Folks,
As a result of the failure of the brushed hub motor on my Evader EV100S, I am in the process of installing a brushless hub motor. The motor is a 1500 watt unit with a controller, all from GreenEMotor.
The motor has been fitted up mechanically and I ran some quick tests today to make sure everything is working OK before I finalize the wiring. While it did run, there were some questions that resulted from checking various voltages during the testing.
The throttle is a 0-5vdc hall-effect type. The supply voltage from the controller is 4.3v and the maximum controller output voltage is 3.7v. Normal? 2 things come to mind. I need the full 5 volts to see the maximum speed of the motor and at throttle wide-open, the output voltage should match the input voltage. That was the way with the brushed motor and controller.
The phase-to-phase voltage was 37v. The input voltage to the controller was 52v (48v nominal voltage). Should the phase-to-phase voltage be this low?
Thoughts and advice appreciated.
Les
Most hall throttles are 1-4V output, so yours is in the right area. Shorting out the positive and signal wires on the controller's throttle connector and seeing if the wheel goes any faster will tell you if the throttle is going full speed or not.
What are you measuring the phase to phase with? A regular DVM won't be able to measure it properly, since power isn't applied to each phase all the time.
The author of this post isn't responsible for any injury, disability or dismemberment, death, financial loss, illness, addiction, hereditary disease, or any other undesirable consequence or general misfortune resulting from use of the "information" contai
I understand a DVM may not be the appropriate tool for reading phase-to-phase voltage. My understanding of motor design is rudimentary, especially BLDC motors.
The voltage from the controller is my main concern. Is it important that the controller output voltage be as close as possible to 5vdc? Or is the motor speed just proportional to the ratio of the controller throttle input voltage and the controller throttle output voltage?
Les
Does someone have some information on how the 0-5vdc throttle works?
I ran a test on my throttle, attempting to see if it may be affecting the motor speed. The motor and controller were not hooked up for the test. 2 AA batteries were used for a voltage supply. The voltage into the throttle was 3.176vdc. The voltage across the output leads of the throttle was 0.816vdc with the throttle closed. With the throttle wide open, the voltage was 2.45vdc. There are no components inline for this test. Are these voltage differences normal?
Les
Hall type throttles are pretty simple. There's a hall sensor that reacts to different poles of a magnet. One pole makes it output a low signal, the other makes it output a high one. There's a curved magnet in the part of the throttle that turns. This passes by the hall sensor and progresses the output signal.
A bit of extra electronics are in the ones with LED indicators, but that doesn't relate to the sensor itself.
Your voltage ranges sound pretty normal. The hall sensor isn't supposed to go to 0V, and outputs only a portion of what's being fed to it.
The author of this post isn't responsible for any injury, disability or dismemberment, death, financial loss, illness, addiction, hereditary disease, or any other undesirable consequence or general misfortune resulting from use of the "information" contai
Ok, I have completed the installation and initial testing of the 1500 watt motor and controller that I purchased from greenemotor. Unfortunately, the performance at this time is disappointing. The acceleration is very slow and the top speed is less than 25 mph, less than the 35 mph quoted. Offhand, this motor does not feel as strong as the 1000 watt motor that was originally in the scooter.
The original scooter is an Evader EV100S. The batteries are 45 amp-hour. The voltage at the start of the testing was above 52 volts, a freshly charged pack. The tires are 3.00 x 10.
The no-load speed of the motor was 682 rpm, lower than I expected for a motor that should have been good for 35 mph. I tested the speed with a laser tachometer.
Any suggestions on why a brushless motor and controller would not perform up to spec? Tests?
I could beef up the shunt to gain more acceleration but that would not improve the top speed.
I am trying to work with the seller of this motor in parallel to solve the problems.
Thanks!
What's the no-load current like? High no-load draw and unusually poor acceleration are typically symptoms of mismatched hall wires.
This post on the ES may be of some use to you.
The author of this post isn't responsible for any injury, disability or dismemberment, death, financial loss, illness, addiction, hereditary disease, or any other undesirable consequence or general misfortune resulting from use of the "information" contai
Thanks for the link to ES. Unfortunately, Fechter and another author only confirm that the motor is likely to be connected correctly. I will still check things but expect to find the wiring is correct. The motor starts with a low growl and accelerates to its operating speed, both with no load and with load (riding).
I will check the current once I have located a shunt to test with.
Thanks!
Whoa, yeah, something is wrong there, and it sounds like halls. I'd also check the halls themselves; a blown one can give the same symptoms as a bad connection.
The author of this post isn't responsible for any injury, disability or dismemberment, death, financial loss, illness, addiction, hereditary disease, or any other undesirable consequence or general misfortune resulting from use of the "information" contai
Hmm, I checked the hall sensors per the methods outlined on this site and did not find a problem. The power supply + lead was loose and I tightened that connector. I have only tested the motor on the center stand but the low speed growl is not as loud.
