Battery Performance

51 posts / 0 new
Last post
AndY1
AndY1's picture
Offline
Last seen: 8 years 12 months ago
Joined: Tuesday, June 17, 2008 - 04:29
Points: 1071
Re: Battery Performance

Thank you!

I need to do 4 more deep-discharges (this was my first one). It's getting pretty cold here (2'C on the way to work, but it was a joyride), so I'll just probably do 1-2 more and leave the remaining deep-discharges for the spring.

AndY1
AndY1's picture
Offline
Last seen: 8 years 12 months ago
Joined: Tuesday, June 17, 2008 - 04:29
Points: 1071
Re: Battery Performance

Yesterday, after I deep-discharged my Vectrix, I waited for 10 minutes and then I plugged it in and started charging. Ambient temperature was 10'C, battery temperature was 38'C. During charging the battery temperature dropped to 36'C. At the very end of EC charging (ambient temperature was 5'C), the battery temperature rose up. In the last 5 minutes of EC charging, the battery temperature rose to 44'C and a battery red telltale turned on. When I turned on the bike in the morning, everything was ok. I guess the battery temperature rose to high during the end of charging.

1. Question: Did I damage the battery and lost capacity?

2. Should I do a second deep-discharge today or should I wait a few charges/days for the second deep-discharge?

Our block-apartment keeper today showed me the location of a plug in the garage :D after he saw the bike. I asked him for a permission to use it and he said it's ok. I'm a happy camper ;-)
That will also enable me to start the charge late at night and not in the evening, because I had to plug it from my balcony extension cord.

Mik
Mik's picture
Offline
Last seen: 6 years 3 months ago
Joined: Tuesday, December 11, 2007 - 15:27
Points: 3739
Re: Battery Performance

Yesterday, after I deep-discharged my Vectrix, I waited for 10 minutes and then I plugged it in and started charging. Ambient temperature was 10'C, battery temperature was 38'C. During charging the battery temperature dropped to 36'C. At the very end of EC charging (ambient temperature was 5'C), the battery temperature rose up. In the last 5 minutes of EC charging, the battery temperature rose to 44'C and a battery red telltale turned on. When I turned on the bike in the morning, everything was ok. I guess the battery temperature rose to high during the end of charging.

1. Question: Did I damage the battery and lost capacity?

2. Should I do a second deep-discharge today or should I wait a few charges/days for the second deep-discharge?

Our block-apartment keeper today showed me the location of a plug in the garage :D after he saw the bike. I asked him for a permission to use it and he said it's ok. I'm a happy camper ;-)
That will also enable me to start the charge late at night and not in the evening, because I had to plug it from my balcony extension cord.

I would stop doing deliberate deep discharges.

What might be happening is that the cells with the least ability to absorb charge will start to heat up at the end of charging.

I do not know if they arrange it deliberately when the packs are assembled, but the worst cell in the Vectrux pack had a temp sensor on it.
The worst cell = lowest capacity cell will get hotter sooner than the others. It would be perfectly sensible to put the temp sensors onto the cells most likely to get hot.

However, it is really time consuming to figure out which cell will likely get hotter than the others, so it might not have been done. But a US$2000.- tool would tell a technician in the factory within seconds what the internal resistance of a cell is, and that resistance is closely related to heating and capacity.
The battery manufacturer most likely has =and uses= this sort of equipment.

So lets assume that the worst battery in your pack has a temp sensor on it and heats up to above 45degC in 5degC ambient temp during charging.

If your worst cell does not have a temp sensor on it, then it will be even hotter than 45degC at that time.

You might not have done much damage yet, but I fear you will if you continue to do deep discharges.

The cells that heat up during recharging are the same that get reversely charged during the end of the deep discharge rides.

A deep discharge is when you notice the first, slightest drop in performance.

In other words, if it cannot go 100km/h any more or if it accelerates more slowly than usual.

When that happens, stop riding (or continue as gently and as briefly as you can) and recharge. The weakest cell will otherwise get stressed from that point on.

