hehe, thought id be fashionabley late, and make my own thread :D
i got my vectrix today.
after having the obligatory ride around the workshop, i got to work pulling it apart.
my fancy current limiting power supplies haven't arrived yet, so i had to come up with another solution to slow charging.
i dug up 2 x old 48v lead acid chargers, and put them in series with my lab powersupply, meaning i now have a 115-170v, 0-1.5A power supply.
i then used alligator clips to temporarily attach them to the battery and powerup.
i thought to measure the pack voltage while i was there, and to my surprise, it was 134v.
the bike had spent at least a week in transit, so Vectrix Australia must have charged before shipping - #^%@$^%#
all the celcsi dots were the right colour, so i thought,y not.
i had planned to charge at 0.35A for 100Hrs (give or take 8 or so, depending when i got back to the workshop).
instead, since the batteries had a decent charge in them, i rode it to get registered.
the VIN tag was supplied loose (that was dodgy), but there was already an imprinted VIN tag on the body.
i couldn't get it registered without the VIN tag attached :(
now its at home, charging with my charging setup at 1.2A
while i had the bike apart, i bodgied a couple of wires to the traction pack, and ran them into the rear storage compartment - because i cbf pulling those panels off again.
the charge connector supplied was an odd one, i haven't got around to putting an Australian one on.
btw - for pre-charge before putting those two aderson connectors back together, i used my power supply, no fancy pre-charge circuit needed :D
Matt
Hip, hip, hurray!
How do you prevent electrocution or fire caused by the new wires which you ran into the boot? Diodes? Fuses? Luck?
I have not properly tested charging currents lower than C/10 (=3A) on Vectric Gold Peak cells, but I was told that weaker charging currents might be ineffective for Gold Peak NiMH cells. That is apparently the advice given by Gold Peak themselves.
What could make low currents ineffective? Could they possibly run through part of a cell, bypassing the rest and therefore never ever achieve a full SOC?
Could small differences in IR between parts of the layered electrolytes cause uneven current flow through some or all cells?
How can you evealuate if or if not the less-than-C/10 charge is effective?
Anyway, another way to do it would be to perform multiple short 3A charges, allowing for cooling down time. That's what the newer software versions in the Vectrix do. Maybe there is a good reason for this?
This information may be used entirely at your own risk.
There is always a way if there is no other way!
presently - luck
ive run two wires, from either extreme of the pack (2 core, double insulated).
its a temporary solution, i will probably re-do with much lighter wire.
i have been playing around with 0.12mm^2 wire for another project, and testing its failure mode when short circuit with no fuse.
at ~48v it just goes open circuit, nothing particularly interesting happens.
at 3.3v, the insultation burns away, then the wire rapidly oxidizes, going high resistance, all in the space of around 10sec.
as far as avoiding electrocution, i may add a diode for a second try :).
I must profess, my knowledge of Ni-MH is not quite what it is with other chemisrtries.
i would suggest, as with other chemistries with secondary reactions, the secondary reaction eats alot of current at high SOC.
at a low charge rate you may never exceed say 80% SOC.
more testing might have to confirm
fortunately, i may have access to a dead vectrix battery, where hopefuly some cells are still ok. i can do some bench testing at my leisure using my CBA II and various power meters.
perhaps the reason for the 3A charge rate, is due to hardware limitations.
The charger may not be able to regulate current accurately below 3A in CC mode.
thus min current is 3A, but pulsed to average at a lower value.
its a possibility, but i don't know for sure (unless i get around to re-programming the charger)
Matt
Daily Ride:
2007 Vectrix, modified with 42 x Thundersky 60Ah in July 2010. Done 194'000km
Yes, in the previous version of software, the TR (15 minutes) and EC (60 minutes) part of charging was at 1.5A.
Nice one Matt :)
Any word yet on the LiFePO4 Vectrix out there?
Is the owner on this forum and willing to share their journey?
no new news re the lithium vectrix.
i suspect they are waiting for the cells ordered to arrive.
i tried to get ym vectrix registered again today, but needed more proof of ownership than i had.
another day wasted :(
i also ran into a new problem,
i forgot the vectrix charger counts the Ah removed on discharge, and puts that much back in during the bulk phase.
tried to charge with the on board charger today, after i knocked this setup out of sync with my initial manual charge. rats.
when you upgrade the firmware, does the vectrix "forget" all the previously measured values (Ah capacity of battery, SOC, etc?)
is there a way to force it to recognise the pack is fully charged?
i know how many Ah i put in, if worst comes to worst, ill just manually discharge some (or let the constant self discharge sort it out)
next time i have the vectrix apart, i will be moving my direct wiring to *after* the hall current sensors.
Matt
Daily Ride:
2007 Vectrix, modified with 42 x Thundersky 60Ah in July 2010. Done 194'000km
mmm let's say 03/2010.
