GoPed Hoverboard LiFePo4 Issues

[bzeller]

I just purchased a used GoPed ESR750 Hoverboard. It looks to have this EXACT set up - https://www.youtube.com/watch?v=d3JZ_GZA1kA&index=34&list=PL76D65CAB28FC1C6E. The guy I purchased it from seemed to be clueless. He first told me it was a LiPo conversion and he also stated that when in run mode, you needed to unplug the Chargery DB8 and only plug it in when charging the unit. After watching the install on the link above a few times, I'm convinced they never unplug the white connector. Now on to my issue...

Unfortunately, I'm running on very limited info and there is no "manual" that came with the unit. I have read through the DB8 manual a couple times and its helped some. I charged the unit last night and then tried to take it out for a run. It's fun, but I get about 1 - 1.5 miles out of thing and the throttle starts blinking the red light of death. I've been pulling the deck to see what the readout is once I'm back. Before I plug it in to AC, the DB8 only cycles through six cells, when there are eight. Once I plug it into AC, it then starts to read out 8 cells. After plugging it in, 1-6 are typically hovering around 3.2V, but the other two cells read out very low. There are also a couple cells that are pretty warm to the touch. At this point, I'm a little confused as to what's going on or where to start debug. It almost seems like I'm only running on a couple cells as the others aren't depleted? With the AC plugged in, the final two cells will slowly come up and balance with the others, but takes bit. If two cells were bad, would it make sense that they would drain quickly and then just shut me down? I guess that could be the case since if two cells lost their charge quickly, it would drop from 24V to under 19V. Help/input is appreciated!

 

[silviasol]

Sounds like the db8 failed at some point and caused a few cells to drop below the lowest voltage and damaged them, or a few cells are dead and others on their way to being dead and he conned you to basically start a possible fire by telling you to run the scooter with a few dead cells at dangerous voltages so it would actually drive more then a few feet when you tested it out before buying it. Are the cells that are getting hot the cells that drop to a low voltage and are those the same cells that the db8 is not reading? The db8 was used with those cells to balance them when charging and shut off the battery if any cell goes below the absolute minimum of 2.8v or cut off the charger if any cell goes above the max 3.6v. There is no protection otherwise unless you are using a battery alarm, the red light on the throttle comes on I think when it is about 21v which would be below the minimum 2.8v of lifepo4.

So I am thinking this. Maybe the db8 does not read a cell when it is too low and considers it a 7s, 6s, 5s... pack instead. Try charging it up completely, each cell to 3.6v, leave the db8 plugged in and see if all 8 cells are reading. Then do some bench testing with a small load on the motor(if you have a rear brake use that or do some burn outs to pull some amps out of the battery). Watch the voltage of each cell. Any cell that drops voltage quicker then the others is a dead one. You may have some good cells so you can just purchase some new ones to replace the dead ones. But at 1.5 mile ride and the higher voltage cells with a 3.2v resting voltage it sound like maybe all of them are toast. I think you can pick up new lifepo4 cans for 10-$15 a piece. Or use that db8 with other pouch lifepo4 if you can solder and wire the bms leads correctly, you will get more range with pouches. Only problem with the db8 is it doesn't allow too high of a amp drain, meaning you can't go up steep hills at a slow speed or similar high amp drain stuff and if you do it will shut off the db8 leaving you stranded or have to open the deck and unplug/plug it back in to reset it.

If you come out with a good db8 and you didn't pay too much for the scooter ($600 or under I hope?) still a good deal. You can't get the db8's anymore. Did you get the silver case motor or the black case motor?

 

[bzeller]

First, thanks for the quick response!

I think the guy I purchased this off of was clueless, but maybe he was a good actor. Since he was running the unit with the DB8 unplugged, my guess is the DB8 didn't fail, but that he ran some cells really low.

Are the cells that are getting hot the cells that drop to a low voltage and are those the same cells that the db8 is not reading?/quote]
I'm not sure how to tell if the cells that are getting hot are the same. Unfortunately, there's no way for me to get a multimeter on any of the cells without scraping off the protective sheath that is covering the terminals. I wasn't willing to go that far yet, but I may need to if I'm going to order new cells anyway.

So I'm a little confused on your statements about LiFePo4. Everything I read on the Headway 38140s is that they're nominal voltage is 3.2V and cutoff discharge is 2.0V - http://www.evassemble.com/index.php?main_page=product_info&cPath=2&products_id=21. I charged all last night and cells 1-6 are reading ~3.3V which I think is to be expected. Cell 7 is all over the place, which looks like it may be one of the culprits and cell 8 was at 3.4V, but it was acting funny as well. I found a wire on one of the packs that i had to resolder last night which is when 7 and 8 seemed to switch places on their "flaky" factor.

I think you might be right on the DB8 not reading a cell when its too low. After charing last night, it shows all 8 cells.

I paid $550 so feel like its a decent deal. I have the silver case motor, or at least when looking at the back half of the motor and heat sink sticking out of the red encasement, its silver!