Recharge fully and always try to watch what happens to the temperature etc. during the recharge after a deep discharge.

I have not come across any information stating that the deep discharges need to be done early in the life of a NiMH battery. Nothing at all. And I have spend considerable time researching the subject. You can always do it later.

Unless you flogged the scooter very hard, the battery temp should not have been 38degC in 5degC ambient temp after riding.

The temp definitely should not rise to warning-telltale levels in such cool surroundings during recharging.

The cells which are getting hot might well be bulging and gassing at that time.

Check if it goes into CP/tr/CP/tr/CP/tr/CP/tr/CC/EC.
That is what usually happens after a BaLPoR. It will happen just the same if there was no "R" (for Reset in the Battery Low Point Reset).

Regarding your earlier post:
The voltage a few minutes after the end of a ride is pretty meaningless.

You can discharge these cells at less than 1A to a cut-off level of less than 0.9V, but the open voltage will still bounce back up above 1V very soon.

The lowest string voltage (when you almost had to push) would have been 108V.

108/101= 1.069V for 101 cells, and 0V for the one being charged in reverse.....

That one cell (or several, of course!) will heat up in the process, too, and bulge, and maybe vent gas.

If that happens, it's capacity is being permanently reduced.

And that is the limiting factor for the capacity of your pack.

Any ride continued until you experience a reduced performance level will be a deep discharge for the lowest capacity cells.
If these lowest capacity cells do not benefit from the deep cycling, (which should result in an increased range before the next time you approach the reduced performance level) then they will simply suffer damage instead.

This information may be used entirely at your own risk.

There is always a way if there is no other way!

marsupi
marsupi's picture
Offline
Last seen: 7 years 8 months ago
Joined: Wednesday, October 1, 2008 - 08:06
Points: 51
Re: Battery Performance

A few cycles of the pack is the usual recommendation from the manufacturer to get full capacity.
this is better explained in the quote below from the mpoweruk.com

"Charge conditioning or Formation

Cell formation is the process of transforming the active materials of a new cell into their usable form. The initial crystal structure of the electrolyte or the electrodes is determined by the manufacturing process by which these components were made and the process of coating the electrodes. This may not be the optimum structure for minimising the internal impedance of the cell and it may not give optimum contact between the electrolyte and the electrodes. The passage of current through the cell and the heating and cooling the cell is subjected to will cause small changes in the microstructure or morphology of the active chemicals.

Formation is essentially the first charge carried out at the cell manufacturer's plant under very carefully controlled conditions of current, temperature and duration to create the desired microstructure of the components and the contact between them.

With some chemical formulations it may take ten charge-discharge cycles or more before the battery is able to deliver its full power or capacity."

So after the initial deep charge/discharge cycles it should be done only for conditioning the pack is absolutely needed. As Mik mentioned previously, the life of the pack is directly impacted by depth of discharge; opportunity charging is best.

I found a cheap microOhmeter that might be useful, initially designed for RC enthusiasts.
RIM1000.jpg

07 Vectrix, Vego 600 mod.48V 800W, Arcade Alu E-Cycle 24VSachs Hub

AndY1
AndY1's picture
Offline
Last seen: 8 years 12 months ago
Joined: Tuesday, June 17, 2008 - 04:29
Points: 1071
Re: Battery Performance

Thank you for the explanation.

That was my first deep-discharge. However, there is no explanation in the manual, what the deep-discharge actually is. Only that I have to make 5 deep-discharges. I considered a deep discharge to be, so that the bike barely moves. I also read it here from some users, that they even pushed the bike to completely discharge the battery.

I ended my ride today with 65km done (from yesterday's 70km (different route though)) and with red battery telltale appearing at 2/17 bars (disappearing bars). I was 100 meters from home so I stopped riding immediately. I left the bike resting in the garage at 15'C ambient temperature and will start to charge it 8 hours after done riding.

If I have damaged the battery pack somehow, I will be disappointed at myself.