I doubt that will make it count the additional Ah gone in.
You can just synchronize it by driving until the battery is empty. You can measure through your wires in the boot (might want to add 15kOhm resistors, then extend them to a DMM on the handlebars for real time readings).
This information may be used entirely at your own risk.
There is always a way if there is no other way!
perhaps a bit off topic:
(again I admit not beeing a specialist in electrics/onics)
I wondered: is it possible to measure the voltage (or overall condition) of a single cell while it is connected to the other cells?
(I also saw the pictures submitted by R where they are measuring in what appears to be that way)
Is it as simple as putting the probes on the +/- of each cell, or is there more?
If so, it would be easy to monitor the state of each cell. (we could measure them from time to time to avoid damaging other cells, or even monitor them constantly??)
"doing nothin = doing nothing wrong" is invalid when the subject is environment
I also admit not being a specialist in electrics/onics, so please don't take me very seriously, hehehe.
Sure, putting the probes on each cell terminals will tell you the cell's voltage. However you will only have easy access to the upper battery layer, There are three layers of battery, so the only way to check the 102 cells is dismantling the layers...
You can monitor and balance them with a battery management with lcd display, however there's no space in the battery compartment to install such devices... a real pity.
are you suggesting the Ah count during recharge is a calculated value supplied by the charger, rather than a measured value from the hall sensors?
its possible, just un-expected, as the accuracy of the measured value from the halls will be far better than the calculated value from the charger.
driving until the battery is empty only tells the guage the capacity of the battery, if the discharge started at a full state of charge, and all intermediate charging is accounted for.
In my case, i have had un-accounted for intermediate charging.
In my opinion, it would be very bad to do a full discharge at this point, as the measured capacity of the pack will be well above the actual capacity, resulting in disappearing bars syndrome.
Fixing it would also be difficult, as the correction after successive full discharges is minor.
i just had the vectrix registered today, so have put some km on it.
today i recharged with another collection of powersupplies from my collection (combined, 115-170v, 2-10A)
tomorrow i shall wire the builtin in charger properly, so i can correct the fuel guage issue.
at present my guage reads 0 bars, 5km est range, and has done so for the past 20km.
definately dont want the guage to recalibrate now :p
Matt
Daily Ride:
2007 Vectrix, modified with 42 x Thundersky 60Ah in July 2010. Done 194'000km
Hi Turok,
yes this is a possibility.
if you *know* which is the weakest cell, you can ride to the constraints of that cell.
It would require dismantling and testing of the pack.
Mik has done this.
I might, if i can get hold of that spare pack with many damaged cells.
having a few spares will make the process far more worthwhile.
Matt
Daily Ride:
2007 Vectrix, modified with 42 x Thundersky 60Ah in July 2010. Done 194'000km
Matt
Do like Mik said. It doesnt matter the state of charge or gauge reading. They will sync themselves.
I do not know, but you will not get the desired result if you only add current through your additional chargers while the stock charger is running.
Be very wary of ground loops if you want to try it for some reason.
You have taken over, and you are batter at it than the stock BMS! It will only be a problem if you suddenly expect the stock system to perform flawlessly although it does not know what extras you have thrown in!
I cannot be sure abut the newer software, but the old software would OVERCHARGE the battery if it is much fuller than expected. If it is not yet fully balanced, then this could potentially damage the best cells in the pack. The voltage would reach the 152V limit (or whatever it says after the "CP") for four times at 10A charge current before it gives up. If half the pack is at lower voltage due to imbalance, then the rest of the pack gets hammered and might vent. The worst scenario is if just a few cells are full, because they are really good cells with high capacity and lower self-discharge rate. They can then get hammered at C/3 for quite a while, until they are no better, or much worse, than the rest of the pack!
This information may be used entirely at your own risk.
There is always a way if there is no other way!
Yeah, but that would only be the first charge after the red battery telltale or BALPOR in the old software. Next charges never charge to the cut-off (CP152, CP151, CP153 (depends on the battery temperature)) voltage.
It can take several charges, depending on the discrepancy between assumed and real state of charge. Each charge add about 2 bars to the display.
This information may be used entirely at your own risk.
There is always a way if there is no other way!
Yep, that was with the old software. The newer software is a bit different.
Old:
- CP 152: cut-off was actually cut off when voltage reading jumped from 152V to 153V: in practice; CP 152 was cut-off at 153V
- TR - 15 minutes: charging at 1.5A
- CC - 1 hour at 3A
- EC - 1 hour at 1.5A
New
- CP 152: cut-off is at 152V. When the voltage reading jumps from 151V to 152V, it is the end of CP
- TR - 15 minutes: no charging, just cooling
- CC - 1 hour at 3A
- cooling for aprox. 30 minutes
- EC - bike is shut down. The only way to observe the reading is to turn the bike on. It reads EC 000
The old software, obviously charged more. Cut-off at CP was 1 Volt higher + 1 hour and 15 minutes of 1.5A more than the newer software.
