Hi Rob, thx, Unfortunately you are right, the cleaned library behaves as the old one. I have no ADC running yet. :) I'm just trying to send a bunch of chars from the PIC to the host in a predictive way with no success. USB serial communication runs only if I receive a char and send a data on each received char, and even in this situation it freezes after a number of chars sent.
PIC18F25k50 runs exactly the same at 5V (3.3V on VUSB), just for the record. On Sun, Feb 28, 2021 at 4:30 PM Rob CJ <[email protected]> wrote: > Hi Vsurducan, > > The updated USB will not solve your problem, it was only cleaned up but > not functionally changed. > > What might be an issue is the conversion time of the ADC. Your call > adc_read(0) > is a blocking call since it waits until the ADC conversion is complete. > This may - not sure - be too long for the usb_serial_flush(). > > If that is the problem then there are - at least - two options: > -) Start the AD conversion yourself and check if conversion is still > ongoing by checking ADCON0_GO. If so skip the reading part until conversion > is complete and then read the ADC value. > -) Use an interrupt based version of the USB driver. I re-started working > on that today and it seems to work for USB serial (tested with a PIC16F1455 > and PIC18F14K50) since then you no longer need the usb_serial_flush() > because the handling of the USB driver is done in the interrupt routine. > > As said I am not yet done with the update of the USB interrrupt based > driver but as soon as it works I will upload it. For now you could try the > other option by checking if conversion is ongoing. > > > With the interrupt based USB driver my main usb_serial sample program now > looks like this: > > const USB_INTERRUPT_DRIVEN = TRUE -- This needs to be set to use the > interrupt driven mode > > forever loop > > -- There is no serial flush anymore 🙂 > > -- check if USB device has been configured by the HOST > if ( usb_cdc_line_status() != 0x00 ) then > if !has_shown_welcome_msg then > has_shown_welcome_msg = true > print_string( usb_serial_data, str_welcome ) > end if > else > has_shown_welcome_msg = false > end if > > -- check for input character > if usb_serial_read( ch ) then > -- echo input character > usb_serial_data = ch > end if > > end loop > > Kind regards, > > Rob > > ------------------------------ > *Van:* [email protected] <[email protected]> namens vsurducan > <[email protected]> > *Verzonden:* zondag 28 februari 2021 10:37 > *Aan:* [email protected] <[email protected]> > *Onderwerp:* Re: [jallib] WIN32CDCDrivers? > > > > On Sun, Feb 28, 2021 at 10:04 AM Rob CJ <[email protected]> wrote: > > Hi Vsurducan, > > To understand this better. The program you mention does not work for that > PIC running at 3.3 Volt, right? I have 2 questions: > 1) Have you tried what happens when running it at 5 Volt? > > > No because it has to work on 3.3V. However I will try it to see if it > makes things different. > > 2) What did you do with the 3.3 Volt generated by the PIC for the USB? I > can imagine that it cannot generate that Voltage if it is powered by 3.3. > Volt. > > > It's within specifications Rob. 3.3V on VDD, 3.1V on VUSB. Tio understand > better, I'm a hardware guy working in electronics from 35years now. > However, software is still not my favorite thing... > > > For your information. Some years ago I updated the USB driver to make it > suitable for a PIC16F1455 but due to a compiler bug I had to implement a > workaround to get that working. This compiler bug was fixed some time ago > so I removed all these workarounds and uploaded the 'clean' USB library > again to GitHub yesterday. > > > Oki, I will test it thank you. > > > Currently I am looking at making the USB driver operate on an interrupt > basis as to get rid of the need for flushing it. So if there are any other > problems with the driver I can have a look but I have to understand your > application. > > I general if you have a problem, try to minimize the code to the > essentials so that it becomes more easy to analyze the