Old BMS Software Thread

[GregsGarage]

I've been needing to do a few adjustments to my master software so this was the perfect opportunity. V29 looks good so far, programmed my master with it just now. I also adjusted the software to work with my 50a allegro sensor. I like the charger on/off just by pressing the button down.

[GregsGarage]

Suggestion for Master V30, how about moving the cell voltage menu to the now unused slot on the first menu. I find I use that quite a lot.

[retepsnikrep]

OK

[retepsnikrep]

Master V30

This has the voltage display moved up one level to the vacant menu1 slot.

Also added a slight delay after charger on/off toggle, if your finger lingered on the button it could turn off again straight away.

[GregsGarage]

I'll give it a try at the weekend. Another thing I was going to suggest was to swap the charge toggle and fault bit display functions. What I am finding is that when you are scrolling through the cell voltages, it is quite easy to toggle the charge bit if you leave your finger on the button too long. So charge would be up and last fault would be down.

[hohisilver]

Now it looks like I'm finally going to be charging more regularly I want to see about adding remote display support to my C software, but I'm puzzled by this line in the PBP code Peter?

serout DriveInhibit, N2400, [85, 85, 85, 85, 85] 'Txd 5 byte preamble to clear Rxd

[retepsnikrep]

The TX/RX pair I used had some demo picaxe code as far as I remember which recommends sending a few dummy bytes to synchronise the TX/RX and clear any garbage out of the buffer. It seemed to work well. I suggest you keep it in.

There are some newer options for the actual TX/RX gizmos now. i have just bought these after good reports on the insight website.

http://rover.ebay.com/rover/1/711-53200 ... MEWAX%3AIT

http://www.insightcentral.net/forums/mo ... post191793

More data on them is spread over several pages of the thread.

[hohisilver]

They do look good, trouble is I already bought the components for the remote display while I was ordering the bits for the main BMS

Another thing I am going to do to my software is inhibit the controller until the reference current reading has been read - I took the scooter out for a run last night and I must have pulled away before it had initialised properly, and all my current readings were skewed by about 15A!

[hohisilver]

Just a quick update:

Remote display built, boxed and working nicely. So nice not to have to keep running out to the garage to check on charging progress

However, due to the seriously bloaty nature of my C software (I could probably have made it much smaller at the expense of its interrupt-driven nature) I am now up to 97.8% of program space in the master

So I've been looking around and the PIC18F2550 looks like almost a drop-in replacement with 16k words of program space and 2k bytes of RAM, so I'm thinking of modifying my master board (again!) to reassign some of the pins. RA7 is no longer (now a dedicated clock input) so I'm also thinking of sticking in a 20MHz ceramic resonator I bought while I'm at it.

I've recompiled my C software for this chip and it reports that it now takes up 41% of program space Of course I've no way of knowing if the software will actually run at this point. The other interesting feature of the PIC18 is the on-board USB so, space permitting (and quite a steep learning curve methinks ) I'm going to see if I can implement logging to a USB memory stick

[retepsnikrep]

Sounds good Ian some pics please of your nicely boxed remote and a look at your C source please.

You can log direct to your house pc into a spreasheet with a serial usb/led from the remote don't forget.

I have looked at logging to sd card etc there are a few routines in pbpro so it is possible i just never got round to it.