Controller Power issue

I have a 50v 18.5amp battery from cell man.
Discharge port of battery shows 55.4v it's charged enough to power a controller.

I connect the battery to the controller, then switch on the ignition,
For a brief 2 seconds my CA displays "low voltage" then the power whimpers down.
I turn it off switch the ignition back on and the same happens accept it powers on for half the time it did the last time.
If I wait 15 minutes or.more I can try again but the same thing happens. The controller does not remain powered.

What could be the issue guys?
Any help is much appreciated!

johnjapan said:

I have a 50v 18.5amp battery from cell man.
Discharge port of battery shows 55.4v it's charged enough to power a controller.

I connect the battery to the controller, then switch on the ignition,
For a brief 2 seconds my CA displays "low voltage" then the power whimpers down.
I turn it off switch the ignition back on and the same happens accept it powers on for half the time it did the last time.
If I wait 15 minutes or.more I can try again but the same thing happens. The controller does not remain powered.

What could be the issue guys?
Any help is much appreciated!

Click to expand...

Hello Japan John,

What is the low voltage cut off(LVC) on your cycle analyst?

Which version of a cycle analyst do you have?

What happens if you hold left on your cycle analyst when turning the ignition on? Same question to holding right?

You might also disconnect the cycle analyst and see if the controller is powered (connected to a motor to test this?) otherwise.

If you have anything drawing power from the cycle analyst (some lights?) or anything potentially drawing power, you might remove it from the equation.

I purchased a battery from EM3EV and it came with pre-charge (three leads), I observed a similar situation (as yours)if I had anything drawing power (such as a light) and trying to turn on the controller to observe the pre-charge voltage rising. Does your battery have three power leads? Do you have all three leads connected to your controller?

bowlofsalad said:

johnjapan said:

I have a 50v 18.5amp battery from cell man.
Discharge port of battery shows 55.4v it's charged enough to power a controller.

I connect the battery to the controller, then switch on the ignition,
For a brief 2 seconds my CA displays "low voltage" then the power whimpers down.
I turn it off switch the ignition back on and the same happens accept it powers on for half the time it did the last time.
If I wait 15 minutes or.more I can try again but the same thing happens. The controller does not remain powered.

What could be the issue guys?
Any help is much appreciated!

Click to expand...

Hello Japan John,

What is the low voltage cut off(LVC) on your cycle analyst?

Which version of a cycle analyst do you have?

What happens if you hold left on your cycle analyst when turning the ignition on? Same question to holding right?

You might also disconnect the cycle analyst and see if the controller is powered (connected to a motor to test this?) otherwise.

If you have anything drawing power from the cycle analyst (some lights?) or anything potentially drawing power, you might remove it from the equation.

I purchased a battery from EM3EV and it came with pre-charge (three leads), I observed a similar situation (as yours)if I had anything drawing power (such as a light) and trying to turn on the controller to observe the pre-charge voltage rising. Does your battery have three power leads? Do you have all three leads connected to your controller?

Click to expand...

I have Cycle Analyst V3
The LVC is unknown since I don't know of any way to check that with out power to the CA
Nothing else is drawing power ie, lights etc. Just the CA
Holding the left or right button down on the CA while switching the ignition on does nothing.
I have the same 3 leads you mentioned with the pre-charge, all 3 leads are connected.
The pre-charge voltage rises like you mentioned when the ignition switch is turned OFF. It goes down when turned ON.

How does one confirm the controller is powered without the CA ?

johnjapan said:

I have Cycle Analyst V3
The LVC is unknown since I don't know of any way to check that with out power to the CA
Nothing else is drawing power ie, lights etc. Just the CA
I have the same 3 leads you mentioned with the pre-charge, all 3 leads are connected.
The pre-charge voltage rises like you mentioned when the ignition switch is turned OFF. It goes down when turned ON.

How does one confirm the controller is powered without the CA ?

Click to expand...

Connect the throttle and the motor to the controller, turn on the controller, turn the throttle.

Alright well, this sounds good, bad but good for troubleshooting.

This could be something as simple as a bad connection, and as complicated as an incomplete solder job. Are your connections between the controller and battery anderson power pole?

I suppose a few more safe suggestions would be to be 100% absolutely certain that all three connections between the battery and controller are good.

