Post by thread:[PIC]: analog pins

Hi all, [A bit sheepishly this time :-)] I though if an anlog pin was set as an output you could use it as an output pin capable of driving whatever. On the PORTA block diagram, I can see how reading the pin is disabled via an AND gate, but no mention is given about driving the pin hi or lo. I just spent a day fooling around with a new prototype PCB for the PicPocket programmer and I could not get the I2C eeproms to work. (Of course I blamed the PCB) I had RA3 controlling the SCL line, and analog enabled with RA0, RA1 and RA3 set as analog. (ADCON1 set as 04h) RA3 was set as an output, not as an input. This did not work. I turned off the A2D and it did work. I've now swapped the SCL pin to RC0 and all works OK with A2D enabled. Anyone know why? -- Best regards Tony ICmicro's http://www.picnpoke.com spam_OUTsalesTakeThisOuTpicnpoke.com -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

Hi Tony, I've used a 16C73A in this mode. The key for me was that you can't use any instruction that does a read-modify-write. In other words, if you write to port a, no problem. But if you do any bit write operations ON ANY PIN of port a, it will clear your "analog" pin. So always use a shadow register. I was using it for the clock line of a 93LC46A and it worked reliably as long as I did what I described. Cheerful regards, Bob > I though if an anlog pin was set as an output you could use it as an > output pin capable of driving whatever. -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

Tony Nixon <.....PICLISTKILLspam@spam@MITVMA.MIT.EDU> wrote: > I though if an anlog pin was set as an output you could use it as > an output pin capable of driving whatever. That's true, Tony. {Quote hidden}> I could not get the I2C eeproms to work. > .... > I had RA3 controlling the SCL line, and analog enabled with RA0, > RA1 and RA3 set as analog. (ADCON1 set as 04h) RA3 was set as an > output, not as an input. > > This did not work. I turned off the A2D and it did work. > > I've now swapped the SCL pin to RC0 and all works OK with A2D enabled. > > Anyone know why? If a pin is configured as an Analog Input, its digital state can NEVER be read... Not by a "MOVF PORTA,W", not by a "BTFSS PORTA,3", and not by the implied reads of read-modify-write instructions like "XORWF PORTA". Whenever you try to read the pin's digital state, you'll always see a "0". What you may not know is that "BSF" and "BCF" are read-modify-write instructions, too; they work by reading the entire port, setting or clearing one bit, then writing the entire port back. Check this out: ; ADCON1 = 0x04, TRISA = xxxx0000 MOVLW 00001000B ;RA3 = 1. MOVWF PORTA ; BSF PORTA,2 ;Read PORTA, set bit 2, then write ;it back out. In the process, RA3 ;is inadvertently cleared. After that last line executes, RA2 will be set but RA3 will be CLEAR: The BSF instruction reads a "0" in bit 3, then writes that "0" back out after setting bit 2. The only safe way to write to RA3 is via instructions which don't read the port first... In actual practice, this usually means that you do all your read-modify-write operations on a "shadow" register in RAM, then just copy that register to PORTA when needed. -Andy === Andrew Warren --- aiwKILLspamcypress.com === Cypress Semiconductor Corporation === Interface Products Division, S.D. === === The opinions expressed above do === not necessarily represent those of === Cypress Semiconductor Corporation. -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

Bob Blick wrote: > > Hi Tony, > > I've used a 16C73A in this mode. The key for me was that you can't use any > instruction that does a read-modify-write. In other words, if you write to > port a, no problem. But if you do any bit write operations ON ANY PIN of > port a, it will clear your "analog" pin. So always use a shadow register. > > I was using it for the clock line of a 93LC46A and it worked reliably as > long as I did what I described. > > Cheerful regards, > > Bob Hi Bob, That would explain it - thanks. Analog pins always read as '0', so '0' gets written back regardless of the output state, and I was using BSF, BCF. -- Best regards Tony mICro's http://www.picnpoke.com .....salesKILLspam.....picnpoke.com -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

Tony Nixon wrote: ......... >Analog pins always read as '0', so '0' gets written back regardless of >the output state, and I was using BSF, BCF. > Hi Tony, I think this is not exactly correct. If you only exercise a single pin uisng BSF/BCF, it will work correctly, but if you do successive BSF/BCFs to > 1 PortA pins, then only the last will be operated upon corrrectly, and all others will always end up being cleared. BSF PORTA, 0 BSF PORTA, 1 BSF PORTA, 2 will leave the port in the following state: xxxx'x100 [either that or it's naptime]. cheers, - Dan Michaels -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

