Vectrix lithium conversion problem. Faking out the battery voltage detection.

20 posts / 0 new
Last post
procrastination inc
procrastination inc's picture
Offline
Last seen: 10 years 10 months ago
Joined: Tuesday, October 6, 2009 - 00:06
Points: 341
Vectrix lithium conversion problem. Faking out the battery voltage detection.

I though this could use a new thread.

Been playing with an electronics applet here: http://www.falstad.com/circuit/index.html to look at this problem

I thought I might have a play with mik's untested circuit (thanks for trying fella, but It doesn't seem to work.)

below is the code to import miks rendition of the circuit on the sensor board

$ 1 5.0E-6 10.20027730826997 50 5.0 50
w 224 224 528 224 0
w 224 224 224 304 0
w 224 224 224 144 0
r 224 144 288 144 0 2000.0
r 288 144 352 144 0 200.0
r 352 144 416 144 0 430.0
r 416 144 496 144 0 2000.0
r 416 320 496 320 0 2000.0
r 352 320 416 320 0 430.0
r 288 320 352 320 0 200.0
r 224 320 288 320 0 2000.0
w 224 304 224 320 0
w 224 224 48 224 0
c 128 224 128 96 0 9.999999999999999E-6 17.10658013536529
c 128 224 128 368 0 9.999999999999999E-6 -17.104580135530824
w 128 96 352 96 0
w 352 96 352 144 0
w 128 368 352 368 0
w 352 368 352 320 0
w 32 368 128 368 0
w 48 96 128 96 0
R 496 320 608 320 0 0 40.0 112.5 0.0 0.0 0.5
g 528 224 624 224 0
w 496 144 528 144 0
w 528 144 544 144 0
R 544 144 624 144 0 0 40.0 40.5 0.0 0.0 0.5

just cut the data above, choose file in the applet and paste in the import dialog box

this is with input voltages for 27 cells at 1.5V and 48 cells at 1.5v

Assuming bike does high resistance measurements of this, output for V1SCL ranges from about 12.8V at 1V per cell to 19.2V at 1.5V per cell above GND
Output for V2SCLD ranges from 35.6V to 53.4V

Now, here is the same circuit with the transistors down the bottom simulating the mid pack inputs.

$ 1 5.0E-6 10.20027730826997 50 5.0 50
w 272 176 576 176 0
w 272 176 272 256 0
w 272 176 272 96 0
r 272 96 336 96 0 2000.0
r 336 96 400 96 0 200.0
r 400 96 464 96 0 430.0
r 464 96 544 96 0 2000.0
r 464 272 544 272 0 2000.0
r 400 272 464 272 0 430.0
r 336 272 400 272 0 200.0
r 272 272 336 272 0 2000.0
w 272 256 272 272 0
w 272 176 96 176 0
c 176 176 176 48 0 9.999999999999999E-6 0.0010222518332160683
c 176 176 176 320 0 9.999999999999999E-6 9.777476750246156E-4
w 176 48 400 48 0
w 400 48 400 96 0
w 176 320 400 320 0
w 400 320 400 272 0
w 80 320 176 320 0
w 96 48 176 48 0
w 544 96 576 96 0
w 576 96 592 96 0
r 272 384 336 384 0 27000.0
r 336 384 416 384 0 48000.0
R 272 384 272 416 0 0 40.0 153.0 0.0 0.0 0.5
g 496 384 496 400 0
r 416 384 496 384 0 27000.0

right click in the applet and add wires to hook up the 112.5 V, 40.5 and GND points.

Maybe, if the bike does read the voltage outputs with very high resistance, we could delete the whole resistor set on the card and just tap a string of resistors to supply GND, 19V to V1SCL and 54V to V2SCLD ?

7circle
7circle's picture
Offline
Last seen: 13 years 9 months ago
Joined: Friday, January 4, 2008 - 09:11
Points: 66
Re: Vectrix lithium conversion problem. Faking out the ...

The first simultaion I think has a error
the 430 shoul dbe 430K

I did the calcs on the other post but thought you guys would have had it.

Interseting Simulation package , thanks

I'm not sure what you expecting to over come.

I would keep the wirein gas similar to the original.

Are you trying to work out how to monitor the Lithium back?

Battery Cct Monoitor from other post.

I woul d keep both strings.
Also the measureing (AD Micro) may have another resistor to 0V(GND) like a 10k or 100K or whatever.. if you fiddle with these string values (well not the top larger ones) then the reading will be out.

So both strings from GND according the circuit (please check with ohm meter)
Pics on ping pong table

oV --> 2KΩ 200Ω T 430KΩ + 2KΩ bat sense point so Factor is bottom Res / (Top + Bottom Res) = 2200/(2200+432000) = 2200/434200 = 5.0667894979e-3 = apprx 1/200 = 0.005 or 0.5%

So if the V1SCL is supposed to pick up the voltage of 3 x 8 1.2V NiMH then 24*1.2 * 5.0667894979e-3 = V1SCL to 0V if all the Batteries are at 1.2V in that base block.

