PIC plus-5110 LCD-GPS-BMP280-HMC5983

jjw · Feb 9, 2018

camerart said:
Hi,
Here's the combined program so far:
The GPS is working ok and showing on the 5110.
The BMP280 shows on the 5110, but the READing is incorrect, and not updating, I recognise those READings from previous tests. I'll have another try tomorrow.
If anyone can spot any obvious errors, please let me know.

EDIT: Programs 050118 1400 and 080218 0900 are the two WORKING programs I'm trying to combine. (Including the two INCLUDES fontsml and get_neo.)
C

The program here xxx -Combined.bas is the BMP-5110 program, not the combined program.

camerart · Feb 9, 2018

jjw said:
Spisend 0x60 should be Spisend 0xE0

You changed the values of control register and config register.
I don't know if they are wrong now.

Seems that there is no return from get_neo ?

jjw said:
The program here xxx -Combined.bas is the BMP-5110 program, not the combined program.

Hi J,
True I'm combining the Working BMP 5110 program with the earlier Working GPS program to make one longer one. (Later I'll do the same with the next
COMPASS program)
C.

camerart · Feb 9, 2018

jjw said:
Is the Yled toggling in the combined program?

Hi J,
I think your on the right track with the LED flashes. The YLED only flashes on start-up, then no more.
I tried YLED toggles inside each of the GOSUBS. get_neo = Y, get_val = Y, calc = Stays on, print = off, cmd5110= stays on, drt_lcd = Stays on.

I've posted a slightly changed program @ #533 ''18LF4520 5110 GPS BMP280 COMBINED 090218 1200'
This is now showing TIM, LAT, LON all updating TMP and PRS are correct on start-up, then the TMP and PRS stop, and the GPS keeps updating.
If I press the RESET button (which re-boots it) it all updates ok, but only once.
C

jjw · Feb 9, 2018

I think the combined program from yesterday might work, if you replace the last goto get_neo with return.
Give it a try.
Edit: and keep Gosub get_neo

camerart · Feb 10, 2018

jjw said:
I think the combined program from yesterday might work, if you replace the last goto get_neo with return.
Give it a try.
Edit: and keep Gosub get_neo

Hi J,
Before I change anything, will you comment on this first?
C

jjw · Feb 10, 2018

If these are from 08.02 version, the last warning might depend on missing return instruction of the get_neo subroutine or the get_val code is inside the get_neo
Others can be forgotten until now.

Try the changes in #544 and move the get_val outside get_neo

camerart · Feb 10, 2018

jjw said:
If these are from 08.02 version, the last warning might depend on missing return instruction of the get_neo subroutine or the get_val code is inside the get_neo
Others can be forgotten until now.

Try the changes in #544 and move the get_val outside get_neo

Hi J,
get_val is outside the get_neo. (As always)
I tried changing LINE 352 from GOTO get_neo to RETURN. It updates, but the PRINT 5110 is all over the place.
I've removed the sections as last 'compiler warning' out of the program, and now get this warning:
C

camerart · Feb 10, 2018

Hi J,
It's working
The get_neo RETURN didn't work first time, so I tried a few things, and re-tried it, and this time it worked, I don't know why.
I tried GOSUB PRINT, which didn't work, and I now realise that that section is CALLed not GOSUB (Am I correct?)
There is an artifact on the 5110 shown along with the latest program in #533. It appears to PRINT the 5110 then the artifact appears in a split second.
All in the 'compiler warnings', have been removed, so no warnings now.
Thanks very much.
C.

camerart · Feb 18, 2018

Hi,
I'm now adding a compass module HMC5983 to the test PCB. This is the last one

As this is another SPI module, I'm unsure whether in good programming protocol, some of the 'old' code can be used to save program length.
Some guidance please.
C.

camerart · Feb 18, 2018

Removed

camerart · Feb 19, 2018

Hi,
Will someone have a look at the attached jpg please?
Bear in mind that I'm pretty sure that 'burst READ' can't be done in Oshonsoft, so the section with automatic incremental pointer most likely won't work [Perhaps test later]

It looks to me that the WRITE BIT %0000001 isn't needed for all of the registers marked READ. Is this correct? SPI MODE.

C.

camerart · Feb 23, 2018

Hi,
I've just got the HMC5983 Compass module to return the X Z and Y register results, shown in the 2x jpg examples.

Reading the D/S, these results are in 2’s complement form.
Can anyone advise me how to do this in Oshonsoft please?
Camerart.

camerart · Feb 28, 2018

Hi,
I'm using SPI to READ the HMC5983. I'm getting results showing on the 5110 screen.

Here's a couple of attachments, showing the section of program with READ x3 plus the output on screen.

For the moment I've commented out the '2.s compliment' equations, just for testing.

