recumpence
1 GW
It must be noted that the use of rc controllers is a misapplication of the technology when used in a bike.
That being said, they work fine if setup properly.
That being said, they work fine if setup properly.
Comparison between RC and ebike controllers?
- Thread starter dirkdiggler
- Start date
ARod1993
100 W
The big issue is that a lot of the credit card-sized RC controllers are typically designed to run near their rated amperages on model airplanes, which are quite different from EVs (and in most ways more forgiving). There are a few big differences, namely that:
-Model airplanes present viscous loads for the motor (that is to say, required motor torque is directly proportional to airspeed), so RC controllers typically need to put out very little current at standstill, and current slowly increases as the motor spins up. Most airplane controllers are sensorless; zero-velocity startups are often performed by bumping the motor phases in sequence until something catches, then spinning up to full RPM. This works fine on an airplane, because the current needed to start a propeller spinning at a couple of hundred eRPM is quite low and the motor catches almost immediately. For an inertial load (like an electric vehicle), attempting to "bump" the motor to lock in position is likely to draw a ton of current and possibly burn out the controller. Similarly, the controller algorithms typically don't function well at low RPM and high load, and so RC controller-powered vehicles tend to cog and rattle at low speeds.
-Similarly, model airplane controllers are designed to sit right behind the motor/prop assembly on the airplane, putting them directly in the path of the prop wash. Since model airplane torque is directly dependent on airspeed, the prop wash provides forced-air cooling directly proportional to the amount of current the controller has to pass; the controller is therefore only pushed to its maximum current rating when under maximum-speed forced-air cooling. By contrast, electric vehicles demand maximum torque at zero speed (and that's assuming your controller is mounted somewhere it'll get fresh air and not inside the body of a scooter-style EV). The high current demand on startup without commensurate cooling has a decent chance of grenading the controller on start. These controllers also don't seem to have the sort of current-limiting functionality that a Kelly or Lebowski controller has, and thus can't protect themselves from accidental overloads.
EV controllers, by contrast, are optimized to provide high torque on start (and shouldn't blow up if you make use of the throttle at low speeds), and aren't designed to require the same massive degree of forced-air cooling as model airplane controllers (so they shouldn't accidentally melt themselves after an hour or so of moderately hard riding).
-Model airplanes present viscous loads for the motor (that is to say, required motor torque is directly proportional to airspeed), so RC controllers typically need to put out very little current at standstill, and current slowly increases as the motor spins up. Most airplane controllers are sensorless; zero-velocity startups are often performed by bumping the motor phases in sequence until something catches, then spinning up to full RPM. This works fine on an airplane, because the current needed to start a propeller spinning at a couple of hundred eRPM is quite low and the motor catches almost immediately. For an inertial load (like an electric vehicle), attempting to "bump" the motor to lock in position is likely to draw a ton of current and possibly burn out the controller. Similarly, the controller algorithms typically don't function well at low RPM and high load, and so RC controller-powered vehicles tend to cog and rattle at low speeds.
-Similarly, model airplane controllers are designed to sit right behind the motor/prop assembly on the airplane, putting them directly in the path of the prop wash. Since model airplane torque is directly dependent on airspeed, the prop wash provides forced-air cooling directly proportional to the amount of current the controller has to pass; the controller is therefore only pushed to its maximum current rating when under maximum-speed forced-air cooling. By contrast, electric vehicles demand maximum torque at zero speed (and that's assuming your controller is mounted somewhere it'll get fresh air and not inside the body of a scooter-style EV). The high current demand on startup without commensurate cooling has a decent chance of grenading the controller on start. These controllers also don't seem to have the sort of current-limiting functionality that a Kelly or Lebowski controller has, and thus can't protect themselves from accidental overloads.
EV controllers, by contrast, are optimized to provide high torque on start (and shouldn't blow up if you make use of the throttle at low speeds), and aren't designed to require the same massive degree of forced-air cooling as model airplane controllers (so they shouldn't accidentally melt themselves after an hour or so of moderately hard riding).
recumpence
1 GW
It must also be noted that large ebike specific controllers tend to blow up too.
It bothers me that every time an RC controller blows, we hear about it endlessly. But when a standard ebike controller blows, that is somehow acceptable.
