Batterypack for speakersystem!

Grymer

New member
Joined
Sep 14, 2018
Messages
18
Hi! My name is Nicolai from Denmark. First thread from mewuhu!

Pre-story

I've build 12v speakersystems using heavy truck-batteries. Sometimes multiple batteries. You can see my greatest creation here(Scroll down to see final product). This speaker weights like 0,7 tonnes.

Situation

Now I'm going to uni, which is far away from home. I'll be living in a VERY small apartment, but I'm not willing to let go of LOUD speakersystems haha. Thus I'm building a system running on light 18650 batteries. My SURE 2x300w amplifier will work up to 52v. Unfortunately the very common13s pack that is rated 48v goes up to 54,6V. So I'm going for a 12s solutioninstead. Choice of BMS

So the pack will be 12s3p -> up to 50,4V with 10,5Ah. :cool:

Questions and doubts

1.Hopefully I can get a hold of a spot-welder. If not - I'll use plastic holders instead. A 18650-cell-lover on youtube have tested them and they can do 4a constant per battery. I'll do 12a peak, so 4a per battery (Remember music is dynamic). If I can get a hold of a spot-welder, i'll use plastic spacers.What do you think of this solution?

2. Is there anything in this setup, which is just plane stupid. I am new to designing such battery, so it wouldn't be a surprise if it was

3. Can you find a good priced charger for a 12s setup. I can find one for a 13s setup, and if that will work on the 12s, then I'll just go with them. REMEMBER: This is not the BMS. The BMS will be for a 12s pack.

Ekstra thoughts:

Funny thought: It's surprising how easy it is to add another parallel layer. And how cheap it is. Right now it'll increase my battery capacityby 33%. If I'm crazy enough I'll increase the parallel number by one or two. It's not gonna increase weight by much. And the BMS is already in place :D

NOTE: If you want to read a bit about the designing-proces:
PART 1
PART 2
FINAL - w/o battery and amp box.

Thanks in advance. And sweet forum you've got here! :D
 
Wow that's a nice setup you got there!
I would recommend going with the LiFePO4 cells because they can handle a much higher amp draw, since you'll be using them for sound systems that is what you probably want to go after

Even though those holders can handle 4 amps, it can't sustain it for a prolonged period of time. The tabs that goes to the circuit board will heat up enough to actually start to melt the plastic.
Running the amp off those cells in a 3P configuration will not give you enough continuous amps to be able to power the amp for very long. The cells will begin to overheat and/or will shorten their life drastically. I would recommend going with at least a 6P configuration.
 
Korishan said:
Even though those holders can handle 4 amps, it can't sustain it for a prolonged period of time. The tabs that goes to the circuit board will heat up enough to actually start to melt the plastic.
Running the amp off those cells in a 3P configuration will not give you enough continuous amps to be able to power the amp for very long. The cells will begin to overheat and/or will shorten their life drastically. I would recommend going with at least a 6P configuration.

Thanks for taking your time to give feedback on my project. I actually think I'll end up buying a spot-welder anyway, if I can't access one.

Now the batteries I have linked to can handle constant draw of up to 10A. And if my peak draw is 4A and average is actually 0,5-1A, then I'll be more than fine! (This is estimates, but since I've build speakers like this for some time, they're not far off)
[img=500x500]https://cdn6.bigcommerce.com/s-t56q...int__74740.1489330179.1000.1200.jpg?c=2[/img]
When looking at the picture above. Imagine that black is power-draw. Only when the speaker is cranked to 100% (Which is next to never) will the sub draw perhaps 240w and the tops 150w during peaks. How often are these peaks?Those tiny black lines that stick out from the rest are peaks. And power-draw is logarithmic, so the middle is actually far less power-consuming than the peaks.

According to my more in-depth speculations, and not just what the amp-specifications says. Peak-draw will be: 400w/44v = 9a. 9a / 3 = 3a peaks pr. battery.
 
Ok, kewls. Just as long as you're aware of the ratings ;)

Out of the listed bms you chose, expect 30A to be the more accurate rating. But if 4A is about the max, you should be fine.