The Resistor network is the same for the other Sense point escept its is at (24 + 24 + 24) x 1.2 NiMH * 5.0667894979e-3 = V2SCLD to 0V = 86.4 x 5.0667894979e-3
V2SCLD = 0.437V
and
V1SCL = 0.145V

They are small signals but the Measuring smarts may be only setup for 1.023 V max value. If its a 10 bit ADC it has a scale from 0 to 1023 (2^10 -1)
A full a battery sense reading of 1.023 / 5.0667894979e-3 = 201V this kinda makes sense.

Well I hope that helps its a bit random but I hope clustered.

Love your work, Did you see Soyachips's Vespa! He might give ya a run for ya lectrons.

7circle
7circle's picture
Offline
Last seen: 13 years 9 months ago
Joined: Friday, January 4, 2008 - 09:11
Points: 66
Re: Vectrix lithium conversion problem. Faking out the ...

some fun with the Simulator,

Hear is my update of the basic circuit.
Not sure what Caps should be maybe 1u would make more sense.
But as there is no ac in there there just fluf at the moment.
http://www.falstad.com/circuit/e-index.html then File - Import into java aplet that was Auto launched.
>>
$ 5 4.9999999999999996E-6 0.46723710703047594 68 5.0 50
w 272 176 272 256 0
w 272 176 272 96 0
r 272 96 336 96 0 2000.0
r 336 96 400 96 0 200.0
r 400 96 464 96 0 430000.0
r 464 96 544 96 0 2000.0
r 464 272 544 272 0 2000.0
r 400 272 464 272 0 430000.0
r 336 272 400 272 0 200.0
r 272 272 336 272 0 2000.0
w 272 256 272 272 0
c 176 176 176 48 0 1.0E-7 -0.4377706126209094
c 176 176 176 320 0 1.0E-7 -0.02533394748963594
w 176 48 400 48 0
w 400 48 400 96 0
w 176 320 400 320 0
w 400 320 400 272 0
w 96 48 176 48 0
w 544 96 576 96 0
w 576 96 592 96 0
v 608 384 608 96 0 0 1000.0 86.4 0.0 0.0 0.5
w 592 96 608 96 0
w 608 384 544 384 0
w 176 176 272 176 0
v 544 384 544 304 0 0 40.0 5.0 0.0 0.0 0.5
w 176 176 144 176 0
w 544 304 544 272 0
w 144 176 96 176 0
w 176 320 96 320 0
O 96 48 80 48 1
O 96 320 80 320 1
w 544 384 96 384 0
w 80 176 96 176 0
w 80 176 32 176 0
w 32 176 32 384 0
w 32 384 96 384 0
<<<

procrastination inc
procrastination inc's picture
Offline
Last seen: 10 years 10 months ago
Joined: Tuesday, October 6, 2009 - 00:06
Points: 341
Re: Vectrix lithium conversion problem. Faking out the ...

I would like the factory voltage monitoring to still be connected to the battery. But changing cell chemistry means the voltage steps in the lithium battery aren't divisable in the same way that the NiMH cells are. So I am worried that the bike will see an imbalance and alarm. I haven't actually tried it though

Thanks for playing with the java applet, it's fun :)

I think your circuit has some problems. your 2 terminal DC sources are opposed, the stock battery isn't like that.

It would be really useful if someone with a stock vectrix could measure the voltage between VS1CL and battery -ve and between VS2CLD and battery -ve (as suggested in the other thread)

procrastination inc
procrastination inc's picture
Offline
Last seen: 10 years 10 months ago
Joined: Tuesday, October 6, 2009 - 00:06
Points: 341
Re: Vectrix lithium conversion problem. Faking out the ...
procrastination inc
procrastination inc's picture
Offline
Last seen: 10 years 10 months ago
Joined: Tuesday, October 6, 2009 - 00:06
Points: 341
Re: Vectrix lithium conversion problem. Faking out the ...

redone with resistor values that I think are right and the battery set up like the stock pack
$ 1 5.0E-6 10.20027730826997 50 5.0 50
w 208 224 512 224 0
w 208 224 208 304 0
w 208 224 208 144 0
r 208 144 272 144 0 10000.0
r 272 144 336 144 0 200.0
r 336 144 400 144 0 430000.0
r 400 144 480 144 0 10000.0
r 400 320 480 320 0 10000.0
r 336 320 400 320 0 430000.0
r 272 320 336 320 0 200.0
r 208 320 272 320 0 10000.0
w 208 304 208 320 0
w 208 224 32 224 0
w 112 96 336 96 0
w 336 96 336 144 0
w 112 368 336 368 0
w 336 368 336 320 0
w 16 368 112 368 0
w 32 96 112 96 0
w 480 144 512 144 0
w 512 144 528 144 0
c 112 96 112 224 0 1.0E-11 0.917592181252777
c 112 224 112 368 0 1.0E-11 -2.5488671701466137
w 112 224 208 224 0
v 608 112 608 176 0 0 40.0 36.0 0.0 0.0 0.5
v 656 176 656 112 0 0 40.0 36.0 0.0 0.0 0.5
v 608 224 608 288 0 0 40.0 40.5 0.0 0.0 0.5
v 656 288 656 224 0 0 40.0 40.5 0.0 0.0 0.5
w 656 112 608 112 0
w 656 224 656 176 0
w 608 224 608 176 0
g 656 288 656 352 0
w 512 224 656 288 0
w 480 320 608 176 0
w 528 144 656 224 0
r 608 288 656 288 0 1.0

procrastination inc
procrastination inc's picture
Offline
Last seen: 10 years 10 months ago
Joined: Tuesday, October 6, 2009 - 00:06
Points: 341
Re: Vectrix lithium conversion problem. Faking out the ...

