Ebike Shuts Down Under Load or High RPMs Without Load

Joined
Aug 30, 2019
Messages
4
EBike Specs:
I have custom built a mid drive ebike with a frame mounted motor and controller (components linked at bottom). I built a 48V 13S battery pack with 3 cells in parallel. A 48V BMS is installed on the battery. The battery is connected to the controller with an XT60 connector. All the battery components including the compatible 48V charger were purchased from VRUZEND.com. I installed a twist throttle that came with the motor and controller.

Issue:
When I ride the bike and open the throttle, the motor drives the rear wheel, accelerating the bike, but as soon as the bike reaches a few MPH all power is lost and the system shuts down. The motor stops and the throttle display fades to black, suggesting that the battery as fully stopped supplying current. If I very lightly feather the throttle, I can sustain the power nearly indefinitely, but if the throttle opens too much, only a fraction of its full range, the bike loses power. When I raise the back wheel of the bike off the ground and open the throttle, the back wheel reaches very high RPMs and can be sustained indefinitely with almost a fully opened throttle, but the bike does shut down in a few seconds if the throttle is fully opened. In the both the test with load and without load, I can use the system again by disconnecting and reconnecting the connector between the battery and controller.

Current Theories and Questions:
Because the power is lost at a certain level of throttle use, it seems the issue could be related to amp draw, suggesting that the BMS may be shutting down the battery when too many amps are being drawn. The BMS is rated for 30A continuous draw while the controller is rated for 33A continuous draw. I have heard that this should not be a major issue, because the BMS should be able to temporarily handle 33A and all 33A are not being drawn continuously for general use. Is this reasonable? What are the most effective methods of measuring amp draw from the battery? Could I install a resistor between the connectors and then measure the voltage drop over the resistor? Will a general multimeter be able to measure between the positive terminal and positive lead of the battery measure the draw?
I have measured the voltage of the battery as the motor is running with the wheel off the ground and the power is lost with the voltage dropping from around 47V to 45-46V which I do not believe is enough for the BMS to shut down the battery, correct?
What are the other variables that may cause this issue? Is there a possibility of the electrical components in the controller having issues? Is it possible that the connectors are not the right gauge to support the current coming from the battery? I would appreciate any thoughts on this issue.

Components:
Motor/Controller/Throttle- https://www.amazon.com/Electric-Brushless-Complete-Controller-Motorcycle/dp/B07G2LZHD8/ref=sr_1_4?keywords=kunray+ray+ebike+motor&qid=1567786124&s=gateway&sr=8-4

Battery Cells- https://vruzend.com/product/set-of-two-2-samsung-inr18650-30q-3000-mah-18650-battery-cells/

BMS- https://vruzend.com/product/48v-13s-battery-management-system-bms/

Charger- https://vruzend.com/product/48v-lithium-ion-battery-charger-3-amps/

Connector: https://vruzend.com/product/xt60-connectors-with-14awg-silicone-wire-leads/
 
I think it's unlikely you are exceeding 30A. In addition to the BMS and Controller, the problem could be a poor connection anywhere between the battery build connections to the controller.

Here are some steps to follow for troubleshooting. Depending on the results, you may find the problem or the results of the testing will help figure out the next steps.

1) Double check all connections starting at the taps on the pack for anything not making good contact. Gonna assume that's all good now.

2) Determine if there is an overload -- Measure the current flow out of the battery. Before the BMS. See the max amps drawn while running full speed. The easiest way to do this is with a DC clamp meter. Here is an example (be careful if you buy one, the less expensive ones only do AC current):

https://www.amazon.com/Meterk-Multimeter-Auto-ranging-Capacitance-Non-contact/dp/B0721MKXBC

If you aren't seeing anywhere near the 30A limit when it stops, something is breaking down under load. If you are seeing near 30A, stop here and post again.

3) Measure the voltage at the outputs from the BMS under load. If that voltage isn't dropping out, the BMS and the pack are probably OK. The controller is suspect. Stop and post the results.