There definitely seems to be a double standard.
Matt
It bothers me that every time an RC controller blows, we hear about it endlessly. But when a standard ebike controller blows, that is somehow acceptable.
There definitely seems to be a double standard.
Matt
gwhy!1
100 kW
recumpence said:
E-bike controllers only tend to blow due to customising them ( hardware or software) to push them to well beyond what they were designed to do and some people that tweek don't really have a full understanding of what may or may not happen if settings ( hardware or software ) don't match the motor that is being used, it is very,very rare for a standard e-bike controller to blow if it have not been tweaked in any way. People don't tend to moan about el-cheapo ( $20 -$50 ) rc controllers that blow up ( similar prices to standard e-bike controllers in price ) but will moan if a $150+ rc controller blows up for no apparent reason.
recumpence
1 GW
I do not agree.
I sell quite a few systems to people who are sick of blowing large ebike controllers, including Ecospeed and other relatively high end stuff.
I do agree that it is tough to take when a $150 to $300 controller blows. But, that is not nearly so common these days.
I have said it before and I will say it again, it is all about proper equipment application and setup. That holds true for e bike controllers and rc controllers.
Matt
I sell quite a few systems to people who are sick of blowing large ebike controllers, including Ecospeed and other relatively high end stuff.
I do agree that it is tough to take when a $150 to $300 controller blows. But, that is not nearly so common these days.
I have said it before and I will say it again, it is all about proper equipment application and setup. That holds true for e bike controllers and rc controllers.
Matt
gwhy!1
100 kW
I know rc controllers can be made to work for ev's but they are not ev controllers, if they were then they would be sold as ev controllers and not as RC model controllers. I stand by what I have said , having had experience with both rc controllers and ev controllers . Im not doubting that the higher end rc controllers are more robust and less likely to go pop as long as the correct steps are followed, but ev controllers are designed to keep the much higher initial currents and heat under control.. thats not to say that ev controller will never go pop but it is less likely, if a higher end ev controller did pop for no apparent reason I'm sure a user will prob not buy another of the same type . There are many people on this forum running cheap modified ev controllers without any dramas and far less running expensive high end rc controllers .
recumpence
1 GW
The amazing thing, from my perspective, is that a 3 inch long device designed for a completely different application can compete head on with a 8 or 10 inch long box that was purpose built for its application.
And that is my point, an bike controller was designed for exactly that--- driving an ebike. But, these little RC controllers that were never intended to move 200 to 300 pounds can (again, if properly setup) put out many thousands of watts reliably in this application.
I will say, however, that RC systems are like running a high revving turbo charged 4 cylinder engine, you can get HUGE power per pound, but you run the risk of problems IF it is not setup properly. However, just like with the turbo charged engine, if it is setup properly, you can compete with huge and heavy normally aspirated big block engines reliably.
For that reason, I do not like the RC versus ebike controller debate. It is an apples and oranges thing. This is kind of like comparing a high revving two stroke engine with a tuned pipe to a thumping four stroke. For certain applications, the bike controller is best, and for other applications, the RC controller may be better suited. They are small, very efficient, generate very little heat, have data logging (the good controllers), and are easy to program. But, they do have their limitations especially if not properly setup.
Matt
And that is my point, an bike controller was designed for exactly that--- driving an ebike. But, these little RC controllers that were never intended to move 200 to 300 pounds can (again, if properly setup) put out many thousands of watts reliably in this application.
I will say, however, that RC systems are like running a high revving turbo charged 4 cylinder engine, you can get HUGE power per pound, but you run the risk of problems IF it is not setup properly. However, just like with the turbo charged engine, if it is setup properly, you can compete with huge and heavy normally aspirated big block engines reliably.
For that reason, I do not like the RC versus ebike controller debate. It is an apples and oranges thing. This is kind of like comparing a high revving two stroke engine with a tuned pipe to a thumping four stroke. For certain applications, the bike controller is best, and for other applications, the RC controller may be better suited. They are small, very efficient, generate very little heat, have data logging (the good controllers), and are easy to program. But, they do have their limitations especially if not properly setup.
