Weird Controller dilemma

Lurkin

100 kW
Joined
Jan 18, 2015
Messages
1,473
Location
Melbourne, VIC, AUS
I'm trying to both work out the wiring diagram and the reason the controller is not working.

Controller details on it:

CE:G3M20211-1592-16
Product Number: LSW969-16F X16050013
Google images shows a bunch of wiring which doesn't match mine.

From the controller I have:
- Black plug supposedly with phase and hall wires coming out of it. This matches the sole plug from the motor.
- two, two wire JST female. Assume this is ebrake cut offs. Nothing plugged into them, both one yellow and one black wire each.
- one 5 pin JST female. 5 wires coloured red, blue, black, green yellow. Assume this is a second set of motor wiring. weird. Measuring across the red and black wires gives 48v when the power supply is switched on. [EDIT] this is the display output.
- one 3 pin JST male. 3 wires coloured red, green black. Had this plugged into a throttle with the same coloured wiring. Unplugging and measuring voltage across red and black wires gives nothing. Should this be 5v+ish? [EDIT] this is not a throttle! it is for PAS!
- one 3 pin JST female. 3 wires coloured white, red and black. assumed this is PAS, nothing plugged into it. Should this be 5v+ish? [EDIT] White is throttle signal wire, red is + and black is GND.
- one 4 pin White weird plug. internet tells me this is for a display, nothing plugged into it. [EDIT] No, this is for programming.
- one red thick wire. Assumed battery positive, plugged into power supply with 48v+ on it.
- one black thick wire. Assumed battery negative, plugged into power supply with 48v- on it.

I've already contacted the supplier, they aren't much help.

Turning the throttle does nothing.[strike]I've tried to measure resistance across the throttle (assumed it worked like a potentiometer) - nada. If its a hall sensor, what should I get here?[/strike]

Either:
- [strike]power supply is not working[/strike] power supply is giving 48v both at output leads and from weird secondary motor wiring.
- controller is not working
- [strike]throttle is not working[/strike] throttle tested below, working.
- I have the wiring wrong (this is most likely the case)
- poor contact on JST connector for throttle?
- motor has not been wired correctly to the controller by manufacturer.

help?
 
Ok tested the throttle with 4.3v on red and black and put the multimeter across green and black. As the throttle turns, increases in voltage to 3+v.
I'll take it the throttle is fine.
 
The 5 pin plug with battery voltage on it is for the LCD/LED display, which contains the on/off switch. The LCD connects the red to the blue to power the controller. I hope you didn't connect it to the motor because the 48v will kill the halls instantly.

You can get analogue and digital displays. There's no way to tell which one you need, except the 5 wire ones are normally , but not always, digital, in which case yellow and green are the data wires, which might also need to be connected together for the throttle to work. If it's an analogue one, the green wire will be the PAS level signal and yellow for the lights signal. If you open the controller, and it says tx and rx where the green and yellow wires are soldered, it's digital.

These controllers often give problems when trying to run without the display.
 
Nope, haven't connected the 5 pin connector to anything, other than to measure voltage across the black and red wires in it with a multimeter. I cant connect it to the motor - its a JST connector and the motor and controller have a matching (non JST) designed plug which has both hall and phase wires within it.

That's bad news on the display front, I'll try to contact the manufacturer again. Guts, because I don't think they actually understand what they are selling themselves.
 
If your controller is wired for a display, there are usually ways to bypass it to get the controller working. That info is usually posted with the controller info.
 
If yours is like my Fusin "1000w" unit (see my thread in the review section here
https://endless-sphere.com/forums/viewtopic.php?f=4&t=39877&hilit=fusin+1000w
) then you can bypass the display to test it simply by shorting two wires (don't remember which two mine were), but it may not operate at full power, or features may not work as expected (or at all) without the display. :(


You may not be getting 5V out for motor halls, throttle, or PAS, if the display is required to "turn on" the controller--that's what I found with mine, then "fixed" it by bypassing the display after some measurement and experimentation, to see if it would run without it.
 