and with the tap points connected to the +ve of the 11th lithium cell from negative. the 31st cell and battery -ve. cell voltage 3.65

$ 1 5.0E-6 10.20027730826997 50 5.0 50
w 208 224 512 224 0
w 208 224 208 304 0
w 208 224 208 144 0
r 208 144 272 144 0 10000.0
r 272 144 336 144 0 200.0
r 336 144 400 144 0 430000.0
r 400 144 480 144 0 10000.0
r 400 320 480 320 0 10000.0
r 336 320 400 320 0 430000.0
r 272 320 336 320 0 200.0
r 208 320 272 320 0 10000.0
w 208 304 208 320 0
w 208 224 32 224 0
w 112 96 336 96 0
w 336 96 336 144 0
w 112 368 336 368 0
w 336 368 336 320 0
w 16 368 112 368 0 begin_of_the_skype_highlighting              16 368 112 368 0 16 368 112 368 0 16 368 112 368 0
w 32 96 112 96 0
w 480 144 512 144 0
w 512 144 528 144 0
c 112 96 112 224 0 1.0E-11 0.9096623722789877
c 112 224 112 368 0 1.0E-11 -2.5635939582407943
w 112 224 208 224 0
v 608 112 608 176 0 0 40.0 36.5 0.0 0.0 0.5
v 656 176 656 112 0 0 40.0 36.5 0.0 0.0 0.5
v 608 224 608 288 0 0 40.0 40.15 0.0 0.0 0.5
v 656 288 656 224 0 0 40.0 40.15 0.0 0.0 0.5
w 656 112 608 112 0
w 656 224 656 176 0
w 608 224 608 176 0
g 656 288 656 352 0
w 512 224 656 288 0
w 480 320 608 176 0
w 528 144 656 224 0

VS1CL is less than 1% different on the lithium pack and VS2CLD is less than 0.5%. If this is correct, then this problem might be trivial :)

Mik
Mik's picture
Offline
Last seen: 8 years 8 months ago
Joined: Tuesday, December 11, 2007 - 15:27
Points: 3739
Re: Vectrix lithium conversion problem. Faking out the ...

...
...
...

VS1CL is less than 1% different on the lithium pack and VS2CLD is less than 0.5%. If this is correct, then this problem might be trivial :)

Unfortunately I don't understand a "word" of all this Java applet gobbledigook....

But I think that it may be true that the data from the voltage divider is not used for anything, anyway, as X-Vectrix wrote.

Remember, I have the 13 worst cells in the Vectux pack arranged into the 27-cell pack at the positive end of the battery - and I never got any warning signals, even when several cells were dropping to zero and there was a marked imbalance that should have run a lot of alarm bells - if the stock BMS was actually watching!

You can probably hook them up to roughly equivalent points on the Lithium battery.

Anyone wanting to measure voltages on the little board: Be very careful, you could cause a major disaster and get seriously hurt! The voltage on the input side of the little board is high enough to kill and there are no fuses.

This information may be used entirely at your own risk.

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

procrastination inc
procrastination inc's picture
Offline
Last seen: 10 years 10 months ago
Joined: Tuesday, October 6, 2009 - 00:06
Points: 341
Re: Vectrix lithium conversion problem. Faking out the ...

sorry for the gobbledygook. Try the applet, it is a really nice tool, quite intuitive.

7circle
7circle's picture
Offline
Last seen: 13 years 9 months ago
Joined: Friday, January 4, 2008 - 09:11
Points: 66
Re: Vectrix lithium conversion problem. Faking out the ...

To show those what the Gobbledygeek produces with the simulator here is an image of the Circuit

