Modular Hobby Cell Battery Pack Design

[winzEracer]

Hello,

I have been building a modular battery box for my eKart, and as the design has progressed I feel that it is something that is easily replicated by others for karts, atvs, and motorcycles. A bit over kill for an ebike though.

A little about the pack, I am building a 36s5p pack with Turnigy Nano-tech 6s 35-70C bricks. Two of the below described modules make up the 36s5p pack. My first design I tried to use bullet connectors and I made bussbars like the one that Jozzer sells on his site with the bullet connectors sticking out of a block of copper. I found that this was hard to electrically insulate (without making it look like a pile of scat!) and due to the mechanical configuartion of the pack was hard to service. So I did a major redesign from the ground up and came up with this modular scalable design.

Here is a screen shot of the design, describes its basic shape and configuration.
View attachment 7

After the model, I decided to use Blue Sea Systems 150 amp continuous bus bars which cost about $20 each ( I used 3 per module) http://www.bluesea.com/products/2722/DualBus_Plus_150A_BusBar_-_1_4in-20_Stud_5_Gang, which fit well and retained a professional quality, although they are technically only rated (or tested) at 48vdc, I attached them with .125" Polypropylene "L" brackets, as seen it pics below, to make sure they were isolated well.
This change will also affect how the Anderson Connectors attach, I will most likely leave a pigtail with an Anderson connector on it.
I also cut off all bullet connectors (What a happy moment that was!) and crimped ring terminals on each of the leads from the 6s Bricks which then lands on the Blue Sea bus bars. Each bank of 6s5p will wire up to one of the dual bus bar units and I will then place series jumpers between each of the Blue Sea units, one of the series jumpers will have a fuse inline for short circuit protection inside the module, a friend of mine does this in his electric drag racing car and recommended it. And since I watched a very expensive pack start on fire because of a dead short (and this fuse would have saved the pack) last week I think I will add this feature.

I designed the delrin spacers and had them laser cut which is really affordable these days. I had 6 pcs cut for $95.00 at Serra Laser in Anahiem CA. Delrin (Acetal Resin) was purchased from McMaster-Carr and was about $50 or so. Attached are the DXF files it anyone wants them, be sure to check all dimensions yourself as they may be different from your battery of choice.

The Aluminium Box is made out of 5052 Aluminium and welded up at a local welding shop. Total cost for 2 boxes with material was about $220, nothing to special here.

BMS is Methods HVC/LVC as shown it Pics, to bad he is out of the scene now, I wish someone would copy the design and sell 'em around here...

I am not finished with construction however here are up to date pics, I will post again when the pack is further along.


View attachment 4





More Pics here http://winzeracer.com/photos.html

Thanks,
Brock

 

[dogman dan]

I'm sure you are smarter than me, but I see a problem.

Won't those narrow spacers that hold the packs chafe or ding the packs on that point where they contact?

Particularly in a crash, all the pressure on the packs will get concentrated on that half inch wide strip. I have just found from experience, that any tight spot will chafe right through the shrink and start rubbing and dinging naked packs. I try to box my packs so all the pressure is very even on the whole pack, and then I see no dings or chafes a few years later.

I just think it's near impossible to make those packs stay still enough to not chafe and rub. It might take a year, but you do eventually see a hot spot that rubbed more on the packs somewhere.

 

[winzEracer]

Dogman,

Yes I think you are right, I can stop them from chafing by shimming them up with foam tape so that they do not move. However in a crash there would be lots of pressure at the point where the delrin contacts the cells. I do have extra spacers so I think in light of this I will use 3 spacers per pack so that the force would be distributed better, however this would still be likely put heavy force on that point of contact. The reason I decided to go with the spacers was to have some airflow between the cells. In my constant current testing of the cells at 8C they got pretty warm. Do you think it is a good idea to have an air gap?

