Home grown BMS ideas !

How do you store and manage your electricity?
Post Reply
User avatar
Johny
Senior Member
Posts: 3729
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Home grown BMS ideas !

Post by Johny » Fri, 03 Apr 2009, 19:28

OK. I just reread weber's PWM pulse idea and I now understand it. Low voltage comparators are cheap, not sure about Op amps.

BUT by the time you have added a timing component (say 555) and a quad comparator, maybe the Tiny25 that Nevilleh is using would be simpler.
Since it could do 4 cells, it sounds way cheaper and 4 times simpler (or more) to make (since there is only 1 for every 4 cells AND it has less components).
Once you get there, why not a serial bus using opto couplers (as per Nevilleh) - and away we go again...sorry Image

User avatar
weber
Site Admin
Posts: 2554
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Home grown BMS ideas !

Post by weber » Fri, 03 Apr 2009, 19:38

coulomb wrote:An LMC555 works down to 1.5 V, so that's a likely contender.
Yes! That's what's been missing from this thing. Everyone knows every DIY circuit must have at least one 555 timer in it, and preferably a 556 dual. And it must use those timers in the most obscure manner possible. Image

Actually, I had this web page up and was brushing up when your message arrived. http://home.cogeco.ca/~rpaisley4/LM555.html Image

Image
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

Nevilleh
Senior Member
Posts: 773
Joined: Thu, 15 Jan 2009, 18:09
Real Name: Neville Harlick
Location: Tauranga NZ

Home grown BMS ideas !

Post by Nevilleh » Mon, 06 Apr 2009, 20:04

Tritium_James wrote: I can almost guarantee you a capacitive coupling scheme for the comms won't work -


Oh well, it was just thought anyway. You've saved me the trouble of even fiddling with it. And, as you say, optos are pretty cheap.

User avatar
weber
Site Admin
Posts: 2554
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Home grown BMS ideas !

Post by weber » Sat, 11 Apr 2009, 07:02

Nevilleh wrote: Lots of interesting ideas here!
But I am going with a simple clamp to stop any cell going too high and using fixed resistors in my original circuit sets the activation volyage at 3.93.
In addition, I use an Atmel ATTiny25 to monitor the voltage of each cell. The Tiny 25 has a 4 channel, 10 bit ADC, so one micro does 4 cells. It also has a built in temperature sensor and I plan to use that as the pcb will be attached to the top of one cell in each block of 4.
The Tiny25 is about $2.50 - 3.00 depending on where you get it from.
Using it's SPI capability lets me run a single line, multi-drop comms setup, at the moment using optocouplers ($1.00 or so each)...


Hi Nevilleh,

I'm getting interested in a microcontroller-per-cell arrangement. I thought the ATtiny SPI was only for programming. I didn't think you could use it under software control. Do you mean the USI?

My thought was to avoid having to give them IDs at all, by daisy chaining them in a ring with the master. So the master prompts the first cell which sends its voltage and temp data to the second cell which passes on the first cell's data and adds its own on the end, until you get to the end where the master gets all the data. Each cell is effectively ID'ed by its position in the daisy-chain. The master might have several UARTs and run several rings to avoid any one ring being too long.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

User avatar
Electrocycle
Senior Member
Posts: 985
Joined: Sun, 19 Oct 2008, 20:23
Real Name: Andrew
Location: Sydney
MSN: dumhed@dumhed.com
Contact:

Home grown BMS ideas !

Post by Electrocycle » Sat, 11 Apr 2009, 16:13

yeah you'd have to watch out for the chinese whispers effect there :)
The Engine Whisperer - fixer of things

User avatar
coulomb
Site Admin
Posts: 3611
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Home grown BMS ideas !

Post by coulomb » Sat, 11 Apr 2009, 16:24

Nevilleh wrote: I tried to upload the circuit diagram, but the image is too big. I'll have to find another way.

You can zip it and upload it as a file; find the File Manager in the Member Control Panel. Then just let us know it's there, and/or post a link to the uploaded file.

Edit: files have a limit of 1 MB; images a limit of 100 kB. So you could upload a 1.5 MB image if it zip compressed to 1 MB.
Last edited by coulomb on Sat, 11 Apr 2009, 16:57, edited 1 time in total.
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, PIP-4048MS inverter, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.