I am guessing you have and know how to use a multimeter based on some of your words. Perform a continuity test, look at the connections, make sure everything looks correct and correctly seated. If nothing looks amiss, I might suggest opening the controller (it's not hard, disconnect the battery first, of course, and just don't touch the capacitors as they may have some voltage in them, a bit of a shock warning) and perform a continuity test between each of the leads between, precharge, ground and power. It should be pretty easy to identify which goes to which. My theory is that, maybe somehow, the power lead (red) isn't getting power to the controller.

bowlofsalad said:

johnjapan said:

I have Cycle Analyst V3
The LVC is unknown since I don't know of any way to check that with out power to the CA
Nothing else is drawing power ie, lights etc. Just the CA
I have the same 3 leads you mentioned with the pre-charge, all 3 leads are connected.
The pre-charge voltage rises like you mentioned when the ignition switch is turned OFF. It goes down when turned ON.

How does one confirm the controller is powered without the CA ?

Click to expand...

Connect the throttle and the motor to the controller, turn on the controller, turn the throttle.

Alright well, this sounds good, bad but good for troubleshooting.

This could be something as simple as a bad connection, and as complicated as an incomplete solder job. Are your connections between the controller and battery anderson power pole?

I suppose a few more safe suggestions would be to be 100% absolutely certain that all three connections between the battery and controller are good.

I am guessing you have and know how to use a multimeter based on some of your words. Perform a continuity test, look at the connections, make sure everything looks correct and correctly seated. If nothing looks amiss, I might suggest opening the controller (it's not hard, disconnect the battery first, of course, and just don't touch the capacitors as they may have some voltage in them, a bit of a shock warning) and perform a continuity test between each of the leads between, precharge, ground and power. It should be pretty easy to identify which goes to which. My theory is that, maybe somehow, the power lead (red) isn't getting power to the controller.

Click to expand...

I connected the throttle and the motor and no power is present, no motor movement whatsoever,
I will confirm the anderson connectors.
Continuity tests ill do last.

Thank you.

Here are the continuity tests I performed with the multimeter after opening the controller.

1. Red wire test
Probe connected from here View attachment 4 to here BEEEP !

2. Black wire test
Probe connected from here to here BEEEP !

3. White wire test * RED *
Probe connected from here to here View attachment 1 BEEEP !

4. White wire test 2 * BLACK *
Probe connected from here to here View attachment 2 NO BEEEP !


Test number 4 failed not sure what that means but no continuity there.

Red and white go to the power of the controller, red is power, white is precharge, black is ground. White and black shouldn't give a beep, I hope this makes sense.

You've checked the voltage of all three leads from the battery for voltage, right? I'd be surprised if you haven't, but I'd prefer to ask instead of think you might not have.

The only other things I might suggest is a discharge test using a watt meter (example: http://www.amazon.com/gp/product/B001B6N2WK/ref=as_li_tl?ie=UTF8&camp=1789&creative=390957&creativeASIN=B001B6N2WK&linkCode=as2&tag=xbnijgbr-20&linkId=UXUAEN6HK7P2DI4S) and something like light bulbs for the load. Also, another power supply of some kind, something that can deliver at least 24v and a few amps, hooked up to the controller to see if this somehow magically powers the controller. There could be something wrong with the battery that I don't understand. Maybe there is something wrong with the BMS, but diagnosing that sort of issue is a mystery to me.

However, maybe there is something wrong with the controller somewhere. I don't know if you can look around at the various soldered points to check if everything looks right, but that'd be hard to know for certain what is what in my mind. In other words, I am a little unsure other than maybe to say the controller itself is broken. I don't know if taking full pictures of the PCB are going to enable anyone to diagnose anything, I'd be skeptical. Maybe someone else has a guess or insight into something that didn't cross my mind.

What is the maximum voltage this pack is supposed to reach?

bowlofsalad said:

Red and white go to the power of the controller, red is power, white is precharge, black is ground. White and black shouldn't give a beep, I hope this makes sense.

You've checked the voltage of all three leads from the battery for voltage, right? I'd be surprised if you haven't, but I'd prefer to ask instead of think you might not have.

The only other things I might suggest is a discharge test using a watt meter (example: http://www.amazon.com/High-Precision-Power-Meter-Analyzer/dp/B00C596UIA/) and something like light bulbs for the load. Also, another power supply of some kind, something that can deliver at least 24v and a few amps, hooked up to the controller to see if this somehow magically powers the controller. There could be something wrong with the battery that I don't understand. Maybe there is something wrong with the BMS, but diagnosing that sort of issue is a mystery to me.