Wotnopetrol
Offline
Last seen: 8 years 9 months ago
Joined: Friday, August 22, 2008 - 01:40
Points: 207
Re: Battery Performance

Hi Andy,

The battery temp after a good hard conditioning ride in which I got 37 miles, got up to 36c.

My usual charging voltage shows 144 however.

I had the red battery telltale come on when I had 4 bars left but I was climbing a big hill at full throttle. I cut back, the lamp went out and didn't come back on. I presume the under voltage cell causing the warning caught up.

My garage is also at an ambient 15c at the moment and as you know, I have just got my Vectrix too. I have one more conditioning full discharge to do now, so we can compare notes as our machines seem to be more or less at the same stage.

Like a complete geek, I'm keeping a log of vital stats at each charging for now, primarily to see when the battery pack starts to rise or fall in performance. Let me know if you'd like a copy of it.

Simon

AndY1
AndY1's picture
Offline
Last seen: 8 years 12 months ago
Joined: Tuesday, June 17, 2008 - 04:29
Points: 1071
Re: Battery Performance

I drive gently - for the distance - no faster that 70 km/h. The temperature outside was 15'C. When I plugged the Vectrix in after an hour, the battery temperature was 38'C! At the end of EC the temperature went up to 44'C and red battery telltale was lit even though the outside temperature fell to 5'C.

It was 15'C outside today too. I drove 20km from work 6 hours ago. I plugged my Vectrix in a few minutes ago. The battery temperature was 24'C. I checked the front battery fans suck with post-it paper and both were very gently sucking. Is that ok?

Because my battery temperatures are way to high - when riding and charging :-( I'm afraid this high temperature has already damaged my battery.

I can't imagine my battery temperature in the summer :-(

I'm afraid this bike will have to go for warranty repair...

AndY1
AndY1's picture
Offline
Last seen: 8 years 12 months ago
Joined: Tuesday, June 17, 2008 - 04:29
Points: 1071
Re: Battery Performance

This is what I was looking at the End of Charge:
//shrani.si/f/v/9e/4OIOtu4R/dsc00533.jpg)

Outside temperature at that moment was 5'C.

Mik
Mik's picture
Offline
Last seen: 6 years 3 months ago
Joined: Tuesday, December 11, 2007 - 15:27
Points: 3739
Re: Battery Performance

A few cycles of the pack is the usual recommendation from the manufacturer to get full capacity.
this is better explained in the quote below from the mpoweruk.com

"Charge conditioning or Formation

Cell formation is the process of transforming the active materials of a new cell into their usable form. The initial crystal structure of the electrolyte or the electrodes is determined by the manufacturing process by which these components were made and the process of coating the electrodes. This may not be the optimum structure for minimising the internal impedance of the cell and it may not give optimum contact between the electrolyte and the electrodes. The passage of current through the cell and the heating and cooling the cell is subjected to will cause small changes in the microstructure or morphology of the active chemicals.

Formation is essentially the first charge carried out at the cell manufacturer's plant under very carefully controlled conditions of current, temperature and duration to create the desired microstructure of the components and the contact between them.

With some chemical formulations it may take ten charge-discharge cycles or more before the battery is able to deliver its full power or capacity."

So after the initial deep charge/discharge cycles it should be done only for conditioning the pack is absolutely needed. As Mik mentioned previously, the life of the pack is directly impacted by depth of discharge; opportunity charging is best.

I found a cheap microOhmeter that might be useful, initially designed for RC enthusiasts.
RIM1000.jpg

But these 5 deep discharges which we are doggedly inflicting on our scooters are NOT done

under very carefully controlled conditions of current, temperature and duration

They are done in traffic, all sorts of terrain, and most importantly they are done to a string of 102 cells.

The Vectrix can only detect an under-voltage in strings of cells 1-27, 28-75 and 76-102.
So you might argue that it's done to a series of 46 and two series of 27 cells.
But it is still the opposite of "carefully controlled".

Thank you for the picture of the milliohm meter, wonderful find.

Here is a link to the manual, EXCELLENT STUFF! Read it even if you will not buy one!

