Custom A123's almost here!

[DeeL2003]

Just got an email from Mike(from Big E-RC) and my A123 packs are ready to be shipped. It's going to be two 8s5p packs in series (16s5p, 80 cells total) to make 52.5v 11.5ah.

I'll be using this to charge them:
http://shop.vendio.com/evan2002/item/818330695/index.html

It should take about an hour or less to charge each pack. For balancing, two Hyperion's LBA10 will be used for each pack. Unfortunately, charging will be cut down to 10A when using the balancers. But no biggie for me. 1-1.5hr per pack is okay.

I'm like a kid waiting for Christmas to arrive so I can open my big present. :lol:

 

[Ypedal]

Sweet !!!!!...

Those babies should be able to handle a good load !!!!

Total weight ?

 

[DeeL2003]

YPedal: Estimated @ around 15lbs total. Will check with Mike. BTW, how do you post pics with the pic info, views, etc.?

 

[knightmb]

more pics:

YPedal: Estimated @ around 15lbs total. Will check with Mike. BTW, how do you post pics with the pic info, views, etc.?

Down here, picture below

 

[DeeL2003]

more pics:

YPedal: Estimated @ around 15lbs total. Will check with Mike. BTW, how do you post pics with the pic info, views, etc.?

Down here, picture below

Thanks! So I guess I don't have to use an image host site, I can get my pics straight from my PC.

 

[erdurbin]

I want a 36 volt pack
I have called the dewalt service centers trying to get used packs. All say they will call if they have one, but not very common.

 

[mcharles13]

I sent this to erdurbin in an email but figure it would be helpful posting it here as well.

There are 2 ways I build packs for people.

1. I source all materials needed for the pack.
2. People purchase packs on ebay, have them sent to me and i provide the rest of the materials. This option is preferred by many because you will typically save about $4 per cell.

A 36V pack is 11S (11 cells in series)

How many Ah do you need? That will determine how many layers parallel
you'll need. We can work out cell count and determine the price and
charging options.

Just FYI, I'm a very laid back person so feel free to ask me any and all
questions you have. Take as much time as you need.

,Michael Charles
http://www.bigerc.com

 

[CGameProgrammer]

A 36V pack is 11S (11 cells in series)

Yes, but it's important to note that the DeWalt "36V" packs are actually 10 cells in series, and the voltage is 33V or less under even the slightest load. I consider it as being 3.2V per cell.

 

[erdurbin]

That is why you have to get a different charger and cannot use the dewalt

 

[tailwind]

Found this very nice video of someone charging their partially discharged 5s2p A123 pack with a Mastech HY5020E power supply (50V, 20A).

http://www.youtube.com/watch?v=q-1fb9_V2II

I am guessing this is the one that DeeL2003 will be using.

He goes through the whole process of setting up the power supply, connecting the batteries, through to the end of charge.
The video also gives a good demonstration of the CCCV (constant current - constant voltage) charging process. I have never really understood how this works but now that I have seen it in action.
I like the fact that the display will give you feed back as to what is going on.

In the video he dials up 18V. His battery pack is 5 in series, so 5 x 3.6 = 18.
When in use he will get 3.3V from each cell, so 5 x 3.3 = 16.5V

So a DeWalt pack is charged to 10 x 3.6 = 36V, in use it is 33V

For DeeL2003 8s5p packs he will charge to 8 x 3.6 = 28.8V, in use 26.4V (2 packs 52.8V).

Greg

 

[mcharles13]

That is the big brother to the charger that DeeL2003 will be using.

The only difference is that one puts out 50v@20A max and DeeL's will put out 30V@10A max.

Regarding voltages. I use the nominal volts which is 3.3 (or 3.2v) when i calculate voltage.

If i were to use the Full charge value of 3.6x11 it could be misleading as someone would think they have a 39.2v pack. As soon as a load is put on it will immediately come down to around 3.2v and hold until the end of the pack.

,mike

 

[DeeL2003]

That is the big brother to the charger that DeeL2003 will be using.

The only difference is that one puts out 50v@20A max and DeeL's will put out 30V@10A max.

Regarding voltages. I use the nominal volts which is 3.3 (or 3.2v) when i calculate voltage.

