Is my controller faulty??

[Smart]

Hi there

I'm a complete noob when it comes to electric bikes but I like to think I'm technically minded so please be gentle.

A friend of my wife's decided to kit out his employees with free electric bikes and decided to purchase a few mid priced bikes to give his staff to test before making a firm decision on which bike was best ( he is not technically minded and also has the luxury of not being limited by a budget ). He didn't do any research and just literally did a quick google then bought 5 bikes from the first suppliers listed on a google search. They all arrived and one bike in particular was troublesome for them to get going so rather than seeking professional help he just put it back in its box and left it in his office.
To cut a long story short, my wife saw the box and asked what it was, he said its a bike and she could have it if she wanted ( he knows I like that kind of stuff ).

So now I'm here to seek assistance as the bikes display unit doesn't stay on for more than a split second and the bike appears to do nothing. I have contacted the manufacturer and the distributor but have yet to receive a response from either of them. This bike cost £950 here in the UK but immediately after a little googling I found it was a rebranded chinese bike that wasn't worth as much as the UK pricing.
I have checked the battery is charged and followed the limited instruction manual that was in the box. I have used a multimeter to test the battery as well, which seems to be fine. I also checked the wires that connect the battery to the controller which is housed under the frame and next to the crank. All of these are showing the expected readings for a 36v battery.
I disconnected the controller unit and checked for continuity across the wires that were originally connected to battery's - + . They had continuity for a split second and would only show continuity if I waited a while before checking again and even then it was a split second of a continuity beep on my meter and then nothing.
Is the controller to blame?
Here is the link to the manufacturer product listing for the bike http://www.guewer.cn/products_en_detail.asp?id=2127&cataid=336
Here is the link to the UK retailer rebranding these bikes http://www.barnesandrobinson.com/electric-phoenix-mountain-bike-104-c.asp
I've also attached pic of the controller

 

[dnmun]

nothing wrong with the controller, the voltmeter is charging up the input capacitor.

 

[Jeremy Harris]

When you checked the battery voltage, did you try checking it with all the wires connected and the bike powered up? It may be that the battery management system (the BMS, inside the battery case) is detecting too low a voltage as soon as any load is applied and is shutting off the power. The BMS may be detecting that the whole pack voltage is too low for an instant, or just one cell inside the pack gets dragged too low under a small load.

The battery voltage for this pack should be a bit higher than the nominal 36V, as it probably has either ten LiCoO2 cells of a nominal 3.7V each (which normally should sit at around 3.75 to 3.8V off load, 4.2V just off charge) or it has 12 LiFePO4 cells of a nominal 3.2V each, (which would normally sit at around 3.25 to 3.3V off load, 3.65V just off charge). In the first case the fully charged (just off charge) pack voltage would be around 42V, in the second case the pack voltage just off charge would be around 44V. The lowest pack voltage you should see off load is around 37V if the pack is made from LiCoO2 cells, or around 38V if it's made from LiFePO4 cells. If you're getting readings that are below 36V then that indicates a potential battery problem.

 

[Smart]

Thanks for the responses.

When I checked the battery I checked it whilst disconnected from the bike and then checked it again whilst it was seated in its battery slot. Both times I got a reading of 41v. I didn't have the controller connected when testing the voltage.
I will test again this evening with controller connected.

 

[Jeremy Harris]

Sounds as if the battery itself may be OK, and as dnmun rightly said above the controller seems to be behaving as expected, so that leaves either a problem with the BMS in the battery (which you should be able to test by seeing if the battery voltage drops when the controller is turned on) or a problem with a connector, switch, or something else.

When you say the "display unit" do you mean the LEDs on the throttle, or does this bike have an LCD display/control panel? If the former then normally these LEDs are just connected to the battery supply and crudely indicate battery voltage, if the latter then there is a possibility that the display unit may have some means of causing this fault (assuming it is able to control the power in some way). Does the bike have a throttle, or is it pedelec only?

I'm afraid I won't be back on ES for a day or two now, but I'm sure others can step in and offer advice.

 

[Smart]

The display is just a series of LED's a power on/off button and a mode button. I gave tried using it without the led display connected but it made no difference.
There is no throttle, its just a pedelec.
I was hoping to retrofit a throttle (as the manufacturer site states it is just removed for the uk market but the controller has the capability) but I need to get it going first.