Matt
windtrader
1 kW
Adding a comment to move this forward rather than push and pull.
I'm a noobie who has done a LOT of research into what I need for my unique terrain and usage pattern.
Many of these posts infer but don't state as Matt has clearly done - DC based drive systems used in ebike are different enough to be considered distinct once you peel back the cover.
"Double standard" - In a sense but more like one common set of standards and.measures but applied to two distinct design and use profiles (use cases).
Matt states numerous times RC technology can work with knowledge how to select, integrate, and tune the components to deliver optimal and reliable power. Off the shelf ebike components have a greater chance of working together with less expertise as they are designed for ebike use whereas RC components are designed for a very different kind of application.
That said I think it still quite useful to publish, measure, and evaluate both types of systems using the same standard measures, just as a turbocharged 4 and Prius both report HP and torque.
I'm a noobie who has done a LOT of research into what I need for my unique terrain and usage pattern.
Many of these posts infer but don't state as Matt has clearly done - DC based drive systems used in ebike are different enough to be considered distinct once you peel back the cover.
"Double standard" - In a sense but more like one common set of standards and.measures but applied to two distinct design and use profiles (use cases).
Matt states numerous times RC technology can work with knowledge how to select, integrate, and tune the components to deliver optimal and reliable power. Off the shelf ebike components have a greater chance of working together with less expertise as they are designed for ebike use whereas RC components are designed for a very different kind of application.
That said I think it still quite useful to publish, measure, and evaluate both types of systems using the same standard measures, just as a turbocharged 4 and Prius both report HP and torque.
gwhy!1
100 kW
I do agree that its is unfair to compare one against the other as both have different running characteristic and precautions need to be taken with both when trying to get the most out of them. But to be fair there is no way that a $300 rc sensorless controller can compete heads on with a well modded $100 ev sensored controller ( except size ), or to put it another way if i was to fit a $300 high end rc controller on one of my bikes it will pop within the first couple of twists of the throttle unless I compromised the performance or control of my bikes. Same way as you can not replace a sensorless rc controller with a sensored ev controller due to motor constraints But it will just not work and it will not pop the ev controller, there are many running 10kw+ reliably using cheapish ev controllers but not so many running 10kw+ using high end rc controllers .
windtrader
1 kW
If you buy the argument these two systems are distinct, then i ask you to take one step forward and agree to allow comparing rather than competing the two different systems. Just as you can compare the top speed of a Prius and some high performance car , they do not compete as they are different types of cars designed for different uses.
recumpence
1 GW
I pull 10 kw from my rc controllers every day. I have not popped a controller in thousands of miles.
Again (and again, and again, and again) it is all about application.
Gwhy,
I would say if you pop a controller within two twists, you are doing something VERY wrong. Or you are merely talking about start up torque from sensored systems. That is where the ebike controllers shine. Where they absolutely stink is in high rpm efficiency if programmed for good low rpm torque.
There is a reason Outrider uses rc equipment and so do I.
This entire conversation is why I was hesitant to get involved with this discussion. I am fed up with people spewing about component limitations when I am proving them absolutely wrong every time I ride any of my bikes. It is aggravating when people tell me something will not work while I am in the middle of using that very product in exactly the application they insist it will not work in........
Amazing........
Matt
Again (and again, and again, and again) it is all about application.
Gwhy,
I would say if you pop a controller within two twists, you are doing something VERY wrong. Or you are merely talking about start up torque from sensored systems. That is where the ebike controllers shine. Where they absolutely stink is in high rpm efficiency if programmed for good low rpm torque.
There is a reason Outrider uses rc equipment and so do I.
This entire conversation is why I was hesitant to get involved with this discussion. I am fed up with people spewing about component limitations when I am proving them absolutely wrong every time I ride any of my bikes. It is aggravating when people tell me something will not work while I am in the middle of using that very product in exactly the application they insist it will not work in........
Amazing........
Matt
gwhy!1
100 kW
windtrader said:
a programable ev contoller can be setup however you would like them to perform ( as long as the software allows it) and rc contollers dont ( mainly because they do not have current limiting ) so there really is nothing to compare as additional harware needs to be introduced on a bike that uses a rc controller to taylor performance and reliability. So for this reason rc controllers can not be compared with ev contoller as regards performance .
