HID USB on 18F87J50

[pedja089]

Does someone have example for this PIC?
I can't get it to work...

[pedja089]

I searched the whole forum and I only found examples for the PIC18f2550 and 4550.
Does PBP2.60 support USB on 18F87J50?
For the second day of trying all possible examples and with all possible fuse configuration but nothing show in device menager, not even device not recognised...
I bought this development system from mikroe, thinking that the is a hardware problem, but still nothing...
What are my other options

[ScaleRobotics]

Should work. I had the 18F67J50 (same family) running CDC. I can try to get an example for it going for HID. By the way, are you using DT_HID260 ? Will try to get something up tonight, or tomorrow morning.

[pedja089]

Yes, I tried that and DT_HID. But nothing happen if cable is plugged or unplugged.
For now, CDC will be good starting point.
I would be very grateful if you could send me CDC example with fuse configuration.
Thank you very much.

[ScaleRobotics]

Well, now I am starting to wonder if I ever had it working on CDC. I had a cdc labeled program for it, but what I had does not work. I had a fairly quick time with the 47J53 in CDC as well as HID, but for the 67J50, I can't seem to get either to work, and I have spent a fair amount of time at it.

These were the configs I used for other programs for the 67J50, which pretty much matches Microchip's HID bootloader software.

'DEFINE LOADER_USED 1
'DEFINE RESET_ORG 1000h ' For Microchip USB Bootloader
#config
    CONFIG PLLDIV = 5       ;for 20 mhz chip
    CONFIG XINST = OFF
    CONFIG STVREN = ON    ;stack overflow reset
    CONFIG WDTEN = OFF    
    CONFIG CP0 = OFF    
    CONFIG IESO = OFF    
    CONFIG FCMEN = OFF    
    CONFIG CCP2MX = DEFAULT    
    CONFIG WDTPS = 32768 
    CONFIG CPUDIV = OSC1  ;DIVIDE BY 1 MODE
    CONFIG FOSC = HSPLL
    CONFIG MSSPMSK = MSK5   
#endconfig

DEFINE OSC 48
clear
OSCTUNE.6 = 1  ' Enable PLL for 18F87J50 family

I will try a few more things, but I wanted to give you a heads up.

[pedja089]

I try cdc from pbp/usb18 and add config, but still nothing...
I use 8MHz crystal, so PPLDIV is 2 is that correct?
What are the chances that the USB module on 3 PIC are broken?

[ScaleRobotics]

Not very high. But (speaking for myself), the chances of the problem being behind the keyboard is pretty high! Just not sure what I could be missing here, doing it two different ways hid, and cdc, and with a USB bootloader, and without. So I am running out of things I can think of to try....

Yes, divide by two would be correct for a 8 mhz.

[ScaleRobotics]

Well, I am not having any luck over here either. Here is my try at using DT_HID260. I have been successful in getting it running on another J chip, the PIC18F47J53, and looking at the two data sheets, the USB doesnt seem to differ much between these two chips. Except on one the USB data memory is below the buffer descriptors, and the other the USB data memory is above the buffer descriptors. The DEV files seem to point to the USB buffer descriptors (67J50 here): USBMEMORYADDRESS Con $400 ' USB RAM starts here

Here is my code (that does not work)

'***************************************************************************
'*  Name    : BasicUSB.pbp                                                 *
'*  Author  : Darrel Taylor                                                *
'*  Notice  : Copyright (c) 2009                                           *
'*  Date    : 7/23/2009                                                    *
'*  Version : 1.x ' Modified by scalerobotics                              *
'*  Notes   : I tried to edit it to work on a PIC18F67J50 and I failed     *
'*          : Worked great on other chips, but I can't seem to figure out  *
'*          : the problem for PIC18F67J50. Board is PIC_LCD3310, which does*
'*          : work with the USB bootloader,so probably not a hardware issue*
'***************************************************************************
'DEFINE LOADER_USED 1
'DEFINE RESET_ORG 1000h ' For Microchip USB Bootloader

