Old BMS General Thread

[retepsnikrep]

Ah, but we are slowing the loop down to around one loop per second at 31khz, so I think the hysteresis is a bit of overkill. It can only switch between on and off at the rate of the program loop and no faster. So even if it triggers for only 1 loop that is still around a second.

OK have a go by all means.

[GregsGarage]

Peter,

Had another idea about reducing the power consumption of the analogue slave boards. Instead of getting the program to run very slowly at 31khz, could we run it at say 250khz but put a sleep command at the end of the program loop for say approx 750ms? What do you think?

Edit: Just tried my idea on version 1.13 running at 250khz but used the command "nap 5" (approx 500ms) just before the "goto main" on the last line. It did slow the program down, but power consumption was unchanged.

[GregsGarage]

_________Programmed________________________________
____________Value__________Ver_1.08_________Ver_1.13_
MaxVOptoOn___375____________3.78_____________3.80___
MaxVOptoOff___360____________3.64_____________3.64___
LoadOn________360____________3.65_____________3.63___
LoadOff________355____________3.59_____________3.61__
MinVOptoOff____250____________2.52_____________2.54__
MinVOptoOn____240____________2.40_____________2.44___

A little while ago I reported on the accuracy of Ver 1.13 of the analogue on/off software. Thought I would have a go at tweaking the values. Having realized that the slow clock speed was making my measurements difficult, because I was having to adjust the power supply and wait for the board to respond, I commented out the line that underclocks the chip. Taking some new voltage measurements, all values were within 10-20mv of the programmed value. I am actually pushing the accuracy of my meter, so I think it would be pointless to try and improve the accuracy any more.

So my recommendation to anyone building/using the analogue slave boards is to get a variable power supply and check each board for full functionality. Make sure both your optos are conducting when they should and that the load switches on and off when it should. If you want to tweak the values for accuracy comment out the line that underclocks the cpu while changing the values. This will also make it easier to download new values as the picaxe software won't reliably connect to an underclocked CPU. Just don't forget to uncomment the line and download it, so you get the CPU back to the power saving underclocked mode.

[retepsnikrep]

Some good work Greg. I like the improved accuracy.

We musn't get too hung up on power consumption. The digital slaves use a whole ma as they run at 8mhz.

Anthing less than 1ma is fine IMO, 1000hrs or 40 days operating for 1ah used.

When I eventually get round to updating the digital slave software so we can uodate values on the fly via the master it will be bit easier.

[ChrisB]

You guys are amazing, having followed the whole thread through its been really interesting

Re-the Faq, while a FAQ is normally more aimed at the newbie (this has been mentioned) which is a good idea covering the basics and having lists of where stuff came from etc , would a "index" to this thread be better ??

Of course both are fine its up you guys really, just thought I'd throw my two pennies worth in

Peter or anyone, if you want to drop back and create an index type thing mentioning page numbers etc I dont mind creating the direct links in said post, I'd do the index but to be honest I wouldnt know really whats important and whats not

Same with the FAQ although I wouldnt really know where to obtain the links for parts etc

With either if you want them as seperate threads and made sticky just shout

Loving ya work

ChrisB

[retepsnikrep]

I'm installed in an umbrian farmhouse and the bms is on the back burner because the pizza and pasta is on the front one

Thanks for the offer of help Chris, all that will have to wait until after my install is complete though

[ChrisB]

No probs , I'm sure in the future this thread will become very helpful to folks so the thanks really should be aimed at yourself

ChrisB

[retepsnikrep]

I'm back now and hoping my charger will arrive tomorrow

Edit

New 28X1 chips have arrived with A4 firmware so hopefully the conflict between charger control and timer is resolved. I'll post them tomorrow.

Anyone else made any other progress or anything else to report?

Edit Again!

Tested a new 28X1 A4 firmware chip and the charger control function seems ok, BMS cranks up over V charge control opto each time through loop when an over MaxV condition exists. It never turns it back down again so once charger opto is on it stays at that level unless it increases further due to repeated over V condition. (Note it is reset by the initiation of a new charging cycle, so charger starts at max output)

I can fine tune the increment value once I see how it works in practise with my Zivan charger.

There are 1024 possible opto drive levels, I'm using a increment value of 10 initially. So it takes takes approx two minutes to go from 0 to 100% opto drive if an over MaxV condition persists. 100% drive = max opto conduction and Max pull down of charger voltage.

The over ABSMaxV function is only used to control/cut the master charger relay if the BMS loses control of the charging voltage using the above system.

Latest version here, this only works with A4 or higher firmware, ditto future releases

www.solarvan.co.uk/bms/MASTER011208_V83_2400BAUD.TXT

Back indoors now as bitterly cold in garage

[GregsGarage]

Hi Peter,

Got the new chip today, thanks. Hope your holiday went well. I haven't done much with the BMS although I did install a electric heater for the car.

I will have to get back to making a shopping list for the rest of the components required to build the master board. I was working on a database but got fed up. Will have another crack at it though.

Any sign of your charger?

[retepsnikrep]

Charger should arrive today

I've been thinking about controling it further than the simple reduction in voltage as a cell goes over MaxV, and will probably implement some current limiting into the software so that charger can start at a lower output as reqd.

I like driving abroad and the full 13A mains power is not always available especially using those european sockets in some remote B&B, so to avoid melting down the local grid I may add a routine which holds charge current at a set lower level on start up by adjusting charger output voltage on the fly.

So you dial in a charge rate between 0 - 15A (15A is max for 144v Zivan NG3) and BMS adjusts software pwm controlled charger voltage opto, which adjusts charger ouput up/down to achieve that current. Obviously as soon as a cell goes over max cell V it reverts to the gradual reduction as now. Have to see how well that works with Zivan alogorithim and if it upsets it.

I used a 110v 2kw american hair dryer bought for £1.00 of e-bay for my van heater/demister. Worked well.

Edit

Charger has arrived might get a chance to work on it Thursday/Friday