Hi Tony. Just a thought I was playing about with this and it may of help to you its based on a summing amp circuit. port x0 via pot to opamp input (1) summing amp. set for low voltage i.e.. 3v7 port x1 via pot to opamp input (2) summing amp. set for high voltage i.e.. 6v7 pot (3) to set working voltage input (3) summing amp set at 5v0 you can then get 4 voltages using 2 output pins and use an AD pin to check voltage. I have just started on this but I have only got it down on paper, beats the last on I was working on it used 8 output ports to AD converter.ect .... Regards Art {Original Message removed}

Dan Michaels <EraseMEPICLISTspam_OUTTakeThisOuTMITVMA.MIT.EDU> wrote: > > Analog pins always read as '0', so '0' gets written back > > regardless of the output state, and I was using BSF, BCF. > > I think this is not exactly correct. No, it is exactly correct. > If you only exercise a single pin uisng BSF/BCF, it will work > correctly.... But it will have side-effects on any other pins within the port which are configured as analog inputs and digital outputs. > .... but if you do successive BSF/BCFs to > 1 PortA pins, then only > the last will be operated upon corrrectly, and all others will > always end up being cleared. Not true. When that happens, it's because of the circuitry OUTSIDE the PIC, not because of any limitation inside the PIC. > BSF PORTA, 0 > BSF PORTA, 1 > BSF PORTA, 2 > > will leave the port in the following state: xxxx'x100 Not if RA0, RA1, and RA2 are disconnected from any outside circuitry... Or if the PIC is running at a low clock speed. -Andy === Andrew Warren -- aiwspam_OUTcypress.com === Staff Applications Engineer === Interface Products Division === Cypress Semiconductor === === for tech support: === @spam@usbappsKILLspamcypress.com -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

Andy, I was referring to the situation where you are attempting to use an analog-configured pin as a digital output, which is what I thought Tony was asking about ??????????? BTW, tell me why I can't ever get my hands on any Cypress FIFOs. Nobody ever seems to have them. Has Cypress stopped manufacturing them? - danM ====================== {Quote hidden}>Dan Michaels <KILLspamPICLISTKILLspamMITVMA.MIT.EDU> wrote: > >> > Analog pins always read as '0', so '0' gets written back >> > regardless of the output state, and I was using BSF, BCF. >> >> I think this is not exactly correct. > > No, it is exactly correct. > >> If you only exercise a single pin uisng BSF/BCF, it will work >> correctly.... > > But it will have side-effects on any other pins within the port > which are configured as analog inputs and digital outputs. > >> .... but if you do successive BSF/BCFs to > 1 PortA pins, then only >> the last will be operated upon corrrectly, and all others will >> always end up being cleared. > > Not true. When that happens, it's because of the circuitry > OUTSIDE the PIC, not because of any limitation inside the PIC. > >> BSF PORTA, 0 >> BSF PORTA, 1 >> BSF PORTA, 2 >> >> will leave the port in the following state: xxxx'x100 > > Not if RA0, RA1, and RA2 are disconnected from any outside > circuitry... Or if the PIC is running at a low clock speed. > > -Andy > > >=== Andrew Warren -- RemoveMEaiwTakeThisOuTcypress.com >=== Staff Applications Engineer >=== Interface Products Division >=== Cypress Semiconductor >=== >=== for tech support: >=== spamBeGoneusbappsspamBeGonecypress.com > >-- >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.

----- Original Message ----- From: Dan Michaels <TakeThisOuToricomEraseMEspam_OUTUSWEST.NET> > Hi Tony, I think this is not exactly correct. If you only exercise > a single pin uisng BSF/BCF, it will work correctly, but if you do > successive BSF/BCFs to > 1 PortA pins, then only the last will be > operated upon corrrectly, and all others will always end up being > cleared. > > BSF PORTA, 0 > BSF PORTA, 1 > BSF PORTA, 2 > > will leave the port in the following state: xxxx'x100 > > [either that or it's naptime]. > Dan, Right: Given A,0 A,1 and A,2 defined as analog and as output: BSF PORTA,0 will SET A.0 and CLEAR A.1 and A.2 BSF PORTA,1 will SET A.1 and CLEAR A.0 and A.2 BSF PORTA,2 will SET A.2 and CLEAR A.0 and A.1 So, when you are done you will indeed end up with a zero on A.0 and A.1, but there will have been a nice pulse of 1 in the meanwhile. Now that gets me to thinking: Here is a neat trick to get a 'push-pull' output from two PORT A pins: 1: define them as analog (ADCON1) 2: TRIS them as outputs. Now, when you BSF either of them the other will automatically be cleared. This would have been useful in a system I built that drove a piezo transducer directly off two PIC pins. Bob Ammerman RAm Systems (contract development of high performance, high function, low-level software) -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

