Conv kit - power cuts at times

Since the problem did not go away with a new controller, the handlebar unit is a likely suspect. And since it goes away if you pick level 2 or level 3 assist, it sure sounds like it's a flaky unit. These units turn on the controller by applying battery voltage to a certain lead, and they turn it off by taking it away.

I've owned several ebikeling kits. Two used this LED display, the 810LED. Ebikeling has two different models of the 810LED. One is analog and was used on their kits back in 2015. Around late 2018, they sold a digital version. They are not compatible with each other, but there are ways to tell.

I would try replacing the 810LED if that's what you have. The analog versions are pretty cheap on ebay or aliexpress. I think only ebikeling sells the digital model.

docw009 said:

Since the problem did not go away with a new controller, the handlebar unit is a likely suspect. And since it goes away if you pick level 2 or level 3 assist, it sure sounds like it's a flaky unit. These units turn on the controller by applying battery voltage to a certain lead, and they turn it off by taking it away.

I've owned several ebikeling kits. Two used this LED display, the 810LED. Ebikeling has two different models of the 810LED. One is analog and was used on their kits back in 2015. Around late 2018, they sold a digital version. They are not compatible with each other, but there are ways to tell.
View attachment 358086
I would try replacing the 810LED if that's what you have. The analog versions are pretty cheap on ebay or aliexpress. I think only ebikeling sells the digital model.

Click to expand...

Excellent! That one pictured is the exact one I have.

How would other ones be compatible or not? If I did want to upgrade to digital / an actual analyst, would any of theirs work? I've never considered interoperability between handlebar unit and controller. What are ways to tell which work with each other?

I'll get a replacement 810 just to make sure that's the problem before upgrading tho. I know they're cheap.

And thanks so much for helping!

Well, I found the culprit and it wasn't the 810 handlebar unit. I purchased one on Amazon, but when i hooked it up it didn't function - would show lights for a half second then go out. Then I realized it had a 48V sticker on the back of the unit, while my setup is 36V. Thanks to these chinese co Amazon listings NOT mentioning it's 48V I didn't even consider they could be different (went back and re-checked the listing and there's no mention at all). So I temp connected my original battery and original 810 unit (which I had cracked open for the heck of it) and it worked. So I set out to do a full battery swap thinking that might be the issue...

When I pulled the battery I found the problem - fused wires on the connector (attached pic).

I guess I was too hasty with the install of this battery that when I got done splicing the wires (had to cut off a JST to splice on XT-90 btw the battery and controller), I didn't solder or cap/protect the twisted leads. I just shoved it all in my homemade ABS wiring & battery compartment and went about my merry way. Looks like the red and black were arcing/shorting in contact with one another which then caused the two to fuse together - melting right thru the insulation. I can't believe this setup worked for YEARS 90% of the time.
Then I had a fun time trying to rig back up the original 810 which I pulled apart two weeks ago. The two halves are glued together (no screws) so I literally had to break it open. The original handlebar clamp from that unit was broken anyway so I had tossed it. It's currently all taped and taped and taped using some black gorilla brand tape to an old reflector mount.


I guess the lesson with these DIY or kit setups for me is to investigate all this sooner. I was reluctant to pull it all apart for a long time because it mostly worked but also because there's a lot of tape and zip ties keeping things from rattling around. The second bottle battery I bought didn't fit tightly enough to the mount and rattled on every bump, so I taped it to it and once around the downtube in such a way that I didn't want to undo it. And unfortunately on my setup I can't unscrew the collar that locks the 4-pin round connector to the battery out (I've seen them called "Aviation" connectors, but idk if there's another name) without pulling the whole battery out.
OR, don't just twist some leads and call it a day!

GOSH DANGIT! Still having this problem.

Replaced the 810 with a SW900. Same issue. Tried to leave it running while on my stand in "walk" mode for 20 minutes and it never cut out. I went to ride to work and after a half mile it cut out. (Power will kick in via PAS and after two seconds it turns off. When I tried thumb throttle it just cut out immediately.)
When I turn on the SW900 it does give me the error 6 (battery watt issue). I am assuming under load the BMS gets triggered with HV or LV and cuts. Whereas the no-load, low speed "walk" test doesn't trigger this.
I tested the battery pins as well as the wires I connect to the controller with a volt meter and they both registered about 35.5 (36v battery).
I'm assuming I need a watt meter and need to start testing with that and see what the BMS and/or cells are telling me.

