[SOLVED] Current drop ?

amund7

1 W
Joined
Sep 6, 2015
Messages
54
Location
Oslo, Norway
Hi all,
First post, sorry if it's in the wrong place.

I built my first E-bike last week:
I put a Magic Pie 3 on a downhill bike, bought a Li-Ion bottle battery from E-bay (48v/13Ah) (ad says 30A continous, 70A peak).

Bike runs great on a full battery, pulls 26A and is truly fun.

But after spending 1 Ah, amps is down to 20A, noticeably slower
After 2Ah - 18A, boring
After 5Ah - 15A
After 9Ah - 10A - I now have to push the bike home. At least pedal a lot.

Voltage is 55v at full and ending at about 46V after 9Ah, and around 2v loss under full load.

To eliminate temperature issues I once used 2Ah, parked the bike outside for 5 hours, then resumed. It pulled exactly 18A.

I was sure this was a battery problem. So today I tested my friend's Hobby King 10Ah 22.2V lipo's, 2 in series. Results are exactly the same, maybe a bit less amp loss per Ah, but very close to my ebay battery.

What's going on? Is there something wrong with my Magic Pie controller? Or is every E-bike like this?


The numbers come from a DealExtreme wattmeter, and from memory, but my problems started before I had any display, and the bike was getting drastically slower with a little use.
 
Where did you buy your controller from and what kind is it?

I would have guessed that your battery that you bought is not what you think you bought. But you say its the same characteristics when you put your friends battery on it.
 
Controller = Magic Pie 3 internal controller. I bought the Pie from BMSBattery.com

I thought the battery was the problem too, but the Hobby King batteries should be dead solid, and should deliver 10 times what I'm asking from them. Still they give the same characteristic.
 
If it's not the battery, then it is probably the controller thermally limiting to protect itself and the motor.

You could install a BBQ thermometer in there (or multiple sensors) that has a long cord for a readout at the handelbars and see whta it goes up to as you ride.
 
amberwolf said:
If it's not the battery, then it is probably the controller thermally limiting to protect itself and the motor.
You could install a BBQ thermometer in there (or multiple sensors) that has a long cord for a readout at the handelbars and see whta it goes up to as you ride.

I thought so too, so I did this:
"To eliminate temperature issues I once used 2Ah, parked the bike outside for 5 hours, then resumed. It pulled exactly 18A."
 
OOps.. I read that originally, then forgot when I was pondering answers. :oops: Sorry about that.

That leaves either a battery issue, or a controller design and/or setting that reduces current based on pack voltage (it's unlikely the controller tracks Ah or Wh used).
 
It has to be something to do with the controller. Are you using the MP one in the hub?

Tell us a bit more about your bottle battery. I've never heard of a 48v 13Ah one. 52 cells is normally the max, which would give you 11.6Ah, not 13Ah. Do you have a link?
 
Yes, it's a stock Magic Pie 3, brand new, recieved last week from BMSBattery.com.

The battery is from E-bay, link: http://www.ebay.com/itm/331623250071?_trksid=p2057872.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT
 
Well, if your top speed is dropping as voltage in the battery drops, then your continuous amps will also drop correspondingly. But your drop seems more than normal for just going slower as the battery discharges. But maybe it is normal, if we know what speeds you are talking about.

26 amps continuous, that would be really hauling ass. But by the time you drop to 22 mph, 15 amps might be just about right, for continuous amps cruising that speed. What seems odd to me is the rate at which you are seeing that drop. Seems like it should be doing better for a bit longer into your ride. But losing 3-4 mph top speed in the first mile, that's totally normal. Then it tends to go to a slower drop for the next 5 miles or so.

But if it's max amps, it should spike close to the same even on a half charged battery, when you do a start from dead stop. Because volts is lower, it will still be lower than when full, but not 5 amps lower till the battery is nearly fully discharged.

Continuous amps, 5 amps lower makes sense to me if voltage is much lower.

Capacity on the other hand, only the very best batteries generally dish out the capacity written on the side of a cell. Some of this is from sag under load, but really good cells won't sag much under load. It does sound like your voltage sag under load was not much. ( voltage off minus voltage under load, at any state of charge is the sag) But you did of course have voltage drop as you reach lower state of charge.
 
do you know if the battery is balanced? do you use a BMS to protect it from over discharge? it only takes one high current over discharge to permanently ruin the lithium ion cells then they have that permanent high internal resistance and storage problem.

but if you had a factory BMS it shsould protect it. but 2C continuous to LVC is really hard on them. i constantly try to get people to recognize the high C is only possible when you are at full charge of 4.20V. when you get to 3.7-3.8V it should be down to .5C imo.
 