User avatar
woody
Senior Member
Posts: 1711
Joined: Sat, 21 Jun 2008, 02:03
Real Name: Anthony Wood
Location: Mt Colah

Home grown BMS ideas !

Post by woody » Wed, 15 Apr 2009, 20:40

Something which came up on the ThunderSky list: if you can join multiple cells monitors onto the same heatsink, then you get a better deal, since mostly one cell will be in bypass at a time, when more than a few cells come onto bypass, the charger should be ramping down anyway.

e.g. Tuarn's BMS board for 20 cells should have 1 or 2 big heatsinks, not 20.

And maybe a small fan.

Another possibility I thought of was a clamp/bypass for the whole cell group, which maybe higher capacity - suitable for high regen - sort of a per module braking resistor perhaps.

discuss/flame/etc. :-)

cheers,
Woody
Planned EV: '63 Cortina using AC and LiFePO4 Battery Pack

User avatar
acmotor
Senior Member
Posts: 3595
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Home grown BMS ideas !

Post by acmotor » Thu, 16 Apr 2009, 07:17

In a way, that is the plan.
The VFD's 750V braking resistor shunt point starts(lots of kW) at 750/220 = 3.4 V/cell up to 850/220 =3.86V/cell so the single large braking resistor takes over once the average cell voltage is 3.4V (90% charged) and depending on the brake resistor chosen and amount of regen, can go up to 3.86V/cell average (98% charge) before controller backs off regen.
The individual cells are still bypassed by 1.5A to 3.5A (depending on resistor) for eq. along the way. This goes into heat. Remember my BMS is designed for regen.

There are no heatsinks on my BMS boards, just ceramic resistors.

I think there would be a lot more wiring for the module based resistor/heatsink ?
If I went that way I would have one big PCB with two wires running to each cell (quite doable). But then those wires should be fused at the battery terminals.
iMiEV MY12     105,131km in pure Electric and loving it !

Nevilleh
Senior Member
Posts: 773
Joined: Thu, 15 Jan 2009, 18:09
Real Name: Neville Harlick
Location: Tauranga NZ

Home grown BMS ideas !

Post by Nevilleh » Thu, 16 Apr 2009, 14:11

weber wrote:
Hi Nevilleh,

I'm getting interested in a microcontroller-per-cell arrangement. I thought the ATtiny SPI was only for programming. I didn't think you could use it under software control. Do you mean the USI?

My thought was to avoid having to give them IDs at all, by daisy chaining them in a ring with the master. So the master prompts the first cell which sends its voltage and temp data to the second cell which passes on the first cell's data and adds its own on the end, until you get to the end where the master gets all the data. Each cell is effectively ID'ed by its position in the daisy-chain. The master might have several UARTs and run several rings to avoid any one ring being too long.


Yes, it's proper name is USI (Universal Serial Interface)and it is configured as an SPI (Serial Peripheral Interface) for programming. What's in a name, I know what I mean! (Sorry if that caused confusion).
Atmel have an App Note on configuring it for asynch comms which is fairly straightforward. It has an 8-bit data register so receiving is OK as the start bit just drops off the end. Tx'ing requires it to shift out 4 bits, then reload the register with the rest of the byte so that it can send the start bit (9 bits in all).
Your "ring" idea dates back to the 'sixties and it is simple and easy, but it gets hard if the master wants to send something to a particular device. Also, the message gets pretty long if you have lots of nodes!
I think each micro having a unique address is more flexible and easier to implement. My solution requires no extra hardware no matter how many micros are on the line. Since I use only a 1-byte address, that is limited to 256 nodes, but you could easily use 2 bytes and have 64K nodes!
I've just finished designing the pcb which mounts on top of a LFP40 and I'm about to try making one using a technique I haven't seen before. That is printing the artwork on a bit of transparency film using a laser printer. Then you "iron" the toner pattern onto the copper using your ordinary household iron. Put a sheet of paper between the iron and the film so it wont melt. The artwork is thus transferred to the copper and you just peel the film off. Dunk it in etch solution and voila! a completed pcb.
Apparently the laser toner is made of tiny plastic particles which provide an effective etch resist. Ink jet ink doesn't work because it is just ink.
Anyway, I hope to have my first one up and running shortly.

