Li battery cutting out

[Deleted member 4366]

From the photos, as far as I can see because they're still a bit blurred, your controller is identical to mine in that corner. To save writing it all again, there's an explanation here in post#5.
http://www.pedelecs.co.uk/forum/electric-bicycles/13365-problem-ku123-controller.html
Looking at your photos, the two pads at ZX are not joined, which makes it 48v. For 36v they need to be joined, which means soldering a blob between them or soldering a short piece of wire across them.
You might want to check that there's nothing joining them on the other side.
The latest KU123s have two red wires soldered to those pads, which have connectors outside the controller so that you can join them for 36v or unjoin them for 48v. I removed them from mine and just put a blob of solder on, but the soldering wasn't as as good as it looked, which is why I had the problem. It worked at the second attempt.

Assuming that this will fix your problem, report back on your new range. I think it'll be going up to about 40 miles.

 

[jhinshel]

OK, ZX pads were already connected with a blob of solder. I replaced this with a tiny piece of wire and resoldered. Then I rode with a charged battery for about 2 miles. The battery completely cut out going up a steep hill and never cut back in. Before it would cut out and cut back in a few seconds later. The LEDs were all illuminated, but the motor could not be engaged. I checked all the connections and charged the battery up again, but the bike is now unresponsive. It seems sensible for me to try a new controller.

 

[Deleted member 4366]

OK, ZX pads were already connected with a blob of solder. I replaced this with a tiny piece of wire and resoldered. Then I rode with a charged battery for about 2 miles. The battery completely cut out going up a steep hill and never cut back in. Before it would cut out and cut back in a few seconds later. The LEDs were all illuminated, but the motor could not be engaged. I checked all the connections and charged the battery up again, but the bike is now unresponsive. It seems sensible for me to try a new controller.

I don't understand. You photos show them very clearly not joined, or were they joined on the back. Could you post a photo of what you did?
Assuming that the pads were already joined, putting a piece of wire across then can't change anything, so the new cut-out could be completely unrelated, because you disconnected everything. Disconnect the brakes and PAS and try it again. Also have a very close look at the thin red ignition wire. Is it still properly energised with 36v?

 

[jhinshel]

The pads were joined from the back with a blob of solder. All i did was replace to remove this solder, and replace it with a small piece of wire which extended from one pad to the other. The copper wire helps to retain the solder. This was all quite neat and I would not expect that this caused an issue. The bike worked fine for a couple of miles.
It seems that the previous issue was with the controller. I would like to acquire another controller to get the bike working. Could you help with this? If so, please you could let me know how to proceed.

 

[Deleted member 4366]

I'd be very surprised if there's anything wrong with the controller. You should do some simple tests before you write it off.

Check whether you have 5v on the throttle or PAS red wire. Did yoy try it with brakes and PAS disconnected?

My new KU123 is £35 including postage if you want it.

 

[jhinshel]

I had a look at the controller - there isn't very much clearance between the rail in the casing and the blob of solder on the underside of the circuit board. It is possible that a connection was made between the casing and the blob. In any event, I cannot get the controller to energise the motor even though there is power to the controller. There is a red flashing led in the middle of the circuit board when the controller is power up. Is this normal.
Are there more tests that you could suggest I try?

 

[Deleted member 4366]

The red light should flash. How many flashes per second is it doing?
With it connected up, but controller out of its box. Have you got 40v on that thin red wire, next to those pads that you soldered.
Next, you need to check whether you've got 5v on the red wires at the other end of the board.

If that blob of solder shorted on the case, it wouldn't work while shorted, but shouldn't do any permanent damage because the two resistors will limit any current. Only 3.6 mA can flow.

You can ppower your motor like that, but be gentle with the throttle.

You still haven't answered my question about testing without brakes and PAS connected! You need to keep an open mind about the cause of the problem until you've found it. Controllers don't just go bad. They can always be fixed.

 

[jhinshel]

I did test without brake wire connected. I dont use PAS anyhow.
I am open to fixing the controller, but am a little frustrated as the bike is very heavy to peddle up and down these hills without a motor.
There is 40v on the thin red wire. The red wire that goes to a terminal labelled 5v has 4.7v.
The led flashes 5 times at a bit faster than once per second then pauses and repeats.

 

[Deleted member 4366]

I've never checked a KU123, but I guess it's the same as the KU63, where the LED flashes continuously at approx 0.8 second interval, so it seems that your controller's trying to tell us something, but I don't know the codes.
One more thing to check. Can you measure the voltage on the ZX pad/s. It should be between 2v and 4v, so use a low scale to get it to two decimal places if you can.

