Lithium Gold Peak Battery Pack for Vectrix

I don't think that picture is of the Vectrix. It says it'a "2003" - Vectrix started in 2007. It also says "Scooters in U.S. and Europe", but never mentions Vectrix by name.

The pdf file is information about a 3.2V battery (like an 'AA' battery).

That would be great if we could update to Lithium ion, but others on this blog (more knowledgeable than me) have said that it won't happen because it's more complicated than simply swapping batteries...
unfortunately.

"Vectrix" started in 1996, I think. The Bike pictured on the GP site is a late preproduction bike.

lots of images here:

http://visforvoltage.org/forum/7853-vectrix-remember-when

I don't think putting Lithium cells into the VX-1 would be that hard. The charger is software adaptable. Some hardware to monitor individual cells and interface that with the charger and motor control seems to me to be all that is needed.

"The pdf file is information about a 3.2V battery (like an 'AA' battery)."

I think you've read the drawing wrong. They are prismatic cells, rectangular 30 x 90 x 176mm at 1.1kg each, not remotely like an 'AA'

Those little 30AH lifepo4 cells might be small enough to squeeze in a 40s2p pack for 60AH. Nice... but complicates battery management.

I have seen the cost for Lifepo4 battery packs on EBAY come down substantially. It would now be possible to assemble a 120V-50Ah pack for under $3000 and a 120V/40Ah pack at around $2000. The former could use 40 of the 50Ah cells (search for: Foxx Power LiFePo4 50a Cell) which can withstand a 250A discharge rate. The latter use the A123 26650 cells. Search for "AAA Battery Geek" he sells 24,36,48,60V packs up to 40Ah. 2 parallel strings of the 60V/20Ah would easily fit into the battery box. Anyone who has access to the Vectrix software have any idea whether the charger can be reprogrammed to work with Lifepo's? How hard would it be to marry a BMS for the Lifepos to the Vectrix control board?

IMO, there should be no problems as long as you're within discharge/charge cut-off limits and you shouldn't need to reprogram charging software at all.

I'm more concerned about those 2 leads, that go in the middle of the pack. What are they for?

I just got a press release from GoldPeak Vectrix:

GoldPeak Batteries Announces the Acquisition of Vectrix

Vectrix has a bright future ahead!

On the 6th of November 2009 GoldPeak Batteries International announced the acquisition of all assets
and certain liabilities of Middletown RI based Vectrix Corporation. Before that, Vectrix Corporation filed
for bankruptcy under chapter 11, US bankruptcy law (for details please see official press release from GP
published Nov, 6th).
Gold Peak Group is a public listed company in Singapore and Hong Kong and is one of the leading
battery suppliers worldwide.
GoldPeak established for this transaction a Delaware based company called New Vectrix and re-named
as Vectrix Corporation. Vectrix Corporation is 100 % owned by GP Batteries through EVB Technology
(HK) Limited. The organizational structure of the new company will be communicated later
The Vectrix Corporation is now working to consolidate all the activities (headquarter, sales, technology
and production). Lean manufacturing is based in Wroclaw, Poland and keep continues operation as
usual
Vectrix continues to produce the Vx1 bikes and provides customers with accessories, spare parts and
technical support.
The company is currently developing new models of high performance, zero emission scooters and will
be introducing them in the near future: By the summer 2010 a new version of the Vx1 including several
technical improvements will be introduced to the market. The new Vx1 will have a GoldPeak Li Iron
Phosphate Battery. For the existing Vx1 customers, an upgrade kit to the Li battery technology will be
offered.
Two smaller electric scooters are also under development: one will be a 50cc equivalent and one will be
a 150cc equivalent. Both scooters will be powered by a GP Li Iron Phosphate battery.
Vectrix will inform all dealers at appropriate time about the technical performance parameters,
availability and price structures of the new bikes.
Vectrix acknowledges all warranty claims in Europe regarding Vx1 bikes which were maintained
according to the operator's manual and which were officially registered for warranty with Vectrix.
Wolfgang Gohl, former Vice president for Operations and new Vice President commented the take over
by Goldpeak: “This move is a triple win for customers, suppliers and for the employees. It provides the
required stability to develop the key market and the right product. The EV market is just starting to
emerge, first policies are under development and Vectrix and his partners are in a prime position to
supply a quality product for various applications.”
Wesley Lau, Head of Strategies Business Unit & Special Project – GP Batteries and New assigned
President of Vectrix commented the partnership: “Vectrix is the leading brand in the electric 2-wheel
market. This is based on a superior product and good positioning in the market. We will work with all
our customers to enhance this position and provide even better solutions to the electric vehicle
market”.
Wroclaw, Nov, 15th, 2009