The digits are changing ok, but don't look correct to me.

(I'll correct the screen artifacts another time)

EDIT: Here's another attachment from Oshonsoft regarding LONGs. I'm pretty sure my problem is something to do with .HB .LB and e,g, .0B

C

jjw · Feb 28, 2018

camerart said:
Hi,
I'm using SPI to READ the HMC5983. I'm getting results showing on the 5110 screen.

Here's a couple of attachments, showing the section of program with READ x3 plus the output on screen.

For the moment I've commented out the '2.s compliment' equations, just for testing.

The digits are changing ok, but don't look correct to me.

(I'll correct the screen artifacts another time)

EDIT: Here's another attachment from Oshonsoft regarding LONGs. I'm pretty sure my problem is something to do with .HB .LB and e,g, .0B

C

The numbers are too big.
They should be 0-4096.
Do you use longs ?

camerart · Feb 28, 2018

jjw said:
The numbers are too big.
They should be 0-4096.
Do you use longs ?

Hi J,
I've found the problem, but not the solution 'yet' It's something to do with where the BYTEs are in the LONG.

Here's my effort so far:
I've been changing it. If I don't use the '2's compliment' section I get 5 digits as expected (I understand it needs the 2's compliment calc!)
NOTE: I've copied the DIMs from earlier in the Program.

'READ X REGISTER
SPICSOn
For i = 0 To 1
compss = 0 'CHIP SELECT COMPASS ON
addr = 0x83 + i 'READ XDATA ADDRESS
SPISend addr 'READ XDATA 0X83 0X84 ADDRESS
SPIReceive data
b(i) = data
'WaitMs 100 '67 MS ????????????????????????????????????
compss = 1 'CHIP SELECT COMPASS OFF
Next i
x_raw.HB = b(1) 'CHECK ORDER???????????????????
x_raw.LB = b(2)

x_hlb.LW = x_raw

If x_hlb > 32767 Then '2'S COMPLIMENT CALC
x_hlb = bit17h - x_hlb
x_hlb = x_hlb + 1
Endif

strxhlb = #x_hlb
''''''''''''''''''''''''''
'Dim x_raw As Word
'Dim x_hlb As Long
'Dim strxhlb As String
''''''''''''''''''''''''''''''

EDIT: There is also something wrong with the READs XZY all show the same result.

C.

jjw · Feb 28, 2018

Are the register addresses right?

camerart · Feb 28, 2018

jjw said:
Are the register addresses right?

Hi J,
I've tried the i2C READ with 0x3D before the READ, and the SPI 0x8x before the READ.
C.

jjw · Feb 28, 2018

x_raw.hb = b(1) -> x_raw.hb = b(0)
x_raw.lb = b(2) -> x_raw.lb = b(1)

What is the value of bit17h ?
Using long x is not needed as was discussed in the other thread.
2's complement:
If x_raw > 32767
x_raw = 65535 - x_raw // also x_raw = Not x_raw should be ok
x_raw = x_raw + 1
endif

camerart · Mar 1, 2018

jjw said:
Are the register addresses right?

Hi J,
I've tried the i2C READ with 0x3D before the READ, and the SPI 0x8x before the READ.
C.

jjw said:
x_raw.hb = b(1) -> x_raw.hb = b(0)
x_raw.lb = b(2) -> x_raw.lb = b(1)

What is the value of bit17h ?
Using long x is not needed as was discussed in the other thread.
2's complement:
If x_raw > 32767
x_raw = 65535 - x_raw // also x_raw = Not x_raw should be ok
x_raw = x_raw + 1dif

Hi J,
I'm working with distant memories , but I'm pretty sure Oshonsoft won't work unless LONGs are used here??
bit17h is BIT17H or %1 0000 0000 0000 0000 or 65536

EDIT: I just searched and found the area where the Oshonsoft problem was discussed. Where I am using LONGs I think I should be using SINGLES for the calculation, am I correct?
See post #408

C

jjw · Mar 1, 2018

The syntax error:
x raw = should be x_raw =
Long is not needed when x_raw = 65535 - x_raw
better use x_raw = Not x_raw
then no need to remember magic numbers.

Singles are needed, when you calculate angles from x,y, unless you have arctan function using integers.

PIC plus-5110 LCD-GPS-BMP280-HMC5983

Active Member

Well-Known Member

Well-Known Member

Active Member

Well-Known Member

Attachments

Active Member

Well-Known Member

Attachments

Well-Known Member

Well-Known Member

Well-Known Member

Well-Known Member

Attachments

Well-Known Member

Attachments

Well-Known Member

Attachments

Active Member

Well-Known Member

Active Member

Well-Known Member

Attachments

Active Member

Well-Known Member

Attachments

Active Member

Similar threads

Privacy & Transparency

Privacy & Transparency