Niu N1S Civic Upgrade

[Molten-Core]

My friend Dimitris said "Knowledge is for everyone", so here is to him, to the people of this forum, to everybody that gave me the smallest of information, i will be sharing back my findings on this work.

I bought a few weeks ago a Niu N1S Civic at a fairly low price, it was a demo unit and partly damaged (scratches and plastic mainly), i tested it and it wasn't up to my expectations, i drive for a now a Honda CBF1000F, i don't need much top speed as i usually cruise at around 55km/h, i need the torque to get rid of anyone behind at the light and for evasive maneuvers (yeah, Brussels is a warzone), and given all the idiocy going around here they are planning to shift the whole city to 30km/h max, my days driving the CBF are numbered...

My goal from this work is to get better torque from this scooter, and i want people to be able to hear me when i want them to, so i may be integrating a soundbox along the way, i also want the scooter to stay reversible, if i ever change my mind i want to be able to unplug everything i added and put back the original hardware.

here are some pictures, they are clickable:


As you can see from this pictures the primary battery doesn't take much space, there is a lot of wasted space there, so i will be using all that space to house a 20s22P battery made with samsung 2900e that tumich sells at a very very affordable price, we are looking at almost 5kwh of energy storage with a 220 continuous amps.

I'm looking forward to working on this project, i'm waiting for batteries, nickel, dc dc converter and a 120amps breaker to arrive.

On these pictures you can see the specs of the bosh hub motor (60v, 1500w nominal, 2400 burst), the original controller, the new sabvoton 72150 controller, and how it will be positioned, it fits really snuggly right there, i drilled 3 holes to maintain it there.



In the first picture you can see the connector going to the controller, and no matter where i look i cant find a place where i can buy a female counterpart alone, it's almost driving me nuts...
In the second picture you can see the integrated alarm, i think i'm gonna be replacing that with a keyless entry system, in the third picture you can see the DC DC converter that outputs both 12v and 5v, i think, but not quite sure though, that the 5v are only for the usb for phone charging, i will be replacing the converter to a more appropriate voltage source one, and i will probably be replacing the 5v part with a wireless charger for my phone.
In the 3rd picture you can see hall wires, one of them is missing, it's actually the motor temperature wire, i think i will open the hub and see if i can add a temperature sensor, if i do so i will probably add some statorade.
The last picture show the hall throttle, if i can get another one like that i will probably do a mod to have a reversible throttle, Brussels is a very hilly city with a lots of going up and down, having the possibility to control my regen brake would help a lot given that i do around a 100 to 150km a day.

 

[Molten-Core]

Today i removed the real wheel/motor and opened it, it looks really similar to a QS Motor as Dui, ni shuo de dui in this forum told me once, it has some markings everywhere, the most interesting i find are 215 35 40h (the same format used by qsmotor)



Given that i dont see any visible temperature sensor (contrary to the super soco), i will add one myself, the reference for sabvoton 72150 is KTY83/122, but it has been discontinued, after some research, this page indicates that it was superseded by AMWEI AM-LPTC1000, both their datasheet confirm that they should be quite a close match (KTY83/122, AM-LPTC1000), the temperature measurments will not be exactly the same, but we will be at 5 degrees drift which doesn't seem to be critical, what i really want is to be able to cutback on motor power at around 100° temperature.

I am also planning to add some statorade to help cooling the motor given that i will probably be driving it at higher currents, now i have no idea how much statorade i have to add, anyone have an idea?


Edit: Actually never mind with that part about the alternative thermistor, i found a KTY83/120, the specs were even closer to the 122 than the LPTC so i ordered one.

 

[Molten-Core]

So, so far i have done a few things, some are idiotic, some less.