Nevilleh
Senior Member
Posts: 773
Joined: Thu, 15 Jan 2009, 18:09
Real Name: Neville Harlick
Location: Tauranga NZ

Home grown BMS ideas !

Post by Nevilleh » Thu, 16 Apr 2009, 14:19

coulomb wrote:
You can zip it and upload it as a file; find the File Manager in the Member Control Panel. Then just let us know it's there, and/or post a link to the uploaded file.

Edit: files have a limit of 1 MB; images a limit of 100 kB. So you could upload a 1.5 MB image if it zip compressed to 1 MB.


Thanks for the tip.
I don't know if anyone is interested, but I'm happy to upload the schematics for my clamp and micro battery monitor and the pcb layouts as well.
They are quite small files done with ExpressSCH and ExpressPCB.


User avatar
coulomb
Site Admin
Posts: 3611
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Home grown BMS ideas !

Post by coulomb » Thu, 16 Apr 2009, 14:54

Weber and I would be quite interested, as I'm sure others would be too.

It will be interesting to see how your ironing technique goes; I assume this is just for prototyping.

Also, I'm interested in how micros (especially the serial ports) behave in an electrically hostile environment. Do you have supply rail bypass capacitors on the board, or just rely on proximity to the battery?
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, PIP-4048MS inverter, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.

Nevilleh
Senior Member
Posts: 773
Joined: Thu, 15 Jan 2009, 18:09
Real Name: Neville Harlick
Location: Tauranga NZ

Home grown BMS ideas !

Post by Nevilleh » Thu, 16 Apr 2009, 17:06

coulomb wrote: Weber and I would be quite interested, as I'm sure others would be too.

It will be interesting to see how your ironing technique goes; I assume this is just for prototyping.

Also, I'm interested in how micros (especially the serial ports) behave in an electrically hostile environment. Do you have supply rail bypass capacitors on the board, or just rely on proximity to the battery?

OK, I'll put the files into a single .zip and see if I can manage to do an upload!
Yes, this pcb technique is just for prototyping and/or small quantities.
I have actually made a board since my last post, that's how quick and easy it is.
I was told to use quite a lot of pressure when ironing, but I think I used too much as the toner on my board blank spread a bit. Also, there are a lot of pinholes, but it worked OK and I regard it as successful for my first try. It'll do to verify everything. I checked all the traces for shorts and opens and it is fine. (see below for finished board but not drilled.)
The circuit uses a 5v regulator with a .33 uF bypass on the input and a .1 uF on the output. This provides pretty good isolation from "crap" on the supply lines. The comms, being async at a fairly low data rate are quite robust and opto-coupling makes them just about bullet-proof. The CRC byte provides sufficient protection against corruption.
I designed a system for connecting poker machines together back in '97 - 2000 and used similar techniques there, although opto-coupled RS485.
The system has been installed in most of the Clubs and Pubs in NSW ("NetCash") and it is very stable in what is really quite a harsh environment. This used the same addressing technique that I described earlier, although the micros were Atmel 8051 derivatives rather than their new(er) RISC chips such as the Tiny25. But running the internal "watch-dog" keeps everything under control.

Image

User avatar
coulomb
Site Admin
Posts: 3611
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Home grown BMS ideas !

Post by coulomb » Fri, 17 Apr 2009, 01:47

Nevilleh wrote: OK, I'll put the files into a single .zip and see if I can manage to do an upload!
Thanks.
The circuit uses a 5v regulator with ...
Err, 5v? From a 3.2V cell? I assume you mean an LM337 or similar adjustable regulator.

The board looks nice, and simple too. Uncluttered. Is it single sided?
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, PIP-4048MS inverter, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.

Nevilleh
Senior Member
Posts: 773
Joined: Thu, 15 Jan 2009, 18:09
Real Name: Neville Harlick
Location: Tauranga NZ

Home grown BMS ideas !

Post by Nevilleh » Fri, 17 Apr 2009, 18:38

coulomb wrote:
Nevilleh wrote: OK, I'll put the files into a single .zip and see if I can manage to do an upload!
Thanks.
The circuit uses a 5v regulator with ...
Err, 5v? From a 3.2V cell? I assume you mean an LM337 or similar adjustable regulator.

