ECITYPOWER 9-FET 350W Controller w/cruise control

[Russell]

Ecitypower.com and their sister site bmsbattery.com sell a few controllers which sound feature rich (http://cgi.ebay.com/ebaymotors/350W...ptZMotorsQ5fCarQ5fTruckQ5fPartsQ5fAccessories) so I figured what the heck and ordered one from E-city's eBay auction. The price for this 9-FET 36V 350W unit was $20+$24 shipping less $1.60 from Bing.com = $42.40. What interested me most about these controllers was the ability to run with hall sensors or without, an auto cruise function which kicks in after 8 seconds, the PAS connector that I wanted to try and of course the low price. I bought it on Sept. 28 and it arrived yesterday, Oct. 17.


The box got beat up a bit and both flanges were bent a bit.


View attachment 5


It's slightly longer than a E-BikeKit controller.





The description on eBay said it was a 14A unit but the rating plate says 22A.





It's sealed with some goop where the wires exit and at the side screws.






Lots of wires but no wiring diagram or instructions were included.



From another post for an E-City controller I thought I had it figured out...for the most part anyway :wink: . From left to right you got your phase power wires, a speed signal (purple), battery power connector, PAS 3-wire connector, 5-wire hall sensor connector, a continuous grey wire loop for which I don't have a clue (EDIT: cruise control) , A standard 3-wire throttle connector, a single brake wire I think but how to use it...?(EDIT: active HIGH) , and finally two white wires which I thought were speed suppressor wires (EDIT: programming wires). The phase wires and power wires were all soldered, a good thing, but I had to cut them off to put on Andersons to match my present configuration.

I mounted the controller on the handlebars of my Raleigh bike in place of a small 36V/15A controller. The bike has a 24V GoldenMotor Mini Motor that I've been running on 36V. I first tried to get the motor to run with just the 3 phase power wires connected but it was a no-go. In two configurations the motor sounded like it was running smooth but the wheel wasn't moving so I assume it was turning backwards, the other 4 combinations produced a hammering affect. I figured it wasn't gonna work so I connected up the hall sensor cable which didn't work properly at first. I ended up swapping the green and yellow wires on both the phase wires and in the hall connector to get it running smoothly. After it was running I did find I could yank the hall cable and it would continue to run, kinda cool I guess, but after cycling power it hammered again.

OK so I figured the thing wouldn't work sensorless , not a huge deal, but then things got a bit confusing. I connected the two white wires together to check what I thought was a speed suppressor function and the wheel started spinning, then it stopped tuning but the motor sounded like it was spinning backwards. The best I can figure is the white wires put it into some sort of program mode where the controller tries different combinations. Instructions would have been really helpful at this point but basically what i did was cycle power and hook up the white wires and when it spun forward slowly but smoothly I disconnected the white wires. After this I found I could run sensorless at all times however it takes a lighter touch on the throttle from a dead stop to prevent the hammering effect. So it does appear that this controller works both with and without hall sensors but it does work better with the sensors. With the hall sensor cable connector I could yank the throttle as fast as I wanted and the motor reacted smoothly.

Now for the COOL part, the auto engaging cruise control. Yep it works as advertised; hold the throttle steady for 8 seconds then let go and the thing just keeps on goin', blip the throttle again and it disengages. 8) Tomorrow I hope to give it a good test out on the road, stay tuned!

-R

 

[floatingdog]

Russell,

Thanks for the feedback. You did a remarkable job of getting it to work! It is really quite amazing that you worked through the wiring and programming issues without instructions.

I ordered a BPM Bafang from BMSBattery but held off on the controller pending your feedback. Perhaps you've read some of the recent feedback on the eCrazyman sensorless controller and the sensorless 12-mosfet Big Box from http://www.lsdzs.com/. Both reported some hammering, but seemed to indicate it wasn't much of an issue. What are your thoughts on this controller? Is the hammering really bad or something that would be mitigated once you got the right touch on the throttle?

Thanks for all your posts they have been very helpful in deciding on the components to select for my first build. I just want some assist and not a rocket. I don't have the speed lust I did in my younger years. You seem to prefer your lower powered controllers. I've ordered the BPM which is rated at 350W. Do you think I should stick with a 350W controller or should go up a notch to a 500W? I do have a CA which, as I understand it, could be used to reign in my controller if I got one that is a little too powerful.

