Post by thread:[PIC]: PIC SBC requirement

I need to toss together a quick one-off for a customer. I need an SBC/prototype board: RS232 in and out (bit bang ok). TTL levels _not_ OK. MAX232 or similar greatly preferred for robustness. Mounting holes (you wouldn't believe the number of small boards that have no way to mount them :-). Small size. 8 pin PIC preferred. 14 pin Ok. 18 pin Ok. Crystal clock. Reasonable price. US source (for reasonably quick delivery). Gotta do this job so it looks 'professional' (my customer's customer has to accept it as 'standard product'). Hence, no wire wrap of prototypish stuff. Otherwise I'd just cobble it together on vector board. :-) Didn't seem to find anything by checking out the usual suspects. Bob Ammerman RAm Systems (contract development of high performance, high function, low-level software) -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! use spam_OUTlistservTakeThisOuTmitvma.mit.edu?body=SET%20PICList%20DIGEST

On Fri, 27 Oct 2000, Bob Ammerman wrote: {Quote hidden}> I need to toss together a quick one-off for a customer. I need an > SBC/prototype board: > > RS232 in and out (bit bang ok). TTL levels _not_ OK. MAX232 or similar > greatly preferred for robustness. > > Mounting holes (you wouldn't believe the number of small boards that have no > way to mount them :-). > > Small size. > > 8 pin PIC preferred. > 14 pin Ok. > 18 pin Ok. > > Crystal clock. > > Reasonable price. > > US source (for reasonably quick delivery). > > > Gotta do this job so it looks 'professional' (my customer's customer has to > accept it as 'standard product'). Hence, no wire wrap of prototypish stuff. > Otherwise I'd just cobble it together on vector board. :-) > If you don't find something (unlikely that you can't), whip up a design and have it made from expresspcb.com. Free software, and if you need a bunch (like 50+) they can get downright inexpensive, even with solder mask & silkscreen. They have a new service that makes qty.2 relatively cheap if you can do it with a fixed size board. I've been tempted to do a bunch myself and offer them for sale, but there seem to be a lot of places already offering a good selection. MELabs has some nice ones, but I don't think they have mounting holes. Dale --- The most exciting phrase to hear in science, the one that heralds new discoveries, is not "Eureka!" (I found it!) but "That's funny ..." -- Isaac Asimov -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! use .....listservKILLspam@spam@mitvma.mit.edu?body=SET%20PICList%20DIGEST

I need qty (1). If worst comes to worst I'll do something up in EAGLE and zap it to APCIRCUITS. Just got my first PC board ever back from them, and I'm quite pleased. Customer will be much happier if I don't have to spend the better part of a day fussing with this, tho'. The code is already written, took about 45 minutes. Bob Ammerman RAm Systems (contract development of high performance, high function, low-level software) {Original Message removed}

Bob Ammerman wrote: > >RS232 in and out (bit bang ok). TTL levels _not_ OK. MAX232 or similar >greatly preferred for robustness. > Hi Bob, Certainly you are correct about using TTL levels for RS-232 - ie, 0-5v. But I was wondering about your feelings regarding bjt RS-232 ports that produce nominal +5v and -12v. Personally, I think having minus-going is pretty crucial. Not too sure you gain a lot using +12 over +5v, however, except possibly for long wires. Big advantage of the bjt ports is low-quiescent power reqs. I have played with these quite a bit and have always had very good luck - using cables to 12 feet and baudrates to 115K. - danM -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

On Mon, 30 Oct 2000, Dan Michaels wrote: > Certainly you are correct about using TTL levels for RS-232 > - ie, 0-5v. But I was wondering about your feelings regarding > bjt RS-232 ports that produce nominal +5v and -12v. Personally, > I think having minus-going is pretty crucial. Not too sure you > gain a lot using +12 over +5v, however, except possibly for > long wires. Dan, Mabe I'm missing something... but why would negative-going levels be important other than meeting the letter of the RS232 electrical spec? I thought it wuld be a major deal for interoperability, but I have yet to see an RS232 receiveer that won't see, say, .4V or below (usually .7 or even higher) as a valid LOW signal. Even the old ones. Is there a large base of devices not using the receiveers I'm looking at (MC148x, MAX2xx, etc) or does it make a difference over longer cables? Noise immunity? I'm not disagreeing with your statement, just trying to figure out why to agree with it... 8-) Dale --- The most exciting phrase to hear in science, the one that heralds new discoveries, is not "Eureka!" (I found it!) but "That's funny ..." -- Isaac Asimov -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