Thanks for the replies. The 26 amps I am talking about is under full acceleration, and not top speed. And that falls to 10 amps at the end of the battery capacity.

I only have about 5 charging cycles on the battery. I don't know if it has BMS, it is not mentioned in the ad so I suppose it doesn't. But anyway, since I've tested with other (known-to-be-good) batteries and they give the same amp drop, I think this battery is good enough.

Yesterday I swapped out some flat lug connections to screw connections (can't find the right word in English, sorry), and the bike felt faster. However, the numbers seem as bad as before, so I am going to write it off as placebo until I have proof.

I will try swapping out more wires and connections to better quality ones as time goes and see if it actually makes any change. But I didn't believe poor wiring or connections could drop the amps, only the voltage under load. However if the volts drop after my watt meter, I won't see the measurement being low... and I guess I'd see good voltage but poor amps?
 
That's a very long bottle battery. I'm surprised that it fitted in your frame. There is a BMS inside it. I'd be surprised if there's anything wrong with it. The specification says 30A continuous, which is plausible for a modern 5S 18650 pack, but they don't mention branded cells, which makes me a bit suspicious. 1C discharge rate would be only 13A. Does it become warm after a long run?
 
How can you tell it has BMS? And what will that BMS do, will it balance the cells (or cell groups) when I charge?

I have a downhill frame, and it didn't fit very well :) Currently the battery is hanging under the front lower tube. Front wheel touches it on hard braking or small jumps, so I need to move it.
 
It has a switch on the base that switches off the BMS. It's not big enough to switch the main battery current on its own. The BMS will look after balancing, etc, so you have nothing to worry about. Just charge it and use it.
 
I think you just drop voltage, and the amps drop is because your max speed is slower. But this pack may be dropping volts pretty fast, if it doesn't really like 2c.

That's what the does it get all hot question meant. If it gets warm, normal. But if it gets truly hot, then the cells are hating it.

If it was the connectors, then most likely you'd see them getting all hot.
 
I am talking about peak amps, they occur somewhere right after starting from a standstill.

Volts are not dropping fast, and the battery merely gets lukewarm.
 
Hi Amund!

You've been running a 48v battery from 55v to 46v

You also tried a lipo which should deliver about 50v to 45v

I would have thought you'd see a change right away from 10% less starting voltage, but a more even discharge amperage (lipo is always higher c-rates)

I am running lipo, and see about a 7% loss in speed and voltage from fresh charge to 'low'. I haven't noticed dimenished current, but I have alot to spare, and 7% wouldn't be missed that much.
Since I'm running 82v, I suppose it would be obvious to expect a much larger percentage of loss on almost half the series'd cells, although amperage really should remain pretty constant, especially on lipo.

If you're positive of similar results on lipo too, I'd look at that pie controller. I've heard they're quirky etc, and would look into that.

Try searching 'magic pie internal' or things like that? search.php?keywords=magic+pie+internal Then hunt down some actual users if you still have questions or need confirmation on your results.
Good luck, I'm curious to see resolution, because that doesn't seem normal imo.
 
nutspecial said:
Hi Amund!

If you're positive of similar results on lipo too, I'd look at that pie controller. I've heard they're quirky etc, and would look into that.

Try searching 'magic pie internal' or things like that? search.php?keywords=magic+pie+internal Then hunt down some actual users if you still have questions or need confirmation on your results.
Good luck, I'm curious to see resolution, because that doesn't seem normal imo.

Thanks, I've been googling this thing hard for a while. And I found a user posting my exact symptoms. Only ONE place in the whole internet. He didn't post the solution of course, but I managed to e-mail him, and he replied back: His controller was shot, and he got a new one.

I've messaged both Golden Motor and BMSBattery, but no reply back from any of them so far. I hope either of them steps up and sends me a new controller, if that is the issue.


I agree with you it's really strange that the amps didn't drop with Lipo, if the voltage is that much lower. I can't remember the voltage reading, but I remember it barely touching 27 amps. And dropping almost as fast as with the other battery.
 
Try asking the question on the Magic Pie forum

http://goldenmotor.com/SMF/index.php?board=15.0
 
I get confused as threads progress. It does the same thing with another battery? Or not? If not, that points to the battery for sure. If it does do the same thing with a different battery, I mean more than normal, then you may have a controller problem. I have had controllers do a slow blow, but that acted different. just weaker and weaker each ride, and never full amps at the start of a ride.

Yes, peak amps drops to 10 amps near the end of the discharge. That is low, and would indicate a lot of sag under load is suspected.

With enough sag, plus the voltage drop from discharge, your battery may only be able put out 10 amps by the end. The rest may be heating up your battery. Your battery when full, may still have a really high internal resistance. You can test that, by measuring sag under load. As you discharge the battery, that internal resistance will increase even more.