Bob Ammerman wrote: {Quote hidden}>----- Original Message ----- >From: Dan Michaels <RemoveMEoricomTakeThisOuTUSWEST.NET> > >> Hi Tony, I think this is not exactly correct. If you only exercise >> a single pin uisng BSF/BCF, it will work correctly, but if you do >> successive BSF/BCFs to > 1 PortA pins, then only the last will be >> operated upon corrrectly, and all others will always end up being >> cleared. >> >> BSF PORTA, 0 >> BSF PORTA, 1 >> BSF PORTA, 2 >> >> will leave the port in the following state: xxxx'x100 >> >> [either that or it's naptime]. >> > >Dan, > >Right: > >Given A,0 A,1 and A,2 defined as analog and as output: > >BSF PORTA,0 will SET A.0 and CLEAR A.1 and A.2 >BSF PORTA,1 will SET A.1 and CLEAR A.0 and A.2 >BSF PORTA,2 will SET A.2 and CLEAR A.0 and A.1 > >So, when you are done you will indeed end up with a zero on A.0 and A.1, but >there will have been a nice pulse of 1 in the meanwhile. > Well Bob, that's 2 of us who seem to agree on this, although AndyW saw it differently. Hmmm, which gets me to thinking - ie, Mchp built some damn many "gotchas" into these thingies that piclist can support 1800 "experts" each with a totally different set of experiences <:-))). ================== {Quote hidden}>Now that gets me to thinking: Here is a neat trick to get a 'push-pull' >output from two PORT A pins: > >1: define them as analog (ADCON1) >2: TRIS them as outputs. > >Now, when you BSF either of them the other will automatically be cleared. > >This would have been useful in a system I built that drove a piezo >transducer directly off two PIC pins. > Guess you could always write xxxx'xx01 and xxxx'xx10, on regular digital pins. And actually, you can drive a piezo quite well using a single pin on a PIC. In fact, as it is, you normally have to use a 100-1000 ohm attenuation R in series, or it will drive you right out of the room. - danM -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

Bob Ammerman <PICLISTEraseME.....MITVMA.MIT.EDU> wrote: > Here is a neat trick to get a 'push-pull' output from two PORT A > pins: > > 1: define them as analog (ADCON1) > 2: TRIS them as outputs. > > Now, when you BSF either of them the other will automatically be > cleared. > > This would have been useful in a system I built that drove a piezo > transducer directly off two PIC pins. Clever; I don't think I've ever seen that. Of course, if you can't configure the pins as analog inputs (because, for instance, the PIC you're using doesn't HAVE an A/D converter), you can do this: MOVLW 00000001B ;RA0 = 1, RA1 = 0. MOVWF PORTA ; MOVLW 00000011B XORWF PORTA ;Toggle RA0 and RA1. XORWF PORTA ;Toggle the two pins again. XORWF PORTA ;And again... Or, if those two pins are the only ones defined as outputs on that port: MOVLW 00000001B ;RA0 = 1, RA1 = 0. MOVWF PORTA ; COMF PORTA ;Toggle RA0 and RA1. COMF PORTA ;Again. COMF PORTA ;And again... -Andy === Andrew Warren - EraseMEfastfwdix.netcom.com === Fast Forward Engineering - San Diego, California === http://www.geocities.com/SiliconValley/2499 -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

