LCD Instrumentation for Cougar Controller

[jackbauer]

Received my boards on monday. Many thanks to all. The tamura hall sensor seems to be out of stock at digikey (surprise , surprise). Just wondering would it be much work to mod the software to use a lem hass 300? What voltage does the tamura output?

[harlequin2]

The Tamura puts out 2.5V (half the 5v supply) with zero field then it goes up for one field direction and down for the other, so it measures current both ways. Chris biases his op-amp to get rid of the negative current so it measures current one way only - also has a bit of gain so the 0-5V result fits the adc range to give better resolution. I would imagine the Hass product is pretty similar although I haven't looked up the specs.
By the way, nice job on the liquid cooled heatsink across the way, a thing of beauty!

[apowers]

I got the LCD in the car. All I am getting is pack voltage and acc voltage.
Both of those are reading low. By 10 volts on the pack. The instructions tell how to adjust that. The acc battery is off by 1/2 volt but dropped as I drove by another volt.


When I had this on the bench it read the correct voltage for the acc. bat.


I am not getting motor amps , battery amps or motor temp. I did not hook-up battery temp.

I will get back at it to find my mistake(s).

Alvin

[Freyguy]

Originally Posted by jackbauer

Just wondering would it be much work to mod the software to use a lem hass 300? What voltage does the tamura output?

Hi Jackbauer. As harlequin2 mentioned, both Tamura and LEM have a 2.5V quiescent output which the op-amp would null out. The remaining difference would be that of gain. Full range positive output for both devices is 4.375V which is 500A for Tamura and 900A for LEM hass 300. You would need to multiply the bAmp value in software by 1.8. The subroutine "fracmult" would let you get close by doing a multiply of 1.8125 (mult by 29 followed by divide of 16). 500 amps would read as 504 but if that's good enough you could use that.

[Freyguy]

Originally Posted by apowers

I got the LCD in the car. All I am getting is pack voltage and acc voltage.
Both of those are reading low. By 10 volts on the pack. The instructions tell how to adjust that. The acc battery is off by 1/2 volt but dropped as I drove by another volt.


When I had this on the bench it read the correct voltage for the acc. bat.


I am not getting motor amps , battery amps or motor temp. I did not hook-up battery temp.

I will get back at it to find my mistake(s).

Alvin

Alvin,
You may want to use a DMM and verify some voltages when installed in the car. What is the voltage on the 5V regulator? What is the current draw from the LCD controller board? Should be in the 120mA range. The ACC voltage changes could be due to other loads in your car causing a voltage drop in the wiring. It only reports the voltage as it appears at Vin. Again, you may want to put a DMM in parallel. It may very well see the same thing.
Motor and battery amps are from the serial message. Is the Cougar controller set to output the message? (rtd-period=200, then save). Motor temp would be the thermistor so that should work. As a starter you could just put a 10K resistor in it's place right on the terminal block.
Keep us posted or pm me if you wish.

-Chris

[apowers]

I changed the setting for the pack voltage and it now reads within .4 volts as the multimeter. The motor temp works . I had a lead wrong. As for the controller settings . It was sent to me preprogrammed and have never adjusted anything in it . Laptop does not connect. I'll keep working on it.

Thanks
Alvin

[jackbauer]

Have the board up and running on the bench. First , for anyone looking to use a lem hass 300 here is a little routine to add to the code to do the multiplication by 1.8 as described by Freyguy above. Insert it at line 1167 :

;************************************************* ******
;Routine here to multiply battery amps by 1.8125 (multiply by 29 , divide by 16) to use lem hass 300s

movff bAmpH, Arg2H ; setup to use fracmult routine
movff bAmpL, Arg2L ;
movlw .29 ;multiply by 29
call Fracmult ;call multiplication routine
movff RES0, bAmpL ;move new result back into battery amp reg
movff RES1, bAmpH ;move new result back into battery amp reg
;************************************************* ******

Last pic assembler i wrote was for a 16f877 almost a decade ago

I replaced the 2.5k precision resistor in the voltage sensing circuit with a 5k 10 turn trimpot. This allowed me to use a cheap 5% 100k resistor and just tweak the pot to give the correct voltage on screen. Seems to be working well.

Next mod will be to use degrees centigrade for temp but miles for distance. I'm a bit of a hybrid guy! This is a really neat piece of kit. I like it a lot. One thing I might look at doing is to cause a reset to 100% soc on detection of absorbtion voltage for a lead pack or cv phase for lithium.

[Freyguy]

Excellent jackbauer! The culprit for the pack voltage error is the HCNR201. The gain transfer error between photodiodes is +/-5%. To be honest I missed this parameter when designing the board and had actually used the HCNR200 which has gain error of +/-15%! Artwork was already on order so I spec'ed in the HCNR201 and implemented the software trim. This is actually an offset correction, but since the working range of a charged to depleted battery pack is narrow relative to its absolute value, I thought it would be good enough.
A trim for gain is better and changing R9 to a 10K pot as you mentioned would do that. Alternately you could change R4 to a 100K pot. If the pack voltage reports low, it could then be a 50K pot.

You probably figured it out by now, but the mod for displaying temp in C but still report as miles is as follows:

In the "line 1" section of the display routine you only need to comment out
btfss Sys_Config, 3.
It will then skip over the part to replace MTE with KTE.

Glad you find this useful

[harlequin2]

Got mine powered up, although not complete (missing 12-way connector) and it all lights up nicely. Resetting SOC on power up gives me 57% not 100%. Any clues?
aVolts is pretty well spot on and connecting same power supply to the battery pack terminals shows 12V (actual is 13.4) so that's not bad. Unless the error is a %age!
Making R4 or R9 a trimpot is not a bad idea.

[harlequin2]

Originally Posted by LouMan

Hey thanks FreyGuy & harlequin2 you guys are the MAN got my boards they other day shame about the silkscreen, still looks like it's going to be a fun project.

Yeah, pity about the silk screen, but its not too bad following the printed overlay Chris did. I only made a couple of mistakes! Where would I be without my trusty solder sucker? I also very nearly soldered the pin socket for the lcd on the component side too. Must be getting old.....