Apologiies for the length of this post but I hope that my (numerous) questions are sufficiently interesting to attract responses...
I wish to use a VESC controller instead of one of the commercially available ESC products normally used in this field. The precise, demanding, application is F3a class competition aircraft for which this controller, for example, is commonly used. Why do I consider VESC?
As I understand it VESC has the following advantages not present in any sensible commercial model aircraft ESC:
- FOC = quieter, more pleasant sound in flight. Can influence the small style/smoothness/gracefulness element of judges' scores
- RPM mode - assists pilot to maintain constant flight speed, which is desirable for a number of reasons, by increasing power in up-lines and providing automatic braking in down-lines (normally only a pre-configured amount of braking is available and only on fully closing the throttle)
- RPM mode - eliminates effects of battery voltage reducing during flight (provided max RPM is not set higher than achievable on nearly exhausted batteries)
- Regenerative braking - makes possible the use of lower capacity (= lighter) battery which would help keep the aircraft under the prescribed 5Kg weight limit, especially as VESC controllers appear to be around 100gm heavier than typical model aircraft equivalents. I will not need BMS as in this application the energy replaced will always be much less than the energy drawn from the battery
- Are my above presumptions correct?
- I have seen advice in boarding forums not to use model aircraft ESCs as they are insufficiently robust, but used in their intended application model aircraft ESCs are very close to 100% reliable. The one I mention above, despite it's small size and light weight, boasts 48 output FETS and claims 99A continuous current capability, yet larger heavier VESCs claim around half that. Comment? Am I safe using, say, a 50A VESC in my application where max current will approach 90A for a few seconds 20 times during a 9 minute flight and average will be 25A to 30A? Battery is 10s lipo and approx 4Ah is consumed in approx 9 minutes.
- I have not yet decided which motor to use but am I correct in thinking that one of those commonly used, a 28 pole, will at just under 7000rpm exceed max eRPM of a VESC - or is it pole pairs which count in which case I should be ok?
- Sensorless or sensored? Required startup torque will be near zero - load being a 21" propeller weighing between 50 and 150gm plus motor inertia.
- Input will be 5v pwm 1mS - 2mS from model rc receiver. Best commercial ESCs have opto-isolated input to prevent interference from motor cables etc getting back to receiver. Should I add opto-isolation? Could I damage VESC by connecting pwm signal from receiver direct when VESC not powered up?
- What length battery cables can I use before I have to add extra capacitance mid-cable?
- What have I missed?
If you've got this far thank you, thank you, thank you. Appreciate your help...
1. Yes - although I doubt regen braking will give you any significant runtime increase.
2. Take amp ratings on ESCs with a grain of salt, they are almost always wrong.
3. Pole pairs x kv x voltage = erpm. So 7000rpm * 14 = 98000erpm. You would be fine with vesc6, vesc 4 only supports up to 60000 erpm before it tends to blow up.
4. Sensorless, sensored is really only required for starting large inertia objects. like eskate etc.
5. VESC has build in BEC so you need to cut the red (+5v) wire on the receiver connection.
6. Mid cable won't do anything much. Needs to be as close to the ESC as possible. I would add more for anything over about 1 meter.
7. A fuse would be a good idea and an anti spark/precharge circuit. The EMI caused by the spark can cause things to go bang.
Thanks Euan, very useful.
1. 10% would be good - would let me go from 4600mAh to 4200mAh
2. For the currents/durations I mention do you think I am in the right ball-park with a VESC rated at 50A continuous cooled by air ducted from a forward facing inlet (avg speed around 70mph)
3, 4, many thanks
5. I won't use the BEC - the receiver plus servos can draw much more than the 1.5A a VESC can supply and in any case an airframe worth a few thousand £ needs a supply which does not rely on an ESC remaining alive!!! Also, modern high-end RC receivers and servos use/need more than 5V, Still need to know - Could VESC pwm input be damaged if input occurs when VESC is not powered up?
6. I'm suprised - model aircraft ESC manufacturers dictate additional ESR capacitance for anything over 200mm
7. We routinely use anti-spark battery plugs which on insertion connect via a resistor initially then, when the plug is fully inserted, make the final full connection. I do not like to use a fuse for it's added resistance, complication and especially weight. I am in a "look after the grammes and the Kgs will take care of themselves" situation.
Further comment more than welcome! Anyone?
A VESC 6 would do the job, but please not that you want to run a 12S system if you can. The motor will not care about voltage so much.
I know that most hobby ESCs run from 6S. Sure you can run a VESC from 6S, but ideally you try to bring down the amps and up the voltage.
The VESC 6 could be run without housing, strapped to a smaller and lighter heat sink sitting in the air stream.
At 98K ERPM you reach the limit of the VESC 6, but it should be doable in FOC.
Battery will be the normal 10S lipo for this class (max allowed is 42.56v (don't ask) off load)
Uncased on a smaller heatsink is indeed my plan, I have accepted that will invalidate the warranty
I might gear for lower rpm to stay safe on eRPM. If commutation is lost due to high eRPM will that likely fry the ESC (or motor)?
No, losing commutation is no issue. You can set an erpm limit to prevent that from happening in flight. 10S is fine. We test the VESC prior to shipping. If you don't do anything stupid, the device will work hard for you for a long time.