I plugged the new DC DC converter to the old harness just by matching colors, well, the universals ain't universal, and the wires did not serve the same purpose, the higher and lower voltage were color inverted, i'm glad i did not fry anything, and given that i did not notice that at first i thought that something is wrong in the harness, which i ended up cutting to pieces and remaking my custom harness, and then i noticed my mistake, i'm glad it just works fo now, rip old harness.

i received the temperature sensor, i went on and about to install it, and then noticed that it will not be an easy task to pull out the old cable to insert a new wire in it, so i'm gonna leave this one out, and after some research i read about a guy losing his magnets due to something in the statorade melting the glue of his qsmotor, given i probably have a qsmotor here i decided to leave that out for now.

I noticed that some laminations are pulling appart, and noticed one that was clearly bent, i put that back with some super glue and an improvised something to hold it together for the night, it seems to have worked, i've seen some traces like if insulation has worn out, i'm not sure what effect it will have on the motor in the future, but for now resistance between all coils is 3 ohms, which seems good.

So i put some Elring Dirko to seal the motor again and put the motor back together, plugged phases and halls connectors to the controller, i was really hoping that the "3 minutes setup" would really hold, but this was not the case, i spent all the spare time i had the last two days to make it work, but i am not able to make a single hall test correctly, the test starts with a big twitch on the motor and then the whole thing goes into a system error, the error is hall fault.

In order to find a suspect i spliced a usb cable to get a 5v output and tested the hall sensors, they all give me either 0v or 5v depending on the position of the wheel, which makes me think that hall sensors are alright, motor side i have a 5v output on the red wire, so it should be good there too.

On the computer interface the Hall status(CBA) varies between 1 3 5 7 depending on the position of the wheel.

I tried changing phases, it felt better because after the big twitch the wheel started rolling slowly just like in the videos, but then it stops and displays hall fault, sometimes when i put too much current in the interface it shows also and overcurrent error.

Right now i am wondering if there a specific type of hall sensor to be compatible with the sabvoton, i also tried correcting the Pole pair number thinking that it was the hiccup, but nope, changed nothing...

As usual here are some photos


Do we agree that there is 30 pole pairs right here?


Those sratches worry me in the sense that i think there may be shorts there, is there anything that can be done to treat those?


The damage that i had to super glue, really hope the super glue will hold, i still wonder how did this damage happen...

 

[Molten-Core]

After much investigation i came to something i did not expect and that i have no explanation for.

As i told befoe, my sabvoton throws a hall fault either at the startup of the system, or if i spin the wheel after starting it up either manually or through the throttle, also, when i spin the wheel the number in the hall status (CBA) is 1 or 3 or 5 or 7, the first 3 numbers seem to be normal and valide hall statuses, the 7 is not.

I checked the halls manually by plugging a 5v supply via USB, and the results where like you would expect from a correctly operating set of hall sensor, either 5V or 0V, and there is always at least one showing 5V or 0V.

Now i went about and plugged my multimeter behind the plug while the halls are connected to the sabvoton, and low and behold, the three sensors go to 5V, but only two go to 0V, the yellow hall goes only to 1.58V.

Right now this seems to be a possible explanation why i have been having trouble with the test, oh, and by the way, i have managed to get a Test Fail3.

Right now i think the next step is to to buy a set of hall sensor and replace all of those just to be sure, according to this post, the right ones seem to be Honeywell SS441A

 

[Molten-Core]

Allelujah, this is a great day

So, it wasn't the sensors after all, i think it has to do with the resistors inside the motor and the eventual pull ups in the controller, all i had to do is bridge the signal coming from the hall sensor to the wire directly bypassing the resistors and all went smooth as butter, i wish i had discovered that before i broke the middle hall sensor.



Just for clarification I had to totally bypass the resistors so that voltage can be pulled to 0 volt, after that i could see all the hall statuses correctly (from 1 to 6), i didnt see hall status 7 anymore which i suppose is a good thing.

Next step is building the battery i guess, i am trying to build a spot welder, we'll see if i can do that correctly.

Edit: just in case, my hall angle is 123°, a bit weird but that is what the hall test gave as a result.