it seems terrible on paper but in practice with the higher P counts (4 and over) its not as bad as it seems like it would be. There’s no free lunch, so if you want more amp hours you lose discharge capacity, if you want more discharge capacity you lose amp hours, and if you try to do both by achieving some kind of balance you get various degrees of voltage sag. That’s my observation anyway.
They’re just old and outdated. There’s better options now that don’t break the bank.
Samsung 30Q from Nkon @ $2.75 if you’re outside Europe. 3Ah per cell, much better power delivery than 25Rs. 25Rs sag as much drawing 5A/cell as the 30Qs do drawing 15A/cell.
Sanyo GA cells are more expensive ($5 a cell). They have 3.5Ah/cell and sag just as much as 25R cells drawing the same amount of current.
Where are you getting those prices and in what quantities? I’m not seeing that kind of pricing anywhere i would actually be willing to buy from. I don’t buy cells from ebay or alibaba, i get them from lionwholesale usually but i try to make sure whoever i get them from are in the US and can actually stand behind a product without getting returns lost in China Post territory.
the 30Q has a max continuous discharge of 15Amp, leaving a 60Amp pack in a 4P config. The 25R has a max continuous of 20amps and a sub-second burst capacity of up to 100Amps. That’s the trade off. You either get more amp hours or more continuous current at a time. I’m betting voltage sag correlates with max continuous current because i keep seeing “better batteries” around that don’t deliver their juice as fast or as hard.
The Sanyo NCR18650GA deliver a whopping 10amps max continuous. Not interested.
So yeah, by limiting their current draw within the cell, they have effectively not allowed you to draw the kind of power that not only results in voltage sag but also results in higher discharge capabilities and a lower amp hour spec.
I guess that makes them better, i don’t know. I build 80 amp packs and use a 60 amp BMSs and try to tune my VESCs to not launch like an out of control rocket, so the voltage sag I am actually seeing on the street just isn’t anything i wouldn’t expect. Out of any cell.
So should I use 15amp cells with a higher Ah rating so that my volt meter doesn’t drop a percent or two while I’m hitting the throttle? Or should i just let the hardware draw what it wants from a set of cells that are willing to give them their all?
It deserves a good pondering i suppose.
he gets them from nkon.nl - a european supplier that also ships to the US apparently.
The current ratings don’t mean anything. You need to look at the discharge curves. You will see that the 30Q don’t just deliver 15A but can go much higher. You will also not drain them continuously at that current so the added capacity and less voltage sag make a big difference compared to the higher temperature increase they see at continuous high current use.
We actually need to get away from this “I build 80amp packs” as your basing this info solely on the rated current. It says nothing about the actual performance of the cells. 30Q would make a rated 60A pack but would kick your 80A pack’s ass 
1 Like
I’m curious about those batteries myself. I might try them out in my next build.
But even on that nkon.nl site, 40 of them still works out to about $3.30 (cause no VAT). I want that $2.75 price 
@longhairedboy I source 30Qs from Nkon and have them shipped to the US. I source GAs from Liionwhesale too.
Stop looking at just the rating. That’s the issue with a lot of people here, you see numbers and take them by heart without understanding the reasoning behind them is. Look up the spec sheets and read them if you have the time. The reason 30Qs are rated to 15A is because if you continuously run them at that current and charge them at 4A, after 300 cycles, you only get 73.8% of rated capacity. That’s where Samsung decided to draw the line. In the same spec sheet they show that for 22A continuous and a temp cutoff of 70C and charge at 4A, after 250 cycles, they get 72.4%.
Similar stuff is done by Panasonic. You can run the cell at higher currents CONTINUOUSLY by sacrificing number of cycles you can run them for. Key word there is continuously. We do not do this unless you’re climbing an endless 15 degree incline at 20mph.
Check these graphs out, I’ve posted them several times:
First is Sanyo GA vs 25Rs: This is continuous current running through them, the vertical difference between curves is voltage sag at that specific capacity. The red curves are GA and blue is 25R. From top to bottom 0.2A, 5A and then 15A. Henrik’s battery holder failed on one of the 15A GA test, that’s why it is cut short. The other one he stopped at 75C.