4) If there is a substantial voltage drop at the BMS output, this clears the controller for now.

5) Now check the voltage at the inputs to the BMS. If the input to the BMS is also dropping, the problem is the pack. Go to step 7). If the input isn't dropping, it's likely the BMS.

6) If the BMS is suspected, temporarily bypass the BMS and feed the controller directly from the battery to see if that resolves the problem. If so, stop and post results.

7) The BMS input voltage is dropping out under load, so the battery pack is suspected. Check the voltage at the battery pack taps. If dropping here, it's the battery pack/build. If not dropping, here the battery pack is probably OK and there is faulty wiring between the BMS and the Pack.

8) Stop and post results.
 
A couple more comments...are you using the new high current version of the VRUZEND caps? You will need those.

All the wire from the pack taps to the controller, and also the phase wires to the motor should be a minimum of AWG 14, but I would be using at least AWG 12.
 
Hi,

I have had the exact same symptoms like you..
After months and months of searching, I found the cause, which you can see in the photo..
Burned because of playing with too many amps..
 

Attachments

  • b754ae42baf89e12719742ee051a0957-756x368.jpg
    b754ae42baf89e12719742ee051a0957-756x368.jpg
    57.7 KB · Views: 4,321
How old is the kit? Did it EVER work correctly, or has this been a problem from day one?

Is that tiny little gear on the motor driving the bike directly thru magic, or are there perhaps some intervening gears and chains and whatnot that just MAYBE could be having an effect on the problem?

Which Vruzend kit are you using, and did you use the large wooden mallet to construct your battery?

Do you know the LVC for both BMS and controller? The fully charged voltage for a 13S pack should be 54.6. Your measurement of 47V starting voltage is quite low. Describing how you got there might provide some useful diagnostic information.

Describing what happens when the battery is fully charged might be even more useful.

Providing accurate data in answer to ALL questions would be ideal.
 
AngryBob said:
How old is the kit? Did it EVER work correctly, or has this been a problem from day one?

I finished custom building this prototype system this past April. When I tested it at that time, this issue was present. I did not work with the project for a while, but upon revisiting it, the same issue was there.

Is that tiny little gear on the motor driving the bike directly thru magic, or are there perhaps some intervening gears and chains and whatnot that just MAYBE could be having an effect on the problem?

Unfortunately, I do not have access to the bike or a picture of the system right now, but I can reply with one later if necessary. I replaced the little gear with a small bike sprocket. A chain runs from this motor sprocket to a large drive chain ring which is bolted the normal front chain ring of the bike. In this way, when the motor runs the motor sprocket indirectly drives the center chain ring of the bike which drives the rear wheel normally. I did some calculations and set a gear ratio with considerable torque advantage. It seems like the issue is electrical because the entire system shuts down before the motor accelerates the bike above a few MPH although I would appreciate any thoughts on hardware involvement in the issue.

Which Vruzend kit are you using, and did you use the large wooden mallet to construct your battery?

I am using the 2.0 kit with the upgraded components. I am not sure which wooden mallet you are discussing.

Do you know the LVC for both BMS and controller? The fully charged voltage for a 13S pack should be 54.6. Your measurement of 47V starting voltage is quite low. Describing how you got there might provide some useful diagnostic information.

I could not find the exact LVC for the components, but the typical LVC for 48V lithium ion batteries is around 36V. For the 54.6V you mentioned, were you looking at the charging voltage on the BMS page? This should just be the voltage at which the battery is charged. Each cell is around 3.7V when fully charged resulting in 3.7 x 13s = approx. 48.1V. I measured the voltage from negative to positive terminal of the battery pack.

Describing what happens when the battery is fully charged might be even more useful.

The situation when running the bike is the same when the battery is fully charged. The charger light turns from red to green although at times it alternates between the two after a while of charging. This could just be a result of the battery hovering around being fully charged, but I am not certain.

Providing accurate data in answer to ALL questions would be ideal.
 