If i were to use the Full charge value of 3.6x11 it could be misleading as someone would think they have a 39.2v pack. As soon as a load is put on it will immediately come down to around 3.2v and hold until the end of the pack.

,mike

Mike,

I caved in and bought the HY5020E. :lol: I later saw a 30v 20a version that was slightly cheaper. Oh well. Either one will be more than enough.

 

[mcharles13]

NICE!

 

[magudaman]

Considering they are 5P they should have put thicker wire than the 12ga wet noodle. You should be able to do 350 amps continuous. Other wise they look awesome. I sure that cost a pretty penny.

 

[mcharles13]

I was going to put fatter cable on there but DeeL said all he needed was 12ga since he would be averaging only 30-40ish amps. Also that made it easier to use the same connectors he already uses.

Thicker wire is definitely available

Regarding prices, erdurbin said it would be a good idea to post the price quote i sent him here. Most people end up sourcing their own cells on ebay to save.

It's really based on cell count so whatever configuration people need it's calculated that way.


,mike

 

[jdh2550]

Hi,

I ride an XM-2000 EV Motor Scooter.

It comes with 38Ah / 60V (5 x 12V 38Ah (20hr rate)). So, by my reckoning I'd need 20s5p (possibly 19s5p) or 100 cells. This would give me 11.5Ah. Is the 2.3Ah per cell a 20hr rate or a 2hr rate? I'm hoping it's a 2hr rate! I'm hoping that 11.5Ah of lithium will give me a reliable 12 mile range (might be close!)

Doing some math on what's posted above I think $8 per cell is a reasonable price to be bidding on ebay? Is that right?

So the pack would cost me $800 in cells plus $515 for Mike to assemble then. Then I also need a charger and a BMS (no way am I spending $800 for cells and then not balancing them and protecting them from overcharge and undercharge)

So I'm up to $1,315 plus ? for a charger and ? for a BMS.

Does anyone have any numbers on the cost of a suitable charger and the cost of a suitable BMS? Currently it takes me 10 hours for a complete charge - so I don't need anything that charges particularly quickly!

Finally does anyone have any dimensions for the above photos?

Many thanks!

 

[DeeL2003]

Hi,

I ride an XM-2000 EV Motor Scooter.

It comes with 38Ah / 60V (5 x 12V 38Ah (20hr rate)). So, by my reckoning I'd need 20s5p (possibly 19s5p) or 100 cells. This would give me 11.5Ah. Is the 2.3Ah per cell a 20hr rate or a 2hr rate? I'm hoping it's a 2hr rate! I'm hoping that 11.5Ah of lithium will give me a reliable 12 mile range (might be close!)

Doing some math on what's posted above I think $8 per cell is a reasonable price to be bidding on ebay? Is that right?

So the pack would cost me $800 in cells plus $515 for Mike to assemble then. Then I also need a charger and a BMS (no way am I spending $800 for cells and then not balancing them and protecting them from overcharge and undercharge)

So I'm up to $1,315 plus ? for a charger and ? for a BMS.

Does anyone have any numbers on the cost of a suitable charger and the cost of a suitable BMS? Currently it takes me 10 hours for a complete charge - so I don't need anything that charges particularly quickly!

Finally does anyone have any dimensions for the above photos?

Many thanks!

Dimensions of each half 8.875" x 5.938" x 2.938".

As for the packs pictured above, I WILL be balancing these packs with hyperion LBA10's. As for charging, the link of the charger/power supply I have at the beginning of this thread provides the quick and safe charging you need. My BMS will be my CA(Cycle Analyst) and a good prayer, j/k. Some of the A123 users bypass the Dewalt BMS during discharge and only use it during charging. RC guys pretty much don't use a BMS and they pull alot of amps. I think there's a guy in the RCgroups forum who's thoroughly testing these cells doing 2C or 3C(?) charging and even discharging them to zero volts with no ill effects. I don't think I would do that :lol:

I think the next evolution of E scoots(I read your thread on the scooter forum)would be to use LiFe packs but the $$ it would cost to add LiFe alone would be a one big hurdle to overcome.

 

[chuck]

jdh,

I have been following these posts on the dewalt cells for the last several months, certainly not an expert on anything lithium. The following link should answer some of the balancing questions. Dr Bass, Patrick Mahoney, oops, is it mcharles? Look at their posts

http://endless-sphere.com/forums/viewtopic.php?t=1996

chuck

 

[usatracy]

OK, I have gone over this a couple of times, you need to get this right if you are going to make a pack for this bike. The controller is tuned to operate with Chinese Silicone batteries.