I have emailed the manufacturer to see if they know what display should have been supplied with it. If they can provide one, I will just buy it and go from there.

In the meantime, I have removed the controller board from the controller housing.

Board has written on it: LSW12G-150Z-FOC-141126 15.50 So I'm hoping the "SW" = Sine Wave, FOC = Field Oriented Controller. No idea about the rest.
(Google doesn't seem to find anything on that).

Wiring description updated to include what it connects to on the controller board (based on what it is marked on the board):

- Black plug supposedly with phase and hall wires coming out of it. This matches the sole plug from the motor. ---> Black insulation opens to reveal thicker Yellow, Blue and Green wires, which appear to be the phase wire connections as expected. Also has skinny wires being white, green, yellow and blue. The white connects to "SS" on the board, green to "SB", yellow to "SC", blue to "SA".

- two, two wire JST female. Assume this is ebrake cut offs. Nothing plugged into them, both one yellow and one black wire each. ---> The yellow wires connect to the board side by side and are both marked "BKL". The black are not, but both are connected and marked "GND".

- one 5 pin JST female. 5 wires coloured red, blue, black, green yellow. Assume this is a second set of motor wiring. weird. Measuring across the red and black wires gives 48v when the power supply is switched on. ---> red to VB+ right next to the battery input for 48v+, Blue to "SW" which is situated right next to VB+. Interesting, there is five slots available for "SW", but only one is used, Black to "GND" right next to the battery input GND, Green is to "TX", Yellow is to "RX".

- one 3 pin JST male. 3 wires coloured red, green black. Had this plugged into a throttle with the same coloured wiring. Unplugging and measuring voltage across red and black wires gives nothing. Should this be 5v+ish? ---> Green to "TA", Red to "5V0", Black to "GND".

- one 3 pin JST female. 3 wires coloured white, red and black. assumed this is PAS, nothing plugged into it. Should this be 5v+ish? --->White to "SF", Red to "VSP5v", Black to "GND".

- one 4 pin White weird plug. internet tells me this is for a display, nothing plugged into it. ---> White insulation opens to reveal four wires. They are red,black,yellow and blue. They all connect at the opposite end of the board to the VB+ and VB-. There is no further indicator of what they are.

- one red thick wire. Assumed battery positive, plugged into power supply with 48v+ on it.---> On controller board = VB+
- one black thick wire. Assumed battery negative, plugged into power supply with 48v- on it.---> On controller board = GND


Other information whilst it is apart:
- there are 12 mosfets. They say "80NF70 CZ080 6 MAR443 ST e3" on them. "ST" appears to be a brand.
- the four larger electrolytic capacitors are 63v
- the main chips appears to say on it "STM32FEB KC6T6 GH20F 93 CHN S46 ST" "ST" appears to be a brand.

From all of that: sounds like I need a digital display. If these can be tricked into thinking they are on, I will use it without. I have a cycle analyst V3 here which I would prefer to use... 8)

Werid, its like the white wire is an analogue display and the 5 pin is a digital display? Could be completely, completely wrong here.
 
Interesting, there are a number of sockets on the board which remain unused. They include:
TB, RP, EBS (assume that's regen braking, fine, because this will be used with a geared motor), AD1, SL, CR,AD2, numerous 5V0 and SW.
 
wesnewell said:
If your controller is wired for a display, there are usually ways to bypass it to get the controller working. That info is usually posted with the controller info.

If your meaning posted with the controller to me - there was absolutely nothing provided by the manufacturer. On request for a wiring diagram, they provided a breakdown of wiring within the Black Motor - Controller connection (which cannot really be changed anyway) and ignored the rest. I've all but given up on asking them anything - I think they are just reselling these from another vendor and don't really know what the deal is.
 
Any ideas of how to work out what display could be used? Just not too sure where to begin with this. I think I have worked out the ultimate manufacturer of the controller - I've emailed them and the vendor I got this from, yet to receive a reply (unlikely to get one until tomorrow or later).

