10v of sag on 72v 18650's

I read through every post I could find about voltage sag but I'm still unsure if I'm experiencing too much. I have this pack from LAebike: https://laebike.com/collections/72v-18650-lithium-ion-battery-packs/products/72v-20ah but I feel like it's not quite delivering the advertised power rating.
I'm using a sabvoton 72150 into a mxus 3k 5T. My sabvoton settings are:
70a continuous
80a max
100a phase continuous
170a max phase

Upon medium acceleration, the battery sags from 80v->70v and honestly, the acceleration off the line doesn't feel very strong. I can't even pull a wheelie although I should be well above 5000w to the motor. The only way I can give it some grunt is to adjust the throttle calibration so that mid throttle=max phase amps. But the issue with this is that it's extremely jerky and dangerous at low speeds due to the instant punch of all the power.

Does anybody have any suggestions or advice to remedy this problem?

Sounds right. You need 200 phase amps in that thing for easy wheelie. Sag sounds about right for a non top tier 1.4kwh 18650 pack. If you want less sag go RC lipo.

No battery in the world performs best at it's absolute maximum continuous current. Look at a discharge graph of any cell on the market and you will see that. They also create an insane amount of heat and live a very short life at max. You always have to overrate batteries because sellers overpromise all the time.

The seller doesn't even list the model of cell ( ICR-18650 is not a model number ), so i can't even google you the datasheet to show you how crappy the discharge is at the 3.5C you're asking of it.

You could probably tune your phase amp to battery amp ratio a bit better in the controller to get wheelies, but the sag will always get in the way.

Thanks for the info!

Neptronix- They are samsung 22p, supossedly rated at 10a continuous

I'll continue to play with the settings. Currently waiting for new brake pads to come in so it's just sitting this weekend. It's almost 100 degrees each day this week in Philly so my ride time will be severaly limited even when I get it back up and running again.

Okay, some quick calcs then.

http://lygte-info.dk/review/batteries2012/Samsung%20ICR18650-22P%202200mAh%20%28Blue%29%20UK.html

You've got a 20ah pack so by division, we get about 9P.
70A / 9 cells - 7.77A per cell.

And whaddya know, lgyte-dk's graph has a 7A discharge ( about 3.5C ) which shows us a very saggy nominal 3.3v, therefore in this condition, your pack is 66 volts nominal if discharged at 70A continuously.
Yeah, basically what i expected.

I always recommend running batteries at half or preferably a third of what they're capable of. a similarly sized pack of Samsung 25R's would have been a better choice for this amount of amps..

As to smoothing the low end torque out, i don't know anything about sabvotons, but the closer the battery amps are to the phase amps, the flatter the torque curve gets. If you want to tune out the low end torque, drop the phase amps.

The link to the battery says 20s7p where is the 9p coming from? 7 cells in parallel is not alot to push 100a peaks so ya 10v sag with 18650 pulling 10a per cell is normal imo

skeetab5780 said:

The link to the battery says 20s7p where is the 9p coming from? 7 cells in parallel is not alot to push 100a peaks so ya 10v sag with 18650 pulling 10a per cell is normal imo

Click to expand...

Each cell is rated at 2150 mah
So 9 * 2.15 ampHr = 19.35 amps

If the pack is 7P with these cells, then the capacity is more 15 Ah

I notice that the cells are labeled 22P M but I haven't been able to find out if the "M" is significant or not. That may account for the discrepancy. If not, there's a real problem with a company selling a 15 ah batter as 20 ah. But lets give them the benefit of the doubt and dig deeper.

Well, it isn't looking good. I found a site showing the 22P M cells and they say they are rated at 2200 mah which gets us to only 15.4 AH for a 7P pack.

Still trying to give them the benefit of the doubt ... is it possible that a 7 got typo-ed as a 9? That would put us at 19.8 AH.
http://www.staringpower.com/product/icr18650-22pm-2200mah/

Each cell weighs 50 grams and the pack supposedly weights 20 lbs. if the pack were 9P, that works out to 19.84lbs. Add some weight for wires and such and it should weigh a bit over 20 lbs. If it is a 7P, then it will weigh somewhere around 15-16 lbs.

At 9P, they are giving a conservative rating for the pack at 70 amps since each cell can run 10 amps continuous. it would probably be a bit less prone to voltage sag. But at 7P they are not being conservative at all and might be more prone to voltage sage if you exceed that 70A draw.

Personally, I'd weigh the pack or count the cells and find out what the heck is going on.

The Borg starts lifting the front wheel above 120A motor current. This with Cromotor and 23.5" diameter rear wheel, Sabvoton 72150 and 18S 32AH from 6S 8AH Multistar bricks in 3S4P.

I get sag, but less than 10V. 80A battery current divided by 4P is 20A per 8AH brick, so 2.5C. Acceleration is excellent. Some people are complaining these days about Sabvoton acceleration, but in some of these cases they have incorrect expectations. The Sabvoton is designed to monitor motor current and produce the set amount, providing there is enough pack voltage. If the pack sags it will limit the max current the controller will produce.