Thanks,
Brock

 

[amberwolf]

The air gap may provide cooling, but I'd generally say that if the cells are getting hot enough to need cooling, when used as you would normally be using htem on your vehicle, you either need better cells or more of them in parallel to reduce the load on them.


If that 8C test is only something you'd see as an occasional momentary burst of power, and they don't get warm in normal use, I wouldn't worry about the cooling.


One other thing to think about is that if you provide a path for fresh air to go thru the pack to cool it, then you are also providing a path for fresh oxygen in the event of a fire.

I can no longer do the test myself (because the defective cells I was going to do the test with were stolen by the cleanup crew at my house, along with the BBQ grill and other stuff), but I had planned to do a test based on other people's testing of similar situations, of a pack enclosed in an ammocan that's almost sealed--only vent being the cable exit, so that pressure can be vented out, but no fresh oxygen/air can come back into the burning pack. I would have ignited the pack by forcing a severe overcharge on the cells, at high current. I did that on some single cells in open air and was able to get one of two to burst into flame from it. I figured with at least a packs worth of cells in the can, enough would reach the same ignition point to be able to test if it would actually burn without external oxygen at the cell itself, or if it would simply vent smoke and stop.

 

[Punx0r]

You could perhaps use stiff plastic sheet (~1mm?) and filament tape to encase each pack if there is enough space in the holders. That would protect the packs and provide compression when they start to swell with age.

 

[dogman dan]

A small redesign may help a lot. Get some kind of thin, but pretty stiff material, and box the cells in groups of two or three. Could be something like fomica sheet, or plastic sheet. Something just a bit harder to dent than the usual shrink. Make it super tight, and all the chafing will happen on the outer material rather than the shrink. You will get a lot more resistance to dents too, from the stiff material.

That shrink is real convenient for them, but it's really not that durable at all. But multiple layers of it might do.

It could be best in the end to do like some have, and just compress the whole pack into one big block with hard corners that won't crush cells on the edge. The compression may help the cells better than the cooling. As AW was saying, if you are really heating up that pack, it's too small for the load. In a race type situation, a fairly hot pack (100F) will put out the juice nice. If it self heats much beyond that, it's too little.

Maybe I'm just being way to fussy, but I have tossed quite a few packs that got chafed to the point of leaking in the box by very small defects in my box designs. None of the batteries I have created an inner box for have chafed and gotten damaged, even in some pretty good crashes.

 

[agniusm]

Lots of expense here but why not trash spacers all together? Better anodise the box and lay packs against aluminuma wall. You will get far better cooling except those cells in the middle. You will get smaller size as well

 

[dogman dan]

That's about the approach I took when I built boxes for the racing bike a few years back. Each row of cells had aluminum plate on two sides. It heat sinked great, but in winter that became a problem instead of a benefit. I had built with a race in 100F temps in mind.

But unprotected soft packs in that box did eventually chafe holes in them. They had a nice tight fit, but one grain of sand finding it's way into that box would become a disaster. Padding the bottom helped a lot with that.

 

[winzEracer]

Ya looks like some redesign is needed. I will likely get rid of the spacers. I designed the pack with Southern California summers in mind, hence the need for cooling.

Perhaps a blend of the ideas here would be to first off have all bricks tightly strapped together with plenty of protection from chafing by using plastic, strapping tape and foam or rubber on the bottom of the aluminum box. Secondly to sink some heat out of the center 3 bricks have a layer of aluminum between them and the outside bricks, protected from chafing but as thermally conductive as possible.

Thank you for the all the suggestions thus far,
Brock

 

[winzEracer]

Dogman,

Do you use anything special for chafing? (Of the batteries that is, haha)

Thanks,
Brock

 

[amberwolf]

Ya looks like some redesign is needed. I will likely get rid of the spacers. I designed the pack with Southern California summers in mind, hence the need for cooling.

I use RC LiPo packs here in Phoenix AZ summers, and even in direct sunlight they're not all that hot, relatively speaking, inside an ammocan--because I'm using them within their limits and not generating heat from within the cells.