The board looks nice, and simple too. Uncluttered. Is it single sided?


No, remember that this thing monitors four cells in series. It derives its supply from the total ie 4 x 3.2v. Although it is bolted to one cell there is a 6 way connector which supplies wire connections to the other 3 cells, plus 2 wires for the comms (1 unused pin).
The board is single-sided but has 2 wire links.
Here's a picture of the assembled pcb:

Image

I've been thinking about the accuracy of the voltage readings. A 10 bit ADC implies that the 2.5v range has a resolution of (+- 1 bit) 2.44 mv, so there's no point in trying for better accuracy. I suppose +- 10mv overall would be quite good enough, especially as I am limiting the charge voltage to about 3.8v per cell and the low end to about 2.7v.
Here's a .jpg of the circuit.

Image
Last edited by Nevilleh on Fri, 17 Apr 2009, 08:47, edited 1 time in total.

Nevilleh
Senior Member
Posts: 773
Joined: Thu, 15 Jan 2009, 18:09
Real Name: Neville Harlick
Location: Tauranga NZ

Home grown BMS ideas !

Post by Nevilleh » Fri, 17 Apr 2009, 19:06

Well, I zipped the Express schematic and pcb files and uploaded the result as BMS.rar - but I don't know where it went or how to find it!

User avatar
woody
Senior Member
Posts: 1711
Joined: Sat, 21 Jun 2008, 02:03
Real Name: Anthony Wood
Location: Mt Colah

Home grown BMS ideas !

Post by woody » Fri, 17 Apr 2009, 19:26

Nevilleh wrote: Well, I zipped the Express schematic and pcb files and uploaded the result as BMS.rar - but I don't know where it went or how to find it!
Thanks for the upload!

It went here :-)

generally: uploads/&lt number>/<filename>

You can find all your uploads in your Member Control Panel (link at top of each page, 2 lines under the logo), in the files section :-)
Planned EV: '63 Cortina using AC and LiFePO4 Battery Pack

Nevilleh
Senior Member
Posts: 773
Joined: Thu, 15 Jan 2009, 18:09
Real Name: Neville Harlick
Location: Tauranga NZ

Home grown BMS ideas !

Post by Nevilleh » Fri, 17 Apr 2009, 21:29

I've done some calculations on the potential errors in my BMS circuit using the nearest preferred value 1% resistors and the worst case is 101mv on cell 3. A bit much I feel. This comes about because the voltage divider resistor of 22K should actually be 22.6K. Probably I should trim these values by making them up from 2 in series ie 22.6K would be 22K + 600 ohms in series.
Just a thought - I'll explore it further, but if anyone wants to make one of these that is something to be careful of.

User avatar
weber
Site Admin
Posts: 2554
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Home grown BMS ideas !

Post by weber » Fri, 17 Apr 2009, 22:43

How about changing the 10k to a 12k, and the 22k to a 27k.

Here's a sorted list of all the ratios (with factors of 10 ignored) available with the E12 series plus "16". You get much better resolution from the E12 series by adding the single value "16" from the E24 series.

So for example you want a 2.26 ratio. You look down the left hand column of each pair below and find the closest, which is 2.25 (4th from the top in the middle pair of columns) and you see, immediately to the right, that it is obtained by 27 and 12. You then adjust by whatever factors of 10 you need.

I wish I knew how to make columns line up, with this web interface. [Edit: Used "pre" BBCode.]