I'm reasonably sure this is meant to be USB programmable, so I've asked about getting a programming cable too. 8) (not to mentioned downloaded the software :D )
 
The only 'bypass' I can think of is to connect the RX and TX together. I.e. rather than the controller communicate with the display and receive a response, it will only be communicating with itself. Thoughts?
 
Lurkin said:
Board has written on it: LSW12G-150Z-FOC-141126 15.50 So I'm hoping the "SW" = Sine Wave, FOC = Field Oriented Controller
Best guess is LSW is more like Long Shen Wong or whatever. ;)

12G probably 12FET

150Z ? dunno.

141126 is possibly design or revision number (possibly date 2014 nov 26th),

15.50 almost certainly revision number.

The white connects to "SS" on the board, green to "SB", yellow to "SC", blue to "SA".
I'd guess SS is speed sensor (like my FUsin geared hub, though it had no halls as it was sensorless; the controller could have run a sensored motor)
Is there no black and red for hall power and ground?



- one 5 pin JST female. 5 wires coloured red, blue, black, green yellow. Assume this is a second set of motor wiring. weird. Measuring across the red and black wires gives 48v when the power supply is switched on. ---> red to VB+ right next to the battery input for 48v+, Blue to "SW" which is situated right next to VB+. Interesting, there is five slots available for "SW", but only one is used, Black to "GND" right next to the battery input GND, Green is to "TX", Yellow is to "RX".
Unliekly to be anything other than the display connector.
48v is power to display, and also is probably like my Fusin, switched by a fet inside the display to connect to another wire on the display connector (probably SW as it can't be RX or TX or GND). So if you jumper from SW to VB+, it would act just like a keyswitch/ignition wire on a typical displayless controller.

- one 3 pin JST male. 3 wires coloured red, green black. Had this plugged into a throttle with the same coloured wiring. Unplugging and measuring voltage across red and black wires gives nothing. Should this be 5v+ish? ---> Green to "TA", Red to "5V0", Black to "GND".
As I noted in previous post, you probably won't get any 5V anywhere until the contorller is powered on by the display, or you jumper it's switch wires to bypass it.



- one 4 pin White weird plug. internet tells me this is for a display, nothing plugged into it. ---> White insulation opens to reveal four wires. They are red,black,yellow and blue. They all connect at the opposite end of the board to the VB+ and VB-. There is no further indicator of what they are.
Nothing on the PCB at all?

Could you post pics of the board inside, both whole board and then closeups of it where wires connect, both sides of PCB?





- there are 12 mosfets. They say "80NF70 CZ080 6 MAR443 ST e3" on them. "ST" appears to be a brand.

ST is STMicro
80NF70 is a common enough FET in controllers discussed here on ES.
https://endless-sphere.com/forums/search.php?keywords=80NF70&terms=all&author=&sc=1&sf=all&sk=t&sd=d&sr=posts&st=0&ch=300&t=0&submit=Search


- the main chips appears to say on it "STM32FEB KC6T6 GH20F 93 CHN S46 ST" "ST" appears to be a brand.
STM32 is one of their ARM-based MCUs. Just on the chip is about the same computing power as an entire 486 personal computer!
 
bunch of photos here:

http://s666.photobucket.com/user/Lurkin000/library/CE%20G3M20211%201592%2016

Its pretty hard to see the writing on top of the controller because its mostly obscured by the components/wiring.
 
amberwolf said:
Lurkin said:
- one 4 pin White weird plug. internet tells me this is for a display, nothing plugged into it. ---> White insulation opens to reveal four wires. They are red,black,yellow and blue. They all connect at the opposite end of the board to the VB+ and VB-. There is no further indicator of what they are.
Nothing on the PCB at all?

Nope, nothing.
 
Sorry, PB doesn't work on my ancient computer and browser. :( (pages come up, sort of, but without the images, except sometimes bits and pieces)

Any chance you can upload to the forum itself as attachments?
 