Some folks want to hit max battery current at zero speed to get really strong off the line acceleration. They don't seem to understand that this requires really high motor currents, well beyond 3x the battery current, it could be 10x the battery current or more. It heats the motor like crazy and may blow FETs. The Sabvoton will not produce more than the rated motor current because it is measuring that current and will not go beyond the design limits. Very simple controllers don't measure the motor current and may produce these really high currents, and sometimes they blow up.

To keep the control loops stable the rate of current increase is probably limited as well. There are tuning parameters, so it may be possible to adjust this, but it may make the controller over respond and hard to modulate.

To get high power requires a high power battery pack. The manufacturer's ratings should be significantly derated for best results. To get high performance requires high performance cells (and plenty in parallel), low resistance interconnects, and high current BMS FETs. The nickel strips used in 18650 packs are not very good for high current. It takes a lot of extra parallel strips to get to 80 amps continuous without losing energy and heating up.

Sounds like they are overrating their packs. The maths just doesn't add up.

For real power from Li-ions you really want a 10p pack made from Samsung 25R or Sony VTC5A cells this will deliver 200A continuous with a modest sag at 70-100A draw.

A 25R sags about 0.4v at 20A which would be about 7.5v on a 19s pack. That equates to a 7.5v drop at 200A on a 19s10p battery. If you pull 100A (10A per cell) it would only sag 3.8v on the same pack.

A 19S7P pack of 22PM cells will sag about 8.5v (0.45v per cell) at 70A when fully charged, best case scenario. If you are getting much more than that then the pack may have some additional resistance issues going on.

lionman said:

Sounds like they are overrating their packs. The maths just doesn't add up.

Click to expand...

Right - the math isn't right. I'm crossing my fingers hoping that the listing is in error and that the pack is actually 9P and not 7P. If it isn't 9P, I'd be pretty upset with them.

wturber said:

lionman said:

Sounds like they are overrating their packs. The maths just doesn't add up.

Click to expand...

Right - the math isn't right. I'm crossing my fingers hoping that the listing is in error and that the pack is actually 9P and not 7P. If it isn't 9P, I'd be pretty upset with them.

Click to expand...

I emailed LAebike about this on July 14th and haven't received a response

That's a bummer.
Well, like many vendors.. they oversold their batteries, and thus you underbought.

Never run a battery near it's maximum. You ideally want to run a battery like that at 1/4th of it's maximum continuous amperage, which would result in 1/4th of the sag, of course.

I'd recommend you throw that up on the sales section and pick yourself up a nice pack of 7P Samsung 25Rs, at the very least.. ideally more like 10p!

That amount of sag is killing your batteries, because they must be getting hot. I draw 150A, with a 31s pack and the most sag I see is 5V, and I have blistering acceleration not remotely possible with a more standard geometry bike like you guys are running. The pack is 22ah 115V nominal with automotive grade cells that never get the slightest bit warm to the touch, and I fully expect to get 8-10yrs of use out of them.

John in CR said:

That amount of sag is killing your batteries, because they must be getting hot. I draw 150A, with a 31s pack and the most sag I see is 5V, and I have blistering acceleration not remotely possible with a more standard geometry bike like you guys are running. The pack is 22ah 115V nominal with automotive grade cells that never get the slightest bit warm to the touch, and I fully expect to get 8-10yrs of use out of them.

Click to expand...

I'm feeling pretty bummed about getting ripped off by LAebike. What type of batteries are you using?

Yermommy said:

John in CR said:

That amount of sag is killing your batteries, because they must be getting hot. I draw 150A, with a 31s pack and the most sag I see is 5V, and I have blistering acceleration not remotely possible with a more standard geometry bike like you guys are running. The pack is 22ah 115V nominal with automotive grade cells that never get the slightest bit warm to the touch, and I fully expect to get 8-10yrs of use out of them.

Click to expand...

I'm feeling pretty bummed about getting ripped off by LAebike. What type of batteries are you using?

Click to expand...

I wouldn't say you got ripped off, it's just the nature of the 18650's. The high capacity cells need a lot more in parallel to safely put out high current, which is why Tesla's have such large battery packs. My main issue is with the wasted space, since any cylindrical cell inherently wastes 22% of the volume it takes up. The cells I have from a Ford are apparently Panasonic prismatic cells with 5mm threaded terminals that bolt together quite easily. I just wish they would produce similar high power cells with dimensions that are more ebike friendly, though with tape compression I was able to squeeze the thickness to fit 4 strings of 5s into an EEB frame with the plastic side covers providing compression too.

John in CR said:

Yermommy said:

John in CR said:

That amount of sag is killing your batteries, because they must be getting hot. I draw 150A, with a 31s pack and the most sag I see is 5V, and I have blistering acceleration not remotely possible with a more standard geometry bike like you guys are running. The pack is 22ah 115V nominal with automotive grade cells that never get the slightest bit warm to the touch, and I fully expect to get 8-10yrs of use out of them.