;--- if you un-comment these, you must comment the ones in the .inc file ---
ASM  ; 18F2550/4550, 20mhz crystal
;   __CONFIG    _CONFIG1L, _PLLDIV_5_1L & _CPUDIV_OSC1_PLL2_1L & _USBDIV_2_1L
;   __CONFIG    _CONFIG1H, _FOSC_HSPLL_HS_1H
;   __CONFIG    _CONFIG2H, _WDT_ON_2H & _WDTPS_512_2H
;   __CONFIG    _CONFIG3H, _PBADEN_OFF_3H
;   __CONFIG    _CONFIG4L, _LVP_OFF_4L & _XINST_OFF_4L
ENDASM

;ASM  ; 18F13K50/14K50  Only 12mhz crystal can be used for USB
;    CONFIG CPUDIV=NOCLKDIV ; CPU runing at full speed
;    CONFIG USBDIV=OFF ; Low speed USB clock diviser disabled, not used here
;    CONFIG FOSC=HS ; Use of High Speed crystal oscillator
;    CONFIG PLLEN=ON ; Enable PLL multiplier, 12Mhz oscilator x 4
;    CONFIG PCLKEN=ON ; Primary clock source is enabled
;    CONFIG FCMEN=OFF ; Failsafe clock off
;    CONFIG IESO=OFF ; Oscillator switchover mode disabled
;    CONFIG WDTEN=ON ; Watchdog timer on
;    CONFIG WDTPS=512 ; Watchdog divider is 512
;    CONFIG MCLRE=OFF ; Disable MCLR pin, enable RC3
;    CONFIG STVREN=ON ; Stack full/underflow will cause reset
;    CONFIG LVP=OFF ; Low voltage programming disabled
;    CONFIG BBSIZ=OFF ; 512kW block boot size
;    CONFIG XINST=OFF ; Extended instruction mode disabled
;ENDASM
#config         ;pic18F67J50 on Olimex PIC_LCD3310 usb board
    CONFIG PLLDIV = 5    ;for 20 mhz crystal
    CONFIG XINST = OFF
    CONFIG STVREN = off    ;stack overflow reset
    CONFIG WDTEN = OFF    ;added
    CONFIG CP0 = OFF    ;added
    CONFIG IESO = OFF    ;added
    CONFIG FCMEN = OFF    ;added
    CONFIG CCP2MX = DEFAULT    ;added
    CONFIG WDTPS = 32768 
    CONFIG CPUDIV = OSC1  ;DIVIDE BY 1 MODE
    CONFIG FOSC = HSPLL
    CONFIG MSSPMSK = MSK5   ;add
#endconfig

DEFINE OSC 48
clear
OSCTUNE.6 = 1  ' Enable PLL for 18F87J50 family

;--- Setup Interrupts ------------------------------------------------------
INCLUDE "DT_INTS-18.bas"     ; Base Interrupt System

ASM
INT_LIST  macro    ; IntSource,          Label,  Type, ResetFlag?
        INT_Handler   USB_Handler
    endm
    INT_CREATE               ; Creates the interrupt processor
endasm

;--- Setup USB -------------------------------------------------------------
INCLUDE "DT_HID260.pbp"

DEFINE USB_VENDORID    6017
DEFINE USB_PRODUCTID   2000
DEFINE USB_VERSION     1
DEFINE USB_VENDORNAME  "Darrel Taylor"
DEFINE USB_PRODUCTNAME "DT_HID"
DEFINE USB_SERIAL      "001"
DEFINE USB_INSIZE      32   ;  IN report is PIC to PC (8,16,32,64)
DEFINE USB_OUTSIZE     16   ; OUT report is PC to PIC
DEFINE USB_POLLIN      10   ; Polling times in mS, MIN=1 MAX=10
DEFINE USB_POLLOUT     10

; --- Each USB LED is optional, comment them if not used ----
;DEFINE USB_LEDPOLARITY 1       ; LED ON State [0 or 1]  (default = 1)
;DEFINE USB_PLUGGEDLED  PORTB,0 ; LED indicates if USB is connected
;DEFINE USB_TXLED       PORTC,2 ;  "      "     data being sent to PC
;DEFINE USB_RXLED       PORTC,1 ;  "      "     data being received from PC