//www.falstad.com/circuit/index.html

And here is the GBG that can remake it (it's a bit long as a changed it with extra detail and wiring)
$ 1 5.0E-6 10.20027730826997 64 5.0 50
w 240 240 240 320 0
w 240 240 240 160 0
r 240 160 304 160 0 10000.0
r 304 160 368 160 0 200.0
r 368 160 432 160 0 430000.0
r 432 160 512 160 0 10000.0
r 432 336 512 336 0 10000.0
r 368 336 432 336 0 430000.0
r 304 336 368 336 0 200.0
r 240 336 304 336 0 10000.0
w 240 320 240 336 0
w 144 112 368 112 0
w 368 112 368 160 0
w 144 384 368 384 0
w 368 384 368 336 0
w 512 160 544 160 0
w 544 160 560 160 0
c 144 112 144 240 0 1.0E-11 0.9006924270452441
c 144 240 144 384 0 1.0E-11 -2.5383150216725623
w 144 240 240 240 0
v 640 128 640 192 0 0 40.0 36.5 0.0 0.0 0.5
v 688 192 688 128 0 0 40.0 36.5 0.0 0.0 0.5
v 640 240 640 304 0 0 40.0 40.15 0.0 0.0 0.5
v 688 304 688 240 0 0 40.0 40.15 0.0 0.0 0.5
w 688 128 640 128 0
w 640 192 656 192 0
w 656 240 640 240 0
w 688 192 672 192 0
w 672 240 688 240 0
x 63 94 101 97 0 12 V1SCL
x 63 417 110 420 0 12 V2SCLD
x 66 230 118 233 0 12 GND (0V)
x 530 151 558 157 0 24 V1
x 530 231 583 237 0 24 GND
x 527 326 555 332 0 24 V2
w 512 336 560 336 0
w 560 160 608 160 0
w 608 160 608 208 0
w 608 208 672 208 0
w 656 224 608 224 0
w 608 224 608 336 0
w 608 336 560 336 0
w 688 320 592 320 0
w 592 320 592 240 0
w 592 240 544 240 0
w 672 192 672 208 0
w 672 208 672 240 0
w 656 240 656 224 0
w 656 224 656 192 0
w 688 304 688 320 0
w 688 320 688 352 0
g 704 352 704 368 0
w 704 352 688 352 0
w 144 112 112 112 0
w 144 240 112 240 0
w 144 384 112 384 0
w 240 240 512 240 0
w 512 240 544 240 0
r 320 464 384 464 0 1000.0
O 112 112 64 112 1
O 112 384 64 384 1
O 672 208 736 208 1
O 656 224 736 224 1
r 320 496 384 496 0 1000.0
r 432 464 496 464 0 1000000.0
r 432 496 496 496 0 1000000.0
w 496 464 496 496 0
w 384 464 416 464 0
w 416 464 432 464 0
w 432 496 400 496 0
w 400 496 384 496 0
w 640 304 640 464 0
w 688 352 656 352 0
w 656 352 656 464 0
w 656 464 656 496 0
w 656 496 496 496 0
w 400 496 400 400 0
w 416 432 416 464 0
O 640 464 592 464 1
w 112 112 112 144 0
w 112 144 16 144 0
w 16 144 16 496 0
w 16 496 320 496 0
w 320 464 32 464 0
w 32 464 32 352 0
w 32 352 112 352 0
w 112 352 112 384 0
w 400 400 528 400 0
w 416 432 528 432 0
O 528 400 592 400 1
O 528 432 592 432 1
c 448 528 496 528 0 1.0E-5 2.535779242429467
c 448 560 496 560 0 1.0E-5 0.8997926344106114
w 416 464 416 528 0
w 416 528 448 528 0
w 448 560 416 560 0
w 400 496 400 560 0
w 400 560 416 560 0
w 496 496 496 528 0
w 496 528 496 560 0
x 708 187 775 191 0 16 Was 28.2
x 709 264 776 268 0 16 Was 86.4
x 226 24 638 30 0 24 NiMh: (24+24+24+24) x 1.2V= 115.20V
x 225 57 604 63 0 24 LiFe: (11+10+10+11) x 3.65=153.50
w 512 336 512 288 0
w 432 336 432 288 0
w 432 160 432 192 0
w 512 160 512 192 0
x 27 139 97 142 0 12 NiMh:667mV
x 41 376 108 379 0 12 NiMH: 1.91V

7circle
7circle's picture
Offline
Last seen: 13 years 9 months ago
Joined: Friday, January 4, 2008 - 09:11
Points: 66
Re: Vectrix lithium conversion problem. Faking out the ...

My Res-Divider values were from wrong values:
The embedded cirtuit schematic didn't show up...
It's on Mik's photo bucket site.

The SM Res was upside down making a 10K (1002) looks like a 200 (2001)
You can see the difference between the two PCBs one with the Short solder blob the other with out it.
http://kastenente.heimat.eu/Vectrix/BMSboardxl.jpg
http://kastenente.heimat.eu/Vectrix/without_Vmon.jpg

"http://s281.photobucket.com/albums/kk217/Mr_Mik/Vectux/S-BMS/?action=view&current=VectrixBatteryandMC41andVoltageD-1.jpg
"Battery Cct Monoitor from other post."

Maybe Mik would want to double check this and update his picture of the schematic.

Old result

...
oV --> 2KΩ 200Ω T 430KΩ + 2KΩ bat sense point so Factor is bottom Res / (Top + Bottom Res) = 2200/(2200+432000) = 2200/434200 = 5.0667894979e-3 = apprx 1/200 = 0.005 or 0.5%

So if the V1SCL is supposed to pick up the voltage of 3 x 8 1.2V NiMH then 24*1.2 * 5.0667894979e-3 = V1SCL to 0V if all the Batteries are at 1.2V in that base block.