The no-load current is 5 amps. Tho I have tested the load under various conditions in the past, I can't for the life of me remember the method. Someone on this forum suggested the method and it worked very well. The method utilized a current shunt to calibrate the system no load current and then... I can't remember.
The motor/controller's supplier is talking with his engineer. Tho the answers were not definitive, it sounds like they normally run the motors with 60v, rather than the 48v I am using at this time. Running with 60v (5 batteries) is a little worrisome since the controller's caps are rated at 63v. The last time I ran a controller with similar caps on 5 batteries, the magic smoke got out. Also, the supplier hinted that they may suggest that the controller's shunt be modified, something that I have done in the past. I am more hesitant to modify stuff while it is under warranty.
Later!
Well, I went through a number of modifications to the shunt and obtained reasonable acceleration, including one extreme modification that exceeded 100 amps. Very sprightly but probably too much. The high amps did not cause a problem but I doubt the system was designed for that many amps. Anyhow, I backed off the shunt mod somewhat and now have acceptable acceleration.
Unfortunately, the scooter is still slow, with a max speed of 27 mph at 48v. The controller supplied with the motor does not appear capable of handling 5 12-volt batteries since the components are rated at only 63v (I opened the controller and checked the capacitors and MOSFETs.
I have worked with Allen of GreenEMotors (GEM) and his partner, Tracy, of Nova, trying to find a controller that will handle 60v nominal and be compatible with this motor. Tracy forwarded 3 controllers, none of which appeared to function with the motor. I appreciate Tracy's patience with me during the testing and thank him for his help. I went through the phase spreadsheet many times and all I can conclude is the possibility that the motor's phase angle is different from the controllers. The best response I could get from any of the controllers was an inconsistent 'bump' of the motor when the throttle was applied during testing. Oh, and by the way, a 10-amp fuse that I was using to protect the system during testing sparks beautifully when too many amps hit it. Allen is sending 1 last controller, a 72v-capable controller that we hope will work. It will likely be the last chance for this motor.
My question is, how can the phase angle be determined? I found one message about the hall sensor on-off sequence for 30 and 60 degree motors. But Kelly talks about motors being either 60 or 120 degree phase angles. Any help out there?
Thanks!
Les
Check that link on determining motor wiring again. There's a new post about building a simple LED indicator that shows when individual hall sensors are on or off.
If there are positions where all three are on/off, it's 60-degree spacing. If there's a max of 2 that are on or off at one time, then it's 120.
The author of this post isn't responsible for any injury, disability or dismemberment, death, financial loss, illness, addiction, hereditary disease, or any other undesirable consequence or general misfortune resulting from use of the "information" contai
GEM sent a final controller to test, mainly to see how the motor would behave at a higher voltage. According to the nameplate, the controller is rated at 72v, 38a and 800w. Since the motor is a 1500w unit and I have seen a peak current of about 80a, I anticipated that I would not be able to ride with the 72v controller but could test with it. Based on checks of the motor's hall sensors, I believe this motor may be have a 120 degree phase angle.
The first thing to get straight was the phase wiring for the hall sensors and the motor leads. This turned out to be a no-brainer since the colors matched up and the motor turned over on the first try. I checked the no-load speed and the speed matched the original controller for 680 rpm at 48v. I then re-rigged the wiring for 60v operation. It took a very light touch on the throttle to accelerate the motor to the maximum speed without the controller tripping. I am assuming the 38a limit involves a sensor, not just a shunt as in most controllers. At full throttle, the motor attained 842 rpm. Proportioning the voltages and the 48v speed of 28 mph, the maximum speed at 60v would be very close to 35 mph, the goal of the project. This 72v controller would be unusable on the street since the slightest acceleration would trip the 38a limit.
The controller will be returned to GEM. The question remaining is whether or not the entire purchase will be returned to GEM. GEM claimed the controller is 60v (nominal) capable and 35 mph capable at 48v (nominal). The controller's internal components are 63v pieces so hooking up to a fully charged pack will likely smoke the controller (+65v). At 48v, the maximum speed with a 3.00 x 10 tire is 28 mph. On the plus side, the motor is mounted to the scooter and an increase in voltage gave a proportional increase in speed.
Alternative controllers are available, especially those capable of operating with either 60 or 120 degree motors. Kelly and Ecrazyman controllers come to mind. At $300 for a Kelly, the controller is pricey when you think that the GEM controller should have fulfilled the intended function. Kelly also has hub motors available at a reasonable price at this time. A short discussion with someone at Kelly indicated that Kelly would supply curves for any motor they sell, something that GEM cannot supply.
I have invested a silly amount of time into this conversion but have learned bunches. Life is interesting!
Les