Newer software works in a narrower SOC range than the old one did. The old one regularly overcharged the battery in the effort to balance the cells with each charge, because the discharge cut-off was so low, that, with unbalanced cells, you could reverse cell/s.
The newer software balances only once in the 12 hour of riding period and prevents polarity reversal with higher discharge cut-off.
The old software would also repeat the CP stage three times if the cutoff voltage was reached before the hourglass display was showing 16/17th.
Each time the cutoff voltage was reached, it would go to "tr" mode with about 1.5A charging for 15min. After three repetitions it would give up and go on to "CC".
I don't know if the new software repeats if the cutoff is reached.
Each repeat "CP" stage was only a few minutes long, but did of course further heat the batteries.
This information may be used entirely at your own risk.
There is always a way if there is no other way!
Im guessing this is something you have tried then.
i was planning on charging with my powersupplies while the bike is on normally (as in ready to ride off), and not charging.
The idea is that the charger/guage assumes its regen.
All my chargers and powersupplies are completely isolated, no ground loops for me.
in other news,
my battery wasn't as charged as i thought it was, and i hit the battery red light $^%#*#
got it back home and charged it.
it charged fine,
start of charge: 123v, 33C
end: 144v, 27C
guage is full.
came back and it was off.
CO 152 was the code when i turned it on with the kill-switch on, and it was counting down.
so looks like i have the latest software already ( i can also tell as power was being cut back before and during the red light came on)
the guage read full, with 65km range est.
rode it about 14km, and lost top two bars, and range estimation said 37km.
i will have to wait till the next red light event to see if the stored battery capacity value is too high (i think it is).
before the red light came on,
i was planning on doing a full charge followed by a two day wait, then discharge to red light for calibration.
the idea is, keep the charges and discharges within the %80 and %20 SOC window for a long service life.
balancing would be at 80%, rather than 100% as with all other chemistries.
i was under the impression in the CP stage the charger just put back the number of Ah pulled during discharge, and stopped there, rather than just continuing charging until it hit the "CP" voltage.
So that approach isn't going to work anyway.
backup plan is to install a CAN bus reader to tell me the Ah pulled during discharge, and i use a manully set timer.
A job for later, for the moment.
Matt
Daily Ride:
2007 Vectrix, modified with 42 x Thundersky 60Ah in July 2010. Done 194'000km
33C! ...! When the battery is so hot you should cool down your battery before recharging.... some would say 9 hours some others 2 hours. It is up to you, but if you keep it charging at 33C without cooling the battery won't last very long...
Only if you discharge before the red battery telltale. If the red battery telltale is lit, then it will do a full charge until the cut-off voltage, that is written by the CP. That cut-off voltage depends on the battery temperature.
Next time, I suggest you use some precharge cooling, because if your charge ended with 27'C, you could have easily precooled it to a lower temperature than 33'C before you started charging.
Remember, battery charge temperature below 30'C is essential to a healthy battery life!
Sing along: "We come from the land downunder, where batteries glow and men plunder,...."
Hey, it just does not get that cool here in summer! Maybe at night in Perth, but generally speaking it's a tad tricky to keep things cool here!
As everyone knows by now (yawn), a 12V 4A power supply (ABCool) gets the cooling impellers going at will. Like at 0300am, when there is some cool air around to suck through those batteries.
This information may be used entirely at your own risk.
There is always a way if there is no other way!
Well, he mentioned, that at the end of charge, battery temperature was 27'C. So it must had been cold enough, so that the precharge cooling would have cooled down the battery under the 30'C.
This information may be used entirely at your own risk.
There is always a way if there is no other way!
actually, up until yesterday, i didn't know i had the new software.
so there was no pre-charge delay.
it started charging straight away at 11A.
i think i shall start using the pre-charge delay, now that i know i have it.
ambient temp at the time was around 20C, and falling.
the pack temperature started falling straight away after plug in.
im a night worker, so it really only gets charged after sundown.
as with my old scooter, i can charge at work (starting at 7pm) and then charge when i go home (a bit after midnight).
since i have much more range than my old scooter, i will probably also start charging only at home.
interestingly,
i have yet to hear the cooling fan going any time other than when charging.
with the new software, at what temp is the fan supposed to operate?
i hadn't realised the max charging temp was so low
one of the reasons im going to try and reduce the SOC operating window, is the majority of damage and heat occurs at the end of charge (as i understand it).
this could be one of the reasons for such a low max charging temp.
thanks for helping to bring me up to speed guys :)
Matt
Daily Ride:
2007 Vectrix, modified with 42 x Thundersky 60Ah in July 2010. Done 194'000km
You could have found out that easily. Just put the kill switch in the off position and pull the left break. If the left LCD display shows Voltage and battery temperature, than you have the new software. If it doesn't, you have the old one.