The takeaway here is that they have less voltage sag than 25Rs, aka they can deliver more power for the same throttle setting at the same discharge level. Also, if you were to throttle to 15A with 25Rs sag down to 3.2V which is most people’s soft cutoff after 1.125Ah consumed. With the GAs, you sag down to 3.2V after 1.55Ah consumed. The gap widens if you bring the cutoff lower.
Next 30Q vs 25R: Same thing as before, Red is 30Q, blue is 25R. Top to bottom 0.2A, 5A, 15A
Here, there is a big difference. The 30Qs drawing 15A sag down as much as the 25Rs drawing only 5A after 1.25Ah consumed (which is 50% capacity of 25Rs). At 15A and 1.25Ah consumed, the 30Q outputs 2070W, the 25Rs on the other hand output 1902W. At 15A, 25Rs sag to 3.2V at 1.125Ah, 30Qs sag to 3.2V at 2.125Ah, that is almost double!
As I said 30Qs are much better at delivering high power than 25Rs. GAs can deliver slightly more power while having 40% more capacity!
I hope this is clear, I’ve been writing this on and off throughout the day while flight testing lol.
I intentionally left out the 20A case. Henrik tested the 25Rs a long time ago and he didn’t measure temperature back then. We cannot say that the cell did not overheat as he didn’t measure this. He didn’t test GA cells at 20A because he thought they would overheat (and he broke his cell holder), rightly so, but again, we do not run continuously. He stopped the 30Q test early at 20A because of heating. He got through 2Ah before he stopped the test because the cells reached 80C. That would be. 6 minutes continuously at 20A per cell.
It seems that Nkon’s prices have gone up since I ordered
3 Likes
That’s a terribly convincing argument. I think you’ve made a believer out of me.
The more people get on board and spread the knowledge the less I will have to repeat myself and the sooner esk8 will leave behind using an overrated and outdated cell! 
Clearly we need a battery FAQ so we don’t have this conversation every week in yet another thread
2 Likes
I wrote this 2 months ago in a battery thread and was wondering why companies like Enertion are still using the 25r. I think it´s just the “we´ve always done it like this and keep doing it like that” thing. Or maybe it´s the price point - 25R are slightly cheaper than 30Q.
well at least he is trying:
1 Like
yep! So I think we have to spread the word of battery gods to the world and after a few months we´ll be having more people with cool packs on the street that also last longer 
I still feel bad about building myself a 25R pack 
1 Like
Like I told you, you had the chance to hopp in our group buy 
But the 25r is still a good battery, it´s not like you´re having a bad pack now
I bet it’ll come at a hefty price increase… @smurf, I’ve been thinking about that but I don’t think I have the time to write it…
he says ~50-70USD - not that bad actually.
What is the price difference per cell. If it’s any higher than that, it’s bs. Lol, I bet you he saw multiple threads showing that 30Qs are better and decided he could make a quick buck. I just checked, the price difference between 25Rs and 30Qs in bulk is 30cents. For a grand total of $12 increase in cost.
That’s what pisses the crap off of me of people like that. Making money on other’s work.
You’re going to be pissed at him a while then. His entire model from the beginning was to leverage the DIY community to fund proprietary development and ultimately develop a product that leaves DIY behind and raises margins. It’s playing out exactly as i predicted a couple of years ago. Its not an uncommon model, and at least one positive result was this community, not to mention a seriously elegant drive system.
So i just checked pricing… NKON definitely has a great price on those 30Qs, so definitely worth switching over to them. Too bad when i loaded 50 of them in my cart they told me they didn’t have shipping options for my address. I was just trying to see what the total would be with shipping from Europe. Weird. I live in a house on a street. I wonder if that’s an issue? lol
Edit: nevermind, they have a seperate site for ordering from outside of Europe. They ship via FedEx and it might take 3 weeks for them to get here but it still seems worth it. The price is still really good. I’m just going to have to stay ahead if i want to reduce build times. One reason i like liionwholesale is because the shit is at my house within a week. That’s not worth the premium, though.
2 Likes
“Proprietary development” /sarcasm
i know, i know… but i might know some things i can’t talk about…