That's why I was asking if it got almost too hot to hold onto. If so, that's meaning your battery is suffering hard, especially near the end of the discharge.
 
amund7 said:
I've messaged both Golden Motor and BMSBattery, but no reply back from any of them so far. I hope either of them steps up and sends me a new controller, if that is the issue...
Did you contact Golden Motor Canada? If not, know that GM has little to no support unless you're contacting Golden Motor Canada.
 
Thanks, I've been googling this thing hard for a while. And I found a user posting my exact symptoms. Only ONE place in the whole internet. He didn't post the solution of course, but I managed to e-mail him, and he replied back: His controller was shot, and he got a new one.

I'm surprised there isn't anyone on the forum? I thought I read about people ditching the internal controllers here before. The search (including advanced and/or googl option) has been working good on ES imo.
I agree with you it's really strange that the amps didn't drop with Lipo, if the voltage is that much lower. I can't remember the voltage reading, but I remember it barely touching 27 amps. And dropping almost as fast as with the other battery.
It's not strange for amps not to drop with lipo, even with 10%less voltage. With the lower voltage, you'd maybe see less top end speed, but lipo has much higher discharge rates than most(all?) of can-cells, so you should see a more even amperage, and the ability to acheive higher amps. Lipo is generally listed 20+ times capacity(ah) for discharge, and can realisticly reliably reach at least 1/4 or maybe half of that.
(10ah of 20-25c <healthy lipo cells> should easily deliver 50-100amps thru the state of charge) So that's why it's great you have lipo to compare with the other pack. If it delivers good amperage, then it would appear the ebay battery isn't up to the 25a draw. But if you're seeing only 10 or 15 amps when the (20c?)10ah of lipo cells are still above 3.7v (44v for the pack), one could rule out the battery.

Good luck!
 
nutspecial said:
I agree with you it's really strange that the amps didn't drop with Lipo, if the voltage is that much lower. I can't remember the voltage reading, but I remember it barely touching 27 amps. And dropping almost as fast as with the other battery.
It's not strange for amps not to drop with lipo, even with 10%less voltage. With the lower voltage, you'd maybe see less top end speed, but lipo has much higher discharge rates than most(all?) of can-cells, so you should see a more even amperage, and the ability to acheive higher amps. Lipo is generally listed 20+ times capacity(ah) for discharge, and can realisticly reliably reach at least 1/4 or maybe half of that.
(10ah of 20-25c <healthy lipo cells> should easily deliver 50-100amps thru the state of charge) So that's why it's great you have lipo to compare with the other pack. If it delivers good amperage, then it would appear the ebay battery isn't up to the 25a draw. But if you're seeing only 10 or 15 amps when the (20c?)10ah of lipo cells are still above 3.7v (44v for the pack), one could rule out the battery.

Good luck!

I can see my last ramblings were a bit... unfocused.

What I meant to report was that the 100-200 amp LIPO batteries showed exactly the same amperage draw, 26-27 on a fresh battery, dropping proportionally with state of charge. And, despite Dogman Dan's predictions, even if the lipo voltage starts out lower than the li-ion, the amperage draw was exactly the same, around 26 amps.

Unfortunately those batteries were borrowed just for a test, we did not have time to run them all the way down, so I don't know if it would go as low as 10A, but it certainly seemed to be heading that direction.


So to sum up:
- yes, the same thing happens with lipo or li-ion.
- no, the battery does not get hot. The lipos got warm-ish, and the li-ion gets merely noticeably lukewarm.
- magic pie forum is where I found the other guy's similar symptoms, 3 years old post, no reply (but he replied to me via e-mail)
- why should golden motor canada help me fix a product that was ordered from BMSBattery.com?
- goldenmotor.com (china) and bmsbattery is not replying. Well actually, bmsbattery first checked if they had sold me the battery, when they hadn't they blame the battery. They are actually blaming the BMS, and when I explain that the lipos have no such thing, and will deliver 200 amps, they stopped replying.
 
If all else fails, it's fairly easy to switch to an external controller. When I did it, the controller was blown, so I unsoldered the hall connector from it and soldered it on to the new one, otherwise you have to solder another one onto the motor hall wires, which can be done, but isn't quite as easy because of the shorter wires. You just take out the controller and it leaves the three phase wires with spade connectors and the halls with the block connector. Run your controller wires through and connect them in the compartment.
 
Got me stumped now. You should see some current drop with any battery, but not usually as much as you seem to have.

Something gets all hot? The motor? The controller? Wiring inside the motor?

If the controller is inside, then it could be related to the controller getting hot, a wire to the phases getting hot.
 
Back
Top