issue. > > > It's the minimum possible: the example "18F4450_usb_serial.jal" has been > adapted for PIC18F25K50. > The only thing which works from this example is the part below: > > forever loop > -- poll the usb ISR function on a regular base, in order to > -- serve the USB requests > usb_serial_flush() > > -- check for input character > if usb_serial_read( ch ) then > -- echo input character > usb_serial_data = ch > end if > > end loop > > > > > Thanks. > > Kind regards, > > Rob > > ------------------------------ > *Van:* [email protected] <[email protected]> namens vsurducan > <[email protected]> > *Verzonden:* zondag 28 februari 2021 07:43 > *Aan:* [email protected] <[email protected]> > *Onderwerp:* Re: [jallib] WIN32CDCDrivers? > > Hi Rob, > Nope, with PIC18F25K50 at 3.3V supply nothing workswith usb_serial library > except the echo part which is useless for me. Anything else is closing the > USB host port and device is seen as unknown. > > About delay, please take a look at one good example of funlw65 (below) > where it has about 1 second delay by flushing in the meantime. However this > example is not relevant for using only the usb_serial because it uses the > bootloader with different fuses configuration and different location for > the code which runs. > On PIC18F2550 at 5V supply and external HOSC, the actual USB library does > not send data reliably as an FT232 would do, an amount of data from ADC for > example would be sent to the terminal with variable speed even if the main > loop has very simple structure with fixed computing time. Sometimes the > transmission stops, to resuscitate it, the port has to be closed and > opened again on the host side... > I do not know why this happens, the library is too twisted for my taste so > perhaps I will modify it if I'll have nerves for that... > > > > -- Project: ADC read and send on USB CDC > -- Author: Vasile Guta-Ciucur (funlw65) > -- License: Code released under New BSD license > -- For jallib libraries, see de licence inside jal.zip > > include freejalduino4 -- FreeJALduino4 pinout layer for the last board > model > > -- include libraries > include usb_serial > include print > include format > include delay > > var dword Value > var word AD_RESULT > var word Volts > -- var byte Millivolts > const byte str1[] = " Volts\n\r" > > > -- ================================================== > -- PROGRAM START ------------------------------------ > -- ================================================== > -- configure pins > enable_digital_io() -- first, all pins set to digital > > > -- now configure ADC constants > const bit ADC_HIGH_RESOLUTION = true -- 10bit resolution > const byte ADC_NVREF = 0 -- no voltage reference > const byte ADC_NCHANNEL = 1 -- how many ADC channels we are using, > -- always starting the count from A0 (RA0/AN0) > -- This will automatically set A0 as analog > input > const word ADC_RSOURCE = 909 -- impendance (R1*R2/(R1+R2)) > -- then, load the ADC library > include adc > -- Initialize ADC > adc_init() > > > -- initialize the USB serial library > usb_serial_init() > > -- start main loop > forever loop > usb_serial_flush() > AD_RESULT = adc_read(0) > > print_word_dec(usb_serial_data, AD_RESULT) > usb_serial_data = " " > > > -- Value = 537 * AD_RESULT > Value = dword(537) * AD_RESULT -- one of the operands must be casted > -- as dword to have a 32bit integer math > -- because AD_RESULT is word and the other > -- operand is a generic one. > > Volts = Value / 100 > > format_word_dec(usb_serial_data, Volts, 4, 2) > print_string(usb_serial_data, str1) > > for 100 loop -- delay 1 second (a little more than) > usb_serial_flush() > delay_1ms(10) > end loop > > end loop > -- > On Sat, Feb 27, 2021 at 9:03 AM Rob CJ <[email protected]> wrote: > > Hi Vsurducan, > > Did you fix this issue? So far I never had issues with the USB driver but > I had always used it with a crystal. Why are you using udc_cdc_putc()? > > About your other post about the USB driver. I saw that you used a delay of > 400 ms in your main program. This may not