The only things that can be compared is cost of owership, weight, size , reliabilty and any additional hardware needed.
Matt I not saying that rc controllers dont work just that more things are needed to make them useable for ev use What would I have to do to make a rc controller survive pulling 10kw without loosing any functionality on my bikes ?. The only requirments for my bikes is I want them to start from a standstill and have a very controllable throttle responce From zero to wot, this is something that sensorless rc controllers are unable to do.
recumpence
1 GW
Low speed throttle response is achieved through programming a very slight delay in the controller. This smooths the response. Starting from a stop is achieved by running a high kv motor with a deep reduction.
Matt
Matt
Hillhater
100 TW
You are both missing the point of the OP....( or rather you are arguing a different issue !).
The OP asked why don't bikes use RC controllers with hub motors and mid drives (bafang and similar) .
Does anybody successfully use a RC controller on a hub motor or Bafang mid drive ?
The OP asked why don't bikes use RC controllers with hub motors and mid drives (bafang and similar) .
Does anybody successfully use a RC controller on a hub motor or Bafang mid drive ?
Dauntless
100 TW
spinningmagnets said:
Hey thanks, I went looking when you mentioned it but without just the right search term it doesn't work.
gwhy!1
100 kW
Hillhater said:
No, Not missing the point of the OP , but giving my opinions ( and facts ) as to why RC controllers are not commonly used on e-bikes with hubs or mid drives.
Addition hardware needs to be incorporated and this can be electronic or mechanical ( this is a must ) if a RC controller is to be made useable and reliable. I think the only place that a rc controller can be used with out much thinking about would be friction drives as the whole nature of a friction drive will act as a slipper clutch.
Gearing down a fast small rc motor mid drive to much introduces its own set of problems with control but it do make it easier for a sensorless rc controller to get going from a stop ( but still not 100% of the time ).
yawstick
100 W
I'm still pretty much a noobie compared to some of you guys but I've had pretty good luck using RC controller on my trike which is RC motor with reduction drive. I'm using about 100.00 controller for 12s. The 6s are much cheaper but not meant for the larger motors. I'm surprised that no one has had a go at customising the software for them. On the copter applications custom software has been developed to improve the response time to changes in throttle request... to help them respond quicker to changes in attitude.
On the softest soft start option I find them pretty good at slow start but the setting also adds a huge delay in throttle response at higher throttle setting. It would be nice to have the best of both worlds.... softest start but changes as RPM increases.
Modification of the speed requested allows for current limiting and also stretching out acceleration time and greatly improves rideability.
I was just looking at some of the new RC ESC's on Hobby King. Some of them are starting to claim to have some current limiting settings and the prices are still coming down.
On the softest soft start option I find them pretty good at slow start but the setting also adds a huge delay in throttle response at higher throttle setting. It would be nice to have the best of both worlds.... softest start but changes as RPM increases.
Modification of the speed requested allows for current limiting and also stretching out acceleration time and greatly improves rideability.
I was just looking at some of the new RC ESC's on Hobby King. Some of them are starting to claim to have some current limiting settings and the prices are still coming down.
Dauntless
100 TW
gwhy! said:
You have to let the thread spread its' wings to deal with the issue. Trying to keep too tight a laser focus misses much, because the application can never have such a tight focus.
If you were going to get Alien or whomever to adapt the basic RC format for a small ebike controller to run an RC motor, what would you have them change?
gwhy!1
100 kW
propper current limiting would be a big bonus that can be adjusted, battery ( and pseudo phase limiting at the very least ) then it will be just like a sensorless e-bike controller in a rc package, this would be a massive improvement as a rc sensorless controller can spin rc motors much faster than sensorless e-bike controllers..
second request would be to add sensor inputs ( with optional sensorless or sensored operation )
second request would be to add sensor inputs ( with optional sensorless or sensored operation )
windtrader
1 kW
What would really help the issues folks have trying to match RC components for an e-bike application is if the RC controller folks believed there was a big enough market to invest in creating an RC controller that is programmed for this use. Recumpence has found this niche and I certainly honor his investment of time and energy and wish him continued success.