As long as you're not doing that, then heat from outside is easy to deal with--insulate the entire outside of the boxes so they don't admit heat from weather into the box. (this of course will also keep all the heat in that is generated, but if you use low-resistnace cells that can actually handle the C-rate you're after without heating, it won't be a problme).

 

[dogman dan]

Currently I carry batteries mostly in a triangle bag, or in a motorcycle type saddle bag. I make up the pack, usually 14s 10 ah, and then make a super tight fitting box from coroplast to fit around them. Gives it a bit of crush zone, and all the chafe is taken by the box. You make that box really tight, so you can't get the packs out without cutting the tape holding the box together. Before the packs go in the box, you do tape up the bottoms of Zippy packs, which have no protection on the bottom.

Right now half my battery is a pair of 5ah hard shell packs, those I can toss in a box or bag with no worries. Just some foam shims to prevent rattling when I carry them in a hard box.


Like AW, since I don't hit the pack hard enough to get hotter than ambient here, I find that I don't need to cool it. It's going to be 100F no matter what in summer anyway. But in cooler months, a bit more insulation helps keep them from being discharged at 40F and losing capacity. Below 40F, I'll drive the car. In the coroplast, and the bag, they will warm to room temp and perform better on a cold day.

If you want to wick some heat from the center of the pack, some aluminum plate should do it well. But if you are discharging the batteries so they self heat a lot hotter than 100F, you need more AH, higher c rate, or less controller. If this is for racing, I'd want my packs at 100F, so they perform more perky. When I took that bike to the Tucson race in October, I wished I had not made that heat sinking box. I could feel it was slow, on that cooler day.

 

[winzEracer]

AW and Dogman,

I am well within the C-rates for the pack, they are 35-70C Nano-techs. However it is a racing kart and will get driven hard. I think the pack will last 15-20 minutes of racing or less. Lets call it 15 minute for a nice round .25hr figure. I have a 25Ahr pack so average currents will be around 100amps (4C), with pulse currents of 200-400amps (8-16C). I think this is withing the margin for the Nano-techs, all packs were cycle tested, aged and capacity matched, so hopefully I will not have any more bad cells. I have 5 6s bricks that did not make the cut. I am not worried about pack longevity, but I am worried about overheating, I have not used a hobby pack at these currents before so I guess I am a bit cautious. OEM's often have cooling and heating on their packs, so I though it would be a good route to go. I am not so worried about running on a cold day, as the packs are quickly removable and I could heat them up with an electric blanket or the like. Others that I have talked to seem to think I am within the margin for these cells, what do you think AW?

I am going to put a few thermistors in the pack, and have them running to my Arduino based vehicle computer, if the temps get to high I will have it retard or cut throttle. Dogman you are right, I defiantly want to keep the packs nice and warm so they perform well and AW seems very confident that I will not not need cooling. So I guess I will leave out the Aluminium sink in the middle of the pack and if I have issues later I can add cooling. Thanks for the tip on the coroplast, I like that idea quite a bit.

Thanks again for the helpful tips Gents! I have a tendency to over engineer things, so I appreciate the feedback

Brock
www.winzeracer.com
http://www.diyelectriccar.com/forums/showthread.php?p=307096#post307096

 

[Ykick]

C rates and thus internal heating get worse throughout discharge curve. An abundance of capacity can be well worth any size/weight penalty.

Pressure switches have often been on my list of techniques to try in order to help identify puffing packs stuffed inside enclosures.

 

[amberwolf]

Others that I have talked to seem to think I am within the margin for these cells, what do you think AW?

Well, for me, a non-engineer that just uses batteries but doesn't pretend to truly understand them , I think that anything that says you can run a cell at a C-rate that causes heat to build up within the cell is over-rating the cell.

