to do it. However, if after plotting the 4 curves they look very similar, you can use any of the 4 individual fits. The PI (Proportional Integral) controller on the ESC will compensate for minor deviations. The error you saw with params not being valid was due to the latest version of voxl-esc tools using new parameter structure. So the solution is (which is what you already figured out) to update the firmware and params to latest. This type of change does not happen too often. We will add this to documentation to make this clear. Alex

VOXL 2 Mini is powered at 3.8V and VOXL 2 is powered at 5V. So, there will be a slight hardware change for the Mini ESC to support VOXL 2. We do plan to have that as an option in the next month or so. It is a PCB stuffing option for 5V

@Alex-Kushleyev Thank you so much for your response. We tested ESC with some other motor and found out RPM and power was exceeding 10000 RPM / 60% power and everything seems to be good with ESC. Our motor must be damaged or there is some problem. I will consider your feedback for motor selection as well. Thank you!

Hi @Eloi , Thanks for the feedback, our website is out of date and the the MCBL-00029 should have been included, let me work with our team and we'll get you a cable in the mail to resolve, I'll send an email to your address here.

Please follow the steps here at https://www.modalai.com/pages/rma

Without propellers you are not calibrating the ESC

Fixed the problem, the mavlink stream name is SERVO_OUTPUT_RAW_0. changing that to 200Hz gives us the RPM feedback at 200Hz on mavros

No flow control.... really just need GND, Rx, Tx

@astro_phil px4 version: 1.14.0-2.0.68

Thanks @Chad-Sweet and @Alex-Kushleyev for the suggestions. Indeed it was a faulty ESC. Swapped it out and all temps are look fairly similar. On idle (room temp 25deg), the ESC temp stabilizes at around 50 odd degrees C.

@Alex-Kushleyev When using RPM control, the mapping between thrust and RPM is usually an almost perfect quadratic, so you should probably use THR_MDL_FAC = 1 to make it quadratic. However, let me double check that last point.. The value of this parameter should not change when you change motor / propeller, since thrust is almost always a pure quadratic vs rpm. Hi @Alex-Kushleyev , did you manage to find out if THR_MDL_FAC should be set to 1 or should i calibrate the thrust curve? PX4's docs specified in https://docs.px4.io/v1.12/en/config_mc/pid_tuning_guide_multicopter_basic.html that if the ESC is a RPM-controlled one, no tuning is required: For PWM controllers, 0.3 is a good default (which may benefit from further tuning). For RPM-based controllers, use 1 (no further tuning is required as these have a quadratic thrust curve). But i noticed from ModalAI's reference params (eg Seeker's https://gitlab.com/voxl-public/flight-core-px4/px4-parameters/-/blob/master/platforms/v1.12/Seeker/Seeker_v1_param_rev_A.params) that the you guys did a calibration regardless # Dyno test shows this model should be 0.90 1 1 THR_MDL_FAC 0.90 9

Hello @DarkKnightCH , the mixer file does not only define the AUX1 and AUX2. Please note this line, which specify the configuration of a quadrotor in X orientation : https://github.com/modalai/px4-firmware/blob/1.11.3/ROMFS/px4fmu_common/mixers/quad_x.main.mix#L1 , specifically R: 4x

Hello stevet, The reported voltage is the voltage measured by the ESC. In fact all 4 ESCs on the 4-in-1 ESC measure the voltage, you can try to run the calibrate script on others and double check the result. The current of 2.9A for a motor similar to EMAX 1306B is quite reasonable, in fact is probably on the low side (not a concern). If you are using a power supply that is capable of outputting 7.4V at much higher amps (lets say 10A), then follow these steps to figure out what is wrong: when ESC reports 2.9A, check to make sure the power supply also reports the same or similar current. run the calibration on a different ESC id to make sure the voltage measurement is similar to your first test use a multi-meter to measure the voltage right at the power input pads of the ESC (may need to solder temporary wires to ESC and connect them to multi-meter in order to do it safely.. use your judgement) the multi-meter voltage measurement should match very closely the voltage measured (under load) in the ESC the issue is most likely in the wiring between the power supply and the ESC (including possibility of a bad connector or bad solder joint at the connector). Alex

Our PWM breakout board does not have the positive rail connected.

Thanks for your suggestions. I ran the pwm test command in the MavLink console and confirmed all motors except the suspect one so looks like I'm going to replace an ESC. I do appreciate your guidance.