 

[d8veh]

You need to determine wether it's the BMS or the controller shutting down. Do what Jeremy suggests. Put your voltmeter into the battery/controller connection while it's connected. Presss the "on" button and see if there's a change in the voltage when it goes off. If no change, then the problem's on the controller side, otherwise it could be on either side. Where about in UK are you?

 

[dnmun]

JH has gone, so charge it for as long as you have time to figure out what to do. when he made up the test for you to do, he was trying to get you to determine if the battery is too low in voltage on one of the cells and it turns off the battery when you try to use power from the pack. try charging as long as you can, see if it will charge more even if the charger light is green.

you need to figure out what kinda battery you have, it is important to know. and measure the output voltage on the charger.

 

[Smart]

You need to determine wether it's the BMS or the controller shutting down. Do what Jeremy suggests. Put your voltmeter into the battery/controller connection while it's connected. Presss the "on" button and see if there's a change in the voltage when it goes off. If no change, then the problem's on the controller side, otherwise it could be on either side. Where about in UK are you?

I'm based in south london.
I'm gonna try both suggestions when I get home from work.

 

[Smart]

Ok I've done as Jeremy Harris suggested and taken a reading while everything is connected and the reading was around the 30v mark. I then pressed the power button on the handlebar display and the lights on it flashed on for a second and the current immediately dropped to 14v and slowly began working its way back up to 30v. I pressed the power button again just to make sure it wasn't a freak reading and the outcome was the same drop in voltage. Does this mean the controller is ok?
Going by Jeremy's earlier comment the battery reading is no longer within the reading it should have. I had a reading of 41v around 20 hours ago.
I will leave it on charge and take the readings again later.
Here's the battery

 

[Smart]

After leaving the battery on charge for 8 hours I was able to get a reading of 41.1v on the battery while disconnected and again while connected. With the battery connected and the controller connected once I hit the power button the reading fell to 14.3v and gradually built itself back up to 41.1v .
The lights on the display flashed for a second and that was it.
I also took a reading from the charger pack and that was 41.7v (the sticker on the pack states 42v output)

Any other tests I should try?

 

[dnmun]

ok, 41.7 and the battery says lithium-ion on the label?

that implies it is 10S of limn2o4 which is what the chinese suppliers now call a 36V pack.

i don't follow what you mean about a power switch, and what is happening. but it sounds like it was not charged originally since it read 30V and now has 41V?

what are the 'indicators' and what does it mean when you say the voltage gradually climbs back up to 41V?

 

[Smart]

The battery has a sticker on it that says its a li-polymer battery.
I have taken readings from the battery 3 times.
The first time it gave a reading of 41.1v after charging for 6 hours with the ignition switch on.
The second time roughly 20 hours later it gave a reading of 30v with the battery ignition on and dropped to roughly 12v once I pressed the power button located on the handle bars. The voltage metre reading then began to climb 0.1v a second until it reached 30v and stopped there.
The third time was after charging the battery again for more than 6hours and the reading was 41.1v with the battery ignition switch on and again once I pressed the power button on the handlebar display unit the voltage dropped but this time it dropped to 14.3v and then began to gradually build up by 0.1v a second until it reached 41.1v

 

[John in CR]

It sounds like a BMS issue to me, so the first step is to determine the charge The battery will have a connection to the BMS that connects to each of most likely 10 levels that add up to the total voltage of the battery. Measure the voltage at each. You will have to disconnect that plug to carefully measure each. What the BMS does is detect when one of those drops below a certain voltage and shuts the battery pack down to protect that cell, because if a cell gets too low it can quickly die.

At full charge the cells should each be above 4V, and they should hold that voltage overnight.

Let us know those voltages, so we can proceed.

Don't worry or be embarrassed or afraid to go at this stuff. The vast majority of us started at your knowledge level, ie little or nothing. The beautiful thing about ebikes is that they are quite simple systems. We don't have to know how the circuits in a BMS or a controller work to be able to troubleshoot them, and the rest of the drive system is quite easy. In no time you can be at a stage where there is nothing that can go wrong with your ebike that you can't diagnose and ultimately fix. The freedom of transportation that brings is quite liberating.

One tool I suggest you get for your arsenal is an ebike tester. It helps make trouble shooting quite easy, though in this case, it's probably a battery issue which only takes a multimeter to diagnose.