> > >Mabe I'm missing something... but why would negative-going levels be >important other than meeting the letter of the RS232 electrical spec? I >thought it wuld be a major deal for interoperability, but I have yet to >see an RS232 receiveer that won't see, say, .4V or below (usually .7 or >even higher) as a valid LOW signal. Even the old ones. Is there a large >base of devices not using the receiveers I'm looking at (MC148x, MAX2xx, >etc) or does it make a difference over longer cables? Noise immunity? I'm >not disagreeing with your statement, just trying to figure out why to >agree with it... 8-) Even the ancient 1488/89 dosen't use negative levels on the receiver. The receiver simply counts everything below about 2V as a 1. (5V is a zero) -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

On Mon, 30 Oct 2000, David VanHorn wrote: > > > > > >Mabe I'm missing something... but why would negative-going levels be > >important other than meeting the letter of the RS232 electrical spec? I > >thought it wuld be a major deal for interoperability, but I have yet to > >see an RS232 receiveer that won't see, say, .4V or below (usually .7 or > >even higher) as a valid LOW signal. Even the old ones. Is there a large > >base of devices not using the receiveers I'm looking at (MC148x, MAX2xx, > >etc) or does it make a difference over longer cables? Noise immunity? I'm > >not disagreeing with your statement, just trying to figure out why to > >agree with it... 8-) > > > Even the ancient 1488/89 dosen't use negative levels on the receiver. > The receiver simply counts everything below about 2V as a 1. > (5V is a zero) I know... that was one of the higher cutoffs. I think the lowest was 0.7V. What I meant was that I've looked at the MAX and MAX-like devices as well as the 148x and none seem to require a negative voltage fo MARK, which was why I was wondering why I should bother. Dale --- The most exciting phrase to hear in science, the one that heralds new discoveries, is not "Eureka!" (I found it!) but "That's funny ..." -- Isaac Asimov -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

> >I know... that was one of the higher cutoffs. I think the lowest was >0.7V. What I meant was that I've looked at the MAX and MAX-like devices >as well as the 148x and none seem to require a negative voltage fo MARK, >which was why I was wondering why I should bother. Dont. While the spec says negative, the 1488/89 steamrollered it. -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

Hi, > I know... that was one of the higher cutoffs. I think the lowest was > 0.7V. What I meant was that I've looked at the MAX and MAX-like devices > as well as the 148x and none seem to require a negative voltage fo MARK, > which was why I was wondering why I should bother. Noise margin. RS232 is voltage switching, not current switching. If there is any induced interference (hum, spikes, RF) the .5 volt (or what ever) is not enough margin. +/- 12 Volts will guarantee clean operation on long wires in noisy environments. It is essential. ........................ Zim -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

> > >+/- 12 Volts will guarantee clean operation on long wires in noisy >environments. It is essential. Well, it was designed to operate in steel mills and such, but in an office environment a 0-5V swing will do 9600 at about half a mile error free in normal telephone wire. -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

On Tue, 31 Oct 2000, Graeme Zimmer wrote: > +/- 12 Volts will guarantee clean operation on long wires in noisy > environments. It is essential. > .. when using long wires in noisy environments. I have yet to see a failure with 0/5V signal levels... but I have not used long ( > 10') cables or operated in extremely noisy environments. Wonder if I could make it fail if I blew the dust off/out of the HF ham station again?? 8-) Dale --- The most exciting phrase to hear in science, the one that heralds new discoveries, is not "Eureka!" (I found it!) but "That's funny ..." -- Isaac Asimov -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

Dale Botkin wrote: > >Mabe I'm missing something... but why would negative-going levels be >important other than meeting the letter of the RS232 electrical spec? I >thought it wuld be a major deal for interoperability, but I have yet to >see an RS232 receiveer that won't see, say, .4V or below (usually .7 or >even higher) as a valid LOW signal. Even the old ones. Is there a large >base of devices not using the receiveers I'm looking at (MC148x, MAX2xx, >etc) or does it make a difference over longer cables? Noise immunity? I'm >not disagreeing with your statement, just trying to figure out why to >agree with it... 8-) > Hi Dale, You are probably correct that <= 0.7v will actually "work" with most devices - BUT my personal preference is to design to meet the "minimal" specs, at the very least. Cover the bases. Plain and simple. And the more recent specs are more stringent than the older, anyways, +/-5v vs +/-3v. But Bob Ammerman was the one who was so adamant about using MAX232, and I was interested in how he felt about +5v/-12v via BJT ckts, ie, within the actual RS-232 spec range - as against +12/-12v via MAX232. - danM -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