 

[jhinshel]

The voltage at ZX is 3.87.
I think I'd like to take your controller - I'll message you my contact details.

 

[jhinshel]

The new controller works fine and the bike easily does a hilly 12 mile trip.
I am not sure what the cause of either the cutting out or the sudden failure of the controller and these two may or may not be related. In any case, it would be useful to get the failure codes for the controller.
i.e. what the 5 flashes mean.
Thanks for all your help Dave.

I wonder if this issue may have been anyway related to riding the bike in such a hilly place - perhaps the controller cant take it. I run on full throttle (cruise mode) and because the hills are not gradual the current must suddenly ramp up.

 

[Deleted member 4366]

I'm glad it's sorted now.
The controller can take it. It has built in thermal protection and over-current protection. My hills are worse than yours, and I've never had a problem.

 

[NZgeek]

Thread Resurrection!

I've recently had a VERY similar problem! So I thought I'd add it here, as it's in a similar vein, and may help in the future... (sounds like I may need a new controller though)

I haven't paid much attention to the details yet... because the sudden loss of power always happens in an awkward spot.

I have a Goldenmotor 36V 12AH LiFePO4 battery, and it used to last MUCH longer. I used to take a different path home, but I forgot to charge at work one day, and it ran out half way home - understandable. It was a demo battery when I got it, but I've only had it a few months, and it's done about 700 kms, and 70 charges. It charges up to about 43.6V happily.

The motor is a 500W direct drive brushless with sensors (though if anyone asks, it's only 300W!). The controller is what seems to be one of those very common chinese ones - as shown, but in a black case. It's from Vpower.hk - standard chinese stuff really.

The probelm started in similar circumstances to this thread.

I was riding home one day - I had always made it home on a full charge previously - but this day I stopped at a gas station to pump up my tyres and left the key on (not as long as the original OP), while I chatted to a guy about the bike for about 20-30 minutes. I did eventually notice that the LED was blinking, though I haven't noted the "code" yet. I honestly have no idea if it ever blinked before - I've never looked!

Since that day, the bike cuts out on the final hill on my way home, EVERY day, within about 100 metres of each time. I've even tried using a bit less assist (Thumb throttle) to save for the last hill, but it hasn't seemed to help.

I have a failry flat ride for the first half of the way home, but there are 3 steep hills. One is a fairly long incline not too steep, then a gentle downhill for a couple of minutes with some assist used here and there. Then there are two STEEP hils, one after the other - only about 10-15 seconds of downhill, with no assist (an image of the first hill is in my introduction thread: http://www.pedelecs.co.uk/forum/introduce-yourself/16082-long-time-lurker-first-time-poster.html.

It's the second of these steep hills it cuts out on.

I'm thinking I may have the exact same problem!

Next time I ride, I'll have to check if ALL power is cut (suggesting the battery) or if the power light stays on, but there is no assist (suggesting the controller). I think the power light stays on... but the caps in the controller keep that light on if I pull the battery connector anyway - for a short while at least.

Also - I THINK it seems to have more drag when not powered than it used to - I might just be used to the assist now though. It still coasts down steep hills at similar speeds, but once it cuts out, it feels like there is a LOT of drag... I might be dreaming that though! When it cuts out on the steep hill, my speed drops from about 15km/h (9mph) to 7km/h (4mph) and Then on the gentle decline, I feel like I'm still pedalling hard. On steep downhills, there seems to be no drag.

Any advice?

Thanks!!!

Allan.

 

[Deleted member 4366]

It's possible that you have a partial phase wire short, so check the phade wire connectors and the wires particularly where they enter the motor.

To solve these sort of problems, you need a wattmeter between the controller and battery. Search on Ebay for the one that costs about £8. It'll show how much current and voltage is coming from the battery in real-time, which wil narrow down the possible causes.

 

[NZgeek]

It's possible that you have a partial phase wire short, so check the phade wire connectors and the wires particularly where they enter the motor.

To solve these sort of problems, you need a wattmeter between the controller and battery. Search on Ebay for the one that costs about £8. It'll show how much current and voltage is coming from the battery in real-time, which wil narrow down the possible causes.

Not sure if I should move this to a new thread now, or continue here? Maybe a Mod can move it if it should be separate (sorry!)

Anyway, Thanks!

I have ordered one of those cheap meters. It'll take a while to get here, and locally they only have stupidly expensive ones! I would like a cycle analyst, but it's quite expensive! The meter will be a good start, and I can always upgrade later.

I just got home from work (took the car today) and turned on the bike - the controller just blinks continuously whenever the throttle isn't applied - no code. I can only assume the constant blink is normal???