The Resistor network is the same for the other Sense point escept its is at (24 + 24 + 24) x 1.2 NiMH * 5.0667894979e-3 = V2SCLD to 0V = 86.4 x 5.0667894979e-3
V2SCLD = 0.437V
and
V1SCL = 0.145V

They are small signals but the Measuring smarts may be only setup for 1.023 V max value. If its a 10 bit ADC it has a scale from 0 to 1023 (2^10 -1)
A full a battery sense reading of 1.023 / 5.0667894979e-3 = 201V this kinda makes sense.
...

New ixed Result
So fixing values:
oV --> 10KΩ 200Ω T 430KΩ + 2KΩ bat sense point, so Factor is bottom Res / (Top + Bottom Res) = (10K+200)/(10K+200+430K+2K) = 10200/442200 = 5.0667894979e-3 = apprx 1/4.2 = 0.2361

So if the V1SCL is supposed to pick up the voltage of 3 x 8 1.2V NiMH then 24*1.2 * 0.2361111111 = V1SCL to 0V if all the Batteries are at 1.2V in that base block.

The Resistor network is the same for the other Sense point escept its is at (24 + 24 + 24) x 1.2 NiMH * 5.0667894979e-3 = V2SCLD to 0V = 86.4 x 0.2361111111
V2SCLD = 86.4V * 0.0230664858 = 1.99V
and
V1SCL = 28.8V * 0.0230664858 = 0.664V

This lines up with the simulation. I hope this helps and maybe why you went to the simulator as the old res values were wrongly drescribed.
For an overcharged Full NiMH back at 1.8V per cell x4 x24 = 172.8V gives a scaled value of 3.98V

V1SCL: 0.667V NiMH
V2SCLD:1.910V NiMH

No Change to Cicruit but with Lithium set as LiFe: (11+10+10+11) x 3.65= 153.50V
V1SCL: 0.901V LiFe
V2SCLD:2.540V LiFe

This is quite different to NiMH levels.
I think the LithiumIron per cell voltage of 3.65V is a bit high compared to NiMH at 1.2V.
So using 3.2V per cell full pack is 134.4V compard to 1.2 x 96 = 115.2V

To make them the same as NiMH might not be whats best.

They need to be at 1/4 and 3/4 of pack voltage.
So when the BMU looks for a fault the pack will look ballanced (even though its actually pick up at 11/42 and 31/42.

Now you have a LiFe pack at 134.4V not 115.2V

So for 134.4V
1/4 = 33.6V
3/4 = 100.8V
4/4 = 134.4V

Best to change Top resistors in divider. so 430K + 2K can be changed or add off pcb in series.

I did try the bottoms for matching NiMh values with two across the bottom 10200ohm string.
V1SCL: 47K + 4.7k for 662mV
V2SCLD:47K + 10K for 1.91V

But using the resistor in series off board toward the Battery side:
33.6 needs : 86K for 662mV
100.8 needs: 75K for 1.91V