OPERATING INSTRUCTIONS

This information may be used entirely at your own risk.

There is always a way if there is no other way!

marsupi
marsupi's picture
Offline
Last seen: 7 years 8 months ago
Joined: Wednesday, October 1, 2008 - 08:06
Points: 51
Re: Battery Performance

07 Vectrix, Vego 600 mod.48V 800W, Arcade Alu E-Cycle 24VSachs Hub

AndY1
AndY1's picture
Offline
Last seen: 8 years 12 months ago
Joined: Tuesday, June 17, 2008 - 04:29
Points: 1071
Re: Battery Performance

Thanx for the link!

Mik
Mik's picture
Offline
Last seen: 6 years 3 months ago
Joined: Tuesday, December 11, 2007 - 15:27
Points: 3739
Re: Battery Performance

I think your right about the BMS, it's only that a Battery Monitoring System, from the looks of it there is no balancing circuit. Looks like there is only a PIC18F877 microcontroller and a MCP2515 CAN chip.

And these chips / controller are not even connected to the voltage measuring circuitry.

I think the 4 resistors and 2 capacitors (in the top half of the PCB shown above) simply reduce the voltage which is being measured. I guess the capacitors dampen voltage fluctuations, but I am not sure about it.

Do you understand how this circuit works?

The circuitry in the lower half of the PCB is connected only to the temperature sensors.

I think I'll have to take better pictures to make it easier to see what is on there, sorry about the poor picture quality!

This information may be used entirely at your own risk.

There is always a way if there is no other way!

Mik
Mik's picture
Offline
Last seen: 6 years 3 months ago
Joined: Tuesday, December 11, 2007 - 15:27
Points: 3739
Re: Battery Performance
So far I believe the Vectrix BMS does a lot less than what is theoretically covered under the patent, but it might actually be shunting energy between 2 batteries of 27 cells each and one battery of 48 cells.

I think your right about the BMS, it's only that a Battery Monitoring System, from the looks of it there is no balancing circuit. Looks like there is only a PIC18F877 microcontroller and a MCP2515 CAN chip.

I had a closer look, it is actually a MCP2515 chip, a MCP2551, and a PIC18F877 (NOT PIC16F677).

Here some more notes I took about the markings on the parts on the front and rear Temp sensor PCB:

PIC16F677 -1/SO (e3) 0712208 -------- (next to temp sensor cables, 10 pins x 2)

MCP2515 I/O(e3) 07145C2 --------- (9 pins x 2)

R100BTB71 --------- (Silver oval object)

MCP2551I SN(Symbolincircle)0710 (Symbol)1Y9 ------- (4 pins x 2)

Do you know how to program these chips - or how to best go about learning it?

http://ww1.microchip.com/downloads/en/DeviceDoc/41262E.pdf

//i281.photobucket.com/albums/kk217/Mr_Mik/Vectux/Battery/Battery%20rework%205%20BMS/th_DSC05844.jpg) //i281.photobucket.com/albums/kk217/Mr_Mik/Vectux/Battery/Battery%20rework%205%20BMS/th_DSC05845.jpg) //i281.photobucket.com/albums/kk217/Mr_Mik/Vectux/Battery/Battery%20rework%205%20BMS/th_DSC05846.jpg)
//i281.photobucket.com/albums/kk217/Mr_Mik/Vectux/Battery/Battery%20rework%205%20BMS/th_DSC05847.jpg) //i281.photobucket.com/albums/kk217/Mr_Mik/Vectux/Battery/Battery%20rework%205%20BMS/th_DSC05849.jpg)

This information may be used entirely at your own risk.

There is always a way if there is no other way!

rgx
Offline
Last seen: 13 years 3 days ago
Joined: Tuesday, November 21, 2006 - 13:01
Points: 137
Re: Battery Performance

The PIC circuit is common and you can easily buy a so called experimental kit, including a programmer (to be connected to PC via USB), a few PIC circuits, and examples and documentation. There are lots of documents, examples and freeware on the internet. You will have to keep track of which PIC model you have because there is an endless number of variants, grouped in so called families. Very early in this process you will learn that a PIC circuit can be locked for reading of the code and/or reprogramming.