The charger and the bms must match on their series. So if the charger is for 13s, the bms needs to be as well. Otherwise you'll have issues.
That charger for the 13s is assuming you charge to 4.2V/cell. If you use 14s, then that'd be 3.9V. This will increase the life span of the cells drastically, by well over 1000 cycles.

Ok, re-read your post. 52V being the max, this makes things a little more difficult. I believe you can use a buck converter that would allow you to use 57V pack (14s) and still have 4A available to you.
There was a pretty powerful buck converter I saw on here the other day. Don't recall who posted it, tho.
 
Korishan said:
Ok, kewls. Just as long as you're aware of the ratings ;)

Out of the listed bms you chose, expect 30A to be the more accurate rating. But if 4A is about the max, you should be fine.

The charger and the bms must match on their series. So if the charger is for 13s, the bms needs to be as well. Otherwise you'll have issues.
That charger for the 13s is assuming you charge to 4.2V/cell. If you use 14s, then that'd be 3.9V. This will increase the life span of the cells drastically, by well over 1000 cycles.

Ok, re-read your post. 52V being the max, this makes things a little more difficult. I believe you can use a buck converter that would allow you to use 57V pack (14s) and still have 4A available to you.
There was a pretty powerful buck converter I saw on here the other day. Don't recall who posted it, tho.

So youre suggesting using a higher voltage and then a step-down, to make sure I dont exceed the max-voltage of my amp?

What charging voltage is needed per cell? Does it exceed 4,2V per cell? That would make me unable to play and charge at the same time :(
 
You use a charger that can go to about 58V, this will make each 14s 4.14V max. If you use a standard lead acid charger for 48V, it can top out at around 56.4V (14.1V per "12V" battery)

So, you'd have:
Charger -> Battery -> Buck Converter -> Amp

Or, another way to look at it is:
Code:
.       / -  Charger
.Battery
.       \ -  Buck Converter -> Amp

You'll be able to charge the battery while using the buck converter. As long as the charger outputs more amps than what the Amp/Buck are consuming, that is.
 
Korishan said:
You use a charger that can go to about 58V, this will make each 14s 4.14V max. If you use a standard lead acid charger for 48V, it can top out at around 56.4V (14.1V per "12V" battery)

So, you'd have:
Charger -> Battery -> Buck Converter -> Amp

Or, another way to look at it is:
Code:
.    / - Charger
.Battery
.    \ - Buck Converter -> Amp

You'll be able to charge the battery while using the buck converter. As long as the charger outputs more amps than what the Amp/Buck are consuming, that is.

Now -Would this be necessary if I get a 12s3pbattery? I actually found a charger for a 12s setup. Its only 2a, but thats just fine for me

And the amp will still do lots of watt for the speakers with a voltage of 44V. More than enough. And I dont want any of unnecessary processes that lower efficiency.
 
Yeah, as long as everything fits good, then go with it. The higher series will just get you longer jamming time ;)
But so will higher parallel count too
 
Korishan said:
Yeah, as long as everything fits good, then go with it. The higher series will just get you longer jamming time ;)
But so will higher parallel count too
Thanks for your interest in the project!

I think 12s3p will give me a VERY good battery for the time that the speaker will be "in the field". The battery will give me atleast 10hrs play-time when max'ed out.

To compare.
A soundboks has a 99kWh battery. Uses 3x 72w amp. Totals about 215w. They claim a 8hrs play-time at full volume. Though that is probably more like 6hrs. (Because real-life parties plays more bass-music than during their testing-methods). So...

99Wh battery/215 amp effect= 6hrs.

The correlation between battery life, play hrs and amp effect is:

Battery Wh/ (Amp effect *0.077).

In my example:600 / (600*0.077) =13hrs playtime.

The example above is very estimating since several parametres will have an impact on the outcome. Most dominant: Music genre and actual amp-output.

Maybe I'll come up with a way of having interchangable battery-modules. This way I can do some battery-harvesting like I've seen you do on this forum. The BMS is not that expensive after all. Then I'll build perhaps 3x 12s3p batteries.
 