The controller will low voltage cut out at 52.5 volts
The stock charger goes to 73 volts
The silicone batteries when new absorb and balance out at 13.45, when older this drops to 13.22
The controller has a high voltage cutoff but we have never seen it activate per the number of flashes that will display for that error, instead, we see 9 flashes which means "48 volt and 60 volt motor have been confused at factory". We see this flash pattern, and the controller disables, when a silicone in the 5 pack string goes bad and has higher resistance and the bike is taken straight off the charger and the key turned off. The indicated string voltage is high due to the bad battery but dropping back to norm rapidly. cycling the key normally clears it, it is that fast, couple of seconds.
The point to that is the controller is sensitive to some unknown high voltage between 67.3 and 73 volts, we have not confirmed the high voltage cutoff point yet, only the low.

The bike pulls up to 50 amps.

If you run a 20 cell DeWalt string, the resulting voltage drop at full speed will soon level off to about 2.575 per cell (my best guess based on graphs, if you have better info on cell volt drop at 50 amp use that).

20 cells at 2.575 is 1 volt below the low voltage cutoff on the controller, so a few minutes into your high speed ride the volt meter will be dead red and the controller will cutoff.

Add 1 more cell, 21, 21*2.575=54.075 or 1.575 above cutoff, this will put you at the left side of yellow on your xm-2000 battery meter going to red

Add 1 more cell, 22*2.575=56.65 and this will put you to the left side of green going towards yellow

Add 1 more cell, 23*2.575=59.225 and you will be where the XM is normally with GreenSaver Silicone batteries during full speed load before the batteries begin to discharge more than 25 % of their capacity. So this is where one might expect the DeWalt also to run at this load.

But with the voltage drop of 50 amps, the silicone actually only drop to 12.6 vdc per battert for the first few minutes and then start a slow linear drop. 12.5 to 12.6 vdc loaded at 50 amps is a good target to get the same fully charged performance or 62.5 vdc at 50 amp load.

Assuming the voltage drop under load is 2.575 at 50 amp, then 24 batteries are needed to maintain performance current.

24 DeWalts charged to 3.3 will be 79.2 volts, well beyond the controller high voltage cutoff, so we will have to start compromising to get the DeWalts in there (all of this assumes a volt drop of 2.575 at 50 amp)

We will limit to 67.3 for full charge string voltage until we can confirm a higher trigger voltage, that is 20 cells, that wont work, low voltage cutoff on full load so we have to add one more cell and jump up to 70.6 charged and pray it does not trigger the "motor confused at factory" message.

This will probably work, but the battery meter will always say the cells are going red and the pack is almost dead whenever you ride full throttle.

If we add one more and cross our fingers, we'll be at 72.6 charged, 0.4 volts below the maximum charge voltage from the string charger, this is real iffy, probably going to see that stupid diag kick in for over voltage.

Yes you can ignore the battery meter, my point is you will be runnnig the controller and motor extensively at lower voltages than we have ever done before, so you will be a test pilot.

But first things first, need to confirm max amp draw for the bike, I think it is really 45, and we need to know the real stats on volt drop and amp load for the DeWalt cell, I know it is here, I haven't looked at all, need that for 45 to 50 amp load ranges.

Then you can start to figure out how many cells you need per string to get the correct or most correct voltage you can.

 

[usatracy]

So, the way I figure it until we get the data on how many amps you pull max, and what the voltage drop of the cell is for that load, for the purposes of a ballpark cost quote, you need either 21 or 22 cells per string.
Thats 2.3 AH
The stock battery is listed as 38 ah but thats 20 hours rate
At 20 amp load it runs 90 minutes new
So thats 30 AH capacity for 1 hour or about 40 minutes full speed which jives with my testing and thats taking them down to 10.5 volts per battery to the 52.5 controller cutout.

So you need 30 AH or 13 strings

273 to 286 cells.

$1,419.60 to $1,487.20 Ebay
$4,914.00 to $5,148.00 "Our Cells" (other post, someone else's cell)

Seems like a lot of trouble to me, and a lot of building and a lot of points of failure.