It's official :-) We will be getting LiFePo4 battery packs.

let's hope ...
then guy from vectrix just told me that the low volatage I can monitor during the discharge is not a reliable criteria of the state of charge and that the gauge vas meriging the capacity of the battery according to the last discharge and according to the internal resistance of the batt ....
it means that I don't know when I will run out of energy ....
what do you think ?
great....

yeh. nise one
music to my ears
all good news thanks for this.. look forward to hering anymore info.
kev

but the replacement pack is only 30Ah.

i reckon i can fit 44 x TS 60Ah batteries in my V (with a little modification by my favourite boiler maker)

Matt

How do you know it's only 30Ah pack?

Matt, How do you plan on fitting 44 TS60 cells?

It looks like you can fit them flat in the bottom of the battery box side by side. If so, 3 stacks of 10 cells, 5 high will fit in the exising space. You'll need to nick space used for the plenum and fans and possibly form a higher step thru hump.

What's the boily for? You don't need to cut the frame do you?

Why 44 cells? I though 40 was a closer. The controller will go low batt at 2.4V per cell for 44, better than 2.7 but looking at the discharge curves there isn't much capacity to be had.

Maybe we need a TS 60 thread...

That is, finally, very good news!!
At least we know that when our NIMHs die we will have a solution, probably at a high cost, but be able to ride our Vecs for many years to come.
They have a possible market of 3000+ Vecs for an upgrade Lithium pack, then there is the new Vecs they will sell with a probable option between Lithium and NIMHs untill these run out of stock.
Anyway, this is Very possitive and GP is an important and reliable consumer Battery Brand, lets just hope that they give, new and old customers, a good long warranty in the Battery Packs because it really is one of the weakest points of the VX-1 and one which worries people most. Just do a poll about "how many of us think that the packs will dure more than 30.000kms" and the probable result will be zero!
If your fuse blows you get worried but know its a small matter, if your software has a glitch same thing but if your batteries die you swallow hard and dry and turn pale, your stomach tightens and you feel very very dissapointed and will curse the Battery(and Brand). I really just hope the new owners understand this and give the best and most reliable Battery pack.
My best Wishes to Gold Peak and hope they listen to the final user, it will be a win-win situation.
Raj

there is something sure : for the moment only the batt manufacturer can proprely run an EV business, in France mr Bolloré, one our biggest industrialists has develloped the blue car that is to be put on the market, and he is making the batt whith his own company and plants.
Small ev companies don't have enough power to struggle against the batt manufacturers, I think that this is what happened to vectrix .
theire pack must have been to expensive for them .
What has been a scandal for me in this story (web master unpublish this post if it is not in the forum ethics frame) is that there price of the packs was not told to me when I asked at the time i have purchased the vec, and now we know that the pack costs at least half of the vec price and has not been available for a long period letting lad out in the cold for a long time with fucked packs .
When we anwser question on the road to petrol bike ridder they often ask us for the price of the pack , and when we say the truth they say they won't buy the vec .
When I say the truth about the high probability to fuck the pack and the impossibilty of getting a new one immediatly, and when I say that if in the future Vectrix manage to sell packs again they will cost half of the bike price for an estimated life millage of 30.000 km in real life (as far as we know reading the forums) they say they will definitly not buy the vectrix .
They all say that the pack should be less expensive and properly warranted .
they all say that if the rider has to pay permanent attention to the way he handles the vec not to burn the pack this means that this bike has not been properly designed .
there is no transparency as for the number of fucked packs out of the number of vec solded.
there is no transparency over the real maximum millages reached so far by ridders .
This is not something that will encourage people to buy vectrix .
Sometimes I also happend to lie, or rather not to say exactly all I know about the necessity of beeing very cautious to manage to keep this bike alive... we all would love that Vectrix survives and manage to issue reliable machines .
Mabye the figures are mutch better then what we think reading the forums posts
Mabye many more bike are working fine like mine at the moment at 15.000 km ,
mabye will it work during all it's expected life .
Nobody knows .
What is sure is that real vectrix ridders are the show-window of theire businesses .
What is sure is that we must be heard as the one that first put a lot of money on those bike making it possible for it to be developped .
what is sure is that they have the higest interest in treating the 3000 first clients as weel as possible .
regards
jean mi

just an evidence of all of this, when I bougth the vec , 5 of my friends were about to buy the vec, but when the have read on the internet all the complains about fucked packs and warranty problem they turn ran away ....

