[mention=4809]vfr400[/mention]

Your logic seems flawed to me ....

 

Obviously I have not explained my idea well enough - let me have another go: (apologies if over simplified )

 

> The KT Controller is located in a separate compartment in the front of the battery box and the battery box is a rear rack mounted affair

> The KT LCD5 display has as part of it's feature set, the ability to enable lighting via one of the KT controller's leads: the 2-pin Julet connector labelled 'Light Output'

> As now revealed, the current available from the 'Light Output' connector is not sufficient for the lights that I have.

 

THE PLAN

> is to use the 'Light Output' as a controlling signal for something like a MOSFET or Solid State Relay and these components are to reside in the battery box's front compartment alongside the KT Controller

> The MOSFET (or SSR)

- will pick up the battery supply, via a suitably rated fuse, from the supply line feeding the controller

- the controlling signal will be disconnected from the 2-pin Julet connector lead at the PCB end and fed to the MOSFET to enable it to provide the switched lighting supply

> the switched lighting supply will be connected to the original (and redundant) 2-pin Julet connector to provide a convenient connection to the front and rear lights

 

So far then, the only additional wiring has been a small amount of wire for the internal connections between the KT Controller and the MOSFET circuitry.

 

ENHANCED PLAN

> Further, the wiring loom from the front mounted LCD5 display includes along with the brake sensor leads, a throttle lead which I do not use. When I have the controller apart, I will look at the possibility of isolating the leads for the throttle and using them to pass the switched lighting supply to the front of the bike for connection to the front lamp. I will of course need to determine if the throttle lead can take the curent for the front lamp.

 

In summary, the only extra wires/wiring will be the internal bits as described above, a short length of cable between the battery box's external 2-pin Julet connector and the rear light, and depending on whether I can usitilise the existing throttle cable or not, some other cable for the front light.

 

The only way it would make sense to me is if you were able to run the lights directly from the controller, but the power provided isn't sufficient to run a decent headlight

.

 

Functionally I think my plan achieves what you say, albeit with the addition of the MOSFET circuitry, which for me is not much of a challenge (retired electronics engineer, albeit many years ago).

Just a heads up to anyone who may be thinking of using a KT LCD3 panel to control lights.

 

Mine is fully working by replacing resistors in pcb and dropping down resistance to 25Ω. Front light is 120Lux, draws 106mA, rear light has brake function so draws around 10mA continuously and 36 mA on auto brake/impact sensor. Both work fine, no smoke no burnt resistors.

 

Still, another issue cropped up. Just realised the brake sensor cuts power to lights when used. Display keeps powered, but front and rear light switch off. Must be sharing same ground or something.

Too tired of taking the goddam box out of the craddle and tweaking with it so I'm just running with no brake cutoff for now.

But if someone knows about this and found a fix would be very grateful.

 

For those who didn't know beware. Controller is KT-6S5-C 2018-1-8

That's a strange one [mention=27403]MSG76[/mention] - I'll re-test my outfit tomorrow and see if I get the same lighting power loss on applying the brakes.

Hi [mention=27403]MSG76[/mention] - so I connected my voltmeter to the 'light output' socket on my KT controller

( T-06S 24V/36V250W Torque Simulation Sine Wave ) with a KT LCD3 display, and got battery voltage at the socket when switching on lights from the display, and the voltage remained when I operated either or both brakes.

 

Going back to your post #64 where you identified the 510 ohm resistor which you thought regulated the voltage - whilst I've not looked inside my controller and have no knowledge of yours either, looking at your picture of the PCB, my first thoughts were that the physical size of the resistor makes me think it is a current limiting resistor.

 

A current limiting function fits in with my own observations in post #67 whereby my light drawing >300mA dragged the voltage on the socket down to 5.5v

 

By paralleling 100 ohm with the 510 ohm you've now got an effective resistance of 84 ohm which has given you the extra current you wanted - but the 106mA you now have may have damaged something in the circuitry - the 510 ohm resistor is that value for some reason - presumably to limit the current to 70mA?

 

Just a thought.

( T-06S 24V/36V250W Torque Simulation Sine Wave ) with a KT LCD3 display, and got battery voltage at the socket when switching on lights from the display, and the voltage remained when I operated either or both brakes.

 

Thanks [mention=25387]Bikes4two[/mention]!

I'll have to look at it. I've reported a malfunctioning controller to the aliexpress store so they have sent another one in about one week. The second one also cuts-off power to lights.

 

Going back to your post #64 where you identified the 510 ohm resistor which you thought regulated the voltage - whilst I've not looked inside my controller and have no knowledge of yours either, looking at your picture of the PCB, my first thoughts were that the physical size of the resistor makes me think it is a current limiting resistor.

 

Yes it is a current limiting resistor

 

By paralleling 100 ohm with the 510 ohm you've now got an effective resistance of 84 ohm which has given you the extra current you wanted - but the 106mA you now have may have damaged something in the circuitry - the 510 ohm resistor is that value for some reason - presumably to limit the current to 70mA?