OK, so it has had the problem from day one.

The large wooden mallet was in the early construction videos for the Vruzend kit. This used a pressure fit ONLY for contact. There was an upgrade, then later bolts were added, which I believe was the Third revision.

Does your battery kit have bolts for retaining contact pressure? Pressure fit alone was just not good enough. Contact was not uniform and not consistent. IMPORTANT to answer whether or not the bolts are present.

On the gearing - not familiar with your chosen motor. Mismatch between motor RPM and torque and selected gear ratio can cause very high amp draw, which could lead to major voltage drop, and trigger shutdown. When such a voltage drop happens, as soon as the load is removed, voltage will quickly rebound, such that using a meter to check voltage AFTER the shutdown is not really useful.

THE CHARGING VOLTAGE - your number of 3.7 per cell when fully charged is incorrect. The accurate number is 4.2V. It is listed in the specs for the charger you have, which you posted. Your battery is NOT repeat NOT fully charged, and the cells likely not all balanced at the same level. Combined with possible not fully seated connections, this may be the problem.

The charger alternating between red and green should indicate a balance charge is happening. Low-power charge is applied while higher voltage cells are drained, in order to get all cells to matching voltage. If this is happening with a metered voltage around 48 and not 54.6, the charger is defective. You can meter the outputs to make certain. If the charger output is indeed 54.6, and the battery measures 48, then you need to let the red-green cycling continue for an extended period of time, WHILE YOU WATCH, possibly for many hours to several days.

Did you test the individual cells for voltage before assembly? You may have one or more dead or severely depleted cells in the group. Hopefully you bought a few extra, and/or the vendor will replace any defective cells. If you have the above situation where the charger output is indeed correct and the battery is severely low, you should probably check each individual cell for voltage.
 
If all the connections are good (motor, controller, battery and bms) throughout the system, then another possible culprit is an incorrect hall/phase wiring configuration which will cause a high amp draw and kick the bms into an over-current shutdown. With the wheel safely and securely off the ground measure the no-load current at full throttle.
 
BlueSeas said:
A couple more comments...are you using the new high current version of the VRUZEND caps? You will need those.

All the wire from the pack taps to the controller, and also the phase wires to the motor should be a minimum of AWG 14, but I would be using at least AWG 12.

I am using the V2 high current VRUZEND kit. I will have to verify the wire gauges, but many wires are of course just part of the controller/motor etc. Thank you for letting me know all of these courses of action to take in order to find and solve the problem. I will take those steps as soon as I get the chance and reply with the outcomes. I need to see where I could access the DC clamp meter or buy it as well as look into some other electrical considerations such as the proper fully charged voltage noted by AngryBob, another person who responded.
 
I have done a few tests and will use a test connector to bypass the BMS and measure the voltage drop of the cells. I also measured the voltages of cell groups when the battery was at rest and again a few hours after a 24hr charge. The measured voltages seem to differ significantly from each other for both sets of measurements. What are acceptable amounts of V differece between the parallel cell groups? Here are the results:

Cell Group V Pre Charge
1 4.082
2 3.58
3 3.588
4 3.272
5 3.389
6 3.594
7 3.548
8 3.53
9 3.163
10 3.868
11 3.358
12 3.78
13 4.085

Total V 46.837

Cell Group V Post Charge
1 4.138
2 3.615
3 3.635
4 3.393
5 3.56
6 3.627
7 3.586
8 3.578
9 3.464
10 3.923
11 3.428
12 3.876
13 4.139

Total V 47.962

pubchart


pubchart


This seems like significant variation between the groups. Is this beyond an acceptable amount? If so, is the BMS not properly balancing? Also, most of the cells are not nearing 4.2V which is what I thought was the fully charged state. The system shuts down at only a couple A at most; does this point to voltage variation or voltage drop being the issue?
 
Your battery is critically unbalanced. BMS would probably not be able to balance them - difference is too much.

