Hi,
I have built a VTOL tricopter with the 2 front motors tilting. I had some HS6975HB servos so used those (8.6kg at 4.8v). I thought 8.6 kg may be a bit marginal so did lots of bench stress tests. All seemed ok. Anyway, yesterday I had about 5 flights in hover to do the tuning. When trying another flight, one servo was dead with a slight burnt smell. On stripping, the power chip is cooked. Very strange it didnt seem to happen in flight.
I have now ordered 2 20kg servos and a 8.4v ubec.
Any thoughts on tilt servo size please?
Plane is 1.8mts and 3.5kg.
I am using this servo Amazon.com
25KG Robot Servo Motor High Torque Full Metal Gear Digital Steering Servos with Mount Brackets
Thanks, I looked at those. I would have been happier if the āidleā side had been a bearing. It looks like its just rotating on a plastic stud?
I used 20kg ones and have designed ball bearing support on the isle side.
I would look at brands like KST, MKS , Align/Savox, or Bluebird These are used by the Heli guys that puts lots of strain on a servo. That basic Hitec was working hard if it burnt the motor drivers.
1 Like
RDS3225SG 25kg Full Metal Gear Servo Suitable for Robotic Arm/Industry, Humanoid, Bipedal and Multipedal Robots, as well as PTZ Monitoring Equipment, Remote-Controlled Vehicles etc.
1 Like
Been looking for tell-tail signs in my previous logs of the tilt servo not behaving, like comparing des yaw to actual yaw. Nothing really obvious showing up. Any recommendations what parameters to check?
The general advice about tilt servo selecting seems to be āget the strongest one you canā. Not very scientific.
What can we do to get somewhere in the ball park?
I decided as its difficult to measure torque, then measuring current draw my be a reasonable indicator. This 20Kgcm servo, when given strong hand force drew about 3A. When tested with a spinning prop it was only about 0.6A. Why did it fail? Do servos have current limiters or do they just keep pushing until the drivers or motor burn out.
I dont think I need a stronger one, just a better one.
I have been using the cheap Emax metal gear digital micro servos (ES08MDII) on a ~2kg Mini Talon VTOL with no failures yet, Though, I have a feeling it may fly better with stronger servos.
If money is no object, look at RC car āracingā oriented servos eg SRT BHX6 HV Brushless servo - - BLS Servo - Servos - Product - SRT
Heewing t2 servo has been great for me. Its a 6-7.4v servo and has Motor mounts. Im using 3215 730kv with 11x8x2 props. Making around 6kg of thrust per motor.
In UK they are not sold separately, just part of a VTOL kit but I like the brushless motor. I think my problems could be with the servo motors.
You can buy them separately directly from heewing. If its not listed just email them for a invoice.
I also built a tricopter which is well and flying, about 4.2 kg takeoff weight. I do not think that the servoes consume a lot of power. I got S3315D 15Kg. If the geometry of the tilt is right, there is virtually no torque needed to keep the motor in position. There is the gyroscope effect when you continuously tilt the servo, like when doing yaw movements, but still, it is not that much. You can measure the current of this effect if you apply full power to the motor with prop, hold the plane, and move the servo motor using some knob on your radio, but i doubt it would be very strong.
Most probably you were unlucky with the servo (or very lucky it did not happened during the flight),ā¦
@Michail_Belov I think you are right that the torque of these tilt rotors is not as much as people think as my last vid showed with very low current draw. However, see this vid. I just got a servo tester in the post that can measure current. During a quick no load reversal test of test of the āgoodā remaining servo it just died. I think its speed that kills it. I will go for a good quality one next time, not just more powerful.
If another servo died without load, than really it is probably a bad batchā¦ The speed of PWM signal change really should not affect any servo.
More bench tests today.
I now have a Savox SB-2267SG installed in the motor unit and have just done some bench testing using a ToolkitRC ST8 servo tester. This device is supposed to show current but I looks like it shows a moving average of the max current (there is a setting to display average but it just shows zero all the time). I also connected a multimeter to confirm.
Findings -
There is a significant difference between the ST8 current and the multimeter. The ST8 shows around 2.5A (I made a mistake in vid and said 4A) while the meter shows 0.5a.
Interestingly there is no significant current increase when the motor is at full power swinging a 15" prop. Who would have thought?
The servo can take 8.4V but the ST8 resets at anything over 6v (see my other vid). In the aircraft its powered by 8.4.
The previous cheap RC truck servo was rated at 20kgm and felt extremely strong. The Savox is rated at 32kgm but stalls much sooner than the truck one. The Savox is probably producing its advertised torque and the truck servo had no protection limit but I didnāt get numbers on this.
The truck servo failed with burned mosfets (which were rated at 8A).
The truck servo was always very rough and jerky from new.
Im quite happy with the Savox servo, so far, and hope to have the plane flying again soon.
Just a random idea that popped in, but I wonder if looking at servos that are designed for helicopters might be worth exploring. I get it that they might not get big enough for some larger VTOLs, but they may be well suited for the speed and the high levels of feedback.
I used to fly medium sized helicopters and as I remember it was all about speed, not really torque. I dont think the first ones failed due to over-torque as all my tests seem to show that the torque required seems to be quite low and is not increased much by the rotating prop. So why did one fail in the plane (thankfully on the ground) and another fail when I gave it rapid unloaded cycles with that tester? Perhaps being connected directly to a vibration is a factor? Perhaps those Amazon truck servos are simply rubbish. What to expect for Ā£20?
I know many people use the Heewing T2 servo and I probably would have got those if they were readilly available in Uk, but anyway I like to do things myself.
@Vince_Hogg Glad to hear you found a savox that meets your needs. It would really be interesting to see someone test many of the servos on the market ( eps gray ones). I suspect the manufactures are close to the specs while gray maket ones are all over the place.
Those are both probably pretty fair comments IMHO.
Iāve just been playing with my T1 and when I look at the filter graphs I keep seeing a spike at 40-50hz. Iāve noticed this on other VTOL logs Iāve looked at. I wonder if this is the servos working during hover. With the load then I wonder if speed is a factor for these.
1 Like