Thanks,

John

 

[Russell]

I went out for my first ride with the new controller this afternoon with the hall sensors disconnected. The hammering as the controller tried to sync up was more pronounced than it was on the test stand. It took a light touch on the throttle to get the motor to run smoothly and because the motor is a freewheeling geared motor (GM 24V Mini@36V) the controller needed to reacquire the position each time I pulled the throttle from the off position. This was very annoying so I stopped and reconnected the hall sensors. A power reset was required before it ran smoothly. The controller might function a whole lot better with a direct drive motor since unless you come to a complete stop the motor is always spinning.

While the sensorless operation did not meet my expectations I did enjoy the auto cruise control function. To activate cruise control all you need to do is hold the same throttle position for 8 seconds then release and the motor continues spinning. It’s not difficult to hold the throttle steady that long which means it can surprise some times. I think I'd prefer a button to activate it. Still once I got used to the idea I found it to be way cool. 8) I played with different ways to set it but found the best way was to accelerate normally with the throttle and pedaling to the desired speed then hold the throttle steady while I continued to pedal, then release the throttle. The whole thing quickly became second nature and the ability to let go of the throttle was quite liberating.

I don't have any E-brake system installed thus there could be a safety concern however in practice it wasn't an issue for me. I have some brake levers with brake cutout switches from an E-BikeKit but they're not installed and frankly I don't know how I'd hook them up to the one wire dangling from this controller. A quick blip to the throttle is all that's needed to disengage the cruise function but I did want to test what would happen in a panic braking situation. I found I could slow the bike with the brakes to the point I could barely balance even as the motor current went to max. In my case the low power motor was a plus but if I had a powerful motor I would re-think the need to implement a brake cut-off function. Once I released the brakes the motor would slowly accelerate back up to the set point.

In other respects the controller worked as expected. I wanted a 15A limit and the description on the eBay ad said this was a 14A controller but in fact the peak current on this ride was 21.06A. The controller also supposedly will do motor braking but as it's impossible to implement on this freewheeling geared motor and I don't presently have my 9C spoked to a rim I can't investigate it further.

-R

 

[Russell]

Getting answers from Ecitypower is proving to be difficult, most likely due to the language difference. They answer my direct questions by simply identifying the wires from the controller. One thing was sort of helpful was learning the grey wire was for cruise, but that's all they said. I cut the loop of grey wire which disabled the auto cruise function then I installed a momentary switch which allowed me to set the cruise instantly. Throttle up, hit the button and cruise is enabled then hit the button again and the motor stops. Now I just have to get a switch mounted more conveniently on the handlebars.




I still have to figure out how to make the single 'brake' wire work :|


-R

 

[johnrobholmes]

try grounding the brake wire and see if it stops the wheel.

 

[Russell]

try grounding the brake wire and see if it stops the wheel.

I touched the lead to the case then to the battery negative lead and nothing happened. I think despite what my rating plate says it may take a 'high' level to make it work.

-R

 

[Russell]

I wanted to see how the controller would run sensorless with a direct drive motor so I dropped my raw 9C motor into the forks of the only bike I haven't converted yet and sure enough it worked much better than with the freewheeling geared motor. This is the motor I wanted to use the controller with anyway so things look promising though E-bike number #3 with the 9C is still on the drawing board at this point :wink:





-R

 

[Russell]

I ordered the 36V model and the rating plate says "36/24V". The main caps are 63V 470 uF and the MOSFETS are IRFB 3607's. Data sheet: http://www.irf.com/product-info/datasheets/data/irfs3607pbf.pdf







-R

 

[dimach]

Hi Russel, I just got a package from bmsbattery myself. It includes 24V battery, Bafang QSWXK and 9 mosfet controller. I don't have a throttle yet.

I'm trying to see if this combo works. I've tried to connect motor and battery to controller and measure voltages on throttle pins. I see no voltage there. I also tried to simulate throttle by connecting two 0.5Kohm resiters, but nothing changed. Am I doing something wrong?

Do I need to connect green wire from power connect to +?

 

[Russell]

Dimach,

If you aren't measuring 5V between the throttle red and black wires then the controller is not powered up. The battery power connector from my controller has 3 leads; red, black and orange. The orange lead is meant to work with a switch to turn the controller on and off. When I replaced the original connector with Andersons I twisted the orange and red leads together before soldering on the pins.

