So, it begins.
My Sky Energy SE040AHA cells from EVworks are in transit and my simple BMS also supplied EVworks has arrived.
I plan to document this little adventure reasonably thoroughly, but I might miss some steps as I go along.
If you have any questions, please ask and I'll do my best to provide more detail. You might also be doing me a favour by pointing out a catastrophic error I'm about to make.
I really like idea of this BMS, for many reasons:
1. It is available
2. Simple, even a dumb fitter like me can understand it and it should be really robust
3. Complete stand alone solution with built in capacity to cut power to the charger if one cell hits the high voltage limit.
More info here: EVPower BMS
I simply love it. Superb. Is there a version for NIMH?
http://www.evworks.com.au/tech/EVPowerBMS/
Looks like the pile of gold you need to hand over for the cells!
How is this part going to work with the Vectrix:
# Throttle control PWM output reduces motor throttle in the event of a system error.
# Main contactor control relay will drop the main contactor in the event of a system error and no response from reducing the throttle.
The throttle may not be a standard type. But I do not understand how it works and maybe it is no problem at all.
And there is no main contactor, unless you manage to install one (which would be a good idea!)
How do they do the 600mA shunting during charging? Does this cause a voltage drop which might confuse the charger?
As in: Half the cells have hit the cutoff voltage and are shunting (or 80%, 90% etc) - what will this look like to the charger?
This information may be used entirely at your own risk.
There is always a way if there is no other way!
I don't believe so.
I think that NiMH chemistry doesn't lend itself to the same control paradigm.
individual cell monitoring would be nice for the stock NiMH pack, but I don't think there is anything available off the shelf to do this.
Missed your post earlier mik.
I think they are actually gold plated :)
I don't plan using the throttle reduction function. I'll let the stock controller do that. The bike should limit discharge to 108V, 2.57 per cell on a balanced pack, module faults at 2.5 and the cells are spec'd to 2.0. Loads of safety there I think.
I only ever expect this system to trigger in emergency situations. Badly out of balance pack allowing one cell to go too low on discharge or high on charging. Red light on the dash and cut power to the charger. The bike doesn't manage max voltage, but the cells can absorb 1/3 more AH than the charger can deliver. I don't expect the stock charger will top out the cells often. The modules will trip and the MCU will cut the supply voltage if they do. I think using a "universal Freddy" for maintenance balance charging and topping off might be useful. Both chargers supplied separately through a single onboard GPO switched by the MCU. Not sure where all that is going yet.
I'm hoping the shunt capacity being so low compared to the charging current, that the change in voltage won't be significant. Hibba hasn't reported any issues using this system on the vectrix.
HUZZAH!
Working nights this week. So expect a solid bout of procrastination :)
So how is the LVC going to work then? Just a red light on the dash? Or will you add in a contactor somehow? Don't forget the 8mA constant-on motor controller drain! It could ruin the Li cells (with the red light on the dash lit) just by letting it stand unused for a few months!
A contactor that disconnects the battery totally will get around that problem. But, a contactor that turns of all power is a hazard during riding. It would typically happen when you give full throttle with a low battery, likely in an emergency!
This information may be used entirely at your own risk.
There is always a way if there is no other way!
all good points mik.
long periods of idleness could be an issue if idle current draw is significant. My setup would need open heart surgury and manual disconnection to store the battery safely.
I don't think it will be an issue normally though. It is pretty much the same for the stock setup, except mine will have a little red light to warn that there is a prob, even when the ignition is off.
But the risk of severe battery damage is much higher when you use Li cells. They will get severely damaged by over-discharge, while NiMH take it as a sort of re-conditioning, deep cycle exercise!
8mA x 24h x 30days x 6months = 34.56Ah
I would not want to take that risk! The battery is too expensive for it!
If you Vectrix was to develop a problem when the battery is close to empty, then the irreversible battery damage could happen within days!
I think LiFePO4 cells should be stored at 40% charged, but I may be totally wrong about this.
This information may be used entirely at your own risk.
There is always a way if there is no other way!
6 months idle? not my V :)
I guess this isn't exactly a production ready design. It'll need care and awareness to protect the cells in these odd situations.
Then again, maybe these lithium cells aren't as fragile as most believe.
Matt's cell abuse confessional:
http://www.aeva.asn.au/forums/forum_posts.asp?TID=2037
The conversion waits.
I have no cell interconnectors yet and she who must be obeyed demands the grass is mowed before I play.
Had to get fuel for the mower
I need an electric mower
What an odd picture : a Vectrix at the gas station ... and more of that, in front of a diesel pump ! Heaven and hell ..
Please don't publish that kind of picture anymore : it hurts me too much ...
April 2010 Vectrix VX-1, 2004 Prius (feeded with E85), Giant Suede (electric bicycle)
Xsus... a vectrix in a gas station... this is scary...