Maybe the controller firmware can be hacked and reprogrammed. Maybe that is what Recupence has already done.
Whichever way it happens, it would be great if there were community mods to these controllers to better adapt the the power demand profiles of ebikes.
If such options exist, then I challenge those to post here or point to links where this is being done to enlighten more of us.
Thanks
Don
Maybe the controller firmware can be hacked and reprogrammed. Maybe that is what Recupence has already done.
Whichever way it happens, it would be great if there were community mods to these controllers to better adapt the the power demand profiles of ebikes.If such options exist, then I challenge those to post here or point to links where this is being done to enlighten more of us.
Thanks
Don
bobc
10 kW
You'll never get a start from standstill of any decency from a sensorless controller. For me, a turn or two on the pedals is small price to pay for the simplicity of a sensorless install. My bike has pedals & I'm not afraid to use 'em! I say again, it is straightforward to make an RC controller tractable and robust by making the pulse width on the control line match motor kV x bike speed/gearing with an extra bit proportional to throttle position. When you do this your rubbish open loop RC ESC becomes a torque controlled nice thing! Honestly, it really is that easy with a wee micro board like the one I'm working on. Here's a picture of it, it uses the atmel tiny10 micro & will include battery current limiting to make it legal when I get hold of some proper shunts. The speed is measured by a reed switch/spoke magnet like on a bike speedo. I'm using a "mystery" 60A ESC because it has a BEC in & powers the whole shebang.
I'll take a better pic when I get it fully operational - shunts are on order so it shouldn't be too long
I'll take a better pic when I get it fully operational - shunts are on order so it shouldn't be too long
Hillhater
100 TW
Bobc,.. what motor are you using on the bike with that controller ?
yawstick
100 W
I've been on the alien power site and saw the controllers with sensor input but seen very little of them elsewhere or heard of anyone using them. I think most of the copter controllers with modified software were actually atmel (arduino) processors. I looked under the hood on my 100 amp HV controllers and they have the C8051F310. There are free compilers for the atmel stuff.... not so sure about the 8051. Well I did just go to the Silabs site and they appear to have a free software package for this controller. I did look at the data sheet a little... it's a 3.3 volt processor... runs up to 25 mhz.
the MCU part of this esc has very few connections to the board with the drivers on it.
the MCU part of this esc has very few connections to the board with the drivers on it.
dirkdiggler
10 kW
- Joined
- Oct 17, 2012
- Messages
- 641
Many, many thanks to those that have offered their advice and knowledge to this thread. I just got back from Memorial weekend and have been reading through the comments again and again. Thanks also for the link to the old RC vs ebike controller thread. Nice to see familiar names from back then. ES is wealth of knowledge.
Yes, I didn't want an argument of which is better, or what people prefer, nor did I want to drag someone into an argument. I would only like to understand why the RC ESCs are not being used for bikes and if not why are they not being manufactured like Alien is trying to do. More so what needs to be modified for them to better suit the needs of a bike. There isn't one single example of an eboard that uses an ebike controller, but hundreds with RC ESCs. There have been very few issues with the ESC part. I know some use well over 1000w on their boards and don't have any issues with cogging or spontaneous combustion. My BBS mid drive only pulls a bit over that on my rides - I think when I was actually measuring 1200w was the highest I'd seen on a Turnigy wattmeter. A skateboard likely doesn't have the rapid acceleration as a bike does, there is no handle bar to hang onto, and maybe that is why they survive. The most common ESC is rated for 6s at 150a. A small survey of riders shows that most never get beyond that 50 amp mark. My contention to the eboarders is that the ESC will survive better on a board than an RC car because we are not going from full stop to full acceleration. This is why we don't see amp spikes, as again its hard to drag race a skateboard. Eboards only really need a 60a or so controller that can take 6s.
A few of the posters that come from RC backgrounds have mentioned that many quad copter ESC's have a modded firmware to change parameters to suit their needs. SimonK firmware and such. They have suggested changing the programming to better suit our needs. As of yet I haven't heard of anyone modding them. This is what I believe Alien is doing with the Flier ESCs.