work since the USB driver needs > to be polled by usb_serial_flush() and this needs to be done quite > frequently (I thought miliseconds). > > I already found this polling a bit anoying myself since it would be nicer > to have it done on an interrupt basis. Some time ago I started to make it > interrupt driven but it did not work immediately at that time. Maybe it is > a good time to pick it up again. > > Kind regards, > > Rob > > > > ------------------------------ > *Van:* [email protected] <[email protected]> namens vsurducan > <[email protected]> > *Verzonden:* woensdag 24 februari 2021 20:33 > *Aan:* [email protected] <[email protected]> > *Onderwerp:* Re: [jallib] WIN32CDCDrivers? > > Thank you Matt and Rob, > I have a non reliable USB communication under both Win10 and Win7 so I'm > suspecting everything... > The following sample is based on 18F4550_USB_serial.jal. PIC18F25K50 has > the same registers as PIC18F2550 except some PLL bits which can be ignored > with a delay. > However, USB works only with the echo. Trying to repeatedly send a byte > from PIC to PC, communication it froze. > > include 18f25k50 -- target PICmicro > -- > -- This program assumes that a 16 MHz resonator or crystal > -- is connected to pins OSC1 and OSC2. > pragma target clock 48_000_000 -- oscillator frequency > -- > > pragma target OSC HSH -- crystal or resonator > pragma target PLLSEL PLL3X > pragma target PLLEN ENABLED -- PLL on > pragma target CPUDIV P1 -- Fosc divisor > pragma target WDT DISABLED -- watchdog > pragma target XINST DISABLED -- do not use extended > instructionset > pragma target DEBUG DISABLED -- no debugging > pragma target BROWNOUT DISABLED -- no brownout reset > pragma target FCMEN DISABLED -- no clock monitoring > pragma target IESO DISABLED -- no int/ext osc > switching > pragma target LVP DISABLED -- low voltage > programming > pragma target MCLR EXTERNAL -- external reset > -- > -- The configuration bit settings above are only a selection, sufficient > -- for this program. Other programs may need more or different settings. > -- > OSCCON_SCS = 0 -- select primary oscillator > -- > enable_digital_io() -- make all pins digital I/O > WDTCON_SWDTEN = OFF -- disable watchdog > > include delay > include usb_serial > include print > > -- constants > const byte str_welcome[] = "JALLIB USB Serial Demo app\n" > > -- variables > > -- interrupts? No thanks > INTCON_GIE = false > > -- setup the USB serial library > usb_serial_init() > > var bit has_shown_welcome_msg = true > var byte ch > > -- main loop > forever loop > -- poll the usb ISR function on a regular base, in order to > -- serve the USB requests > usb_serial_flush() > > -- check if USB device has been configured by the HOST > if ( usb_cdc_line_status() != 0x00 ) then > if !has_shown_welcome_msg then > has_shown_welcome_msg = true > print_string( usb_serial_data, str_welcome ) > end if > else > has_shown_welcome_msg = false > end if > > -- check for input character ; this works > if usb_serial_read( ch ) then > -- echo input character > usb_serial_data = ch > end if > ; none of the following works, the USB is frozen... > ; usb_cdc_putc ("B") > ; format_word_dec(usb_serial_data, 0x3FC , 4, 0) > ; print_byte_dec(usb_serial_data, 0x04) > > > > On Wed, Feb 24, 2021 at 9:02 PM Matt Schinkel <[email protected]> > wrote: > > Hi, it is in jallib under tools/usb > > Sent from my Android device. > > ------------------------------ > *From:* [email protected] <[email protected]> on behalf of > Rob CJ <[email protected]> > *Sent:* Wednesday, February 24, 2021 1:41:57 PM > *To:* [email protected] <[email protected]> > *Subject:* Re: [jallib] WIN32CDCDrivers? > > Hi Vsurducan, > > I do not know but I think you do not need them. When I updated the serial > library some time ago I could test it with a JAL sample program. Windows 10 > recognized the usb device without the need for driver. > > If you use an older version of Windows I do not know if this is also true. > > Kind regards, > > Rob > > ------------------------------ > *Van:* [email protected] <[email