However, maybe there is something wrong with the controller somewhere. I don't know if you can look around at the various soldered points to check if everything looks right, but that'd be hard to know for certain what is what in my mind. In other words, I am a little unsure other than maybe to say the controller itself is broken. I don't know if taking full pictures of the PCB are going to enable anyone to diagnose anything, I'd be skeptical. Maybe someone else has a guess or insight into something that didn't cross my mind.

What is the maximum voltage this pack is supposed to reach?

Click to expand...

The Maximum voltage is 58.8v but because this battery has been used for just over a year a full charge gets to 58.0v right now its 75%-80% charged

Here are photos of voltages taken from the battery

Charger Port


Pre Charge and Red


Red and Black


Pre Charge and Black


Can you see anything wrong ? Does this look normal ?

Great pictures, I have to spare a moment to note my appreciation of your skills, will and ability involving several things here.

Getting voltage from the leads as we see suggests that things, to some degree, are functioning as they should with the battery. I think a load test with a watt meter might yield much more definitive results for the battery and BMS. The difference in voltage between the charge and discharge leads seems unusual to me, I just checked my em3ev packs leads but got the same voltage from both sets. It's hard to say that isn't normal, but it really feels like that should not be that way, but I can't come up with a reason as to why the variation in voltage would be that way. Does your pack have balance leads? Have you checked the balance of the cells of your pack?

I imagine you have other power supplies sitting around that you could feed power to the controller. Performing tests on either one of these will give much more confidence in rounding up which is the culprit of this problem. If you power the controller with a known working power supply and then you are unable to get the controller to turn the motor, then you might have a fair level of confidence in knowing the issue relates to the controller. I imagine it's a safe guess that you don't have any other controllers to test with this battery and motor? Swapping the controller and finding everything else works as it should would give an obvious statement.

Nothing looks or smells burned concerning the controller? I have to admit, I was under the impression this was a brand new setup for some reason, silly me. So, maybe some high quality pictures of the PCB may be useful to someone, especially if something looks or smells burned or anything to that effect.

Thanks bowlofsalad appreciate the compliment, I'm far from my goal of being an Ebike technician.


No PSU or other power source to test the controller
Even if I did, what minimum functional voltage do I need for a.Controller purely for testing purposes?
I don't have the device you mentioned to test battery workload over time.

Will a 12v 400w psu do?

johnjapan said:

Thanks bowlofsalad appreciate the compliment, I'm far from my goal of being an Ebike technician.


No PSU or other power source to test the controller
Even if I did, what minimum functional voltage do I need for a.Controller purely for testing purposes?
I don't have the device you mentioned to test battery workload over time.

Will a 12v 400w psu do?

Click to expand...

I don't know the minimum, but trying 12v probably won't hurt anything.

You could probably get some alligator clip type connectors from a hardware store and wire up a link between the anderson powerpole connectors from your battery to a strand of some lights, maybe totalling around 300w worth of lights, but much less would be fine (even just a 60w incandescent bulb would be good enough for this). Let that run for 30 seconds or so and see how it goes. You might also put your multimeter somewhere in there to observe the voltage. If the lights continually work while connected can probably assume that the controller is at fault. Be sure to do this discharge test with the red and black leads involving the set of three leads. A stand alone, plug into the wall, type lighting with a lightbulb or two would be good for this, you could use the plug as it is to hook the alligator clips to. I am assuming you aren't an idiot when I give you this suggestion in that you aren't going to try and lick the leads or something. If you are worried at all about getting shocked, dielectric or rubber gloves would go far in protecting your hands. I don't think there is any concern with polarity needed for this setup.

If the battery is good, you might consider simply buying a new controller, you could continue to try and troubleshoot and fix the problem with the current controller as well, maybe you'll fix it and have a spare.

http://www.amazon.co.jp/gp/aw/d/B00456XC6Y/ref=mp_s_a_1_6?qid=1415356149&sr=8-6&pi=AC_SX110_SY165_QL70

Bought this power supply to test the controller.

I got a hold of Paul cell man and he suspects it's a BMS issue in the battery so I have to run some tests by checking.voltages
Of.each of.the cells.