There are some lithium powered mowers...
have a look:
http://www.bosch-pt.com/productspecials/garden/li-ion/es/es/rotak/highlights.htm
It would be like riding the vectrix... ;-)
AAAAAAAARGH ! How could you do that to your unfortunate Vectrix !?! Don't be surprised if it commits suicide (fried IGBT, short-circuit in battery pack, etc.) after such an humiliation !
"We don't inherit the Earth from our ancestors, we borrow it from our children." - Antoine de Saint-Exupery.
Funnily enough, the RH indicator stopped working after this...
And it started again when you threatened it with calling in Vectrix Australia!
You got a very smart bike!
This information may be used entirely at your own risk.
There is always a way if there is no other way!
KITT's little brother?
I don't want to be scary, but I was thinking more of something like this :
April 2010 Vectrix VX-1, 2004 Prius (feeded with E85), Giant Suede (electric bicycle)
The cell interconnectors arrived yesterday, so I've had a little play
the 4 cells took about 25Ah to fill
The cell top Control Modules (CMs) light up a green LED if the voltage is between 2.5V and 4.1V. A red LED indicates the cell has reached 3.65V and is shunting current around the cell so the other cells can catch up.
On the first cycle, I noticed one cell hit the shunting voltage about 1.5Ah before the others. I watched individual cell voltage over a couple of hours and saw that the full cell went to 3.8V while the others caught up from about 3.4V. With the charger pushing out a CV of 14.4V (EVPower recommends 14.6V for a 4s array, 3.65V per cell), the array was still pulling about 2A. The CM seem cool enough. It was interesting to see the red LED get brighter as the cell voltage went up. I was worried about this current as they are only rated to .6A so I spoke with EVPower he advised that the safe limit for the CM's is 1.0A.
When the pack is close to balance, these CM's seem to keep the balance tight suppling constant voltage to the cells with an average 3.6V per cell while still keeping them under the 1.0A. But it seems that if the pack is miles away from balanced (like when you build a new pack) there is a risk of over working the CMs on the full cells. I think I'll need to build a "Universal Freddy" (low current charger) for the initial charge and possibly for pack maintenance and top up later. For now, I'll trickle 6s strings with this programmable R/C charger and limit current to 1.0A
If one CM happens to fry, the built in signaling switch will go open circuit,the Master Control Unit will stop charging and throw up an alarm
Really pleased with this BMS
Oh, I just noticed in the pic that the front CM is installed under the interconnect. This isn't recommended as it bends the the CM over the cell vent. The others are done correctly
Do these CMs have some sort of buzzer when it detects to low voltage on the cell it's watching over (in case, if cells have different capacities)?
IIRC, they have a single "wire" going through all of them, and each module can interrupt the series so that the circuit opens.
You could do whatever you want with that signal, a buzzer is an option, or a light, or opening the main relay and stopping the vehicle!
This information may be used entirely at your own risk.
There is always a way if there is no other way!
Excuse me for jumping in with a silly question, but those brass plates with the plastic centers are the auto shunts right?
Thats exactly what Im looking for, how much are they and where did you buy them from please?
Thats such a more logical way of balancing than having charging wires to each cell!
Gold! it is! Not brass...I like it, too!
This information may be used entirely at your own risk.
There is always a way if there is no other way!
the first time balancing is always a pain.
you could just have the master continually disconnecting the charger when a cell goes over 4.2v for balancing, until the pack becomes fully balanced.
It's certainly much less work than building a charger for a one of equalisation.
Thats how Rod balanced out the cells (all 100 of them) in his Jumbuck.
it does mean you want have all your range right away though.
they come from here:
http://www.evworks.com.au/index.php?category=6
Matt
Daily Ride:
2007 Vectrix, modified with 42 x Thundersky 60Ah in July 2010. Done 194'000km
1. So, these CM modules will only balance the pack when it is fully charged by shunting the cell?
2. Maybe I'm wrong, but when the pack is reaching the full charge level, one CM after another will start shunting, because some cells will be more fully charged than others. But doesn't that mean, that when Vectrix' charger will still charge with 150V, less and less cells will get charged and that's why all other cells, which won't get shunted yet, will be charged with increasingly higher voltage per cell, because there will be less cells charged with the same overall voltage (150V)?
What I mean is, that this kind of charging is only safe at constant charging current, not at constant Voltage or Power charging.
If it's constant current charging, the charger will decrease the voltage, when cells are being shunted one after another, keeping the current the same and charge power per cell the same.
But when you charge with constant Voltage or Constant power, Vectrix' charger won't know, that cells are being shunted on the 'per cell' level and it will keep pushing that 1.5kW at 150V into the remaining cells, that haven't been shunted yet. The last cell in the pack, that hasn't been shunted yet, will receive the full 1.5kW of charging power.