I don't know about the other circuits but some googling will tell you what they are. They might not be microprocessors, nor programmable. The silver, oval object looks like a crystal - determining the processor speed of the PIC. It is in fact likely that all characteristics you want to change are determined by the PIC program.

The board might have a connector for in-circuit programming and debugging, if you are lucky. You might be able to make a connector (similar to what the experiment board has). Otherwise it's programmed via the CAN-bus, which is much more complicated. The system might not have the reflashing code included in the firmware (for many reasons, one being that you don't want the circuit to be changing its own code by accident if it runs off track). So when you want to update the firmware, you must first load a flash program into RAM, then boot that program, then continue downloading the firmware and it will be flashed permanently into the circuit. After a reset the reflash code disappears from RAM and the circuit will be running the new firmware.

You will of course need to know the protocol and commands to be sent over CAN to do all this. (The CAN interface itself is open and described on can.org.)

There might also be locks and codes in the circuits to detect or prevent alternative software to be run, so even if you only change one byte in the program it might be detected and bike shut down. Not sure if Vectrix have something similar implemented though. For cars it is required for emission-related systems, so called anti-tampering measures.

So how do you reengineer all this? One way is to collect and analyse the CAN communication when a dealer is updating the firmware. But I do suspect that many people who are working with alternative software downloading have had help from inside to figure it out. Example: one car manufacturer found their own code in a tuning company's software. (They made an agreement so the tuner could continue tuning.)

In any case will be a lot of work and I doubt that it's worth it. Some of the steps are also in a grey area in many countries, from a legal point of view. I am not suggesting you do any of the above.

Mik
Mik's picture
Offline
Last seen: 6 years 3 months ago
Joined: Tuesday, December 11, 2007 - 15:27
Points: 3739
Re: Battery Performance

Thanks, rgx! Much appreciated.

Seems like it's time to learn a bit of programming.....

I've seen the USB / PIC experimenter kits, that should be kind of interesting, anyway.

I'll have to try to log the can message traffic for a while somehow, that might help to figure out some settings.

One way is to collect and analyse the CAN communication when a dealer is updating the firmware.

The problem is that at that time the only access to the Canbus is used for the updating.
But I guess one could install another logger somewhere, which could then automatically log all traffic. But by the time the warranty has expired on someones Vectrix they will probably have introduced encryption of some kind.
And the Vectux is not likely going to get connected for an update, I prefer the devil I know!

This information may be used entirely at your own risk.

There is always a way if there is no other way!

HarryS
Offline
Last seen: 10 months 6 days ago
Joined: Wednesday, October 8, 2008 - 20:06
Points: 342
Re: Battery Performance

I had the bike in the garage for the past 2 weeks because of the weather. Just took it for a ride and there was about a 20% loss of charge. I had 16 of the 17 bars when I turned her on but lots one more right out of the driveway. The next 3 bars also went much more quickly then they used to. By 18 miles I had 5 bars left, by 25 miles it was essentially over. I typically get 32 miles comfortably under same riding conditions and can stretch it to about 36miles. So beware of the self discharge. Evene 2 weeks of inactivity will set you back considerably.

moccasin
moccasin's picture
Offline
Last seen: 7 years 9 months ago
Joined: Friday, July 11, 2008 - 19:05
Points: 494
Re: Battery Performance

Mornin' Harry!
Mine will have been down about two weeks by the time I get to ride it again. I think I will charge it the night before and see if it takes a full charge without having been ridden.

kevin smith
kevin smith's picture
Offline
Last seen: 7 years 11 months ago
Joined: Monday, December 29, 2008 - 04:57
Points: 446
Re: Battery Performance

hi simon what speed was you travviling at to get the range that you said 49.3 miles and what type of regening was you doing and was there loads of hills thanks kev

Mik
Mik's picture
Offline
Last seen: 6 years 3 months ago
Joined: Tuesday, December 11, 2007 - 15:27
Points: 3739
Re: Battery Performance

I was wrong with my assumption that the Vectrix does not calculate or otherwise measure self-discharge.