Resistor ratios
1.067     16/15          2.200     22/10          4.556     82/18
1.125     18/16          2.200     33/15          4.615    180/39
1.182     39/33          2.206    150/68          4.667     56/12
1.191     56/47          2.250     27/12          4.681    220/47
1.200     12/10          2.353    160/68          4.700     47/10
1.200     18/15          2.438     39/16          4.756    390/82
1.205     47/39          2.485     82/33          4.821    270/56
1.206     82/68          2.519     68/27          4.848    160/33
1.214     68/56          2.545     56/22          4.853    330/68
1.220    100/82          2.553    120/47          5.125     82/16
1.222     22/18          2.564    100/39          5.455    120/22
1.222     33/27          2.600     39/15          5.455    180/33
1.227     27/22          2.611     47/18          5.467     82/15
1.250     15/12          2.647    180/68          5.556    100/18
1.333     16/12          2.679    150/56          5.556    150/27
1.375     22/16          2.683    220/82          5.600     56/10
1.424     47/33          2.700     27/10          5.641    220/39
1.436     56/39          2.750     33/12          5.667     68/12
1.444     39/27          2.857    160/56          5.732    470/82
1.447     68/47          2.938     47/16          5.735    390/68
1.463    120/82          3.030    100/33          5.745    270/47
1.464     82/56          3.037     82/27          5.893    330/56
1.467     22/15          3.077    120/39          5.926    160/27
1.471    100/68          3.091     68/22          6.250    100/16
1.500     15/10          3.111     56/18          6.667    100/15
1.500     18/12          3.133     47/15          6.667    120/18
1.500     27/18          3.191    150/47          6.667    180/27
1.500     33/22          3.214    180/56          6.667    220/33
1.600     16/10          3.235    220/68          6.800     68/10
1.688     27/16          3.250     39/12          6.818    150/22
1.697     56/33          3.293    270/82          6.829    560/82
1.741     47/27          3.300     33/10          6.833     82/12
1.744     68/39          3.404    160/47          6.912    470/68
1.745     82/47          3.500     56/16          6.923    270/39
1.765    120/68          3.636    120/33          6.964    390/56
1.773     39/22          3.704    100/27          7.021    330/47
1.786    100/56          3.727     82/22          7.273    160/22
1.800     18/10          3.733     56/15          7.500    120/16
1.800     27/15          3.778     68/18          8.000    120/15
1.829    150/82          3.830    180/47          8.148    220/27
1.833     22/12          3.846    150/39          8.182    180/22
1.833     33/18          3.900     39/10          8.182    270/33
1.951    160/82          3.917     47/12          8.200     82/10
2.061     68/33          3.929    220/56          8.235    560/68
2.063     33/16          3.971    270/68          8.293    680/82
2.074     56/27          4.024    330/82          8.298    390/47
2.103     82/39          4.103    160/39          8.333    100/12
2.128    100/47          4.250     68/16          8.333    150/18
2.136     47/22          4.444    120/27          8.393    470/56
2.143    120/56          4.533     68/15          8.462    330/39
2.167     39/18          4.545    100/22          8.889    160/18
2.195    180/82          4.545    150/33          9.375    150/16
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

Tritium_James
Senior Member
Posts: 683
Joined: Wed, 04 Mar 2009, 17:15
Real Name: James Kennedy
Contact:

Home grown BMS ideas !

Post by Tritium_James » Fri, 17 Apr 2009, 22:51

Ah, why not just pick from a higher series count, eg MRS25 0.6W 1% from Farnell, then you can just choose your 22.6K directly...

And even with the 'wrong' value in there, can't you just calibrate it out in software, it's a known error after all.

Nevilleh
Senior Member
Posts: 773
Joined: Thu, 15 Jan 2009, 18:09
Real Name: Neville Harlick
Location: Tauranga NZ

Home grown BMS ideas !

Post by Nevilleh » Sat, 18 Apr 2009, 13:15

Thanks for that guys, those ratios are a great help. Be even more useful to do voltage divider ratios!
And of course the s/w can compensate for known errors.
We are a bit limited here in having only what Jaycar stock. Farnell are OK for mail orders, but it goes against my Scrooge mentality to fork out $20 for postage for $5 worth of resistors!
Last edited by Nevilleh on Sat, 18 Apr 2009, 03:25, edited 1 time in total.

Tritium_James
Senior Member
Posts: 683
Joined: Wed, 04 Mar 2009, 17:15
Real Name: James Kennedy
Contact:

Home grown BMS ideas !

Post by Tritium_James » Sat, 18 Apr 2009, 14:58

OK then, Jaycar it is! I was wondering why you weren't using surface mount bits, it makes life a lot easier with homebuilt boards, as you don't need to drill all those holes.

Nevilleh
Senior Member
Posts: 773
Joined: Thu, 15 Jan 2009, 18:09
Real Name: Neville Harlick
Location: Tauranga NZ

Home grown BMS ideas !