Try to use as long precharge cooling as your time permits, so that when you leave, you have the bike charged. It will stop cooling when the battery temperature reaches 24'C or it's within 3 degrees of ambient temperature. When the precharge cooling delay expires, then it starts charging and turns on the battery fans.
How is it running so far?
Can you give an estimate of the useable battery capacity, yet?
The big question about the remaining new Australian VX-1s is of course how much battery damage they have.
With a careful break-in (the opposite of what Vectrix recommends) the batteries might be very good; but it is also possible that they have too many bad cells. Of particular interest is how many cells are still in excellent or very good condition.
Could an Australian Vectrix owner reasonably expect to make a very good battery pack out of a damaged (used) pack and the pack out of one of these "new" bikes? (Or shall we call them "matured"?)
If 90% of the cells were still good, it might make sense to buy one to fix the battery and to secure spare parts.
This information may be used entirely at your own risk.
There is always a way if there is no other way!
Hi Mik,
battery capacity until i notice reduction in power is around 15Ah, based on how long the CP stage goes for at 11A.
even at that stage, 100kmh is still possible, i just dont feel the good acceleration at 70kmh when i whack the throttle wide open.
hopefully i can rig an Ah counter to give more accurate data by using the current measurement from the CAN bus
i haven't pulled my pack apart, so i can't know for sure the capacity of the individual cells.
all the celcsi dots i could see were nice and white.
however, temperature hasn't been a problem for me, even on 33 deg C days at deep discharge.
so that would indicate the cells are all reasonably similar capacity, with no reversals.
the higher low voltage cut off also helps with this
i get disappearing bars syndrome, however, this could be poor calibration of the "fuel" guage due to my charging with external charger and the discharging to battery light (unless capacity calibration occurs on the charge cycle?)
a bit less than half the bars equates to around 15Ah, so having the last few disappear is no surprise.
when i get the source code, i will know which ID to change so the assumed capacity is correct, no relying to discharge measurement for me.
The battery in my "matured" bike has the characteristics of a good battery.
a damaged battery from a used vectrix will probably have half good cells.
the cells that had the best cooling, and were never over-charged or over-discharged will be fine.
the low voltage cut-out would have been to low in the original versions, but the lower cells would have protected the higher cells.
likewise on charge, the higher and lower capacity cells would have protected the lower cells.
the mid-range ones should be fine, the outliers are history.
im trying to get access to a old dead battery from the guy changing to lithium, but it looks like i will have to wait till early next year before that happens.
personally, if you were to buy one of the cheap "matured new" bikes, just use that one, and use your old one for spares.
the new bikes do appear to come with good condition batteries.
Matt
Daily Ride:
2007 Vectrix, modified with 42 x Thundersky 60Ah in July 2010. Done 194'000km
I would be interested in buying a damaged Vectrix battery which needs work and analysis, for cheap! (Or a whole Vectrix, for that matter!)
But of course, you might want to do this yourself.
My new CBA III should arrive any day now, the new laptop is working, all ready for action! I've got a USB Isolator and the "Pro" version of the CBA software, and a "commercial" warranty. I hope I'll be able to use it a lot!
At the moment I don't think I'd buy another new Vectrix, unless they get dirt cheap. I'd like to find AU$ 3500.- because I have an offer to buy a Prius MK1 in working order + another identical vehicle with some chassis damage for spares. That would b e nice for those days when it's too hot for the Vectrix, or when I need more range. And, it would allow me to build and test reconditioned batteries and ECUs (Electronic Control Units) for the Prius MK1. I think there is a market.
If you figure out how to put a Lithium pack in the Vectrix, then I might be interested in a group cell buy.
This information may be used entirely at your own risk.
There is always a way if there is no other way!
a bit of an update,
almost got stranded at work this morning.
pulled outta work with 2 bars over half remaining, and battery tell-tale came on.
silly me ran the fans the whole time i was at work but didn't charge it :(
a combination of fan discharge, memory effect, and self discharge screwed me.
after full recharge, i still get full capacity (well i took it down to last 4 bars and pack voltage was still 128v, so im pretty sure i do anyway).
anyway, just testing the waters,
how much would you pay for a used vectrix battery that after full charge and balancing still gets:
more than 30Ah with the 70A to 108v testing method,
and around 30Ah with the discharge at 1-2A till 1v step drop testing method (tells you capacity of lowest cell without pulling the pack to bits)?
Matt
Daily Ride:
2007 Vectrix, modified with 42 x Thundersky 60Ah in July 2010. Done 194'000km
Pages