 

The result of the parallel 100Ω resistor along with the 51Ω smd was 33.4Ω (measured). I realised now that the calculation was mislead by me, apologies for the typo, because the code of the smd is "510" but not 510Ω - resistance of 510smd is 51Ω.

Current raised to 120mA +/- and power dissipation increased to 5W from the new ceramic parallel resistor.

 

In this second controller a through hole 51Ω was soldered so made my job easier and snipped it off replacing it by two 50Ω 5w resistors in parallel. I now have a very high wattage resistance of 25Ω!!! (2x 5w - overkill I know: https://uk.farnell.com/ohmite/45f50re/res-50r-1-5w-axial-ceramic/dp/2448958)

And a current of around 200mA. Both my front light and rear where only fitted with this new controller and the brake sensor still cuts-off power to lights when using it.

 

 

***

Actually just received a reply from technicians. Factual reason is my brake sensor is using 2 wires and controller needs 3. I made a 3 to 2 wire adapter from spare cables. So I'll have to purchase a replacement brake lever for this to work fully with controller.

 

Schematics here for future reference:

 

 

 

 

Just a thought.

Edited September 25, 20205 yr by MSG76

OK, [mention=27403]MSG76[/mention] -firstly, my 'bad' - I thought the PCB resistor was 510Ω not 51Ω !

 

So to be clear, is your 'lights going out on applying the brakes' issue down to the controller requiring 3-wire brake sensors, or a controller fault?

 

And not being clued up on brake sensors, why is it not workable to use your 2-wire brake levers - looking at the circuit diagram, I'm assuming that if you wire up the red and white to your 2-wire brake levers, that might work - (an assumption which needs confirming or clarifying by others as I'm a newbie at this game).

Edited September 25, 20205 yr by Bikes4two

Hey [mention=25387]Bikes4two[/mention] ,

yes it's down to controller requiring 3-wire brake sensor, but I thought exactly the same thing and that wiring both red and white could make it work. Although it may be a 2 way switch so when cutting off brake signal conducts 5v and vice versa. Having both wired together may just equal both power to light and brake motor signals on simultaneously.

Will have to investigate but less worried about it.

The main reason why I wanted a new controller was to have light control on both front and rear from battery/lcd to increase security. I've been riding without brake cut-off and actually don't feel a diference. Motor cuts off quite rapidly when coasting and I still haven't found myself both pedalling and braking at the same time

Hey [mention=25387]Bikes4two[/mention] ,

yes it's down to controller requiring 3-wire brake sensor, but I thought exactly the same thing and that wiring both red and white could make it work. Although it may be a 2 way switch so when cutting off brake signal conducts 5v and vice versa. Having both wired together may just equal both power to light and brake motor signals on simultaneously.

Will have to investigate but less worried about it.

The main reason why I wanted a new controller was to have light control on both front and rear from battery/lcd to increase security. I've been riding without brake cut-off and actually don't feel a diference. Motor cuts off quite rapidly when coasting and I still haven't found myself both pedalling and braking at the same time

I'm not sure what you're saying, but if you have a three wire brake connector and you want to connect a 2-wire brake, it should be connected to the signal and ground wires, not the red 5v.

 

The brake signal wire is connected to the controller's CPU. It's held at 5v by a pull-up resistor, and the cpu will give power as long as it's high. When you operate the brake, the switch shorts it to ground, which pulls it down to 0v. When the cpu sees a low signal, it cuts the power. It's as simple as that - a binary signal on one of the cpu's digital inputs.

I'm not sure what you're saying, but if you have a three wire brake connector and you want to connect a 2-wire brake, it should be connected to the signal and ground wires, not the red 5v.

 

The brake signal wire is connected to the controller's CPU. It's held at 5v by a pull-up resistor, and the cpu will give power as long as it's high. When you operate the brake, the switch shorts it to ground, which pulls it down to 0v. When the cpu sees a low signal, it cuts the power. It's as simple as that - a binary signal on one of the cpu's digital inputs.

 

 

Yep, it's currently only connected as ground to ground (black) and signal to signal (white). The red 5v was left off disconnected.

If I ever connect the red 5v to brake signal it won't cut off as 5v will always flow, right?

So with current 2 to 3 wiring brake works but when cutting off to 0v also cuts power to light circuit. This indicates a 2 way switch may be operating when shorting brake signal to ground by opening the 5v red connection back into the circuit thus feeding throttle and lights.

What I may do is search for a bypass in the controller at the point cables are soldered on the board. But will have to look into it

Edited September 26, 20205 yr by MSG76

Yep, it's currently only connected as ground to ground (black) and signal to signal (white). The red 5v was left off disconnected.

If I ever connect the red 5v to brake signal it won't cut off as 5v will always flow, right?

So with current 2 to 3 wiring brake works but when cutting off to 0v also cuts power to light circuit. This indicates a 2 way switch may be operating when shorting brake signal to ground by opening the 5v red connection back into the circuit thus feeding throttle and lights.