It's really common for just about any battery seller (not necessarily, or even usually, the makers of the cells in the battery!) to over-rate their batteries in C-rate and capacity. Some are honest enough about the capacity. Many are honest enough to advertise a C-rate that the cells *can* achieve, without major voltage sag, but still causing excess heat to build up because of their internal resistance.

In a lot of the applications these RC packs are meant for, it doesn't matter. The pack will be exhausted so quickly, then removed from the device it powers and left to cool while another pack is installed and the device used for a few minutes more, etc., that it could cool off enough to not worry about much.

But when they are used outside their original intended application above, like we do, in packs that allow much longer (usually) operation because of many paralleled packs, and then cramming those packs together, then they can't shed that heat and so using them at the advertised rates doesn't seem realistic or probably safe.

So, it depends on what your personal judgement of what a real C-rate rating means: Does it mean the pack *can* achieve a specific current draw thru it without severe voltage sag, regardless of heat generated due to high internal resistance? Or does it mean that pack has a low enough internal resistance to not generate the heat in teh first place when used at the specified current draw?

That's up to the end-user to decide...but for me, if it's being used at a high enough rate to heat up, it's not designed for that level of current draw.


So: If the cells are being used at a rate high enough to cause a lot of internal heating, you probably do need a cooling solution to get the heat out.

If they are not heating up a lot internally from use, then you don't need one.

 

[dogman dan]

To my way of thinking also, if that pack is getting much hotter than a hot day in my climate on discharge, it's getting hammered harder than you would like for lifespan.

I saw this plenty in a small way with my dirt bike, and a small 72v 10 ah pack. Most riding did not heat the pack above ambient even on a hot day. I never used thermistors or really knew my cell temps, but my routine involved removing the packs to store and charge after the ride. So I could tell that 20c packs heated a lot more than the 30c packs. When I ran a 5 ah pack, the pack got very hot, like about as hot as the hottest water you could stick your hands in. 140F? Somewhere in there. Those packs treated like that did not last 50 cycles btw. But WERE being used within the stated C rate. Always, the pack would get all hot and bothered if I discharged it deep, no matter what the size of it. For sure, I was always discharging the small pack deep. But the small pack got warmer before the end of the ride too.

But on the track, your needs and desires will be totally different. 50 cycles may be fine for you. Running that pack at 120F might be what your want. But maybe you don't want it hot from high internal heating and race losing voltage sag. Hot and perky fine, hot and saggy not fine.

I'm even less an engineer than AW, but I think if you need to cool the pack because it makes itself hotter than 100-120F, then you have too much sag as well. This may only happen when the pack is nearly empty, but you will lose every race if you lose all your volts in the last lap.

Build so you can cool off an overheated pack makes sense. You may need to get it cooler between heats. I don't think you want it to run or charge at 140F. But without getting over heavy, you do need to build large enough so you don't run out of power, sagging and heating up that pack in the last lap.

 

[docnjoj]

Tesla liquid cools their cells, and Allcell uses some "phase change" goop that Justin says is OK. We used to cool our Nicads after high power electric model flights by tossing them into a bucket (plastic) of Icewater. it really helped cause Nicads got hot! But they were cheep.
My point is ingenuity can be used in this area. We can't keep using bigger batteries to overcome poor C-rates. There's gotta be a better way.
otherDoc

 

[dogman dan]

For sure, some kind of heat management becomes imperative when a pack reaches a certain size. It may be that this kart will have a pack that big. I'm not sure if that's the whole pack in the picture, or just one section of it.

And you can't just keep making the packs bigger and bigger infinitely. Cost gets too high, in money, and particularly in weight in a race. But I still think that if you want to win a race, you will need to use a pack size that doesn't drop your speed 50% in that crucial last lap. I'm not saying you need to end that race with a cool pack at all. But you can't have it hit bottom, go over the voltage drop cliff, and use that last wh of the pack to finish the last lap. That's when a pack really heats up, when it's resistance gets so high in the last few % of a 100% discharge. Plus, you lose when you get passed on the last lap. You can't have 4v of sag under load turn into 15v of sag in the last lap and win.