You can get 50 AH LiFePo4 cells and build it out with only 20 cells and be done and it only weighs 60 pounds. And for a whole lot less, including BMS and charger, but it wont be the holy grail A123, who cares, its LiFePo4, and it comes prebuilt plug and play.

And it fits your bike.

When all said and done this built as A123 will weigh 44 pounds and be 29.9 AH. 50 AH LiFePo4 already built only weighs 16 pounds more and costs less and is a lot less hassle.

Maybe I am missing the big benefit here for the enormous cost and hassle.

Clue me in.

 

[jdh2550]

Clue me in.

Now there's an invitation!

However, the short answer is I'm simply exploring possibilities. I see that I was naive in assuming 20 cells because I didn't consider voltage drop at 50A (the max I've seen is 48A). As far as triggering the controllers over voltage cutout - I wonder if that's modifiable in some way?

A123 made quite a splash with their patented nano-technology based cells. Is their special sauce worth anything? I don't know. But we do know that all LiFePO4's are NOT created equal. Some by intent (i.e. purposely offer a different discharge characteristic) and some by lack of quality control (horror stories about the Thundersky cells).

So, I'm still learning and still looking at possibilities. At the end of the day the DIY DeWalt route doesn't seem like it's worth it for a motor scooter sized application.

But unless I ask I don't know.

 

[usatracy]

Oh, I understand all that, I looked at the DeWalt Ebay route 3 months back, too much trouble.

I'll wait until they rent the Prius with the HyMotion battery pack in it (A123) at Budget rental cars sometime this Spring and have a convenient car jacking

Those are going to the the really BIG 123 cells

I rounded to 50, specs sheet says 48, I have only seen 45 inductive so your 48 is probably more accurate.

Point is you have to be fairly accurate in engineering the packs no matter what cells you use, we are both fortunate that we have some data from usage to fall back on.

Modify the controller, probably, but not by me, my other concern is what will happen if the system is run at different voltages than it had been run at thus far, reduced life, no impact at all ?

 

[jdh2550]

My only intent for a different voltage would be to allow for slightly more headroom to cover the sag. I wouldn't want to run the motor at a higher voltage (well, actually I guess i'd be tempted to!)

Stop picking on me! Wah, wah! if I want to ask half-assed questions that show I'm dumb then let me...

You, as always, made good points. I think I slept through the last class when you "went through all this before".

Don't worry though - I've GOT IT NOW... (maybe...)

(for those who may not know I consider usatracy a friend and while his style is sometimes a little prickly I generally learn enough to be worth the price of entry).

As far as the high Ah A123's I dare say they will be well beyond the means of you and I. Also, I'd love to get my hands on some of the Altairnano batteries - they allegedly have a 12,000 cycle life.

 

[usatracy]

Oh yes, I am very knowledgable and use only the most advanced technology available.

Like today, my worst silicone finally got down to 1 3/4 miles at full speed.

I went to advanced auto and got 5 U1 360 CCA flooded Lawn and Garden batteries for 25 bucks each, they are installed now and I am charging each at 4 amps slow before we start breaking them in tomorrow on the string charger.

Wont be going far on them though, is just a stop gap measure til I get the new batts in from China, but could be weeks before that happens and I have to be able to go to 7-11 in the morning with my wife and her bike to get coffee

Can't beat some cheap flooded to keep you going, but it is a violation of state law to operate an EV on public roads with them...

As for the 123 and sag, I have to recalc at 48 amps, does anyone have REAL data for voltage drop on 123 at a 48 amp load ?

I had to guess based on their chart and I admittedly err'd on the lower side, it coud be 2.575 to 2.65 at 50, and these will go up slightly if we stick at 48 max.

But if you put a bigger motor on that XM, or go to a Z, this pack would not run it probably. Z will have the same high low cutout at some voltage and the pack would have to be engineered for that and Z is 2500 watts not 2000, but you knew all this.

Beware of one size fits all universal packs !

Unless they are mine

 

[usatracy]

And does 123 own the patent, I thought the nano-phosphate production was under exclusive license to A123 from MIT, the actual patent holder and inventor, and that this was done to prevent a recurrance of selling the completed technology to the highest Saudi petrol bidder who would just remove it from the mobile use market like GM did with the EV1 batteries.