Next, the controller needs to see a low throttle signal on power-up so your fixed resistors won't work. If you can't wait for a throttle you can trot on down to your local RadioShack and pick up a 5K linear pot. Hook up the left lead to RED, the center wiper lead to GRN/BRN and the right lead to BLACK.

Place the motor in the bike's forks then turn up the pot. If the motor does not rotate in the proper direction or doesn't rotate smoothly turn the pot back down then connect the two white wires together, this puts the controller in program mode. The controller will try the different phase combinations and when the correct order is found the motor will turn slowly and smoothly in the proper direction; disconnect the white wires as soon as this happens.

Good luck

-R

 

[dimach]

Thanks Russel, connecting orange and red in the power connector did the trick.

Unfortinatly, I would have to get a new fork before I can try the motor again.

 

[Russell]

After 3,000 miles some hall wires and a hall sensor were damaged on my Bafang geared motor. I removed the hall wiring then locked up the clutch so it would not freewheel (details here; http://endless-sphere.com/forums/viewtopic.php?f=4&t=10458&start=45#p212576 ) and have been using this controller in sensorless mode ever since. It works great requiring only the slightest forward movement of the wheel to start perfectly. I think it would make a great low-cost replacement for folks running Aotema motors or any other direct drive motor for that matter. Also while I specified the 36V version I have been running it with a 48V pack (63V caps, 75V FETs).

-R

 

[317537]

I got the ecrazyman version of this but in the 1000 watt model.

The hammering on the non GM geared motor still happens under load.

How is the throttle response on your controller? Does it have some soft start happening.

With my controller

Once you get used to it and get the starts happening smooth the rest is gliding down the road smooth as silk. This seems to give awesome efficiency and my pack comes home with a much higher SOC than usual. Like where I would do 60% DOD I now only do 75% dod. Strange really as the newer controller is rated at 40 amps.

The response is fast to the throttle . From much reading, the soft start had been removed or toned down substantially the firmware of my controller. Being used to a slow soft start controller the change had me putting on the gas too fast and this is where the hammering kicks in.

I feel if the human jerkyness was taken out of the mix with an appropriate incremented soft start this would help a lot and this controller would be a much better device.

Another way to get this baby moving off the mark is to allow the hub to do its wiggle. I do the by lowering the seat and putting the feet to the ground and taking my ass off the seat. The wheel may take a small step back and off she goes. I still need to go light on the throttle though.

 

[317537]

Hi Russel, I just got a package from bmsbattery myself. It includes 24V battery, Bafang QSWXK and 9 mosfet controller. I don't have a throttle yet.

I'm trying to see if this combo works. I've tried to connect motor and battery to controller and measure voltages on throttle pins. I see no voltage there. I also tried to simulate throttle by connecting two 0.5Kohm resiters, but nothing changed. Am I doing something wrong?

Do I need to connect green wire from power connect to +?

I found that I needed a proper hall throttle or resistor throttle for my controller to work.

The old test of just joining the hall input and output wires results in a clunk and nothing. I'm not sure what's going down with the no voltage on the pins though.

 

[Russell]

I got the ecrazyman version of this but in the 1000 watt model.

The hammering on the non GM geared motor still happens under load.

How is the throttle response on your controller? Does it have some soft start happening.

With my controller

Once you get used to it and get the starts happening smooth the rest is gliding down the road smooth as silk. This seems to give awesome efficiency and my pack comes home with a much higher SOC than usual. Like where I would do 60% DOD I now only do 75% dod. Strange really as the newer controller is rated at 40 amps.

The response is fast to the throttle . From much reading, the soft start had been removed or toned down substantially the firmware of my controller. Being used to a slow soft start controller the change had me putting on the gas too fast and this is where the hammering kicks in.

I feel if the human jerkyness was taken out of the mix with an appropriate incremented soft start this would help a lot and this controller would be a much better device.

Another way to get this baby moving off the mark is to allow the hub to do its wiggle. I do the by lowering the seat and putting the feet to the ground and taking my ass off the seat. The wheel may take a small step back and off she goes. I still need to go light on the throttle though.

The only time I get any "hammering" now with the Bafang clutch locked up running sensorless with this controller is if the wheel is completely stopped or rolling backwards. All it takes for a smooth launch is the slightest forward movement of the wheel.

