72v 1500W 40A Hua Tong CA08 XM06AD_P04.1 controller

def not cutting off due to lvc (67V or higher when controller was acting up), its clearly sagging more at wot (65V) but no choppy behavior there.
i would like to see 24s on an rc motor though :mrgreen:

mauimart, any updates on the choppiness? also what happens if you use wot right from the start?
 
cohberg said:
def not cutting off due to lvc (67V or higher when controller was acting up), its clearly sagging more at wot (65V) but no choppy behavior there.
i would like to see 24s on an rc motor though :mrgreen:

mauimart, any updates on the choppiness? also what happens if you use wot right from the start?

As far as I can tell the choppiness is not a result of LVC, throttle input, or hall sensors. I think it's simply a mediocre algorithm implementation in the control section. I modded my controller over the weekend by decreasing the resistance of the current shunt. It seems to have smoothed out the jerkyness a bit but at the same time it has accentuated the power surge at roughly 25% throttle input. Low rpm starts are smooth but once you hit that 25% threshold the controller commands a rather violent surge in power that will throw you off the back of the bike if you are not ready for it. I did do a run where the controller peaked at about 3000W input power and it seems to handle 2000W continuous without much heat output. Still not bad for the price if it could only be tamed :?

[youtube]8f5BbskodCk[/youtube]
 
mauimart said:
I modded my controller over the weekend by decreasing the resistance of the current shunt. It seems to have smoothed out the jerkyness a bit but at the same time it has accentuated the power surge at roughly 25% throttle input. Low rpm starts are smooth but once you hit that 25% threshold the controller commands a rather violent surge in power that will throw you off the back of the bike if you are not ready for it. I did do a run where the controller peaked at about 3000W input power and it seems to handle 2000W continuous without much heat output. Still not bad for the price if it could only be tamed :?

[youtube]8f5BbskodCk[/youtube]

yeah thats sounds similar to what was happening to me. it gets a lot worse at 5-6kw. to smooth out the throttle response try a variable potentiometer between the throttle output and controller or since you have a CA use current throttle.
 
The Mighty Volt said:
Is this just because we've got a change over from the "normal" hub to the outrunner with the added sensors??

To get the thread back on topic. lol.

Well yeah. mauimart is running his outrunner at much higher rpms than the standard hubs. I still think that its a hall sensor issue: the outrunner gains and looses revs so quickly + the diy halls setup (not inferring that they're poorly done just that they're not glued [thus optimal spacing and field strength calcs were purportedly done] onto the stator like the rest of us lol) that the controller is having trouble locking on for a couple revolutions during acc., then when the motor slows down it gets lock again (thus little to no problems during no load tests). Also, the controller algorithm is shitty to start with: the high rpms probably just exacerbated the problem. And Gordo's inductance comment might have merit despite the rewind: its still much higher than a hub motor.

I think with a more logarithmic throttle this could work really well! Mauimart, do try the inline throttle resistor or CA current throttle!
 
Cheers Cohberg- one of MauiMarts excellent youtube videos shows him running the motor and losing synchronization at the very very high RPM.

Here it is

[youtube]OL4XvbuKoYk[/youtube]


Maybe another re-wind down again might help??
 
The Mighty Volt said:
Cheers Cohberg- one of MauiMarts excellent youtube videos shows him running the motor and losing synchronization at the very very high RPM.

Here it is

Maybe another re-wind down again might help??


Well i don't think that with his current gearing he is hitting those rpms. Sensored and sensorless controller alike have trouble with these outrunners at high rpm. Thus lfp's controller is something like ~$1.2k, i think, and takes 3 pros to tune it optimally lol.

But sure a slower wind and different gearing might help with the lock issue, if it is a lock issue that is =P
 
What is the inductance of this outrunner? If we can see the HALL's firing on a scope at slow speed, we can make a WAG as to the accuracy of their placement. Then, as we increase rpm, if we continue to watch the HALL's on a scope, we should be able to see the trace if it gets lost? Then we can make an informed comment on what is happening.
 
Just to clear up any confusion regarding my setups with the Hua Tong...
I have two Turnigy 80-100 outrunners:

One is a stock 130kV (delta wind I think). This one does not have internal halls installed but rather external halls at 17.4 degrees, or at least as close to 17.4 as I could do. I have powered this motor to roughly 7700rpm unloaded before the controller lost sync. Theoretically I should be able to hit 9100 rpm (70V x 130kV). I also loaded this setup with a 30 inch prop with smooth operation.