I do know that Hobbyking does currently sell sensored ESCs. Only as high as a 6s 150a version. They have sensored motors, but at much higher KV than would be useful. Most over 1900kv. I've not ventured into installing halls sensors on RC motors, hopefully we will see some larger motors manufactured with sensors soon.
http://hobbyking.com/hobbyking/stor...nsored_Brushless_Car_ESC_PC_Programmable.html
I may just be the guinea pig and purchase this to see how it turns my hub motor. Again, though stuck on the 24v issue.
In the thread from 2011 it was mentioned that the ebike controller is produced in China without good access to the schematics and Lyen and others are just modifying them to increase amps and doing minor mods. ie increasing FET, shunts and such. It seems that there is a lack of innovation in the design. I personally see many advantages in the RC ESC. Well see if anything changes. Hopefully someone like Bobc can change things and combine the good of the RC and the good of the ebike.
Yes, I didn't want an argument of which is better, or what people prefer, nor did I want to drag someone into an argument. I would only like to understand why the RC ESCs are not being used for bikes and if not why are they not being manufactured like Alien is trying to do. More so what needs to be modified for them to better suit the needs of a bike. There isn't one single example of an eboard that uses an ebike controller, but hundreds with RC ESCs. There have been very few issues with the ESC part. I know some use well over 1000w on their boards and don't have any issues with cogging or spontaneous combustion. My BBS mid drive only pulls a bit over that on my rides - I think when I was actually measuring 1200w was the highest I'd seen on a Turnigy wattmeter. A skateboard likely doesn't have the rapid acceleration as a bike does, there is no handle bar to hang onto, and maybe that is why they survive. The most common ESC is rated for 6s at 150a. A small survey of riders shows that most never get beyond that 50 amp mark. My contention to the eboarders is that the ESC will survive better on a board than an RC car because we are not going from full stop to full acceleration. This is why we don't see amp spikes, as again its hard to drag race a skateboard. Eboards only really need a 60a or so controller that can take 6s.
A few of the posters that come from RC backgrounds have mentioned that many quad copter ESC's have a modded firmware to change parameters to suit their needs. SimonK firmware and such. They have suggested changing the programming to better suit our needs. As of yet I haven't heard of anyone modding them. This is what I believe Alien is doing with the Flier ESCs.
I do know that Hobbyking does currently sell sensored ESCs. Only as high as a 6s 150a version. They have sensored motors, but at much higher KV than would be useful. Most over 1900kv. I've not ventured into installing halls sensors on RC motors, hopefully we will see some larger motors manufactured with sensors soon.
http://hobbyking.com/hobbyking/stor...nsored_Brushless_Car_ESC_PC_Programmable.html
I may just be the guinea pig and purchase this to see how it turns my hub motor. Again, though stuck on the 24v issue.
In the thread from 2011 it was mentioned that the ebike controller is produced in China without good access to the schematics and Lyen and others are just modifying them to increase amps and doing minor mods. ie increasing FET, shunts and such. It seems that there is a lack of innovation in the design. I personally see many advantages in the RC ESC. Well see if anything changes. Hopefully someone like Bobc can change things and combine the good of the RC and the good of the ebike.
Hillhater
100 TW
DD,. A few comments..
Alien sell several large (80100 etc) RC out runners with low kV (80) and sensors fitted.
Many of the RC suppliers sell high voltage , 12 - 14 s RC controllers also. Try Alien , Castle Creations , etc
You won't be the first to try that RC car sensored controller on an Ebike, but
carry on, self learning is always the most convincing.
If you keep searching and reading this forum, you will find many threads on Ebikes with RC motors, many of them with sorry tales of burned RC controllers also.
Look up the threads from Luke (LFP) who has run many tests on RC controllers .....
Alien sell several large (80100 etc) RC out runners with low kV (80) and sensors fitted.
Many of the RC suppliers sell high voltage , 12 - 14 s RC controllers also. Try Alien , Castle Creations , etc
You won't be the first to try that RC car sensored controller on an Ebike, but
carry on, self learning is always the most convincing.If you keep searching and reading this forum, you will find many threads on Ebikes with RC motors, many of them with sorry tales of burned RC controllers also.
Look up the threads from Luke (LFP) who has run many tests on RC controllers .....