Matt, How do you plan on fitting 44 TS60 cells?

It looks like you can fit them flat in the bottom of the battery box side by side. If so, 3 stacks of 10 cells, 5 high will fit in the exising space. You'll need to nick space used for the plenum and fans and possibly form a higher step thru hump.

What's the boily for? You don't need to cut the frame do you?

I was planning on doing two layers of 2 x 11 cells.
for these cells, the plenum and fans are un-necessary (though the fan for the controller is still needed)

The boily is to weld the mounts and hold downs for the second layer.
being made of aluminium, its well beyond my skill level.

Why 44 cells? I though 40 was a closer. The controller will go low batt at 2.4V per cell for 44, better than 2.7 but looking at the discharge curves there isn't much capacity to be had.

the max on charge voltage of the charger is 160v (will need to muck with the firmware to get it to go this high). 44 * 3.6 = 158.4v
The speed at which you hit maximum power point is very dependant upon pack voltage.
a stock vectrix makes max power at around 75kmh when the pack has sagged to 110v
44 TS60 cells at 275A is around 115v. thats 31.6kw at the battery at around 80kmh (minus losses)
I actually have some A123 cells on order that was going to go to another project to reduce sag some more.

the battery side current limit of the controller is 275A (or so i have been told)
the battery side low voltage limit can also be changed with firmware (actually im more looking to tie with a BMS)

it will be quite some time before i do any of this.
for the moment im waiting on:
1) the original pack failing
2) getting my hands on the source code

Matt

Hi Matts,
the key point as a matter of fact is too be able to get the soft, but is it necessary to get the source code ?
maybbe would it be enough to get the current soft to modify the charging dischargin parameters no ?
regards
jean mi

it depends how adept you are at coding in machine code (one below assembly)
well beyond my skill set aswell.

at the source code level, you do much more with much less knowledge of how it all works

to go to lithium, the behaviour at end of disharge, and end of charge will have to be changed.

at end of discharge, BalPor trigger will have to be a flag from the BMS, while below a specific current, for a specific period of time.

at end of charge, charge termination will have to be a flag from the BMS (actually more like BMS flags one cell is high, then go to a small constant current if top balancing. if working from bottom balance then the charger should just shut down)

actual pack voltage limits on either charge or discharge would only be there in case of BMS failure.
so max charge voltage would be 160v.
minimum discharge voltage would be 115v.

all this is relatively straight forward (on paper at least) with the source code, but a PITA with just the machine code from a firmware update.

having said that, if anyone could get hold of the firmware update to go with the K2 BMS for the Vectrix, then life is even more straight forward :D

Matt

...
for these cells, the plenum and fans are un-necessary (though the fan for the controller is still needed)
...

Be very careful there! Have a good long look at the impeller housing and the spaces on each side on the back of it. I believe it might divert a portion of the cooling air stream to re-circulate through the battery, after cooling the MC board from the inside!

Matt, it would be nice to have the TS cells so well integrated, but I'd be happy to settle on a less sophisticated system sooner.

Here's what I'm thinking

With my old software, I believe the low batt level is 108V. spread across 40 cells, that's 2.7 volts, a little above the 2.5 minimum

from the TS60 spec sheet at the bottom of this page. If you hit 2.7V at 1C or higher, that should trip the batt low to reset the "fuel gauge" and still leave 5 AH or so to trickle home.

If the charger is trying to push 153V at the pack, that'll give 40 cells about 3.8V, well under the 4.2 max and a touch over the 3.65 ideal.

I'm hoping these: http://www.evworks.com.au/store/datasheets/EVPower_CM_datasheet.pdf will cope with that over voltage and be able to shunt enough in CC mode to balance the pack nicely.