protected]> namens vsurducan > <[email protected]> > *Verzonden:* woensdag 24 februari 2021 19:17 > *Aan:* [email protected] <[email protected]> > *Onderwerp:* [jallib] WIN32CDCDrivers? > > Hi, where can I find a copy of the Win32CDCDrivers.zip ? > The links posted in the usb_serial_demo are broken. > thanks, > -- > You received this message because you are subscribed to the Google Groups > "jallib" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/jallib/CAM%2Bj4quYAsdKiqVSCM%2BV6%2BaPFLvyeGj5WWfCgGnETB-e%2BW8wwQ%40mail.gmail.com > <https://groups.google.com/d/msgid/jallib/CAM%2Bj4quYAsdKiqVSCM%2BV6%2BaPFLvyeGj5WWfCgGnETB-e%2BW8wwQ%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > -- > You received this message because you are subscribed to the Google Groups > "jallib" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/jallib/AM0PR07MB6241E6D8D505B288C9B322B1E69F9%40AM0PR07MB6241.eurprd07.prod.outlook.com > <https://groups.google.com/d/msgid/jallib/AM0PR07MB6241E6D8D505B288C9B322B1E69F9%40AM0PR07MB6241.eurprd07.prod.outlook.com?utm_medium=email&utm_source=footer> > . > -- > You received this message because you are subscribed to the Google Groups > "jallib" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/jallib/YQXPR01MB41496334B29E540ECA2A9766DE9F9%40YQXPR01MB4149.CANPRD01.PROD.OUTLOOK.COM > <https://groups.google.com/d/msgid/jallib/YQXPR01MB41496334B29E540ECA2A9766DE9F9%40YQXPR01MB4149.CANPRD01.PROD.OUTLOOK.COM?utm_medium=email&utm_source=footer> > . > > -- > You received this message because you are subscribed to the Google Groups > "jallib" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/jallib/CAM%2Bj4qtwEeyURARHT7%3Dc2%2BqE-w7AtFX0Hmyg%2ByJAOL7q%3DGCEAQ%40mail.gmail.com > <https://groups.google.com/d/msgid/jallib/CAM%2Bj4qtwEeyURARHT7%3Dc2%2BqE-w7AtFX0Hmyg%2ByJAOL7q%3DGCEAQ%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > -- > You received this message because you are subscribed to the Google Groups > "jallib" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/jallib/AM0PR07MB6241C7A557FE8743121FB378E69C9%40AM0PR07MB6241.eurprd07.prod.outlook.com > <https://groups.google.com/d/msgid/jallib/AM0PR07MB6241C7A557FE8743121FB378E69C9%40AM0PR07MB6241.eurprd07.prod.outlook.com?utm_medium=email&utm_source=footer> > . > > -- > You received this message because you are subscribed to the Google Groups > "jallib" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/jallib/CAM%2Bj4qvwzWBSCXT1RQEYEucBtW%2BSLYOinp%3DxSt6HZgKTnZ5ECA%40mail.gmail.com > <https://groups.google.com/d/msgid/jallib/CAM%2Bj4qvwzWBSCXT1RQEYEucBtW%2BSLYOinp%3DxSt6HZgKTnZ5ECA%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > > -- > You received this message because you are subscribed to the Google Groups > "jallib" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/jallib/AM0PR07MB6241E61D143816CC5B793407E69B9%40AM0PR07MB6241.eurprd07.prod.outlook.com > <https://groups.google.com/d/msgid/jallib/AM0PR07MB6241E61D143816CC5B793407E69B9%40AM0PR07MB6241.eurprd07.prod.outlook.com?utm_medium=email&utm_source=footer> > . > > -- > You received this message because you are subscribed to the Google Groups > "jallib" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/jallib/CAM%2Bj4qvBKoEeVO7-_tOxhUGFcd9nQ714g1Ufs%3DzS8%2BgmfnTqtg%40mail.gmail.com > <https://groups.google.com/d/msgid/jallib/CAM%2Bj4qvBKoEeVO7-_tOxhUGFcd9nQ714g1Ufs%3DzS8%2BgmfnTqtg%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > > -- > You received this message because you are subscribed to the Google Groups > "jallib" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/jallib/AM0PR07MB624163DA01D4AE936236F2DDE69B9%40AM0PR07MB6241.eurprd07.prod.outlook.com > <https://groups.google.com/d/msgid/jallib/AM0PR07MB624163DA01D4AE936236F2DDE69B9%40AM0PR07MB6241.eurprd07.prod.outlook.com?utm_medium=email&utm_source=footer> > . > -- You received this message because you are subscribed to the Google Groups "jallib" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/jallib/CAM%2Bj4qt0xk%3DLtekZRyYCjKe-EyjoH2JdiiSoHwf_AXWwC9um-g%40mail.gmail.com.