Post by Nevilleh » Sat, 18 Apr 2009, 15:50

Tritium_James wrote: OK then, Jaycar it is! I was wondering why you weren't using surface mount bits, it makes life a lot easier with homebuilt boards, as you don't need to drill all those holes.


You are quite right and I wish I could get surface mount bits more readily. But I have a small, high speed drill press so the holes aren't too bad for small boards.
I plugged all the resistor values from the E24 range into a spreadsheet and re-worked the voltage dividers to get the values shown below.
The worst error is now only 2.6 mv which is near enough the 10-bit ADC resolution, so that will do just fine.

Using 1% resistors gives an potential error of +- 40-odd mv, so maybe there's a case for going to .2% if you can get them.

Image

User avatar
weber
Site Admin
Posts: 2554
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Home grown BMS ideas !

Post by weber » Sat, 18 Apr 2009, 17:37

To convert a voltage-divider ratio to a resistor ratio you merely subtract one.

A one-decade table of voltage-divider ratios would be a multi-page table, because there are a large number of possible ratios slightly greater than 1, using resistors from widely differing decades.

In contrast, the one-decade resistor ratio table (for E12 plus "16") is compact enough to stick to the wall above your workbench or resistor storage.

For example, let's say you want a voltage divider ratio of 1.63. We subtract 1 to find we need a resistor ratio of 0.63. We scale that by tens until we get a value between 1 and 9.999... So that's 6.3 which we look up to find that 6.25 is 100/16. Therefore 0.625 is 10/16 and so a 10k above a 16k will give us a 1.625 voltage divider, close enough to our desired 1.63.

The full one-decade E24 resistor ratio table is of course 4 times as big as the E12, but you can probably still fit it on an A4 sheet. You can leave off all the pairs of identical resistors that give a ratio of 1, as I have done above. You might also leave off resistor pairs involving E24-only resistors when an E12 pair give a ratio that is the same (or within say 0.5%).

The idea of just adding one E24 value to the E12 series was in regard to what values to stock for prototyping. All those drawers take up a lot of space. So I spreadsheeted all the ratios available from E12 alone and looked at where the big gaps were, and which E24's were most useful for filling in those gaps. It turns out 16 is the most useful. Adding the entire E24 series does not reduce the size of the largest gap any further than does adding this single value, although the average gap size is of course reduced.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

Nevilleh
Senior Member
Posts: 773
Joined: Thu, 15 Jan 2009, 18:09
Real Name: Neville Harlick
Location: Tauranga NZ

Home grown BMS ideas !

Post by Nevilleh » Sat, 18 Apr 2009, 20:48

weber wrote: To convert a voltage-divider ratio to a resistor ratio you merely subtract one.

lot of space. So I spreadsheeted all the ratios available from E12 alone and looked at where the big gaps were, and which E24's were most useful for filling in those gaps. It turns out 16 is the most useful. Adding the


That is really quite interesting and useful, I'll revise my own parts stock accordingly, thanks.

I have a problem with your 'merely subtract one" though.

My voltage divider ratio will always be less than one as it is a divider!

Take R2 as the top leg and R1 as the bottom leg, then the division ratio, Dr = R1/(R1+R2). A bit of algebra and re-arranging gives me R1/R2 = Dr/(1-Dr). So if my Dr is .63 say, then the resistor ratio is .63/(1-.63)= 1.702. from your table, the nearest ratio is 1.697 or 56/33. Putting R2=33 and R1=56 gives an actual ratio of .6292 which is pretty close.

So what have I missed here?

Perhaps "division factor" would be more exact than "division ratio".
Last edited by Nevilleh on Sat, 18 Apr 2009, 10:52, edited 1 time in total.

User avatar
acmotor
Senior Member
Posts: 3595
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Home grown BMS ideas !

Post by acmotor » Sat, 18 Apr 2009, 22:04

I used a circuit very much like this a while back and one thing I noticed was that the voltage drop across the battery straps / terminals (particularly by the fourth cell) needs to be corrected for when the cells are under high load (3 to 5C).

The error was typically in the 10's of mV so not gross.
This is not too much problem at the BMS master as it can measure the current and apply a correction (perhaps along with any minor corrections of the voltage dividers)
iMiEV MY12     105,131km in pure Electric and loving it !

Post Reply