Grymer, it sounds like you can use the Headway 38120 cells instead.

Google it, like you have this community here that mainly specialises in powerwalls, there's also a community that's into audio.
Mostly for car audio but since you want to work with battery it's the same.

The Headways are even Lifepo4 cells as Korishan suggested.

Nice project anyway.
 
yoeri_w said:
Grymer, it sounds like you can use the Headway 38120 cells instead.

Google it, like you have this community here that mainly specialises in powerwalls, there's also a community that's into audio.
Mostly for car audio but since you want to work with battery it's the same.

The Headways are even Lifepo4 cells as Korishan suggested.

Nice project anyway.
Hi Yoeri

Thanks for your contribution to my thread!

I looked up 38120 and the wouldcost me a lot more. I can get the cells I'm looking at for 3,95 pr cell, and since these cells have a good reputations and are waaay within my margin of needs.Both in terms of max discharge and capacity. I'm not entirely sure why I would bother finding another type of cell?
 
12 x 3 cells = 36 cells x (3.6V nominal x 3.3Ah) = 427Wh
9.25 Hours using your calc...

Plus, that is if you are charging the cells fully upto 4.2V... and discharging to 2.5V

With 4.2V charge and 3.3V discharge (2.8Ah per cell) you have 50.4V max and 39.6V min with the time that 50.4V is present is only when charging. The 3.3V per cell drops your power available down to 362Wh or 7.8hrs playback.

Moving up to 13s4p would give you a better working voltage range and energy without having to charge the cells up to 4.2V, reducing some of the issues for BMS selection..


26650 x 5Ah cells approx $3.20... direct from China, 500 cycle life at full charge discharge, 20A peak discharge rate.
https://lygte-info.dk/review/batteries2012/LiitoKala INR26650-50A 5000mAh (Black) UK.html
These are re-badged pbbattery...
 
completelycharged said:
12 x 3 cells = 36 cells x (3.6V nominal x 3.3Ah) = 427Wh
9.25 Hours using your calc...

Plus, that is if you are charging the cells fully upto 4.2V... and discharging to 2.5V

With 4.2V charge and 3.3V discharge (2.8Ah per cell) you have 50.4V max and 39.6V min with the time that 50.4V is present is only when charging. The 3.3V per cell drops your power available down to 362Wh or 7.8hrs playback.

Moving up to 13s4p would give you a better working voltage range and energy without having to charge the cells up to 4.2V, reducing some of the issues for BMS selection..

26650 x 5Ah cells approx $3.20... direct from China, 500 cycle life at full charge discharge, 20A peak discharge rate.
https://lygte-info.dk/review/batteries2012/LiitoKala INR26650-50A 5000mAh (Black) UK.html
These are re-badged pbbattery...

Perhaps I went a little too quick with the calculations haha :D

Those cells in your description: 26650 cells. I have been adviced to go for cells that come from a reliablesource -> which these cells do. It's just to ensure that the cells are good. Perhaps Ihave been ill adviced? I kind of hope now that I see I can save a fair bit of money! :D

And I don't see how your calculations with 3.3V makes sense. The voltage varies, so the total capacity will be a result of the average voltage. Not lowest voltage. And I think thecapacity of my cells aretypically3,45Ah = 447Wh. = 9,6hrs Still longer than one would expect. And if the battery goes flat I can build external batteries that I can connect and thus extend the battery life. I'll try and build a battery using older harvested cells. Perhaps make it 12s5p (or 6p).

And the thing about playtime is,my amplifier will output less wattage as voltage goes down, so if I crank the amp all the way up, then it'll still play for 9.25hrs since max output will be a result of in-going voltage. It's the same with the SOUNDBOKS. It goes from 12,7V to 11,0 (Not entirely sure

Now if I go for 13s I won't be able to charge and play at the same time. Which is muchneeded. And when fully charged. The cells will be 4,2V (If I understand correctly), so when fully charged, I won't be able to use the speaker either.

