LCD Instrumentation for Cougar Controller

[Freyguy]

Mora,
I wouldn't waste any more time with that old display. It drew 700mA because something was reverse polarity. It gave up the ghost after a few seconds.
And there's no guarantee that one or more of the I/O pins interfacing to the PIC at the time aren't compromised. If it was me I'd start over with a replacement PIC and correct display.

When I started this project I had a 40char x 4 line LCD in perfect working order I planned to use. I accidently connected it to the power supply with odd and even pins swapped. Vcc & ground became reversed. I had current limit on ps set to 300mA, which it hit immediately and I quickly shut it down.
I corrected and tried it again but it was dead to the world.
Good thing because it pushed me in the right direction- I re-evaluated my LCD display needs and decided on a 16 char large font LCD

-Chris

[mora]

Yes, I've already ordered correct display and I'll get another PIC just in case the current one went bad.

[BadDog98]

ianbartie,
Just wanted you to know my PCBs arrived today in the post. Everything looked to be in great shape too! (Long Trip)


Thanks Again, Mark

[1-ev.com]

ianbartie,

My PCBs arrived today in the post too. Everything looked an OK.

Thank you
-Y

[DJBecker]

Originally Posted by jackbauer

Little update on the bmw fuel gauge. I had some play time today so i hooked up a 500r pot and made some measurements. At 500r the wiper of the pot read 3.75v and the gauge read just over full. Reduced the pot to 3v then 2.5v. Gauge stayed the same. Recycled the ignition switch and at 2.5v gauge read just over half full. Reduved to 1v. Still no movement then recycled ignition and gauge reads about 1/8th.

So it looks like i can "fill" the tank as quickly as i like but it won't "empty" very fast. So it looks as if the stock gauge is useless for soc as its the opposite for an ev. Fast drain and slow fill. I'll be getting rid of the bmv600hs battery monitor soon now that the lcd module can measure charge and will just fit a normal resistive fuel gauge in the pod.

That's the expected behavior for software-driven BMW fuel gauges.

The instrument cluster computer has special code to make a better guess at the actual fuel level. It appears to have been added after BMW had a problem with the senders getting a dead spot (from corrosion?) in the presence of alcohol-containing fuels if you overfilled the tank. The software "knows" that you can't empty the tank instantly, and ignores any quick drop in the sensor reading. When the firmware update was introduced people complained that the gauge remained on 'F' until four gallons or so were used -- a side effect of the software work-around for the hardware problem.

[jackbauer]

Thanks DJ. Confirms what i suspected. Its funny cos at times the gauge seems to work. I had a mad day today and actually rolled the car to a stop.At 23% soc the gauge started working then swung back and forth a few times over the full range. I'm just going to bung a resistor on it to keep it on F and install a normal restive type gauge in a pod.

[ianbartie]

Hi Everyone,
I hope someone can pint me in the right direction because I am at a bit of a loss right now.
My LCD does not want to show any serial message info from the controller.
The display seems to work fine, shows aux V OK, pack V OK, SOC, KTE & motor temp all OK.
The motorA, batteryA & controllerT refuse to display properly. The firmware was set to receive serial mesages (00101100).
RTD Explorer show rtd_period as 0100. Changing it to 0200 & saving it shows as 0200. Close RTD & connect LCD. Display of serial message info shows 0 amps & will not change. Opening RTD Explorer shows rtd_period back at 0100 again. It does not seem to stay saved.
Have made a test cable to go straight to the control board but still no luck.
Have double checked the component placement & all look OK.
Any suggestions please.

[mora]

I'm still waiting for my new LCD to arrive but I guess I'll face the exact same problem. I noticed rtd_period not staying at set value even after saving settings. Could it be some Cougar firmware issue?

[apowers]

Mine did not get the info from controller at first. It was because my firmware was from fall of '09. Adam told me to load the new firmware from his web site and that fixed it. I think the RTD explorer sets the rtd_period at 100 only when RTD explorer is connected. And your saved 200 should be there at other times. I think
Alvin.

[Freyguy]

The firmware handles the received serial messages as follows:
From start of string, character position is counted, with the target fields being stored in temp registers. This data doesn’t get used until the linefeed character is detected. The linefeed must be the 67th character or else the comm error flag is set and the string is ignored & overwritten by the next transmission. This ensures that only the status message data is used.
Format below (from the wiki)

TR=000 CR=000 CF=002 PW=000 HS=0315 RT=0716 FB=00 BA=000 AH=000.0

Other received data includes the firmware version on powerup, as well as the confirmation data sent back when you change a cougar parameter. Since the linefeed is at a different position these strings are ignored.
Framing errors and receive queue overflows also set the comm error flag so any string occurring during these events will also be ignored.

Older versions of Cougar firmware that do not follow the above status message format are incompatible. The new 1000amp controller follows the same format, with a new pack voltage field appended to the end. To be compatible, this only requires a change to one line of code indicating the new position of the line feed character.

-Chris