This has been an annoyance, but it seems to be fairly common with certain batteries (this is a Chinese bottle style battery from Amazon - can't remember the manufacturer name) as I've seen a lot of threads here with similar issues when searching for solutions.

I'm hoping the indexed battery testing threads provide some guidance. But if you see this and have any ideas, I'd appreciate it as always!

In cheap kits, the part where they really cut costs is on the electronics. And this is where we also have the most problems.
I will generally refuse to use components like these, more hassle than they are worth, had enough electronics failures in the field that i prefer solid, well supported, and more expensive stuff.

Can't help anyone diagnose this stuff since i don't touch it.

A watt meter would be an extremely good diagnostic tool for ruling out if the battery is cutting out or not, and assessing it's health, etc. You are totally in the dark if you don't have at least that.

I would recommend a cycle analyst standalone ( $150 ) or turnigy watt meter ( $25 ) for these purposes. The turnigy watt meter will require some extra long battery cables. I used to run one.

Example of mounting one next to the headtube:

neptronix said:

In cheap kits, the part where they really cut costs is on the electronics. And this is where we also have the most problems.
I will generally refuse to use components like these, more hassle than they are worth, had enough electronics failures in the field that i prefer solid, well supported, and more expensive stuff.

Can't help anyone diagnose this stuff since i don't touch it.

A watt meter would be an extremely good diagnostic tool for ruling out if the battery is cutting out or not, and assessing it's health, etc. You are totally in the dark if you don't have at least that.

I would recommend a cycle analyst standalone ( $150 ) or turnigy watt meter ( $25 ) for these purposes. The turnigy watt meter will require some extra long battery cables. I used to run one.

Example of mounting one next to the headtube:

View attachment 361508

Click to expand...

Yeah I had seen some other posts recommending the turnigy. I looked it up and none of the pics on the site that was selling it showed the connectors - only red/black coming out of it.
What are the connectors and also: do I connect this in line between the battery and controller?
Thanks!

You need to solder the connectors of course.
What connectors do you currently have?

If you can't find duplicates, XT60 is a good choice for <2kw. Hobbyking will sell you those too!

I can solder! Just bought the little clip stand thingy too. Excited to use it and not fumble around as much as before.

Currently my battery and controller connect via XT-90s. I have a few short (4 inch?) XT-90 to red/black wires lying around (both m and f ends).

So does the watt meter go between battery and controller OR between the battery out (4 pin aviation connector) and the XT-90?

About to open the battery to see if there's anything obvious.

On my ride home today, after the first power cut, the battery would not power up. The on-battery LED battery level would not show when i pressed the button. This is despite it being charged all night and not really being drained (see above) on my ride to work. Got home, plugged the charger in and it went from red to green in about 10 minutes. Powers on now.

It had done this one time before, but it at least took an hour or so to "charge" that time.

Definitely seems to be LVC, BMS, and/or loose wire related. We shall see.

Yeesh, where to begin?

The connection is as such:
Battery - > watt meter -> controller

Hey if you use xt90's already, great. Standardized connectors across the fleet is good

What you can do to test the battery without something like a RC Balancer to discharge graph the cells is pretty limited, however, you can measure each cell group with a multimeter using that 11 pin connector that goes into the BMS, starting at the negative lead ( cell 1 ).

If you see big discrepancies then you could have a pack with some dead groups..
Really cheap BMSes will do dumb things like draw all their power from cell 1 or cells 1-4, eventually killing the pack because the first cells get too weak for any kind of discharge, and the BMS trips the low voltage cutoff

If the pack is worn, with the watt meter, you'll see massive voltage sag when the motor is getting started. That's a hint that the battery is either too small, or too worn!

So, IDK if I'm doing this right, but I pulled the 11-pin connector and measured each of them.
The top one (in the pic) is black, so I kept the neg lead on that from my volt meter. Then I moved the positive one to each slot in sequence from there. It's odd because it's nearly just adding 4.1 to each as I go along (with some variance):
1- 4.1
2- 8.2
3- 12.3
4- 15.5
5- 19.7
6- 23.8
7- 27.9
8- 32
9- 36.2
10- 40.3

Did I do that right? Removing the batteries from the aluminum bottle package is going to be a project. This thick rubbery white glue they used to secure it in place is ridiculous. I don't think my heat gun is going to melt it without damaging wires/components. I've already pulled and cut pieces of it out and it's still not budging. Might be easier (but not safer) to cut the aluminum casing vertically on opposite sides and crack it open that way. So stupid.

So I guess group 4 has a problem since it's only adding 3.2 volts rather than the 4.1 or 4.2 that every other group adds as it goes along?