Will get back on any progress Thanks bowlofsalad.

To me, it sounds like the battery's BMS is cutting off output for some reason. Often it does not cut off input when this happens, so you can put a load (controller) on the charger input of the battery and see if it shows significant sag, or other problems. (the A123 EM3EV pack I have has separate charge and discharge connectors, and the one is unaffected by the state of the other, AFAICT).

The first thing I'd test is that the charger is actually charging it up fully, becasue I've had enough connector problems on chargers to cause this kind of problem to suspect it first.

Next would be testing teh cells, measuring at the BMS sense wires at the input to the BMS board, if you have access to that. If there are any cells with a significant difference in voltage from all teh others, it could be a bad cell or group, or a problem with the BMS channel stuck in drain mode and draining a cell down, if the problem happens over a longer time just sitting there not hooked up or in use, but doesnt' happen right off the charger.





FWIW:

A CA's min voltage is often around 15V. YOu can change taht in it's menu, if you like.

But really, all you need is two 9V block batteries, which you can connect together in series, offset physically by plugging the - of one into the + of the other, leaving the other terminal of each unconnected. Then use alligator clips (or your meter's test leads, unplugged from teh meter) to connect themto the CA's power leads.

The 20V or so you'll get out ofthat shoudl turnon teh CA just fine, and let you access it's settings, if you need to.

However, it's probably not enough to run your controller (definitely not enough to test a motor), as the current output of 9V's is not very high, and the controller's LVC is probably much higher (40V+).




BTW, to use that IBM PSU, you may have to put a resistive load that will cause a draw of up to half an amp on it's outputs to get it to turn on properly. Sometimes it can be on teh 12V line, somtiems on the 5v or 3.3V. On the motherboard ATX connector, you'll also need to short the green wire to any black wire, if it usesthe standard colors, to turn it on at all. There are Instructables and other pages out there that tell how to turn a computer PSU into a lab PSU; One of them'll say how to do it if yours doens't work the way I suggest.

I think you're making the diagnosis more difficult than it is. What about connecting your battery directly to the controller without the CA? Put a voltmeter on across the battery wires. If it goes to zero volts as soon as you connect, you've probably got a short in your controller. If it goes to zero volts when you switch on the controller, it's most likely a battery problem.

Some batteries will show voltage on the output wires when the BMS is switched off. This is only minimal leakage charge, which can't provide any current, so as soon as you connect it to anything, it goes to zero. When it's like that, you normally see a slight difference in the voltage between the charge and discharge connections - assuming your BMS has separate charge FET/s.

When your battery is fully charged, it should more or less match the charger voltage if you measure it immediately after disconnecting the charger. If it's more than 2v difference, you can expect to find a bad cell in your battery, which could cause the BMS to shut off discharge.

Well ive been busy as hell which is why i haven't been on here in awhile but Ive made some progress.
Here is my IBM power supply


I cut the 12v wires yellow and black here and attached the alligator clips.


Then attached the alligator clips on the other end to the black and red controller connectors.


The end result booted the CA and displayed LOW V
With the battery fully charged I couldn't even get this far.



With the battery already fully charged and left on the charger to give the battery a chance to balance the cells etc, I checked the voltage of the discharge ports on the battery and got a reading of 53.2V Waaaaay to low. Fully charged it should be around 58V

The issue is no doubt the battery guys, Paul has sent me some documents on how to check for BMS issues will get back to you guys once I have done that thanks for the assistance.

PS: I must admit guys, I am terrified of even touching unscrewing or checking a BMS on a battery which is what I have to do but it looks like I have no other choice, wish me luck.

I had a similar problem with my Adappto controller and 6s lipo packs, which happened after I washed the bike and water got into the battery box, Im not sure if some of the batteries got a mild short, as the voltage was low, but not below the low voltage cut off, and the controller is sealed so shouldnt effect that, and was no where near the water anyway, but after leaving it to dry overnight it worked fine the next day, so not sure if you have some shorted cells that arnt giving the right current under load or something along those lines.

you should not have to unscrew anything on the BMS to check the cell voltages.

use 20V DC scale, and put the voltmeter probes into the back of the plug to make contact with the metal clip inside and two adjacent pins on the plug will give you cell voltage for the cell it spans.

dnmun said:

you should not have to unscrew anything on the BMS to check the cell voltages.

use 20V DC scale, and put the voltmeter probes into the back of the plug to make contact with the metal clip inside and two adjacent pins on the plug will give you cell voltage for the cell it spans.

