Post by thread:[PIC]: Strange RS232 receive errors on 18F252

Hi, I've been developing a firmware driver for an impact-type miniprinter. Problem is, I'm having trouble with data corruption. Basically, if you send a string of repeating characters, e.g. "XXXXXXXXXX", every second character gets its MSbit set, even though the data was transmitted with the MSbit low... Has anyone seen this happen before (and found a solution)? Thanks. -- Phil. | Acorn Risc PC600 Mk3, SA202, 64MB, 6GB, spam_OUTphilpemTakeThisOuTphilpem.me.uk | ViewFinder, 10BaseT Ethernet, 2-slice, http://www.philpem.me.uk/ | 48xCD, ARCINv6c IDE, SCSI ... I'm not nearly as think as you confused I am.

Hi Phil, I posted an answer to this last month :) Using Hyperterminal? Using a USB-RS232 adapter? Then don't look at your PIC for the problem. Either get a USB adapter that isn't based on the same chipset, or use something besides Hyperterminal. Cheers, Bob > I've been developing a firmware driver for an impact-type miniprinter. > Problem is, I'm having trouble with data corruption. Basically, if you > send a > string of repeating characters, e.g. "XXXXXXXXXX", every second character > gets its MSbit set, even though the data was transmitted with the MSbit > low... > Has anyone seen this happen before (and found a solution)?

Absolutely sure about the bitrate? Is it possible the stop bit is being interpreted as the msb? I've run 18F252 at 500kbps with no problems... Harold > Hi, > I've been developing a firmware driver for an impact-type miniprinter. > Problem is, I'm having trouble with data corruption. Basically, if you > send a > string of repeating characters, e.g. "XXXXXXXXXX", every second character > gets its MSbit set, even though the data was transmitted with the MSbit > low... > Has anyone seen this happen before (and found a solution)? > -- FCC Rules Updated Daily at http://www.hallikainen.com

Philip Pemberton wrote : > Hi, > I've been developing a firmware driver for an impact-type > miniprinter. > Problem is, I'm having trouble with data corruption. > Basically, if you send a string of repeating characters, > e.g. "XXXXXXXXXX", every second character gets its > MSbit set, even though the data was transmitted with > the MSbit low... Let's see... It's the PIC that sees this data as corrupted, right ? Have you been able to verify the data as it it on the RS232-line ? Can you move the cable from the PIC to something else that can display the data, such as an(other) PC with some terminal emulation program ? Or just monitoring the line with a scope (or better using some RS232 line logger) ? Is the PIC osc crystal based ? What freq ? What is the calculated baudrate error ? First thing would be to figure out *where* that MSbit gets "set"... Jan-Erik.

In message <.....63292.207.177.227.29.1117744824.squirrelKILLspam@spam@kauko.hallikainen.org>> "Harold Hallikainen" <haroldKILLspamhallikainen.com> wrote: > Absolutely sure about the bitrate? Is it possible the stop bit is being > interpreted as the msb? I've run 18F252 at 500kbps with no problems... Checked, doublechecked, clocked the crystal with a frequency counter.. The bitrate is bang on 9600 baud with a 11.0592MHz xtal. Later. -- Phil. | Acorn Risc PC600 Mk3, SA202, 64MB, 6GB, .....philpemKILLspam.....philpem.me.uk | ViewFinder, 10BaseT Ethernet, 2-slice, http://www.philpem.me.uk/ | 48xCD, ARCINv6c IDE, SCSI ... BREAKFAST.COM Halted... Cereal Port Not Responding.

In message <EraseME19688.66.52.191.4.1117744261.squirrelspam_OUTTakeThisOuTwebmail.sonic.net>> "Bob Blick" <bblickspam_OUTsonic.net> wrote: > Using Hyperterminal? Using a USB-RS232 adapter? Then don't look at your > PIC for the problem. Either get a USB adapter that isn't based on the same > chipset, or use something besides Hyperterminal. [searches archives] Gah. That'll be it. I've been using a Manhattan USB-serial converter with a Prolific PL-2303 chipset *and* Hyperterminal. Talk about two wrongs :D I've switched to Realterm and my PC's onboard RS232 port. The problem seems to have vanished. Methinks it's time to bodge an RS232 port onto one of the spare blanking plates. There's a COM2 connector built onto the motherboard, but the tightwads at Asus decided not to include a connector plate for it, even though they included a FireWire/USB plate. Grrrr... If the PL2303 acts up *once* *more*, I'm going to perform some Widlar Optimization methods on it with a really big hammer... then replace it with one of the FTDI converters that JELU are selling. I've been meaning to buy a few MP3 CODECs anyway :) Later. -- Phil. | Acorn Risc PC600 Mk3, SA202, 64MB, 6GB, @spam@philpemKILLspamphilpem.me.uk | ViewFinder, 10BaseT Ethernet, 2-slice, http://www.philpem.me.uk/ | 48xCD, ARCINv6c IDE, SCSI ... Beam me up, Scotty, but leave the others here.

> >If the PL2303 acts up *once* *more*, I'm going to perform some Widlar >Optimization methods on it with a really big hammer... then replace it with >one of the FTDI converters that JELU are selling. I've been meaning to buy a >few MP3 CODECs anyway :) I widlarized some Belkin units, after I got my Edgeports. I LOVE these things. Industrial-grade.