Here is GBG for simulator http://www.falstad.com/circuit/index.html:
$ 1 5.0E-6 0.2954511527092107 72 5.0 50
r 240 160 304 160 0 10000.0
r 304 160 368 160 0 200.0
r 368 160 432 160 0 430000.0
r 432 160 512 160 0 10000.0
r 432 336 512 336 0 10000.0
r 368 336 432 336 0 430000.0
r 304 336 368 336 0 200.0
r 240 336 304 336 0 10000.0
w 144 112 368 112 0
w 368 112 368 160 0
w 144 384 368 384 0
w 368 384 368 336 0
c 144 112 144 240 0 1.0000000000000001E-7 0.6629738274056652
c 144 240 144 384 0 1.0000000000000001E-7 -1.907520572407698
w 144 240 240 240 0
v 640 128 640 192 0 0 40.0 32.0 0.0 0.0 0.5
v 688 192 688 128 0 0 40.0 32.0 0.0 0.0 0.5
v 640 240 640 304 0 0 40.0 35.2 0.0 0.0 0.5
v 688 304 688 240 0 0 40.0 35.2 0.0 0.0 0.5
w 688 128 640 128 0
w 640 192 656 192 0
w 656 240 640 240 0
w 688 192 672 192 0
w 672 240 688 240 0
x 63 94 101 97 0 12 V1SCL
x 63 417 110 420 0 12 V2SCLD
x 66 230 118 233 0 12 GND (0V)
x 530 151 558 157 0 24 V1
x 530 231 583 237 0 24 GND
x 527 326 555 332 0 24 V2
w 560 160 608 160 0
w 608 160 608 208 0
w 608 208 672 208 0
w 656 224 608 224 0
w 608 224 608 336 0
w 608 336 560 336 0
w 688 320 592 320 0
w 592 320 592 240 0
w 592 240 544 240 0
w 672 192 672 208 0
w 672 208 672 240 0
w 656 240 656 224 0
w 656 224 656 192 0
w 688 304 688 320 0
w 688 320 688 352 0
g 704 352 704 368 0
w 704 352 688 352 0
w 144 112 112 112 0
w 144 240 112 240 0
w 144 384 112 384 0
w 240 240 512 240 0
w 512 240 544 240 0
r 320 464 384 464 0 1000.0
O 112 112 64 112 1
O 112 384 64 384 1
O 672 208 736 208 1
O 656 224 736 224 1
r 320 496 384 496 0 1000.0
r 432 464 496 464 0 1000000.0
r 432 496 496 496 0 1000000.0
w 496 464 496 496 0
w 384 464 416 464 0
w 416 464 432 464 0
w 432 496 400 496 0
w 400 496 384 496 0
w 640 304 640 464 0
w 688 352 656 352 0
w 656 352 656 464 0
w 656 464 656 496 0
w 656 496 496 496 0
w 400 496 400 400 0
w 416 432 416 464 0
O 640 464 592 464 1
w 112 112 112 144 0
w 112 144 16 144 0
w 16 144 16 496 0
w 16 496 320 496 0
w 320 464 32 464 0
w 32 464 32 352 0
w 32 352 112 352 0
w 112 352 112 384 0
w 400 400 528 400 0
w 416 432 528 432 0
O 528 400 592 400 1
O 528 432 592 432 1
w 416 464 416 528 0
w 400 496 400 560 0
w 400 560 416 560 0
w 496 496 496 528 0
w 496 528 496 560 0
x 708 187 775 191 0 16 Was 28.2
x 709 264 776 268 0 16 Was 86.4
x 226 24 638 30 0 24 NiMh: (24+24+24+24) x 1.2V= 115.20V
x 225 57 585 63 0 24 LiFe: (11+10+10+11) x 3.2 =134.8
x 27 139 97 142 0 12 NiMh:667mV
x 41 376 108 379 0 12 NiMH: 1.91V
r 304 192 368 192 0 4700.0
r 304 192 240 192 0 47000.0
w 240 160 240 192 0
w 240 192 240 240 0
w 240 336 240 304 0
w 240 304 240 240 0
r 240 304 304 304 0 47000.0
r 304 304 368 304 0 10000.0
r 752 496 752 464 0 100.0
174 688 464 720 464 0 100.0 0.5792 Resistance
w 720 464 752 464 0
w 688 464 640 464 0
w 704 448 688 448 0
w 688 448 688 464 0
s 688 496 720 496 0 1 false
w 752 496 720 496 0
w 656 496 688 496 0
x 300 85 502 91 0 24 (35.2+32+32+35.2)
w 512 160 512 192 0
w 560 160 560 192 0
w 560 336 560 304 0
w 512 336 512 304 0
r 512 304 560 304 0 75000.0
r 512 192 560 192 0 86000.0
c 496 528 432 528 0 1.0E-11 -1.905614957450248
c 496 560 432 560 0 1.0E-11 -0.6623115158897757
w 432 560 416 560 0
w 432 528 416 528 0
x 345 599 577 605 0 24 Measure Cct example

7circle
7circle's picture
Offline
Last seen: 13 years 9 months ago
Joined: Friday, January 4, 2008 - 09:11
Points: 66
Re: Vectrix lithium conversion problem. Faking out the ...

Hang on

Few issues:
Why are you using 24+24+24+24 for 96cells when the
link below shows 27+24+24+27 for 102 celss
(link: front battery of 48 cells and the rear battery of 54 cells)

xxxxx

So
If you have fitted 11+11 in Back and 10+10 in front with Sky Energy SE040AHA x 3.2V is 5376Wh @ 134.4V.
Compared to Standard Vectrix with 27+27 in Back and 24+24 in front with standard GP NiMH 30Ah cells x 1.2 is 3672Wh 122.4V

Also
Nominal NiMH Voltage ?? 1.2V
Nominal LiFePO4 Voltage ?? 3.2V

Charged NiMH Voltage ?? 1.45V
Charged LiFePO4 Voltage ?? 3.65V

Max NiMH Voltage ?? 1.8V
Max LiFePO4 Voltage ?? 4.0V

What do you recommend to use?

And that makes the simulations are out.
So resistors need fix up from last post.

mikemitbike
mikemitbike's picture
Offline
Last seen: 11 years 5 months ago
Joined: Saturday, September 20, 2008 - 13:07
Points: 310
Re: Vectrix lithium conversion problem. Faking out the ...

Hi 7circle,
the java-aplet is great, I searched for a programm to check small circuits and this applet seems to work
well. In my case the link ditn´t work as shown above, so I used Circuit simulator (Java Aplet)

Nominal NiMH Voltage ?? 1.2V -> 125v pack voltage, so nominal should be about 1,225 v
Nominal LiFePO4 Voltage ?? 3.2V -> some manufactuers say 3,3 some 3,2v

Charging cutoff voltage:
Charged NiMH Voltage ?? 1.45V -> the vectrix cells are designed for about 1,5 v cuttoff for charging
with 1C and about 1,47 v with 0,5C
Charged LiFePO4 Voltage ?? 3.65V -> should be the correct voltage

Max NiMH Voltage ?? 1.8V -> should not exceed 1,52 v per cell at 1C Charge and 1,48 at 0,5C (a spare cell
reached 1,47 v when it was full, falling down to 1,46 when I continued to charge
and started to heat up in this phase.
Max LiFePO4 Voltage ?? 4.0V -> some say must not exceed 3,8 v per cell

Greetings Mike

7circle
7circle's picture
Offline
Last seen: 13 years 9 months ago
Joined: Friday, January 4, 2008 - 09:11
Points: 66
Re: Vectrix lithium conversion problem. Faking out the ...