OutCount    VAR  WORD : OutCount = 0
Value       VAR  WORD
X           VAR  WORD

;--- Setup ADC -------------------------------------------------------------
DEFINE ADC_BITS 10 ; Number of bits in ADCIN result
'       Configure ADC 
'       =============
'   See thread http://www.picbasic.co.uk/forum/showthread.php?t=11128&highlight=18f67j50
'   for more information about setting analog, and the WDTCON.4 on this chip
WDTCON.4 = 1   ' Switch to Alternate SFR
ANCON0_ALT = %10011110
           '  -xx------ Not Implemented
           '  1--1111-- AN7, & AN<4:1> = Digital
           '  -------0- AN0 = Analog

ANCON1_ALT = %11111111
           '  11111111- AN<15:10> = Digital

WDTCON.4 = 0    ' Back to Default SFR
 
ADCON0 = %00000001
       '  0-------- VCFG1, Vref- 0 = AVss (1=AN2)
       '  -0------- VCFG2, Vref+ 0 = AVdd (1=AN3) 
       '  --0000--- CHS<3:0>, Channel select -- 0
       '  ------0-- Go/Done, don't care for now
       '  -------1- ADON, Enable ADC
                  
ADCON1 = %10000111
       '  1-------- ADFM, Right justified result
       '  -0------- ADCAL, Normal A/D converter operation
       '  --000---- ACQT<2:0>, A/D Aquisition time = 0 Tad
       '  -----111- ADCS<2:0>, A/D conversion clock = FRC
;--- The Main Loop ---------------------------------------------------------
TRISD = 0
PORTD.0 = 1   'turn on g sensor
PORTD.1 = 0   'select scale for g sensor
PORTD.2 = 0   'select scale for g sensor


TRISA = %00000111   'set porta.0, 1, and 2 as inputs
TRISB = 0
CM1CON1 = 7   'CM1CON in PBP3 as CM1CON1
CM2CON1 = 7   'CM2CON in PBP3 as CM2CON1
TRISC = 0
TRISE = 0
pause 100
;--- The Main Loop ---------------------------------------------------------
Main:
    FOR X = 0 to 1000           ; Check for incomming USB data while pausing
@     ON_USBRX_GOSUB  _HandleRX
      PAUSE 1
    NEXT X
    
    ADCIN 10, Value
    OutCount = OutCount + 1
    ARRAYWRITE USBTXBuffer,["AN0=",DEC Value," -",DEC OutCount,"  "]
    IF Plugged THEN GOSUB DoUSBOut     ; only send when USB is connected
GOTO Main

;---- This just sends the received packet back to the PC -------------------
HandleRX:
    ARRAYWRITE USBTXBuffer,[STR USBRXBuffer\USBBufferSizeRX]
return

[pedja089]

I try similar thing, but same result.
So is it possible there are some bug in PBP for this PIC family?

[Darrel Taylor]

I'm looking into it too.
Trying to build a board.

But first I have to ask ...
Do you have VDD (3.3V) tied to VUSB?
There's no internal regulator on those chips.

[cncmachineguy]

I can't speak for Pedja's board, But I am pretty sure Walter is using the PIC_LCD3310. If so that DOES show Vusb tied to 3.3V on the schamatic.

[ScaleRobotics]

Thanks Darrel. I had some weird behavior in my computer. PBP3 kept defaulting to MPASMX, no mater how I set it, I'd check back, and it was set to MPASMX. So after removing the Microchip directory, and re-installing MPLAB 8.76, I was able to get PBP to settle on C:\Program Files\Microchip\MPASM Suite . Now CDC and HID work for the 18F67J50 with PBP3.0! Thanks for routing that out for us!

[ScaleRobotics]

I have installed MPLAB8,76. Now I get CDC to work on 2.60C, but I don't get the same file to work with PBP3.

But I am wondering where PBP3 is looking for MPLAB or MPASM? 2.60 allowed you to select the location for PBP, and it also allowed you to show it where the assembler was located. In PBP3 I get this option:



How am I supposed to enter them both? Or are they really supposed to be in the same place??

[Darrel Taylor]

In the Start menu, go to All Programs > PBP3 from melabs > MPASM-MPLAB setup.
It allows you to set the path to MPASM.