 

[silviasol]

Hmm. Ok I just watched the go-ped install video and I see it is wired just like they show in video which completely bypasses the undervoltage and over amperage protection of the db8. This is just beyond stupidity on gopeds part, goes against every safety warning that goes with lithium ion batteries. Either they wanted to create a fire risk for anyone buying their kit or they just are plain stupid and did not know what they were doing. The way it is hooked up/how they showed it to be hooked up in the video you could wot the throttle while stalling the motor and in about maybe 5 minutes the cells would be toast or start on fire from pulling so many amps from them. This must be why they discontinued the esr1000 and do not sell this kit on their site.

But anyways yea the way it is hooked up should work fine charging it, during discharge it will bypass the db8 all together(you should not need to unplug anything) so if any cell goes under the 2v then it will go under the 2v(I didn't see that these cells have a different min voltage then other lifepo4 cells do) and would fail permanently or if it was charged again could start on fire(if the db8 was hooked up correctly it would cut the power when it gets below 2v). I would check all the connections to make sure they are getting voltage from each cell, both at the multicolored wire end and at the black end that plugs into the db8(another just stupid moronic move by goped using small balancing wires to parrallel connect the two packs together :? ). You should be able to disconnect the jst connectors then test on each pin to make sure you get voltage from each wire, if any did not make connection those cells would not get balanced or would not charge completely or even over charge.

Now if you want to correct goped's mistakes and hook up the db8 correctly it would be done like this. This would give you undervoltage and over amperage protection. But like I said before the over amperage protection doesn't allow much power, it says 25 amp nominal and 100 pulse but 25amps isn't much since average amp drain is 20-25amp for the silver case motor. So going up a hill or something for more then 15-30 seconds while drawing more then 25amps the db8 will cut the power off(this could be why goped chose to go the fire risk way by setting it up like this)

 

[bzeller]

Hmmm, not exactly sure what you're proposing in that block diagram. They cut the charger to controller board red and green in the video. Hooked the charger out to the P+/P- and the controller in to the B+/B-. Correct me if I'm wrong, but I believe you're saying to do it right I should remove the controller in B+/B-, cap them off and then disconnect the both pairs of battery terminals from the large red/black wires off the controller. From there, connect the large red/black battery terminals to B+/B- and connect the red/black from the controller to P+/P- as well?

At this point, I'm trying to decide if I should just purchase some new Headway cells and wire it up like it is (easiest, medium expense), do a total rebuild of the system the right way with a better BMS (hardest, expensive, but maybe worthwhile), or return to SLA. I love my old Goped ESR750 SLA, but purchased this newer hoverboard in hopes of solving some of its deficiencies. Namely, the pneumatic wheels (way smoother than my 6" GoActive on the old) and I was hoping the LiFePo4 would greatly improve run time as well as last a long time.

Anyone have any suggestions on maybe a fool proof set up that might support 50A+ continuous?

 

[silviasol]

The way goped hooked up the bms will work fine there is just no undervoltage or over amperage protection other then the controller cutting off at the 18v or whatever voltage it is, these cells are safe down to 16v so there is nothing to worry about unless one cell goes bad which would be rare until they get a few years old. The headway cells themselves are fine for normal riding or even doing hard offroading, the max amp draw with the silver case motor you will see is under 50amps and I think these cells are 25amp/ 100 amp burst, with two cells parralled that is 50amp/200 amp burst. You may see 80-100 amps if you really push the motor hard, like if you try to go up a steep incline from a dead stop which these motors are capable of doing. The main concern is if the balance wires are working well still. If not it will not charge correctly and surely over charge one of the cells over time. There was a really good youtube video of the guy who first built a goped with these headway cells a three or four years back, it worked so well and was so cheap that was the main reason they built the esr1000 and offered this kit. The thread was on gopednation.com which goped shut down because of spam and I can't seem to find the video on youtube. I remember he had an issue with one of the cells coming unscrewed and a cell shorted out from the bad connection, that pack has some good shrink wrap around the corners and the middle connections look tight so that probably isn't an issue but with all the vibrations these scooters make in the pan and if the pack would bounce around some in the pan or the fact he said he was always unplugging the bms(the jst connectors can damage over time if you unplug them often) I would think the balance wires may have become damaged.

The bms should be hooked thru the cable that goes from the battery to the controller, that way it can shut the power to the controller if there is any problems with the cells. Or will cut the power if the amps get to high which like I said sucks if you are in the middle of a ride. The way it is now is strictly for balance charging.

 

[bzeller]

I was just about to pull the trigger on 16 new Headway cells, however, it would be a real waste to replace them all if some are actually still good. What would be an easy way to test an individual cell to see if its still OK?

 

[silviasol]

What voltage is each cell after you charge them all up? Do the stress test like I explained in other posts. If one or two cells are bad it is best to just replace them all. Chances are they are all just about dead.