Daisy chained to this: http://www.evworks.com.au/index.php?product=BMS-MCU-TS90-EV2 this system could detect a cell over voltage and turn of the supply to the charger. a cell under voltage on the run could throw an audible and visual alarm to alert me that I had a problem. I guess the worst cell would sag the most under load, so the alarm would sho first the, and backing off would protect the cell for a short while, recharging would protect it for longer until I could sort some way to trickle charge the weak cell, or the whole pack to balance it up again

I was thinking of laying the cells on their largest face in 3 groups, 2 stacks 6 high/2 stacks 7 high/2 stacks 7 high.

I still have work out how to incorporate the factory temperature and voltage sensors so they atleast don't throw up faults when everything is ok. Mik, help?

From EV works website 40 cells ts-lfp60, 40 modules, 40 terminal straps and the BMS box come in at a shade under 5k, plus shipping

If that will all work, I'm keen, these NiMH's are pissing me off. 4km push at 2am this morning

ah yes,

ive been using those BMS modules for ages, great piece of kit.

the modules have enough balancing capacity to do the intial balance (though it does take 4 days)
the master will disconnect the charger no worries

causing a BalPor is trickier.
causing controller cut back at end of discharge is also trickier, but necessary.

if you could monitor current (RETAMPI comes to mind) it might not be so bad.

basically, you need to keep reducing discharge current to keep the lowest cell above 2.5v.

for my setup, i was looking at re-using one of the sub-pack voltage sensors for a good/bad signal for the controller, so current back off can be automatic.
im not sure how this would work without changing the firmware.

its unfortunate you had to push.
when i had an outta juice problem with my V, i happened upon a construction site power box.

did you get disappearing bars syndrome on your usual commute?

Matt

...

I still have work out how to incorporate the factory temperature and voltage sensors so they atleast don't throw up faults when everything is ok.
...

Have a look at the stock BMS here: http://visforvoltage.org/book/ev-collaborative-hand-books/6747

You need to simulate that a battery is connected to the sensors like that, somehow.....

Voltage monitoring occurs through this voltage divider:

The temp sensors should not be much of a problem, just stick them somewhere so they report the same temperature.

Maybe the voltage sensor tab cables could be used to connect the VX-1 to a LV shutoff signal from the BMS of the replacement pack: Insert some device (which has a close to zero resistance when not triggered) into the cable. When the Low Voltage Cutoff signal is received by the device, it's resistance increases so that the stock-BMS sees a strong imbalance in the pack. With a bit of luck that might cause the crawling mode,but it could also just turn on the battery telltale or shut the motor down.

I may be missing something regarding the physical dimensions on these cells. each is 61mm thick. How do you get stacks of 6 or even 7 in a battery box that is only 330mm high?

you don't.

I haven't pulled the bike apart to measure and confirm yet, but I think it might fit if you remove the top cover and fans from the battery box, the plastic body work over the battery box between the seat and the ignition switch and fabricat new items to fit over the cell and mount the seat. New fans or relocating the old ones might be necessary as well as rejigging the cooling air flow.

the TS40 cells look like they will just about fit in the existing space, but that isn't a huge range improvement for about $3,800. an extra 1200 would get me double my existing capacity (if my NiHM cells work as advertised...better if they don't)

Thanks Mik, I thought I had read about you being into that part of the bike. Thanks for your excellent input into this forum :)

I'd really like a functional description of that module and what the trigger points are on the controller. I wonder if it would tolerate 10/10/20 cell LFP split. Maybe it is simply a matter of resizing the resistors to balance the outputs for the different group voltages.

As you say, the NO relay contact on the http://www.evworks.com.au/store/datasheets/TS90-Installation_instructions.pdf might be used to switch in an extra resistor on one leg of the circuit to unbalance it. Maybe the same switched power could supply the visual and audible alarm. I don't know if the internal relay is isolated enough to do that safely.

it depends how adept you are at coding in machine code (one below assembly)
well beyond my skill set aswell.

at the source code level, you do much more with much less knowledge of how it all works

to go to lithium, the behaviour at end of disharge, and end of charge will have to be changed.

at end of discharge, BalPor trigger will have to be a flag from the BMS, while below a specific current, for a specific period of time.