Lurkin said:
- one 4 pin White weird plug. internet tells me this is for a display, nothing plugged into it. ---> White insulation opens to reveal four wires. They are red,black,yellow and blue. They all connect at the opposite end of the board to the VB+ and VB-. There is no further indicator of what they are.

Then I would guess that this is a programming connector (which might also work with a display meant for that purpose, like the one Dillenger has an ad/contest for in the for-sale section).
 
20160619_151646_zps24cbiwzq.jpg

20160619_151609_zpslogrhjqs.jpg

20160619_151552_zpsiellphfx.jpg

20160619_151530_zpsteke7glk.jpg
 
I think your right about it being a programming cable fitting. it was the only plug to have a silicon covering sock over it (preventing it from being exposed at all). I think that means our other 5 pin connector is the display plug.
 
Have you tried the suggestion of jumpering 48v to SW?

If you want to verify if SW is indeed a "keyswitch/igniton" wire then you can trace it's path on the PCB. I'll bet it goes to the input of whatever low-voltage power supply that makes the 12v and 5v for the rest of the board.


FWIW, I can see placeholders for images in your previous post but there are no actual images, probably because of the Photobucket code that doesn't work on my browser.

If you upload the images as attachments to the actual post, they will work for anyone that can view images at all.
 
As attachments. PB version is more than 10X the quality.
 

Attachments

  • 20160619_151530.jpg
    20160619_151530.jpg
    28.9 KB · Views: 6,370
  • 20160619_151552.jpg
    20160619_151552.jpg
    75.3 KB · Views: 5,141
  • 20160619_151609.jpg
    20160619_151609.jpg
    83.9 KB · Views: 5,141
  • 20160619_151646.jpg
    20160619_151646.jpg
    82.7 KB · Views: 5,141
No, I haven't tried the SW to +VB. I think I might wait until tomorrow to see what, if any, response I get from the vendor.

Now that I'm more aware of what this unit should be able to do, I've become a lot more fond of it :lol: If it can do what I think it can do (or should be able to do), its a very, very capable controller for the price and what I want it for.
 
The attachments are good enough to see a lot; much better than none at all (I've tried many ways to make PB work; my computer is just too old for browsers that will work reliably with it, apparently).

Anyway, SW has a trace on the back that goes to an area that looks like a small SMPS, common in controllers these days (in place of the old giant resistors and LM317s/7812-7805s).

So it is almost certainly what you would jumper the 48v wire to to turn on the controller (though keep in mind that if it isn't, it could be possible to damage the controller).



BTW: You've never posted (here in this thread at least) where it came from, or what manufacturer you think it is. That's information that might help us help you figure it out. ;)

(essentially, any detail might be the one thing that helps someone recognize it and then know to give you the info they have...so I'd recommend posting every bit of detail and info you have about it, inside and out, including all pictures you can take of it inside and out).
 
Its a Li Shui controller. I think they use King Meter displays

https://bmsbattery.com/ebike-kit/543-36v250w-lsw-675-sine-wave-controller-with-lcd-ebike-kit.html
 
Yes, d8veh thats pretty much where I had wound up.

If that's the case (which I think it is) blue SW = switch. I think you might be right Amberwolf, that appears to be the on switch. If I connect that to 48V+, hopefully it should turn on. Do you think it needs to be touched on, or held on?
What to do with the displays RX/TX connections? Ignore?

I'm 50/50 on whether to just go ahead to see if it will work, or wait and buy one a KM529 and test. Its a tricky decision - try it and potentially fry, buy a KM529 (which I don't actually want given I have a cycle analyst I would rather use) just to get it going.... hmmm.
 
Lurkin said:
Do you think it needs to be touched on, or held on?
Held on, because if it's like most controllers, without 48v at that point on the PCB, there's no power to run the SMPS that provides 12v & 5v to the rest of the controller, and nothing can work. ;)

What to do with the displays RX/TX connections? Ignore?
Probably.
 
Back
Top