The wrong impression was due to erratic behaviour when the three main fuses failed and when the motor controller board needed repairs twice.

Whenever my first or second Vectrix were functioning (about 50% of the time for the first 4 months) I was riding them almost daily. That was why I bought it for AU$16,000.- ...to use it!
Therefore I never had a chance to observe what it does when unused but not broken.

Until now:

After 3+1/2 months of battery analysis and M-BMS developmwnt I plugged a (manually) fully charged and balanced battery back into the Vectux, running Feb or March 2008 firmware: The display dropped to 5/17th when I turned it on. The battery had been removed also in a fully charged state 3+1/2 months earlier.

The time of day on the clock display was about 10 minutes slow, it is always falling behind a minute each 1-2 weeks.

What it all means is that the Vectrix Stock-BMS (Ver 03-2008) has an internal power backup and calculates self discharge according to time elapsed, at least partially irrespective of the actual SOC.

I also observed a drop from 17 to 16 or 15 bars after 3 or 4 days of inactivity recently. The charging behaviour was then appropriate: Short CP, followed by tr, long CC and 59 min EC.

After the battery work I rode the Vectux for 30km, 20km of it with Zero bars and Zero range announced by the S-BMS. The weakest cells were then empty. After plugging in the S-BMS did what it has always done after a BALPOR: It tries to get a full charge into the battery and over-charges the cells with reduced capacity in the process. Of course I did not let it do this: stopping the charging process (when the weak cells are shown to be full by the M-BMS) leads to re-synchronisation between battery SOC and S-BMS display after 3 or 4 charges.

This information may be used entirely at your own risk.

There is always a way if there is no other way!

Mik
Mik's picture
Offline
Last seen: 6 years 3 months ago
Joined: Tuesday, December 11, 2007 - 15:27
Points: 3739
Re: Battery Performance

I was wrong with my assumption that the Vectrix does not calculate or otherwise measure self-discharge.

The wrong impression was due to erratic behaviour when the three main fuses failed and when the motor controller board needed repairs twice.

Whenever my first or second Vectrix were functioning (about 50% of the time for the first 4 months) I was riding them almost daily. That was why I bought it for AU$16,000.- ...to use it!
Therefore I never had a chance to observe what it does when unused but not broken.

Until now:

After 3+1/2 months of battery analysis and M-BMS developmwnt I plugged a (manually) fully charged and balanced battery back into the Vectux, running Feb or March 2008 firmware: The display dropped to 5/17th when I turned it on. The battery had been removed also in a fully charged state 3+1/2 months earlier.

The time of day on the clock display was about 10 minutes slow, it is always falling behind a minute each 1-2 weeks.

What it all means is that the Vectrix Stock-BMS (Ver 03-2008) has an internal power backup and calculates self discharge according to time elapsed, at least partially irrespective of the actual SOC.

I also observed a drop from 17 to 16 or 15 bars after 3 or 4 days of inactivity recently. The charging behaviour was then appropriate: Short CP, followed by tr, long CC and 59 min EC.

After the battery work I rode the Vectux for 30km, 20km of it with Zero bars and Zero range announced by the S-BMS. The weakest cells were then empty. After plugging in the S-BMS did what it has always done after a BALPOR: It tries to get a full charge into the battery and over-charges the cells with reduced capacity in the process. Of course I did not let it do this: stopping the charging process (when the weak cells are shown to be full by the M-BMS) leads to re-synchronisation between battery SOC and S-BMS display after 3 or 4 charges.

This information may be used entirely at your own risk.

There is always a way if there is no other way!

Pages

Log in or register to post comments


Who's online

There are currently 0 users online.

Who's new

  • tomjones
  • kirkthomas7
  • Minja_Taria9
  • JVale
  • Eom bikes

Support V is for Voltage