Size that pack so you end the race with 3.75v per cell, and you won't cook the pack near as much, won't need as much heat sinking, won't have to wait to put it on a charger to be ready for the next go. And won't be as slow at the finish line.

 

[Ykick]

Compared to ICE racing, electric power certainly differs.

Gas powered burns off weight and towards the end of the race the vehicle is lighter than when it started and at it's fastest.

OTOH, electric is the same weight it began the race but voltages have dropped over the course and it's now slower than when it started.

Pretty much obvious stuff but some new readers might want to consider for EV racing applications.

 

[winzEracer]

I do expect the cells to get fairly warm while racing, I am not sure how much. This project is an experiment for me, to build a smaller race car (read less cost), so that when I build a full size car I will have learned a ton and will not make expensive noob mistakes. So this pack while expensive is and a ton of work will pail in comparison to a full size race car pack, this is where I want to figure out things like did the pack get to warm? did it need more cooling? did it have mechanical defects? ect...
From my testing and the advice of others, I believe that the batteries are well suited for the burst current draws that they will see, YES they will make more heat than your average non-performance pack, which will shorten their life considerably. I would be happy with 100 or so cycles. I am also of the school that getting rid of heat versus adding more battery (for performance applications) is preferable. As far as if the pack has enough energy to make a whole race, I am not sure as there are hardly any sanctioned E-kart races. I would like to take it to Refuel 2014 at Laguna Seca this year, as far as I know that is the only race there is, other than getting a group together and doing a bit of DIY race organization.

So back on topic, I am going to perform a test with a 12s1p pack strapped together and insulated to simulate conditions. This will fall short of a perfect test, but it will provide better information. I will take it up to 16C bursts, which for a 5ah pack would be 80amps and see how warm it gets. I will be out of town for a couple of weeks, so I will report back when this is done.

Thanks,
Brock

 

[Hillhater]

A race kart can run average speeds up at 50+ mph, so the potential for cooling ducted air into the pack is an option.

 

[dogman dan]

Yeah I agree, you should be well in the c rate zone to not have any more heat than you would desire from the pack, until it goes below 3.65v. That's what I'm saying, the pack must not be so small that it does that in a race. Off the cliff towards 3.5v, you will have more excess heat, more sag, more damage, and lose too. That's all I meant, don't let it be so small it goes to there. Gotta try to finish the race with 3.7v or so or you will really slow down in that last lap.

But if the pack makes some heat, and gets to an equilibrium max temp during the run of 120F or so, I think that will just make it more perky in the last lap. I don't say you must run a cool pack, just that you don't want it overheating because you over discharged. Too big, too heavy, and cold, you will not gain a thing on the track from that. If you need a huge pack just to finish, it could be so big you decide you need to pre heat it for a race. You want it pretty warm to perform in a race.

Lots of thermometers in the pack will tell you which way you need to go. Like goldilocks, you will want that pack at a temp that is just right.

 

[winzEracer]

Hey All back from vacation and got the kart running so that I could test some batteries!



I ran a 12s1p pack 5ah at 50v. I ran 2 tests the first at current limited to 100 amps (20C) and did a few laps in the neighborhood, hottest temp battery temp was 28 Celsius with a low ambient and low starting temperature. Then I did another test today limited to 150 amps (30C) with warmer temps, pack started at 22 Celsius and topped out at about 43 Celsius. Though the controller did cut me off due to voltage sag. I will perform one more test at 200 amps max just for fun maybe a vid of that. I do not think that I will need airgaps, however I am still leaning toward a small aluminium heat sink for the three 6s bricks in the center of the pack, just for good measure. Again my max current will be 16C so from what I am seeing, keeping it simple and tossing the spacers is probably the best solution.



Brock

 

[Arlo1]