First guess:
It can indicate you have voltage drops in the connections that is both root cause to the cutouts and the unbalancing.
Or
BMS is faulty causing both cutouts and unbalancing

Take each cell group and carefully discharge to the lowest voltage any group has. Charge it like normal. you need to watch voltages during charging as the bms might be faulty.

Try again😀
 
The BMS might be able to rebalance them, in a few weeks sitting on the charger, assuming it's the type of charger that resumes once the BMS re-enables charging after it shuts off once a cell reaches HVC.

(this is because a common BMS can only drain high cells down at a very slow rate, as they are only intended to keep a good pack good, not fix a bad one. So the cycle goes like this: Charger runs normally, filling up all the cells at the same rate. Cells are at very different states of charge, so at least one group reaches full, HVC on hte BMS shuts off charging current and drains that high cell down just a tiny bit, reenables charging current, and the process starts over and over and over, as the lower voltage cells get closer and closer to full. It can take a very long time for cells as unbalanced as yours are for this process to finish.

However, if they are that imbalanced, the cells are probably of such different characteristics from each other that it's likely to unbalance again every time it's used, especially if ran down closer to empty than full. If this is the case, then the cells should probably be replaced entirely with cells that are tested and matched with each other.

It's also possible that the kit itself is simply not making correct connections between cells in the groups, so some of the cells aren't even being used by the system.

That can be tested for by disassembling it and measuring the voltages on each cell that had been grouped in parallel. If any are different within a group, those different ones weren't connected to the rest (or else they would have been the same).

If that's happening, either it wasn't assembled correctly, or the parts themselves are defective and unable to make connections correctly.


If you want the balancing to finish quicker, you would need to manually charge all the cells to the same voltage, either one at a time or in groups taht are already at the same votlage as each other. If you're doing that, I'd recommend actually testing each cell for it's internal resistance and capacity, etc. Then you will know if any are actually bad to start with, before rebuilding the pack.
 
That's a huge imbalance as others have said.

You must either charge low cells to match high cells or discharge high cells to match low cells.

To charge low cells, you need a 4.2 V capable charger. Discharging is easier. You can probably use 100-150 feet of 18 gauge wire shorted across high cells until they match low cell voltage.

Once low and high are similar recharge the entire pack.

The pack may still be toast...even after all this if the cell voltages don't line up after further charging and discharging once equalized.
 
With that much imbalance, I'd first figure out why it happened, then cut up the pack and start over with new cells. Maybe salvage a few of the stronger cells for 4S use. A fire isn't worth it, and I suspect that's where this pack is headed.

Is your pack soldered or welded?
 
THe vruzend kit is a compression-contact kit, there's no permanent cell connections.

That is probably the reason for the problems, if the cells themselves started out good / balanced.
 
Ah I see. Well, maybe it's salvageable. I say don't push the limits, because group 9 has probably been sucker punched a bunch of times.

I think asking 10A per cell from that kind of connection is probably a mistake.
 
Life is a series of learning experiences. You failed to do your research, and you bought a piece of crap from a shyster. It may well be a total loss, BUT - you can learn.

The charger needs to be tested for correct output voltage. Your two highest cell groups are just short of 4.14. All the rest are under 4.0. Either the BMS is shutting down charge early, or the charger is defective or incorrectly spec'ed for your battery. Also, you either have multiple defective cells, multiple poor connections with a high degree of variability across the pack, a defective BMS, or some combination of these. When you purchase one piece of crap from a shyster, you may well find you have purchased MULTIPLE pieces of crap from that same shyster.

Note that the above is assuming that the voltage measurements used for analysis were taken either during or IMMEDIATELY after charging.

Both design and implementation of the Vruzend kit are incorrect and ineffective. Study the multiple design changes. Next, find the multiple pack designs detailed on this board using compression methods for connection. Study the design changes, materials used, and most especially note reports of long-term successful usage. No such reports have surfaced here using the Vruzend kit.