Spinning the wheel by hand, either fast or slow doesn't seem to offer any more resitance than normal, and seems even all the way around (not like a direct short, which can be felt). I';; try to take some measurements later.

Like I said - the drag etc could EASILY be something in my head! I can't be 100% yet.

My biggest concern is the battery and if it's failing

I'll report back anything that might be interesting!

Thanks again,

Allan.

 

[Deleted member 4366]

You can keep posting in this thread. It' keeps similar problems in the same place for anybody else searching. If you search back for the keyword "wattmeter", I've twice recently explained how to wire up one.

The wattmeter will show whether it's the battery or controller cutting the power, and under what conditions.

The blinking light is normal on many controllers. Steady blinking is OK. Any pattern is a fault.

Did you ever rise it with it switched off befor you had the problem? If so, is there a clear difference. If you get stuck without power again, disconnect the phase wires, which should make it much easier to pedal, unless there's an internal short,

 

[NZgeek]

I've left the battery charged, but disconnected for 24 hours now, and it's sitting at 42.2V today (lifepo4).

I had ridden it switched off, and the drag was minimal. To be fair though, it was just near the house, and it wasn't after riding most of the way home, so it's not a very good comparison :-(. Spinning the wheel by hand seems to have no resistance.

I did some measuring, and all seems "normal"/acceptable except one thing.

Between the 3 wires to the motor (unplugged), was .6 ohms (probably normal)

Between the controller>motor outputs and the positive battery input is open.

The result I wonder about is between the negative battery input and the motor outputs: 2 measure 10K ohms, but it's only 8K ohms on the green cable.

I'm not familiar with the inner workings of the controllers, and don't know if this is significant - it's quite different to the other two connections though.

A new controller (or fets) sounds a LOT more appealing than a new battery!

Cheers,

Allan.

 

[NZgeek]

I've been sick for 2 weeks, and on call last week, so no cycling! I did get my watt meter though! I may go for a ride tomorrow, and see what happens.

I set the wattmeter up so it runs of an external 9v battery, so will keep running after the battery shuts off. It just plugs in between my existing battery connectors so I can take it on and off when ever I want. I have a couple of metres of cable on the negative side up to the meter, but I used 4mm2 cable for the run, to reduce voltage drop (It's medical grade, for EP/Poag cables).

I'm starting to think that I must just use the assist TOO much! The few days after my last post, I tried to reduce the assist I used on the way home, and then rode up and down past my house to flatten the battery - Sometimes it took quite a while to use it up - somedays I did another Km without pedaling.

Only thing I have found so far with the Watt meter... If I run it up to speed on a stand in the garage and apply the brake, I managed to get it to read over 1000W! I wonder if climbing the hills at only about 10 MPH but basically full throttle was a bad idea On the flat it will assist me up to almost 20mph.

Anyway, I'll report back when I have some actual info

 

[NZgeek]

I used my new blue watt-meter

Of course some of the numbers aren't recorded, because I arrived home to my wife trying desperately to phone me as the front door had been smashed in at the house and all our stuff rummaged through Not much stolen though.

Anyway!

To work (never a problem):
39.54 Vm (recovers to 40.2V - I checked about 30 mins later)
33.13 Ap
1184.3 Wp
2.36 Ah
80.7 Wh

Home (didn't record all results, bue to break in):
37.13 Vm ( I saw about 35V on the screen at one stage though, it must do some averaging)
34.32 Ap
3.49 Ah

When it cut off, I stopped and voltage was close enough to 0 (I used a 9V battery as a supply to the meter, using the extra inputs to keep the display running)

So it seems that the BMS is cutting the power because a cell voltage is getting low. I'm certainly nowhere near the 12Ah capacity. These two hills are very close together and very steep, so it may be a voltage sag issue??? It usually cuts out at about the point the video stops - often within about 50 yards of the same spot! Here is a video of the hills: [video=youtube_share;NyDD60mRSCY]http://youtu.be/NyDD60mRSCY[/video]


What I found REALLY interesting is after the BMS shut off the battery, the voltage dropped (duh!), but the faster I cycled, the higher the voltage went on the meter - I'm not supposed to have regeneration, but it does seem to feed power back to the battery. I wondered if this would upset the BMS??? Or is it just stray, because power is fed back from the wheel turning, but the controller has no actual power supplied to it from the battery???

I guess the trick is to go easy on the battery on these two hills? I've tried that recently, and made it home, but I used to just go for it, and made it home too!

Thanks!

 

[Deleted member 4366]

Could you provide some more details of your battery? Does it have the large blue cylindrical Headway cells? If not, 33 amps is a lot for a 12aH LiFePO4.