at end of charge, charge termination will have to be a flag from the BMS (actually more like BMS flags one cell is high, then go to a small constant current if top balancing. if working from bottom balance then the charger should just shut down)

actual pack voltage limits on either charge or discharge would only be there in case of BMS failure.
so max charge voltage would be 160v.
minimum discharge voltage would be 115v.

all this is relatively straight forward (on paper at least) with the source code, but a PITA with just the machine code from a firmware update.

having said that, if anyone could get hold of the firmware update to go with the K2 BMS for the Vectrix, then life is even more straight forward :D

Matt

Wouahw !!! working on this terrible RISC microchip ASM code ... a real nigthmare ...if you manage to do it you are the best .
I use to program this kind of pic's using their assembly ... I wanted to make pure programs...but my basic compiler generated shorter and unbugged programs then me in assembly even paying a great attention to the soft architecture .
I hope you will be able to reach your goals my friend, but it will not be a piece of cake .
good luck (this not ironic)
regards
jean mi

42 cells are a much neater fit for the standard charger and low batt cut off

153/42=3.64V charge (recommeded 3.65) worried that the balancing won't work though

108/42=2.57, low cell alarm switches on at 2.5, close to fully utilized capacity at low batt and will alarm if cells un balanced.

back to considering the 40AH cells, no body work mods , keep all factory components including fans, in stock now...

with 40Ah cells, you can fit at most 34 cells, without bodywork mods.

body work mods are just about unavoidable.

another issue is higher continuous discharge rates reduce battery service life.

on my emax, my 40Ah cells at 55A continuous, 200A peaks gave equivalent 500 cycles to 80%dod before capacity reached 70% of original.

60Ah cells will last disproportionately longer.

Matt

btw, EVworks don't have 42 cells of anything in stock atm (they do have one or two in a few sizes)
large cell orders won't be around till after christmas.

all the TS vendors i know of on your side of the pond only have 90Ah and 160Ah cells, since theyre aimed at cars.

i have about 25 40Ah cells floating around though, and another 74 90Ah cells. its a real shame i can't fit 90Ah cells in my vectrix, since i already own enough :(

on the upside, the pricing for more than 16 cells is somewhat better than one or two.

pricing is around AUS$ 1.45/Ah/cell, changing daily with exchange rate.

Matt

I thought I might be pushing the TS40s for discharge rate...

EVworks web site show the 40s "in stock" and the 60s "order now". Might be time to update the website inventory :)

These 40s, layed flat should fit. 2 x 116 wide is close enough to 230 ( nothing a linisher couldn't fix) 7x46 is 322 (well under 330) 3 x 190 long gives about 45 mm clearance to fit all the connectors and control modules. Should be a doddle. Probably a moot point if the current draw is too much.

to fit the 60s in the same configuration, I have to find another 100 mm in height. terminal fit will be a lot tighter too.

No news on the lithium conversion over your way?

Here's the data sheet for the new GP LifeFo4 pismatic batteries. They are actually ideally sized for the Vectrix battery box. As you can see, they are 45Ah and 46 would fit in two rows of 23 side by side terminal facing up with lot of room to spare.
This lloks like a perfect arrangement. either 39 or 40 of these would give a 45Ah pack at just 64kg!

http://liionbms.com/pdf/goldpeak/GP45EVLF.pdf

Type : Rechargeable Lithium Ferro Phosphate Model No.:GP45EVLF
Prismatic Batteries
Nominal Dimension : T = 30.0mm (max.)
L = 225.0mm (max.)
W= 95.0mm (max.)
Nominal Voltage : 3.2V
Capacity : Min. 45Ah at 22.5A (0.5C) discharge to
2.0V at 20℃
Charging Method : 22.5A (0.5C) to 3.65V for 2.5hrs at 20℃
Charging Termination
Control
: Taper Current 0.05C (2.25A) at 3.65V
Max. Discharge Current : 450A (10C)
Service Life : >1000 cycles
Weight : ~ 1.6kg
Internal Resistance : < 2mΩ
Ambient Temperature
Range
: Charge : 0 to 60℃ (conditionally)
Discharge : -30 to 70℃

They also have a 30Ah prismatic set. see here

http://liionbms.com/pdf/goldpeak/GP30EVLF.pdf