Hello everyone,
after some years of absence from the portfolio of the esc VR, I present the new design.
We made choices very conservative to have 27 and 50 amp ESC from 4s to 6s ... we are doing tests and validations necessary because this is a very delicate component.
Here are the first photos and video of the first test.
The firmware supported are the Simonk and BHely, given the complexity of the topic we did not want to reinvent the wheel but to make a quality product that can be certified and has the ability to be installed on a dedicated distribution board.
Attached are photos and here is the video
original blog post : http://www.virtualrobotix.com/profiles/blogs/vr-esc-2-0-tech-preview
Comments
@All @Quadzimodo Thanks for suggestions about the quality improvment. Tomorrow we 'll discuss with Diego about some upgrade as rf shield and form factor and voltage.
@naish about the can bus i agree with you CAN could be a good choice , my idea is to implement a can subsystem for manage multiple esc at the same time with a double can bus. With this subsystem could be possible add some advanced feature as temperature monitoring , current per single esc , rotation speed and vibration iusse. So the complexity of basic function of esc could be the same , but is possible to add advanced functionality to motor board.
Ciao Roberto,
I will suggest you to move form PWM to something more professional if you are planning to create a top quality product. CAN bus is a really good candidate, and this days any new autopilot is supporting it.
I will also like to agree with other people about the LED's and the Removal to the BEC.
An other good safety feature will be to have an emergency stop of the propellers when they hit soothing, as the AR drone does. They implement that just by sensing the current load on the ESC.
1. Integrated BEC perhaps is unnecessary weight and heat generation in most multirotor applications, maybe you can ditch that from the design and save space and heat energy.
2. Voltage desired by DIY mini builders is heading north fast, maybe 4S might be too low for a 27A rated ESC. 6S would be better, but higher (like 8S+) in a similarly small package would be ground breaking (but initial demand for such specialized item would be pretty limited).
3. The cylindrical electrolytic capacitor is cumbersome and can complicates installation. Can this be replaced with Tantalums on the board instead? So to better maintain a rectilinear form factor? Sure it would have higher manufacturing cost, but the increase in price would be worth it.
4. Creating narrow double sided board of about 10-12mm seems to be the best approach. This obviously allows for installation into variety of tubes, but also suits surface mounting on most multirotor arm designs better. Most small 250-300mm size multirotors mount the ESCs on flat arms, so narrow design provides more streamlined design which presents less drag.
5. The RF shield featured on uBECs like the HobbyWing 3A or ESCs like the Sunrise 40A is extremely useful. It not only helps to minimise RF noise, it also makes it easier to mount to a flat surface and helps minimize risk of failure due to improper installation - generally caused by board deformation (over-tightened zip ties) or shorting of components against CF frame material. It has also help draw heat away from the devices and dump it out to the frame via the larger and flatter surface area, yet the added weight is very minimal. Also, in order to keep failures to a minimum I suggest providing customers with instructions on how to avoid problems caused by bad installation.
6. KISS 18A is certainly the standard at the moment. I agree that the solder pads are very small, but is is tolerable. It is perfect size and offers excellent performance, especially for the smaller acro setups where performance requirements are very high. Many people are eagerly waiting for the release of the new 30A 6S compatible version of the KISS ESCs, but it is apparently 35x21mm (wider than I'd like it too be). Maybe VR provide similar performance specification in a smaller better package?
There is also a 40a version which I expect to be equally good.
I hear good things about this one too:http://www.himodel.com/electric/Sunrise_BLHELI_Multi_Series_40A_2-6...
Regarding being tube: this is why they are rated to 40a while in practice loading is no more than 12-15a on each ESC. Also, the tube itself is a good heat conductor, so you can feel how warm your ESCs are just by touching the tube. I speak out of experience here. The problem with all ESCs on a center boad is that they all work in 400-600mhz envelope the same as our FCs grequencies as well as 433mhz telemetry/uhf frequency. By putting them inside the CF tube we effectivelly schielding most of their RF noise. Look at DJI s1000 they chose to put thei ESCs as far away from the FC as spossible and air flow has little to do with it, since they are stackng it right under the motor, so airflow there is minimal, however there is an additional source of heat -motor. Right above it.
Roberto, very interested in your update on that last subject, exciting!
Yes could be interesting ... we are working on a new telemetry board that support the RGB high intensity led and OSD . News in the next week :)
@Roberto,
these are green and red high intensity LED. I think DJI only use them to help getting the orientation in flight. But you could as well add extra intelligence for exemple by :
- powering ON these LEDs only when copter is armed
- blinking when the low voltage threshold is triggered by the ESC
- add a blinking code on startup that briefly shows the main configuration of the ESC instead of the usual BIP-BIP
- ...
Hi Johann ,
i think that Artem intend to use CF tube because it's conductive so you can put some conductive part up the FET that intercept the tube.
best
Roberto
-
1
-
2
of 2 Next