You never answered if your kit is the third iteration using additional bolts for compression. After studying designs that actually work, you may note the improvement offered by using a non-flexible plate or board across all the endcaps to evenly distribute the compressive force from the bolts across all cells, as in the fourth design change. If you have done your research, you should notice what is STILL missing from the design, and what additional changes are needed to achieve a reliable connection in an e-bike pack. Basically, the entire connection apparatus should be discarded.

Your pack is dangerously unbalanced, and represents a very real threat to those around it. It would not be in my home, and would be forcibly removed immediately.

Disassemble completely, individually charge each cell, test extensively, repeat this process, then start all over with a workable connection method. Test the charger, you may need a new one. You will likely need to replace some or all of the cells. You may need a new BMS.

The good news is that you may become slightly less likely to purchase another poorly designed piece of crap from another shyster in the future.
 
On a completely seperate issue, though possibly related, it is quite likely you have aggravated the problem with inadequate gearing for the motor. If your replacement for the motor drive gear was much larger than the original gear, as a bicycle sprocket would be, unless connected to a truly humungous gear on the crank, this will need to be changed as well. That thing was designed for a kick-scooter with teeny-tiny wheels.

Multiple threads on functional gear reduction systems, also info on improper ratios causing tremendous amp draw and necessary corrections made.

What you WANT is not important, what the motor can handle, IS.
 
I just have same issue, bike works well off ground, very well, or very light speed, or another battery pack 36V 12Ah.
I tried bypass BMS, nothing changed.
I would say it's battery pack issue, as all works well with another battery pack.
I'll try open it and I'll report back.
 
faxavier said:
I just have same issue, bike works well off ground, very well, or very light speed, or another battery pack 36V 12Ah.
I tried bypass BMS, nothing changed.
I would say it's battery pack issue, as all works well with another battery pack.
I'll try open it and I'll report back.

If the battery is too weak for the load you're putting on it, then it could be sagging lower than the controller's LVC, and the controller is cutting it off. Is the battery old or unhealthy? Or very small?
 
I have 1 battery 36V, 9Ah, 10s 3p, Vruzend kit, measures 42V charged.
Bike works very well with this pack.
To increase range I built another pack, with a box that fits in the lower frame. It's 36V, 12Ah, 10s 4p, spot welded, measures 41.7 V charged. Both with same brand and type of batteries, 3000mAh, 25R.
I just checked, changed for 2 different new BMS, all still the same, wheel off ground runs well, with me riding, it tries to run PAS, 2 sec, goes off, with throttle it does not work at all.
It's not BMS, I did try bypass BMS, shunting both + cables, nothing changes. Then I got 2 new BMSs nothing changed. I'm trying to avoid going into batteries, it's a big job.
Besides this, there's a XT60 connector, I'm going try bypass it, 1 box switch, and nothing else.
So it might be something to do with battery pack.
 
A just had a good idea, as battery pack is new, never used, 36V, 12Ah, 41.7 volt charged, I'll try ride bike and measure battery pack voltage.
If it drops a good bit, it's battery pack.
 
faxavier said:
A just had a good idea, as battery pack is new, never used, 36V, 12Ah, 41.7 volt charged, I'll try ride bike and measure battery pack voltage.
If it drops a good bit, it's battery pack.

A 48V controller usually cuts off when the battery drops to 40V.
 
I just had a big surprise, I upgraded, my nr1 pack from 10s 3p to 10s 4p, just like pack nr2, 10s 4p, which is shutting down on the bike.
Now pack nr1 has the same issue. Shutting every 2 sec on PAS, not running at all with throttle.
I have to think about to find a way out.
1 idea- try no s connection for 4th p, just parallel connections.
 
Problem solved.
I was not happy when I started thinking about battery pack as my problem.
So I checked more, and I started thinking about a XT90S connector I installed to prevent sparks when doing or undoing connections.
It was not fully inserted so it showed 8 ohm resistance between both sides.
That was it, after full insertion, male to female, 0 ohm, no more issues.
Beautiful.
 
Back
Top