John

 

[Smart]

Here's the contents of the battery pack, I'm presuming the candy bar shaped sticks you can see in the hole of the blue shrink wrapping are the actual cells.
I'm not sure I'm competent enough to replace all the protective outer wrapping should I decide to test the voltage of each cell. Or is it dead simple and I'm over thinking it?

 

[d8veh]

I don't believe that the problem is in the battery. A test you could try is with the battery switched on, connect it to the controller whilst measuring the voltage. You should get sparks at the connector and maybe a slight dip in volts for a very short time. I think that there's a short in your controller, because the difference in current between when you operate the power switch and not is minimal and wouldn't be enough to trouble the BMS, but the switch activates the controller, so if there were a short there, it would trip the BMS. I'd open your controller rather than the battery, but first, try disconncting all connectors except the panel one and the battery, and then see if the lights stay on.

 

[Smart]

I don't believe that the problem is in the battery. A test you could try is with the battery switched on, connect it to the controller whilst measuring the voltage. You should get sparks at the connector and maybe a slight dip in volts for a very short time. I think that there's a short in your controller, because the difference in current between when you operate the power switch and not is minimal and wouldn't be enough to trouble the BMS, but the switch activates the controller, so if there were a short there, it would trip the BMS. I'd open your controller rather than the battery, but first, try disconncting all connectors except the panel one and the battery, and then see if the lights stay on.

Thanks I'll give it a go

 

[Smart]

http://s1301.beta.photobucket.com/user/smartstradamus1/media/VID_20121012_103827.mp4.html
Here's a vid of what happens when carrying out your test with only the display and battery connected to the controller.

 

[Jeremy Harris]

OK, I'm back from a short break now. Looking at the way the battery voltage drops with just a small load I'd say there is definitely a battery problem somewhere, as it looks very much to me as if one or more cells in the battery pack have developed a high internal resistance. This is exactly how these cells fail when they are either allowed to discharge beyond the point of no return, or if they have a manufacturing fault. It seems probable that the failure either occurred during the fairly long shipping time over from China or it may well have been shipped with a faulty battery (not at all uncommon, QC on some of the Chinese made products is far from good).

I'm afraid that your going to need to do some open-heart surgery on the battery pack to get to the bottom of the problem and may well have to replace the all the original cells with new ones, as I doubt that you'll be able to obtain spare cells of exactly the same type as were originally in the pack. This isn't as daunting as it sounds, though, and there are a few of us here who can talk you through the process.

First thing to do is to remove the packaging around the cells (just the blue heatshrink sleeving, NOT the silver wrapping around the cells themselves) so that you can get at the individual cell connections. You need to then measure the voltage of each cell in sequence, ideally when the battery is disconnected from the bike and also when it's connected, turned on and the overall battery voltage is being loaded down. I strongly suspect that you'll fine several cells in the pack that drop to a very low voltage under load, so causing the big voltage drop you're seeing.

 

[Smart]

Thanks for the advice Jeremy. I will try to build up the courage to dismantle the battery fully and check it over.
I'm torn between finding the fault and just getting a new battery.

 

[dnmun]

that is curious. to me, finding what is wrong would be more interesting than having a new battery. plus it will cost $300.

 

[Smart]

I also want to know what is wrong. It pains me to part with money for no reason, but If taking the battery apart would end up rendering it useless anyway as I won't be able to put it back together and seal it properly then I'm at a loss either way.
The bike frame has a slot that the battery slides into perfectly so I guess I will need a direct replacement.
Right now I'm just confused..
I will probably change my mind later.
What the hell... I'm gonna go for it
I'll report back later this evening on my findings

 

[dnmun]

there is nothing difficult about taking the wrapping off of the pack. i always use scissors to cut the plastic so that i can be pulling the plastic away from the inside while it cuts instead of cutting through with a knife.

just be careful of shorting out the battery on metal stuff, but that is usually just startling, not serious. we all have tools with big holes burned in them.

but you will see the BMS right away, post up a picture of it and we can give you advice if you get spooked.

 

[Smart]

Should I give it a full charge before I take it apart?

 

[Jeremy Harris]

Should I give it a full charge before I take it apart?

Probably not worth it, as the battery should still be fully charged, or as close as it can get to it in it's present condition. In some respects, measuring the cell voltages in it's present condition (i.e. not after another charge) might well tell us more about what's going on.