And yet again. Welcome to my thread. I am thankful for your time and interest in the project!
 
New problem: The amp is actually 48V. I thought it was 52V since the amp-chip goes to 52V, but the amp itself is only rated at 48V. Thus I need to use 11S, but I can't find a BMS 11s from china, so I might actually have to follow your advice and get a small step-down converter. Perhaps one that just regulates to 47,5V at all times. Can someone link to one, since I am not sure what to search for. I tried, but didn't have any luck.


EDIT: I did find a new 11s, so now I'm think I might go for a 11s3p or perhaps (If I have the money for it) 11s4p.

When the amp is at 48V, I can get about 245w in each channel whilst distortion is under 1% and 300w with distortion at 10%. If I have a nominal voltage of 39,6V with the 11s pack using 18650cells with nominal 3,6V. Then I'll have 247.5w with less than 10% distortion and 202w with distortion less than 1%.

And actually. These numbers will be a bit highere since they're calculated using speakers at 4ohm, whereas mine is actually varying from 3,2ohm to 4ohm.

And 200w for a subwoofer is actually quite good. I will be able to produce 123dB 50hz outdoor and about 133dB indoor. And indoor reflections are not calculated. These are preeeeeetty good stats tbh.
 
Have a look at the chart on page 2 of the panasonic cell specification you linked in the first post.

The link I sent has actual cycle test results for the cells, I use the site often if I am looking for cells because they are tested independently, consistenly and seem quite accurate. Cycle life is a separate issue and story, don't cycle at high rates and to maximum voltage an you will get a long life out of them. I have 40 x 26650 x 5Ah cells in an electric skateboard (4 wheel drive !) and it works ok, pulls my fat ar*e up a 20% hill.. 1.4kW..

You can charge and play at the same time, just don't charge up to 4.2V because between 4V and 4.2V you gain relatively little in terms of actual cell capacity. Again check out the chart in the specification.

If your getting 10% distortion at 39.6V and charge to 4V per cell (52V), discharge to 3.3V (42.9V) and that would give you approx 365Wh with 13s3p (2.6Ah per cell X 3.6V nominal X 13x3 cells). Go with 4p and you get approx 486Wh.

One test that may be useful is to buy / borrow a plug in kWh meter (plug type) and run your music system through it for 3 hours at typical noise levels and check the energy used.
 
I checked out some other sites to get a consensus result, and the max voltage seems to be 50V. It probably can handle 52V, but things might get really warm at that voltage.

So with 50V / 4.1V = 12.1. So a 12s setup would still work. This will also allow your cells to get longer cycle life.

A good way to check the rating truly, is to see what the capacitors are rated for. That'll help a lot in diagnosing it. Then you can look at the voltage regulator and its specs to get even better results.
 
Korishan said:
I checked out some other sites to get a consensus result, and the max voltage seems to be 50V. It probably can handle 52V, but things might get really warm at that voltage.

So with 50V / 4.1V = 12.1. So a 12s setup would still work. This will also allow your cells to get longer cycle life.

A good way to check the rating truly, is to see what the capacitors are rated for. That'll help a lot in diagnosing it. Then you can look at the voltage regulator and its specs to get even better results.

The retailer adviced me to stay below 48V. I wrote an email to them and they said it would be wise to find another way to tackle the problem. And I think I'm going to stick below 48V. I don't need 100% of the avaliable wattage.

What if I charge them to 4.0V? That is precisely 48V:D

If I want to charge my cells to 4.0V, do I just set my charger to 4.0V?

Maybe this will do (48V, 5A) https://www.ebay.co.uk/itm/AC-110-2...d6:m:m8E6cn2PnwT0FRiIeFrt5pQ&var=661042246179

completelycharged said:
Have a look at the chart on page 2 of the panasonic cell specification you linked in the first post.

You can charge and play at the same time, just don't charge up to 4.2V because between 4V and 4.2V you gain relatively little in terms of actual cell capacity. Again check out the chart in the specification.

