Old BMS Software Thread

[GregsGarage]

Been thinking about the extra memory available on the eeprom for version 2 of the master. How about a setup/configuration routine. When the master is first powered on it loads cell ah capacity and number of cells from eeprom. If they are zero then it jumps to the configuration routine so the system can be setup for the vehicle. This would mean one program for everyone, no need to compile the program for your specific vehicle. The configuration routine should also be accessible for later tweaks, say adjusting ah capacity as the cells age, or setting the unit to display capacity as ah or percent, temperature in C or F, miles or km. I think you get the idea. What do you think?

[retepsnikrep]

Lots of work there Greg!

But an Idea as you say. I do need some sort of menu system.

The Master only uses 1650 bytes or so out of a possible 4000 at present so we have lots of space even now for a configuration routine.

Although the Master software works well now and is reasonably laid out I may well re-write it completely for Master V2.00. There is quite a lot of tidying and optimisation I could do to make it run a bit faster, look neater and be easier to follow/ammend.

I could increase the Master speed to 16mhz for a doubling of performance to start with. In fact I'll probably use a 16mhz resonator on my Master V2 board. This needs a change in software/option to select speed when programming as it effects quite a few commands.

With the 64kb eeprom and data logging there will be a couple of methods of transferring data to a PC.

1) Via the programming lead plugged into the wireless remote unit. 2400baud

2) Via the programming lead plugged into the Master. 9600baud (Upto 19,200 baud at 16mhz.)

[GregsGarage]

The key would be to just do little bits at a time. So just do the Cell count and ah capacity as a basic configuration. Then add other features later.

[retepsnikrep]

I tried the remote link at 4800 baud, as expected that did not work reliably. (The radio modules data sheet does say 4k max to be fair.) So that is limited to 2400 baud. I'm now using the remote display in the house each time I charge the vehicle. No errors yet!

The % Soc seems to work OK as does trip meter after 40 mile test drive yesterday.

Interestingly yesterday I found when using pretty much maximum amount of added electric power over the 40 mile trip in hilly terain that 1% of Soc = 1 mile travelled for 125mpg fuel consumption. Meaning I had enough power for approx 100 miles at that level of drain. My pack has 6400wh (6.4kwh) of available power, I was discharging at 64wh/mile

One small change to Master today to improve screen layout by adding a leading zero to cell number in HiV & LoV display lines.

www.solarvan.co.uk/bms/MASTER_SERIAL_28 ... 00BAUD.txt

Edit 31/08/09

Small change to Master added a leading zero to individual Cell V display and corrected a few spelling errors in my code comments!!

www.solarvan.co.uk/bms/MASTER_SERIAL_28 ... 00BAUD.txt

[retepsnikrep]

Master V1.42

Added current (Amps) to data sent to remote display.

www.solarvan.co.uk/bms/MASTER_SERIAL_31 ... 00BAUD.txt

RemoteDisplay V0.11

Added current (Amps) to display data and PC data logging output.

www.solarvan.co.uk/bms/RemoteDisplay310809_v11.txt

[retepsnikrep]

Master V 1.48 16mhz

Master speed now increased to 16mhz with external resonator. Software versions from now on won't work on 8mhz Masters unless lots of changes are made by the end user.

Added quite a bit over last few days including much improved Soc 0-100% counting to two decimal places, which now updates every second. Thanks to Jeremy and the Picaxe guys for help with the maths/ideas.

Power in kw being used/generated at any one second displayed. (Amps x Pack Voltage) Distance now updates every 1/10th mile and includes wh/mile although that calculation is not quite right yet.

Speedo now a bit more accurate. Almost everything updates now every second. Seems to work well.

www.solarvan.co.uk/bms/MASTER_16mhz_SER ... 00BAUD.txt

[retepsnikrep]

Master 1.51

Wh/Mile and Distance remaining now work

Corrected an error with Alarm logic.

Odometer now updates in whole miles (Max 65535)

Tripmeter updates in 10/ths Mile.

Wh & Distance remaining update after each mile.

www.solarvan.co.uk/bms/MASTER_16mhz_SER ... 00BAUD.txt

Might be going away this weekend so hopefully can give it a bit more of a thorough test.

[retepsnikrep]

Interestingly the change to 16mhz has given me a timing problem with the serial video data which drops characters when the picaxe is cold.

The data should be OK and is when the temp of the picaxe is more than 15C but if less the resonator/picaxe is drifting and upseting the 9600 baud comms. I'll try another resonator later or picaxe or drop back to 8mhz for my particular board.

Software is working well otherwise, went on 114 mile trip today. No Alarms. One of odometers is off by about 2% according to my Insight trip meter, I don't have sat nav to check both/which, but that's not bad.

www.solarvan.co.uk/bms/BMS050909.jpg

This is a big pic file 700kb so you can zoom into the displays. Soc and wh/distance remaing calculations all working and in line with my top of the head estimates. You can see we did 134mpg for the 112/114 mile trip.

The battery wh/mile was 51 and we know my pack is 6400wh, then dividing 6400 / 51 gives 125 miles. You will note on BMS display trip distance is given as 112 travelled with 13 remaining or 125

The Soc shows 10.55% remaining so I used 89.45% for 112miles. Again 112 / 89.45 gives 1.25. 1.25 x 10.55 = 13 miles

I have been trying a version of the software which is quicker and does not use the 128 byte scrachpad ram, this means the number of cells is unlimited, but 168 is a more reasonable maximum giving around a 650V pack. You can't look at the individual cell voltages yet though.

www.solarvan.co.uk/bms/MASTER_16mhz_SER ... 00BAUD.txt

The one below is the normal one allowing cell V viewing. I haven't tried it yet.

www.solarvan.co.uk/bms/MASTER_16mhz_SER ... 00BAUD.txt

[GregsGarage]

Great job Peter. I particularly like the miles remaining info as well. And the remote boards in the hardware thread also look good.

[retepsnikrep]

I just need to do a few measurements now to see what wh/hrs used by the Zivan charger comes out to, then I can add that to the wh/hr/mile routine in the Master to get a true mains to road figure whr/mile figure.

I hope to get the Mega8 Master board pcb layout done this week.

Hopefully I then can order some Master boards mk2/mk3 and the led/buttons boards as well. When I get the new Master boards in stock I will send them out to you both (Greg/Rick)

Another 100 mile run on Sunday gave similar figures to Saturday but a bit heavier on wh/mile as more hills and only managed 130mpg

As I am now getting quite a lot of data, I can see quite clearly on the Master screen when driving and charging that cell 26 has a higher IR than the others It reaches Max voltage first when charging and drops lowest under heavy loads. It doesn't go out of the acceptable range but it's one to watch. I might move one of my temp sensor and put it next to it just to see what it is doing from a temp POV as well. I have got a spare cell so may change it if/when I next have the pack out.

I'm still working on getting data into excel as well need to try and install

http://www.parallax.com/tabid/393/Default.aspx