Click to expand...

Are you sure?

Let me get back to you with a picture of Paul cellmans battery.

dnmun said:

you should not have to unscrew anything on the BMS to check the cell voltages.

use 20V DC scale, and put the voltmeter probes into the back of the plug to make contact with the metal clip inside and two adjacent pins on the plug will give you cell voltage for the cell it spans.

Click to expand...

What I mean by unscrewing is this


This is a view of the back of the battery, it is the only accessible area that shows screws.
I assume I have to remove the yellow packaging on the back and remove the screws shown here to access the BMS but I am not absolutely sure, that's what I feel uneasy about.
I think you're referring to messing around with the cells and what not, I haven't had any results that would merrit any testing at that phase.

Paul has instructed me to check the voltages from the bms and my problem is how do I safely access the BMS which screws do I remove etc. I'm assuming its the screws I circled above.
There aren't instructions for that.

that looks like the heat sink on the top of the BMS. the sense wire plug is on the end. the yellow stuff is kapton tape i assume or similar type of wrap. you have to cut the black shrink wrap away from the BMS to get it out. no need to unscrew those screws, which won't unscrew anyway because they have glue on the threads.

you wanna measure the cell voltages on the sense wire plug.

More progress has been made,

Here is how the battery is oriented so you can get an idea of how I'm looking/ working on it.


Here is a screenshot of the multimeter settings used.


And here is a zoomed in pic of the BMS showing the balance wires where you measure individual cell voltages.
Keep in mind I charged the battery pack as much as I could and after it was charged full I left it for an additional 8 hours.

In the picture you'll see the balance wires numbered.
Touching the red probe on 15 and the other on 14 I measured 4.2 or 4.3V
I progressed to the next pair 14 and 13 and got the same results
I repeated this method all the way until 5 and 4 that's where I ran into problems.

The yellow circle indicates all the voltage measurements 5-4, 4-3, 3-2, and 2-1 all read 3.5V which is significantly lower than the others.

Where I go from here I am not sure...
BMS issue or bad cells ?

you will have to use the 20V DC scale to measure so you can measure to .01V in order to know what the BMS is doing. you will have to expose the BMS so we can see how it works to tell you where to measure. it should charge all of the cells to the same voltage if your charger puts out a high enuff voltage.

if you have no intention of exposing the BMS you can leave it on the charger for as long as you have patience and see what happens.

dnmun said:

you will have to use the 20V DC scale to measure so you can measure to .01V in order to know what the BMS is doing. you will have to expose the BMS so we can see how it works to tell you where to measure. it should charge all of the cells to the same voltage if your charger puts out a high enuff voltage.

if you have no intention of exposing the BMS you can leave it on the charger for as long as you have patience and see what happens.

Click to expand...

Here is a picture of the BMS.


I need to measure using the 20V setting?
Does it have to be that accurate?

The measurements i am getting are to the nearest 10th of a volt do I really need the nearest 100th of a volt ?
The results I gave below for the voltages should say enough about what the problem is.
I always start with 200V for safety precautions but if i need to do the measurements all over again ....

I have re measured the voltages using the 20V setting on my multimeter here are the results starting from pin 15

4.19,
4.18,
4.18,
4.19,
4.18,
4.18,
4.18,
4.18,
4.18,
4.18,
3.46,
3.47,
3.47,
3.47,

Pretty much the same result as last time
Those last for cells seem low still.
Any ideas?

the BMS is shutting off the charge during the balancing because the higher cells are forced up to the HVC which shuts off the BMS charging mosfet.

it is hard to tell what you have since it looks like the BMS is for 16 cells but only 14 channel measurements.

to force the BMS to allow the uncharged cells down at 3.5V you will have to put a power resistor across the battery on all of those high voltage channels in order to drain them all down to the same level as the uncharged group.

this will take a long time and require some care.

i recommend you find some of the big ceramic power resistors and solder them into a series capable of handling the discharge or if you have a radiant space heater or oil filled electric radiator type heater you can use to discharge the higher cells. it is best to discharge them all together so you have the maximum voltage to force more charge out of the high cells.

if you decide to do this then we can show you how to connect the heater or resistors, (the type and size to use), to the cells in the battery.