[ScaleRobotics]

Thanks!

Yes, I am using Olimex's PIC_LCD3310. Making a board for this family is a little bit of a pain. Darrel, do you want me to send you the Olimex board? I can ship it out tomorrow.

Walter

[Darrel Taylor]

Thanks Walter, that's a nice offer.

But I ordered one from sparkfun. They're only 50 miles away so I should have it by friday, shipped ground.
That's sure easier than building one.

I'm going to try a 26J50 too.
The 87J50 may still be a question mark.

[ScaleRobotics]

Now that's what I call support! Thanks Darrel.

I can vouch for the PIC18F46J50 working with DT_HID260 and PBP3. I have Micochip's mini demo board http://www.microchip.com/stellent/id...cName=en540669 going. But the 67J50 is elusive for me. Not ruling out my own operator error here, but this operator is running out of ideas ..

[pedja089]

Darrel, thank for support.
I have 3.3V om Vusb... That was first thing to check...
I can send you PIC. Just give me adress.

[Darrel Taylor]

Thanks Pedja,

We have some of those chips, but no boards to put it on.
At least none that would be easy to use.

I think the 67J50 should be good enough since it's in the same family as the 87J50.
We shall see this weekend.

[ScaleRobotics]

I found this on page 81: But I guess you found that already.

Note: Addresses, F40h through F5Fh, are not
part of the Access Bank, therefore specifying
a BSR should be used to access these
registers.

See, its a feature, not a bug!

[mister_e]

[Darrel Taylor]

I found this on page 81: But I guess you found that already.
See, its a feature, not a bug!

I tried to head off that question in my earlier post.
I understand that it's outside of the access bank, but that only affects the "banked" opcodes (movf, movwf, bcf, bsf, iorwf etc). Banked opcodes are the only ones that work in that range on the 67J50.

MOVFF and indirect addressing are not affected by the BSR register.
They use the full address of the register.

From the MOVFF Instruction definition ...

Location of source ‘fs’ can be anywhere
in the 4096-byte data space (000h to
FFFh) and location of destination ‘fd’
can also be anywhere from 000h to
FFFh.

[ScaleRobotics]

I think it is sort of interesting that in both the PIC18F47J53, and the PIC18F46J50 (both known to work with PBP USB) it states:

Note: The SFRs located between EB0h and
F5Fh are not part of the Access Bank.
Either BANKED instructions (using BSR) or
the MOVFF instruction should be used to
access these locations. When programming
in MPLAB® C18, the compiler will
automatically use the appropriate
addressing mode.

However, for the PIC18F67J50, they leave the MOVFF part out of this statement. I wonder if Microchip knows its errata and just didn't add it to the errata list. Or it's just a coincidence.

[Darrel Taylor]

Hmmmm, very interesting indeed.

[pedja089]

I was out of town, now I'm reading posts, and i tested hardware with HID boot loader, and it works.
When can I expect solution?
Is this will be corrected in PBP2.60(which I currently use), or I must upgrade to PBP3?
I really appreciate your efforts.
Thanks to all

[mister_e]

There's no good reason to not upgrade to PBP3... forum is free... now say to Melabs & Crownhill you appreciate the forum because it saved your butt a couple of times... get the upgrade

[pedja089]

Yes, you're right, I'll get update

[ScaleRobotics]

I didn't get very far on the Microchip forum. http://www.microchip.com/forums/tm.a...mpage=1#597082 They wanted me to try the simulator, and see if the simulator thought that should work or not. I couldn't get far enough in Mplab to find out. Though I am not sure how useful that would be. Does a simulator always sim what it should?

[mister_e]

MPSIM worked with PBP 2.x, will try with PBP3. It has to.

And NO a simulator do not always give you the good results, some are worst than others. As a guideline they are usually decent tool or pure toy, not much.

[Darrel Taylor]

What did you try in MPSIM?

I'm finding that it's reversed in MPSIM.
MOVFF and FSR's work, but banked instructions don't.

[mister_e]

I haven't tried this specific PIC yet. I try to stay away of Sims

[Darrel Taylor]

OK, I'm getting closer.
I think I can make USB work on it.

What a screwy chip.