David, > Well, it was designed to operate in steel mills and such, but in an office > environment a 0-5V swing will do 9600 at about half a mile error free in > normal telephone wire. If you could bias the center point to +2.5 Volt, the margin would then be +/- 2.5 Volt which would probably work ok. But if the center point is at 0 Volts you will be certain to get angry customers.... ................ Zim -- http://www.piclist.com hint: The list server can filter out subtopics (like ads or off topics) for you. See http://www.piclist.com/#topics

>> RS232 in and out (bit bang ok). TTL levels _not_ OK. MAX232 or >> similar greatly preferred for robustness. Any product that I designed to interface to an EIA-232 device would have the negative swing. Besides noise margin, there are a few devices out there that will not accept 0 volts as marking. I've worked with DEC, HP, IBM, Sun, and other serial ports for a hair over 25 years now. The couple failures that I've seen were when a negative supply in the "RS232" transmitter failed and it was putting out 0V and 12V. It's rare but some devices will not work at 0V and 5V. > But I was wondering about your feelings regarding bjt RS-232 > ports that produce nominal +5v and -12v. Personally, I think > having minus-going is pretty crucial. Not too sure you gain a > lot using +12 over +5v, however, except possibly for long wires. The EIA-232 and previous RS-232 standard call for a voltage at the transmitter pin of -15 to -5 volts or +5 to +15 volts. The receiver pin is supposed to accept -25 to -3 volts as marking and +3 to +25 volts as spacing. So the standard allows for 2 volts drop due to resistive loss in the interconnecting cable. Lee Jones -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

>Dont. While the spec says negative, the 1488/89 steamrollered it. Wrong the 1489 conforms to the RS232 spec which requires an open circuit input to correspond to a negative voltage. hence the threshold is put at a slightly positive voltage so with the input held to ground by an internal 3k resistor, when the input goes open circuit it does what the RS232 spec says. -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

>Hold on... your saying that the RS232 spec includes an option for +V / open >circuit as well as +V / -V levels? What copy are you reading? No, the spec makes provision to intelligently protect the receiving end from nuisance break conditions when the line goes open circuit. It just so happens that the implementation of this on most chips will handle TTL type levels very nicely thank you. But I would only use this on the desk top - if going over any other distance then the >+/-3V swing is a must for noise immunity. -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

Bob Ammerman wrote: >Yeah, I'm comfortable enough with transistor driver ports. They would be >especially ok for this application where the RS232 connection can me >measured in inchces. Just don't want to have to come up with a -12 supply. > That's why you steal it from the RXD line. Easy enough to do. Makes for a nice 0 quiescent current RS-232 port. My BJT ports typically run at 38400 over about 16-18' of RS-232 cable - two 6' cables to a switch box, plus another 5-6' cable to the controller board. Using +5v/-12v, you get plenty of noise margin. 0-5v, I wouldn't use myself. - danM ====== ---- Original Message ----- {Quote hidden}>From: Dan Michaels <.....oricomKILLspam.....USWEST.NET> >To: <EraseMEPICLISTspam_OUTTakeThisOuTMITVMA.MIT.EDU> >Sent: Monday, October 30, 2000 9:43 PM >Subject: Re: [PIC]: PIC SBC requirement > > >> Bob Ammerman wrote: >> > >> >RS232 in and out (bit bang ok). TTL levels _not_ OK. MAX232 or similar >> >greatly preferred for robustness. >> > >> >> Hi Bob, >> >> Certainly you are correct about using TTL levels for RS-232 >> - ie, 0-5v. But I was wondering about your feelings regarding >> bjt RS-232 ports that produce nominal +5v and -12v. Personally, >> I think having minus-going is pretty crucial. Not too sure you >> gain a lot using +12 over +5v, however, except possibly for >> long wires. >> >> Big advantage of the bjt ports is low-quiescent power reqs. >> I have played with these quite a bit and have always had very >> good luck - using cables to 12 feet and baudrates to 115K. >> >> - danM >> >> -- >> http://www.piclist.com hint: The list server can filter out subtopics >> (like ads or off topics) for you. See http://www.piclist.com/#topics >> >> >> >> > >-- >http://www.piclist.com hint: The PICList is archived three different >ways. See http://www.piclist.com/#archives for details. > > > > > > -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.