CMo44 said:

Then I bought a spare battery via Amazon (which we know are not cheap), wondering if that would solve it if it was a voltage problem or something. That did not stop it either.

Click to expand...

CMo44 said:

So I guess group 4 has a problem since it's only adding 3.2 volts rather than the 4.1 or 4.2 that every other group adds as it goes along?

Click to expand...

It appears to be an issue with this battery, but seems like a big coincidence for both batteries having the same issue. How did the voltages read on the other battery?

That 3.2 volt cell is probably the worn one that's been killed by the BMS.
Would take ages to balance that pack.. and it would probably get disbalanced very easily.
The 3.2 volt cell will quickly dip during discharge to below 3v and lead to the BMS cutting off the entire pack.

How many calendar years are on this battery?

E-HP said:

It appears to be an issue with this battery, but seems like a big coincidence for both batteries having the same issue. How did the voltages read on the other battery?

Click to expand...

No idea, as I have that battery sitting around but about to be used on another project. Still acts as if it charges full and I also hooked it up to a project motor and it ran fine. Never took that one apart.

neptronix said:

That 3.2 volt cell is probably the worn one that's been killed by the BMS.
Would take ages to balance that pack.. and it would probably get disbalanced very easily.
The 3.2 volt cell will quickly dip during discharge to below 3v and lead to the BMS cutting off the entire pack.

How many calendar years are on this battery?

Click to expand...

probably 5 years?

If it's at the end of it's functional life, are there any options? I'm thinking: What if I remove that 3.2 group entirely? (Provided I can get the damn thing out of the bottle...cursed white rubber glue!) If the other 9 groups are still working fine would the battery get some more life?

I could potentially use the rest of this battery on my proof-of-concept project and get a new battery for this bike which I use more. The project one won't be run much - it's to show functionality and then will probably be deconstructed and parts repurposed somewhere down the line.

You could replace this group, but you'd end up with a battery that is disbalanced the other way. The consequences wouldn't be as severe as what you've got going on.

The problem is matching the other cells in the pack, IE getting the same internal resistance and capacity roughly as 5 year old aged cells, but with new cells. The ideal replacement series group is the same cell of the same vintage.

You're going have to invest in a spot welder etc to replace these cells.

personally, i'd rather invest in one of these:
Infinite: The repairable & universal Ebike battery

neptronix said:

You could replace this group, but you'd end up with a battery that is disbalanced the other way. The consequences wouldn't be as severe as what you've got going on.

The problem is matching the other cells in the pack, IE getting the same internal resistance and capacity roughly as 5 year old aged cells, but with new cells. The ideal replacement series group is the same cell of the same vintage.

You're going have to invest in a spot welder etc to replace these cells.

personally, i'd rather invest in one of these:
Infinite: The repairable & universal Ebike battery

Click to expand...

Very cool product/project! seems worth it!

But, what I was asking wasn't about replacing that series group - it was: what if I just REMOVED that group and didn't replace it? I'd have less capacity, but it just wouldn't trip the BMS then, correct? I think I'd be ok with that in the short term.

It depends on your BMS. Some models will be ok if you remove a group and bypass it with a jumper. More advanced models are going to see that as a dead group and kill the output. You may need to remap the balance wires so the unused one is at the positive end and connected to pack positive.

fechter said:

It depends on your BMS. Some models will be ok if you remove a group and bypass it with a jumper. More advanced models are going to see that as a dead group and kill the output. You may need to remap the balance wires so the unused one is at the positive end and connected to pack positive.

Click to expand...

ah-ha. I had a feeling it wouldn't be that straight forward.

I grabbed my other battery (the original one that came with the kit - even older, but still holds a charge well) and see if it could at least get me to ride for the next few months until the Infinite suggested above would ship. Even if this one does what it originally did, I can live with it (just need to charge it every time and run it at the higher power). I just plugged it into my system and it gave me the same ERROR 06. But riding conditions are the real test. If it works for getting to my workplace and back without totally shutting off all the time I can live with it short term.

So, I'm going to buy the Infinite - I love that idea. And I'm going to use the one that I've been mostly talking about on this thread (the one I took apart partially and tested voltage) and see if I can remove that group and use it for my other project. I can learn about rewiring and remapping balance wires on a battery as I go.

Thanks all a TON for the help!

And actually once I receive that battery, I can cannibalize these two old bottle batteries to see if I can make one good pack from the cells that are still decent. Like I said, it's a good learning opportunity. Plus I don't want to just get rid of them if there's life in them.