My other motor is the rewound Turnigy (9T wye). It contains internal hall sensors glued in between stator teeth at 120 deg. It's currently installed on my bike. Under no load (bike tire off the ground) acceleration is smooth over the entire rpm range. It's only while under load that it exhibits the jerky behavior and surges. I can not definitively rule out a hall sensor issues but because of the smoother performance I get with the Kelly controller (with the same hall sensors) I'm inclined to think it has something to do with the Hua Tong. If only I could find an extension cord long enough for my scope so that I can probe the hall sensors while riding the bike under load... :D

I plan to do some more testing with this controller and the outrunners in the future but for now I'm content with the Kelly. I may look into an optical solution for the sensors (Burtie's setup) and see how that goes. I have a feeling that once I get my larger battery pack and start pushing the Tunigy, the heat generation might have its way with the internal halls. We shall see.
 
mauimart said:
If only I could find an extension cord long enough for my scope so that I can probe the hall sensors while riding the bike under load... :D

That is where having Arlo1 with his dyno, just down the road comes in handy. :mrgreen:
 
Gordo said:
mauimart said:
If only I could find an extension cord long enough for my scope so that I can probe the hall sensors while riding the bike under load... :D

That is where having Arlo1 with his dyno, just down the road comes in handy. :mrgreen:

Nooooo, I want to see the video of him duct taping a big ole Tektronix to his bike and go WOT until the cord bungies him back... :twisted:
 
pelle242 said:
Gordo wrote:
mauimart wrote:
If only I could find an extension cord long enough for my scope so that I can probe the hall sensors while riding the bike under load...


That is where having Arlo1 with his dyno, just down the road comes in handy.


Nooooo, I want to see the video of him duct taping a big ole Tektronix to his bike and go WOT until the cord bungies him back...

lol.

Something more practical would be to use one of those bicycle trainer (modulating the load to simulate real life scenarios) to confirm good hall sensor resolution and that the hua tong's control algorithm is just lousy.
 
P1050928.JPG
cohberg said:
Something more practical would be to use one of those bicycle trainer (modulating the load to simulate real life scenarios) to confirm good hall sensor resolution and that the hua tong's control algorithm is just lousy.

You mean something like this?
 

Attachments

  • P1050923.JPG
    P1050923.JPG
    146 KB · Views: 2,624
mauimart said:
cohberg said:
Something more practical would be to use one of those bicycle trainer (modulating the load to simulate real life scenarios) to confirm good hall sensor resolution and that the hua tong's control algorithm is just lousy.

You mean something like this?

Yup, make sure to modulate the load as rough as you can get it (so like while at mid throttle: hit the back brakes, release the load, engage the load, etc.) The kelly controller probably samples the halls a lot faster than the hua tong: so to replicate the problem, you should simulate uneven uphill terrain.
 
Has anyone actually made this controller work in sensorless mode?

There are all types of statements, like this one:
http://www.endless-sphere.com/forums/viewtopic.php?f=9&t=35848&p=520563&hilit=sensorless+hua+tong#p521310

But I haven't found a thread where someone actually made it work in sensorless.
 
Skippic said:
Has anyone actually made this controller work in sensorless mode?

There are all types of statements, like this one:
http://www.endless-sphere.com/forums/viewtopic.php?f=9&t=35848&p=520563&hilit=sensorless+hua+tong#p521310

But I haven't found a thread where someone actually made it work in sensorless.

I find Scotty's post confusing. I am guessing the controller he sold was capable of sensorless. I see no one claiming ANY of the Hua Tong's are sensorless.
 
The 72v 1500W 40A Hua Tong CA08 XM06AD_P04.1 controller is NOT a sensorless controller. Unfortunately some people are posting about other controller models in this thread.
 
In my case it was around 25mph on 22S.
 
Does anyone know if it is possible to modify this 72 volt controller to run on a 64 volt LiFePO ( 20s) battery or is there a controller that is a similiar price with the same attributes?
 
So, according to the ebikes.ca simulator, with your 9c 2807 at 22s, the "low speed" of 25mph correspond to 50% throttle. Interesting...
 
I've bought one and I'm very satisfied. Running at 24s no problems yet.

What I would like to change on it is the way the throttle controls it. It's like the throttle sets a speed you want to go and the controller does everything it can to go that speed.

For example if you go 50% throttle, it tries to go 25mph. If you are at 26mph it turns the motor off completely, if you are moving at 22mph it goes to full power. So say you are going 25mph constant on flat, but uneven surface. The vibrations you get from little bumps on the road jiggle your throttle a little, enough to set the desired speed +/-1mph and the motor is going from no thrust to 100% with every bump.

Therefore I would like to find on the controller what sets the PWM and control that directly: 0-10% throttle = PWM 0%, after that map 10-100% throttle to 0-100% PWM.

The million dollar question:
Where is the PWM controlling on the board?
 
Back
Top