(Edit - more resitor mistakes!!!
Please confirm that both resistor dividers are R10 and R13 were 10K

R14 + R8 T R9 + R10
10K + 200 T 430K + 2K

R15 + R11 T R12 + R13
10K + 200 T 430K + 2K

Bloody checks
I Updated all bits below)

Looking at the expected error now considereing the setup for NiMH

V1SCL should be at % of pack 27/102 = 26.47%
V2SCLD should be at % of pack 75/102 = 73.52%

1.2V NiMH 27+24+24+27
740.0 mV
2.060 V

3.2V LiFe 11+10+10+11
803.94mV % error to NiMH 8.640%
2.270V % error to NiMH 10.39% 10.194%

The LiFe full pack is 3.2V LiFe 11+10+10+11 = 134.4V
With no changes to R divider
So % of full pack: 11/42 : 26.19%
& % of full pack: 31/42 : 73.80%

Probably not worth fixing as that is very close to NiMH
27/102 = 26.47%
75/102 = 73.52%

But Resistors are shown on simulation. Use either V1&V2 or bottom not both if you want a very close approx to Original Measurement.

And heres a update to the Simulation if it helps.
$ 1 5.0E-6 1.7725424121461644 72 5.0 50
r 256 160 320 160 0 10000.0
r 320 160 384 160 0 200.0
r 384 160 448 160 0 430000.0
r 448 160 528 160 0 2000.0
r 448 336 528 336 0 2000.0
r 384 336 448 336 0 430000.0
r 320 336 384 336 0 200.0
r 256 336 320 336 0 10000.0
w 160 112 384 112 0
w 384 112 384 160 0
w 160 384 384 384 0
w 384 384 384 336 0
c 160 112 160 240 0 1.0000000000000001E-7 0.7404130495813
c 160 240 160 384 0 1.0000000000000001E-7 -2.0564649785307165
w 160 240 256 240 0
v 656 128 656 192 0 0 40.0 32.0 0.0 0.0 0.5
v 704 192 704 128 0 0 40.0 32.0 0.0 0.0 0.5
v 656 240 656 304 0 0 40.0 35.2 0.0 0.0 0.5
v 704 304 704 240 0 0 40.0 35.2 0.0 0.0 0.5
w 704 128 656 128 0
w 656 192 672 192 0
w 672 240 656 240 0
w 704 192 688 192 0
w 688 240 704 240 0
x 79 94 117 97 0 12 V1SCL
x 79 417 126 420 0 12 V2SCLD
x 82 230 134 233 0 12 GND (0V)
x 546 151 574 157 0 24 V1
x 546 231 599 237 0 24 GND
x 543 326 571 332 0 24 V2
w 576 160 624 160 0
w 624 160 624 208 0
w 624 208 688 208 0
w 672 224 624 224 0
w 624 224 624 336 0
w 624 336 576 336 0
w 704 320 608 320 0
w 608 320 608 240 0
w 608 240 560 240 0
w 688 192 688 208 0
w 688 208 688 240 0
w 672 240 672 224 0
w 672 224 672 192 0
w 704 304 704 320 0
w 704 320 704 352 0
g 720 352 720 368 0
w 720 352 704 352 0
w 160 112 128 112 0
w 160 240 128 240 0
w 160 384 128 384 0
w 256 240 528 240 0
w 528 240 560 240 0
r 336 464 400 464 0 1000.0
O 128 112 80 112 1
O 128 384 80 384 1
O 688 208 752 208 1
O 672 224 752 224 1
r 336 496 400 496 0 1000.0
r 448 464 512 464 0 1000000.0
r 448 496 512 496 0 1000000.0
w 512 464 512 496 0
w 432 464 448 464 0
w 448 496 416 496 0
w 416 496 400 496 0
w 656 304 656 464 0
w 704 352 672 352 0
w 672 352 672 464 0
w 672 464 672 496 0
w 672 496 512 496 0
w 416 496 416 400 0
w 432 432 432 464 0
O 656 464 608 464 1
w 128 112 128 144 0
w 128 144 32 144 0
w 32 144 32 496 0
w 32 496 336 496 0
w 336 464 48 464 0
w 48 464 48 352 0
w 48 352 128 352 0
w 128 352 128 384 0
w 416 400 544 400 0
w 432 432 544 432 0
O 544 400 608 400 1
O 544 432 608 432 1
w 432 464 432 528 0
w 416 496 416 560 0
w 416 560 432 560 0
w 512 496 512 528 0
w 512 528 512 560 0
x 242 24 671 28 0 16 NiMh: (27+24+24+27) x 1.2V= 32.4+28.8+28.8+32.4= 122.4V
x 241 57 614 61 0 16 LiFe: (11+10+10+11) x 3.2 =35.2+32+32+35.2=134.8
x 43 139 123 142 0 12 NiMh:740.0mV
x 57 376 131 379 0 12 NiMH: 2.060V
r 320 192 384 192 0 15000.0
r 320 192 256 192 0 100000.0
w 256 160 256 192 0
w 256 192 256 240 0
w 256 336 256 304 0
w 256 304 256 240 0
r 256 304 320 304 0 82000.0
r 320 304 384 304 0 15000.0
r 768 496 768 464 0 100.0
174 704 464 736 464 0 100.0 0.5396000000000001 Resistance
w 736 464 768 464 0
w 704 464 656 464 0
w 720 448 704 448 0
w 704 448 704 464 0
s 704 496 736 496 0 1 false
w 768 496 736 496 0
w 672 496 704 496 0
r 528 304 576 304 0 42000.0
r 528 192 576 192 0 39000.0
c 512 528 448 528 0 1.0E-11 -2.054410567962754
c 512 560 448 560 0 1.0E-11 -0.7396733762050951
w 448 560 432 560 0
w 448 528 432 528 0
x 361 599 593 605 0 24 Measure Cct example
x 186 545 238 551 0 24 hello
w 576 160 576 192 0
w 576 304 576 336 0
O 704 128 752 128 1
w 400 464 432 464 0
w 528 160 576 160 0
w 528 336 576 336 0
w 384 336 384 304 0
w 384 192 384 160 0