If your getting 10% distortion at 39.6V and charge to 4V per cell (52V), discharge to 3.3V (42.9V) and that would give you approx 365Wh with 13s3p (2.6Ah per cell X 3.6V nominal X 13x3 cells). Go with 4p and you get approx 486Wh.

One test that may be useful is to buy / borrow a plug in kWh meter (plug type) and run your music system through it for 3 hours at typical noise levels and check the energy used.

It's good to see the panasonic cells in anactual tests. Seems like it delivers around 3,0 - 3,1Ah. Which is still fine for me. I'm sure that the same unreliability applysfor the SB battery as well. They're probably just saying (max voltage * Ah) for marketing! But I know for a fact that they run for atleast 6hrs w/o a problem.

And distortion doesn't depend on input-voltage alone. It is also a matter of wattage. I could feed it 35V and still achieve 100w w/o distortion.
The plan is to order everything today, so I can build my speaker and have it ready before mid-November. I'm ordering everything tomorrow. And I'm not really going to measure kWh since it will heavily depend on the musical genre. I'm pretty sure I can drain the batteries in an hour if I play the right music. But since most dance/party music is less bassy and more kicky, I'll have a lot better battery lifetime.

Do you reckon I should order extra cells to be sure, or can I just go for 36cells (12s3p setup) ?
 
There are multiple retailers, and they will probably all tell you something different.
I was also going off the manufacturer of the IC and they state about 52.5V is max. Mouser.com (Texas Instruments IC)

But that's your choice which one you will agree with. It's your money and equipment ;)

Charging to 4.0V is just fine. Going with a standard power supply will not give you your full capacity on the cells. Even if it charges the cells to 4.0V, you won't have full capacity. This is because of the nature of Lithium cells. They really like the switch from CC (constant current) to CV (constant voltage) switch over in the last part of their charging. It's sorta like a saturation phase for the cells. If you were to liken it to sand in a bucket, it'd be like shaking the bucket slightly to be able to fit a little more sand in. You still have a full bucket after the first quick fill. But you can squeeze a bit more in with a little shaking.
It's not "bad" to not do this, just isn't the best way.
 
Korishan said:
There are multiple retailers, and they will probably all tell you something different.
I was also going off the manufacturer of the IC and they state about 52.5V is max. Mouser.com (Texas Instruments IC)

But that's your choice which one you will agree with. It's your money and equipment ;)

Charging to 4.0V is just fine. Going with a standard power supply will not give you your full capacity on the cells. Even if it charges the cells to 4.0V, you won't have full capacity. This is because of the nature of Lithium cells. They really like the switch from CC (constant current) to CV (constant voltage) switch over in the last part of their charging. It's sorta like a saturation phase for the cells. If you were to liken it to sand in a bucket, it'd be like shaking the bucket slightly to be able to fit a little more sand in. You still have a full bucket after the first quick fill. But you can squeeze a bit more in with a little shaking.
It's not "bad" to not do this, just isn't the best way.
If there is a better charger that charges that extra, then please show me! Im not very sure what to search for!
 
Grymer said:
Korishan said:
There are multiple retailers, and they will probably all tell you something different.
I was also going off the manufacturer of the IC and they state about 52.5V is max. Mouser.com (Texas Instruments IC)

But that's your choice which one you will agree with. It's your money and equipment ;)

Charging to 4.0V is just fine. Going with a standard power supply will not give you your full capacity on the cells. Even if it charges the cells to 4.0V, you won't have full capacity. This is because of the nature of Lithium cells. They really like the switch from CC (constant current) to CV (constant voltage) switch over in the last part of their charging. It's sorta like a saturation phase for the cells. If you were to liken it to sand in a bucket, it'd be like shaking the bucket slightly to be able to fit a little more sand in. You still have a full bucket after the first quick fill. But you can squeeze a bit more in with a little shaking.
It's not "bad" to not do this, just isn't the best way.
If there is a better charger that charges that extra, then please show me! Im not very sure what to search for!

And also. Won't the powersupply stay 48V when battery reaches 48V. Thus resulting in constant voltage CV?
 
Back
Top