I don't know if it has a soft-start feature or not or if the E-BikeKit/Infineon controller I used with the motor before had it. When I was running the motor with hall sensors I could of course pull the throttle from a complete standstill and often have the front tire spin as I grabbed the bars and unloaded the front end as I pedaled away from a stop. Now if I'm completely stopped I put in a stroke first before pulling the throttle so while the tire does slip on occasion it isn't as much as before. So I would say performance off the line hasn't been enhanced even though this controller peaks around 21A Vs 15A previously. Performance up grades however does feel stronger due to the additional current.

I doubt I am getting any increased efficiency, in fact I would expect the opposite from a sensorless controller, however it's a tough thing to check this time of year. I use more power as it gets cold out for a number of reasons and yesterday I used the highest power per mile ever. I also no longer have a motor which freewheels so for the times I am off the throttle, and that's not often anymore, I do get a bit more drag.

I do like running sensorless since now without the halls as a possible failure point I feel I can push the little motor harder. I'm now running peak power of a bit over 1,000W whereas I used to limit it to around 760W. The only time I'm reminded I am running a sensorless controller is if I get anxious at a stop light and yank the throttle before I at least get the bike rolling forward.

-R

 

[317537]

Thanks for the response.

Yea I recently I replaced the halls only to have them go out 2 mths latter.

I run mine on on a direct drive gear less hub and free-wheeling is totally working with no cogging what-so-ever. My old controller was an older GM where I couldn't or wouldnt disable regen and I was pickling them up for $30au delivered. It seemed that every time a hall went regen or the cogging would take a channel out. Maybe this sensorless might be much more reliable.

Freewheel vs regen and efficiency is possibly some reasoning for the improved efficiency but the 18km ride I did where my middle aged 48v SLA stack came back with 12.93v per SLa seemed to be unusually good. I feel there is less cogging resistance and a more advanced timing in my sensorless than any of the sensored units ive gone through. The speed is slightly better with lower noise coming from the motor indicating this could be true.. The extra weight on my ebike due to the heavier SLA's doesnt help much with the start ups IMO.

If your controller had a Softstart that may help with the controller you would definitely notice the delay in the throttle.

I am working with a spice model with and external softstart put on the throttle. I need some resistance readings from the circuit. I really hate pulling stuff apart when its all hooked up though and getting any values for something like this here has been like getting blood from a stone.

See how I go.

 

[Russell]

I run mine on on a direct drive gear less hub and free-wheeling is totally working with no cogging what-so-ever. My old controller was an older GM where I couldn't or wouldnt disable regen and I was pickling them up for $30au delivered. It seemed that every time a hall went regen or the cogging would take a channel out. Maybe this sensorless might be much more reliable.

I don't understand what you are running. If you have a direct drive hub motor then you aren't freewheeling and will always have some cogging torque. If you have a geared hub then most of them will freewheel when not using the throttle unless you do as I did and lock-up the clutch.

-R

 

[ebikerider]

Hi,
These people sell garbage - hear my story.
I communicated with this company through email and told them I need a 24V 20A LiFePO4 continuous discharge battery for my bike, a charger and a 500W controller. I paid $423 for it online through PayPal and was eagerly awaiting my new LiPo battery. It It took 8WEEKS to arrive - 4 weeks "processing" and 4 weeks "shipping". When I received the box, it had a 250W controller, not 500W. Pretty much useless controller for my purposes.
I emailed the guy and he agreed that it was his mistake and said he would send me another one - but it has been 3 months and no replacement came.
The charger stopped charging after TWO charges - next day, pretty much. It just puts out 19V and does not charge. I connected to the company through Skype and for half an hour SHOWED the guy on camera all the testing of the broken charger. He promised to send another one - and guess what, NOTHING CAME.
The battery burned out in 2 weeks. I used it on Izip electric bike, which still rides nicely with simple SLA that comes with it. But this company's battery is completely burned (see picture). Needless to say, they promised to send another one and then stopped answering my emails. It's been 3 months and my $423 IS GONE....
I thought I'd save a few bucks but I lost all the money. And have a pile of junk in my apartment.
DO NOT BUY FROM BMSbattery/eCITYPOWER or your money will be wasted!
ebikerider