procrastination inc
procrastination inc's picture
Offline
Last seen: 10 years 10 months ago
Joined: Tuesday, October 6, 2009 - 00:06
Points: 341
Re: Vectrix lithium conversion problem. Faking out the ...

"The LiFe full pack is 3.2V LiFe 11+10+10+11 = 134.4V
With no changes to R divider
So % of full pack: 11/42 : 26.19%
& % of full pack: 31/42 : 73.80%

Probably not worth fixing as that is very close to NiMH
27/102 = 26.47%
75/102 = 73.52%"

that's what I'm thinking.

Mik
Mik's picture
Offline
Last seen: 8 years 8 months ago
Joined: Tuesday, December 11, 2007 - 15:27
Points: 3739
Re: Vectrix lithium conversion problem. Faking out the ...

Back to basics, please:

Are these schematics of the stock-rear-battery voltage divider correct or not - and if not, what needs to be corrected?

(Click to enlarge)

//i281.photobucket.com/albums/kk217/Mr_Mik/Vectux/S-BMS/th_VectrixBatteryandMC41andVoltageDivi.jpg) //i281.photobucket.com/albums/kk217/Mr_Mik/Vectux/S-BMS/th_VectrixBatteryandStockVoltagedivide.jpg) //i281.photobucket.com/albums/kk217/Mr_Mik/Vectux/S-BMS/th_DSC05897.jpg) //i281.photobucket.com/albums/kk217/Mr_Mik/Vectux/S-BMS/th_DSC05896.jpg)

I think there is indeed an error: R 14 and R15 are labelled "1002" (not "2001") - so they are 10000ohm instead of 2000ohm.

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: 8 years 8 months ago
Joined: Tuesday, December 11, 2007 - 15:27
Points: 3739
Re: Vectrix lithium conversion problem. Faking out the ...

How about this one:

//i281.photobucket.com/albums/kk217/Mr_Mik/Vectux/S-BMS/th_VectrixBattery31withvolagedivider.jpg)

Photobucket

This information may be used entirely at your own risk.

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

procrastination inc
procrastination inc's picture
Offline
Last seen: 10 years 10 months ago
Joined: Tuesday, October 6, 2009 - 00:06
Points: 341
Re: Vectrix lithium conversion problem. Faking out the ...

i think reading the codes on the resistors is the problem.

I get 10k, 200, 430k, 10k... but I can't be sure, I can't read mine and rely on the pics posted.

Measuring the voltage on a live circuit is the only way we'll know for sure

marsupi
marsupi's picture
Offline
Last seen: 10 years 1 month ago
Joined: Wednesday, October 1, 2008 - 08:06
Points: 51
Re: Vectrix lithium conversion problem. Faking out the ...

Howdy,
great find this little applet
i agree with you guys, the voltage taps can be left alone, the 11/42 and 31/42 split is plenty close to the original voltage taps.
the only question, but I seems to remember reading this somewhere in this posting or other, do the voltage taps simply light up the battery and temp light or does it limit limit the performance. anyone can confirm?

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

AndY1
AndY1's picture
Offline
Last seen: 11 years 4 months ago
Joined: Tuesday, June 17, 2008 - 04:29
Points: 1071
Re: Vectrix lithium conversion problem. Faking out the ...

If I remember X Vectrix correctly, he said, that you can just delete the mid pack voltage monitoring wires.

Log in or register to post comments

Buy Ecotric bikes, get free accessories!


Who's online

There are currently 0 users online.

Who's new

  • eric01
  • Norberto
  • sarim
  • Edd
  • OlaOst

Support V is for Voltage