Click to expand...

I'm feeling pretty bummed about getting ripped off by LAebike. What type of batteries are you using?

Click to expand...

I wouldn't say you got ripped off, it's just the nature of the 18650's. The high capacity cells need a lot more in parallel to safely put out high current, which is why Tesla's have such large battery packs.

Click to expand...

I wouldn't say he got ripped off until we hear from the the vendor. But until then, there appear to be significant inconsistencies with what they claimed to be supplying and what they've actually supplied. It doesn't add up. The difference probably isn't the full (or even primary) source of his sag problems, but its still a difference worth complaining about and getting and explanation for IMO.

Pretty common for there to be discrepancies between claims and actuals. That's marketing. My shop vac says 4hp on 120V, for example, whereas the electrical code only allows 12 amps continuous. A lot of things are measured in different ways from what one would expect to produce these claims.

Alan B said:

Pretty common for there to be discrepancies between claims and actuals. That's marketing. My shop vac says 4hp on 120V, for example, whereas the electrical code only allows 12 amps continuous. A lot of things are measured in different ways from what one would expect to produce these claims.

Click to expand...

The price of amp hours being what they are, I'd say an overmark of 25% capacity is beyond fudging for marketing and would not be acceptable. I do understand that fudging goes on with certain consumer items. For instance the voltages of the battery packs that are sold for some "higher voltage" power tools. But even those packs are usually sold with pretty accurate amp-hour or watt-hour specifications and they do operate at the voltages claimed.

As an example of this kind of consumer targetted BS/hype/misleading crap, I was beyond pissed when I got home with my 60-watt Weller soldering that I purchased to solder some XT60 connectors only to find that it was really an 18 watt iron that was labeled as "60 Watt performance". The 60 Watt bit was LARGE and the "performance" bit was small. And trust me, it doesn't have 60 watt performance given its 18 watt wall-wart power supply. It only has 60 watt "performance" perhaps in some very narrow/specific mealy-mouthed marketing way. Their "marketing" is dancing on the edge of being outright fraud and earned them my unofficial @ssh0les of the year award. I will avoid Weller henceforth. But even so, their claims come down to matters of opinion, and attention to detail (read the fine print) not fact. I'd bet you'll find some fine-print disclaimer on your shop-vac somewhere. I don't see how you could have a similar situation regarding the amp-hour capacity of an e-bike battery.



But just to be clear, I'd reserve judgement on the battery pack until the vendor has a chance to explain (we coulda missed something) and/or until the pack is actually tested so we know more precisely what it is. That said, their lack of response isn't helping matters.

Thank you to everybody who has contributed to this thread. LAebike has had a message on their website for months that they are moving storage facilities and will not be able to respond to emails very often. I will be taking apart the battery pack this weekend to discover if they lied about the AH rating on the battery or not.

Yermommy said:

Thank you to everybody who has contributed to this thread. LAebike has had a message on their website for months that they are moving storage facilities and will not be able to respond to emails very often. I will be taking apart the battery pack this weekend to discover if they lied about the AH rating on the battery or not.

Click to expand...

You would probably have to do a discharge capacity test on the cells to know for sure.

What that soldering iron is a toy especially the way the tip plugs in like an headphone Jack. A 60 or 80 watt weller fat tip iron. Yeah the old-fashioned one were there copper tip is almost as big as your little finger. Let it heat up good and hot before you start. Yeah I've been trying to get a good 20s battery that can handle 80 amps so I'd like to be made to handle 100 hundred 20 amps. That means Samsung 25r.

999zip999 said:

What that soldering iron is a toy especially the way the tip plugs in like an headphone Jack. A 60 or 80 watt weller fat tip iron. Yeah the old-fashioned one were there copper tip is almost as big as your little finger. Let it heat up good and hot before you start. Yeah I've been trying to get a good 20s battery that can handle 80 amps so I'd like to be made to handle 100 hundred 20 amps. That means Samsung 25r.

Click to expand...

Actually, the iron is good for smaller electronics type of jobs where you need to do a fine soldering job and/or with thin wires - jobs where you need a small spot of heat and not a ton of it. Weller should have simply labeled their iron for for what it is. Its fine for certain things. But, "18 Watts" should have been prominently displayed somewhere on the package - especially since all the other Weller irons had their real wattages displayed in circles just like the one on the package I bought. I, of course, I ended up buying a REAL 60 watt iron with a nice wide chisel tip for soldering my cables. It works well. It wasn't a Weller. It'll never be a Weller for me. Ever.
Seriously ... ever.
(Yes, I can hold a grudge because our markets give me options.)
For ever.

Well well weller hater.

Update: I opened the battery this weekend to discover that there are indeed 9 parallel groups of 20s 2150mah batteries in here. This means LAebike did not lie about the AH rating, but the 20s7p/140cell count was wrong. Lesson learned here- when spending $1000 on a battery, reach out to a recommended builder rather than an unknown vendor.