I may be completely wrong. Please correct my assumptions if I'm wrong.
the shunts burn off the power that would have otherwise gone to the full cell. It isn't bypassed to the next cell, so that one won't be given excess voltage.
That's the 1 Amp you mentioned earlier?
Hello all
I'v been reading this forum for a year now since i bought my vectrix, i want to say thanks to you GURUS, thank you all for the great literally life saving (:Hazard lights) info here. Thanks form Israel.
Proc' You have stated something i'v been waiting for and dreaming about from my firt days on the vec... finally!
so, I decided to try and contribute with my personal expirience and knowledge to help you and others here.
Im an Electric eng' and doing PWM and high voltage/current for work so e-bikes is a great hobby/future business (-:
SO, to Those Diamonds of yours (yes, its illigal to import batteries this size to Israel, you have to buy these local and expansive) but i have a chance to test the TS-LFP40AHA here (nominal 3.3v) i dont have enough, i have only 24 pcs and a charger and thats enough for testing discharge and power.
How much do you have from these?
My vec has battery problems and my dealer won't fix it under warranty, so im pushing him with pingpong. anyway for now, he gave an E-rider LiFEpo4 to go.
OK, before it went in, i tested the vec with the charger disconnected from the battery (under the headlight to the right, the back plug). and i connected my trusty meter to the plug. this plug it has 3 ping, the 1 marked with the earth sign is the PLUS, other in same row is minus.
the 3rd, i couldn't figur out expect 10mho to the minus. maybe its the current sensor and works only if there is current, didnt theck.
i found out that if the plug is disconnected, the charger is dead, dispite the big plug from the CANBUS. is it powered from this plug?
the SOC metter turns into a volt meter just like described in the posts here, and the red batt-low light never comes on even on 110-108v.
This is where i tought...the battery is low, what if i connet a voltage source to trick the MC into 140-150v? so i did, i connected a rectified 145v/1500w transformer from work, with a big cap to DC it out. the vec went on my hydraulic lifter.
when the key is on, and the battery at 122v and crawling current, i cennected the transformer to the charger plug... power it up, and WALLA! the wheel spined at 100-102kph on the lift for about 15 seconds and the big rectifier gave it, it slowly dropped to crawling again as the part ov the voltage went to the battery so it drained out... still no batt-low light! (so good to see that).
what this tought me is that we can put any source we want, as long as the voltage is higher than 120v, the MC will not slow down the motor no metter the SOC of the battery, we CAN use ANY KIND of battery as long as the charger is not sending SOC/LIGHT/TIMER CANBUS commands everywhere... i had entire range trips (yeah...29km) with no charger and it actualy went longer until the volgate-cutoff.
we defenatly need external charger(s) for a pack this size. i see no way (reprog?) the VX-1 charger will charge anything that it doesen't know. all the other parts will work for sure from other sources.
i think 44 cells will be the best choice (i dont know about space inside the vec...) the TS-LFP40AHA im testing pretty much fit their specs at 2-3c the voltages goes from 3.6v to 2.8-2.6v. so its 114v-low to 158v-high. i think the vec can handle the 158v full. (up to 175v as i recall from other posts. 42cell will work i think but you will have to be very careful with the low-end (2.6*42=~109), an invisible cliff in the middle of nowhere.
as i see it, the stock VX-1 minus stock CHARGER + your great battery and BMS setup + Li charger = 70km RANGE DREAM!!!
Please keep posting, this is the best thing that happend here in the forum.
ho, and Thanks again for the brains.
SonicheD
In "Back To The Future II" There are flying cars in the year 2015. Its only 4 years from now! lets make it happen! or at least lets have MR. Fusion for our EVs...
15,000km 2009 Vectrix VX-1.
Voltage IS the way to go.
Hibba hellfire on this forum has already done a TS40 conversion
http://visforvoltage.org/forum/8427-jealousy
There are also many other threads discussing Lithium conversion ideas and rumors of factory conversions nearing reality
http://visforvoltage.org/search/node/vectrix%20lithium
TS40 and SE40 test data
http://www.evworks.com.au/tech/ThunderSkyVsSkyEnergy/
That's the maximum safe limit that these can shunt. They are at risk of failure over that.
Before installing the cells, I'll have them all full and balanced. I'm thinking that the final charge with my little RC charger might be 1A on a 6s7p array... 280Ah 18V pack :D
Thats a very interesting point Andy! has made, in that the shunts do need to be a similar resistance to a battery so the last ones dont get charged at a different rate. (the bat controls the votage!)
eg if shunts are 10ohms and batts are 20 ohms internal resistance, the last batt would in theory get double the charging voltage. -But that doesnt really happen exactly because the battery pulls it down to its own voltage, so it will just charge a bit quicker
I use these ones on my car: http://www.cleanpowerauto.com/MiniBMS.html
They also have LVC, works great and are quite cheap, I payed 12$/pcs +30 for the head end board.
Pages