What I may do is search for a bypass in the controller at the point cables are soldered on the board. But will have to look into it

I don’t understand your logic. No current flows (you said voltage). Its just voltage sensing. It's very simple. The signal wire is connected directly to the cpu with a resistor between for protection. The cpu senses whether it's 5v or 0v. The switch only has 2 wires, which are ground and signal. There's a resistor between the signal and the 5v rail that holds it at 5v when the switch is off.

 

In all the recent controllers that I've studdied, the lighting is worked by data. Data comes down the Tx line from the LCD into the Rx on the cpu. The cpu interprets that request and switches one of its digital outputs from 0v to 5v. The 5v then switches a transistor that switches on the lights. When that output goes low, the transistor switches off and the lights go off.

 

I've not seen one with a separate pcb like yours has. Without being able to see where the wires go, I can't comment on it. All I can figure out from the limited info is that the transister at the bottom is switching the mosfet at the top, which is most likely powering your lights, in which case it should be able to supply all the power you need without modification, i.e. 1 amp at 36v.

Edited September 26, 20205 yr by vfr400

Reading through this thread it seems as if you guys are trying to run your lights directly or indirectly through your controller which imo is a bad idea even given they may well have that function built-in.

If you have a problem or short circuit with you lights it could screw your controller, likewise a controller fault could render you lighting system useless. You'll be far better off running lights directly from the battery independent of any controller interaction.

In my own case I fitted a throttle which had one of those naff key switches not needed as I already have a separate power switch. I removed the key mechanism and replaced it with a push button switch that had the bonus of a LED inside so I could see if my lights were switched on accidently during the day. I used a thin multi-core cable doubling up the leads to handle any high current, the indicator LED works on it's own circuit direct from the battery via a dropper resistor and utilising one of the switching poles inside the switch itself.

The lighting circuit and the LED both have their own earth -v return hence the multi-core lead (don't use the negative on the throttle a mistake many make ) and it all runs completely independent of the bike control system. Works perfect.

The controller has a separate mini pcb circuit with a current regulator. It won't affect the main board. The current regulator issue was already superseded with a higher current high power capable solution

As that mini pcb board is internally connected to the controller circuits it could still cause a problem if a short were to develop. The only way to stop that happening would be by opto isolation which wouldn't be worth the trouble, besides the regulator should give short circuit protection anyway but can't be guaranteed to work.

A dedicated line from battery to lights as vfr has also pointed to is safer & more efficient all round.

A direct line to the battery is equally potentially dangerous depending on the bms and max discharge of battery. Let's say both battery max discharge and controller max current match. A cyclist, unaware, following your advice may add a power line to draw extra amperage from the battery and cause overheating or explosion of the battery cells. So in terms of potential danger we both can go as far as the imagination lets us.

In practice and looking at facts, the controller in question with a separate pcb circuit is not a liability. If for any reason get's damaged I know how to fix it, so again, it's not a liability.

I must confess I've already shorted my lighting circuit by accident and it damaged nothing, a simple fuse would of course sort out any overload problems but I haven't bothered going down that route myself.

There's no reason why having a lighting control on a separate board would be a liability but it's obvious (to me at least) that Kunteng added it as an afterthought otherwise it would be on the main board, room could be found I'm sure. Another advantage of having power from battery to light via a switch on the bars is I have full potential right to hand that I can use for any device in the cockpit as it were. One idea I had was to fit micro LED's inside my gear changers on the bars so I can see what gear I'm in when it's dark, I've seen that done before but I could do it much better with SMD LED's. Now there's a use for a lighting circuit !

That's a really good idea [mention=21727]RossG[/mention] !

When I finally get round to doing it I'll be using wire like this, it's very fine even more so in the flesh. On my bike which is a folder all the cables are covered or concealed so it doesn't matter what colour the wire is it can't be seen, I'm guessing you can get it in black and attach it to any existing cabling by wrapping or gluing. Current drawn is very low so a tiny resistor to bring the voltage down could be soldered to the LED.

Surface mount LED's are tricky to handle so I'd suggest using an LED from some of those bead lights you can get cheaply from pound stores and I would go for, red less in your face.

My bike already has a USB 5v power socket for charging a phone but that's another thing you could use a supply for.

 

https://www.ebay.co.uk/itm/Dolls-house-50FT-Double-Wire-Electric-Wiring-Lighting-DIY-Miniature-Lights-LGW/392872875865?hash=item5b790c5759:g:awcAAOSwfi9ZrFVX

I've just fitted a feed from the battery output, replaced the female bullet connectors with twin ones, a 3amp micro blade fuse in rubber holder and used Wago connectors (available in 2, 3 and 5 wire, so can add accessories on the handlebars.) I've fitted a USB socket and a cheap and cheerful light/horn.

Running a bbs01b with HL battery.

 

I also took advantage of the British Cycling offer I came across here, join and get a free set of see.sense ace lights, I did and they are amazing. I'm not sure I'd have paid £70/£80 for them. I reckon I still need a battery-powered headlight of some sort though, for dark canal towpath etc. Hence fitting the power feed. I might splash out on one of the proper non-dazzling headlight though, anyone have any recommendations?