Dan Michaels <RemoveMEPICLISTEraseMEEraseMEMITVMA.MIT.EDU> wrote: > I was referring to the situation where you are attempting to use an > analog-configured pin as a digital output, which is what I thought > Tony was asking about ??????????? Dan: That IS what he was asking about, but the problem with analog-in/digital-out pins has nothing to do with "successive" writes; that's an entirely different issue related to capacitance and the fact that I/O Writes happen at the end of an instruction cycle while I/O Reads happen at the beginning of the cycle. Tony's problem will occur even if many minutes elapse between writes to a port. The problem you described will happen whether or not the pins are configured as analog inputs, but only under certain circumstances, and then only if the writes occur in very rapid succession. > BTW, tell me why I can't ever get my hands on any Cypress FIFOs. > Nobody ever seems to have them. Has Cypress stopped manufacturing > them? Don't ask me, man... I just work there. Seriously, though, the entire semiconductor industry is feeling the supply-and-demand squeeze right now; nearly every vendor (with, perhaps, the exception of Microchip Technology) has put its parts on allocation, which means that customers who want only a small quantity of parts may have to wait a LONG time to get them. It's a cyclical phenomenon in the industry; I wish it didn't happen, but the reality is that a new chip fab costs over a billion dollars (WAY over a billion if you're doing anything state-of-the-art like 300-mm wafers or 0.13-micron processes), and a vendor will lose customers if his parts cost just pennies more than someone else's. No vendor, therefore, can afford to have any unused production capacity... And since you can't build half a fab, or just ten percent of a fab, or lease a fab -- not when everyone else is ALSO at full capacity, anyway -- we all do the tightrope- balancing thing, striving for exactly 100% utilization of our fabs at all times. Naturally, each of us occasionally falls off the tightrope. Cypress seems to ba able to meet its customers' demands better than some companies -- not to name names, but have you tried to buy an AVR lately? -- but I'll honestly admit that we're not perfect. All I can suggest is that you place your order early and wait it out. Sorry... -Andy P.S. Of course, the opinions expressed in this email do not necessarily represent those of Cypress Semiconductor Corporation; as I said, I just work there. === Andrew Warren - RemoveMEfastfwdspam_OUTKILLspamix.netcom.com === Fast Forward Engineering - San Diego, California === http://www.geocities.com/SiliconValley/2499 -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

Dan Michaels <RemoveMEPICLISTTakeThisOuTspamMITVMA.MIT.EDU> wrote: > > BSF PORTA,0 will SET A.0 and CLEAR A.1 and A.2 > > BSF PORTA,1 will SET A.1 and CLEAR A.0 and A.2 > > BSF PORTA,2 will SET A.2 and CLEAR A.0 and A.1 > > > > So, when you are done you will indeed end up with a zero on A.0 > > and A.1, but there will have been a nice pulse of 1 in the > > meanwhile. > > Well Bob, that's 2 of us who seem to agree on this, although > AndyW saw it differently. Dan: No, I didn't see it any differently. You described a certain sequence of events and I explained WHY that sequence occurs. I wasn't arguing that it DIDN'T occur; if there was any misunderstanding, it was over your use of the phrase "successive writes to an I/O port", which is the language invariably used to explain ANOTHER issue with PIC I/O pins. > Hmmm, which gets me to thinking - ie, Mchp built some damn many > "gotchas" into these thingies that piclist can support 1800 > "experts" each with a totally different set of experiences > <:-))) Yeah, I guess you're right about the "different set of experiences" thing... I mean, I've been on the PICLIST for five years and I would've SWORN that there were only about 18 of us experts here. -Andy === Andrew Warren - EraseMEfastfwdspamspamBeGoneix.netcom.com === Fast Forward Engineering - San Diego, California === http://www.geocities.com/SiliconValley/2499 -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

> > Hmmm, which gets me to thinking - ie, Mchp built some damn many > > "gotchas" into these thingies that piclist can support 1800 > > "experts" each with a totally different set of experiences > > <:-))) > > Yeah, I guess you're right about the "different set of > experiences" thing... I mean, I've been on the PICLIST for five > years and I would've SWORN that there were only about 18 of us > experts here. > > -Andy > Its easy to reconcile this latest difference in perception: There are 1800 "experts", but only 18 experts (actually probably a few more than that). Bob Ammerman RAm Systems (contract development of high performance, high function, low-level software) -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

> >Now that gets me to thinking: Here is a neat trick to get a 'push-pull' > >output from two PORT A pins: > > > >1: define them as analog (ADCON1) > >2: TRIS them as outputs. > > > >Now, when you BSF either of them the other will automatically be cleared. > > > >This would have been useful in a system I built that drove a piezo > >transducer directly off two PIC pins. > > > > Guess you could always write xxxx'xx01 and xxxx'xx10, on regular > digital pins. Sure: But here I only have to write 1 pin and the other 'magically' switches state. Saves me at least one instruction. Gotta have some use :-) > And actually, you can drive a piezo quite well using > a single pin on a PIC. In fact, as it is, you normally have to use > a 100-1000 ohm attenuation R in series, or it will drive you right > out of the room. Out of the room was the whole idea. Actually, I used this trick to get two volume levels: for key-click prompts and warning tones I held one output fixed and toggled the other. For the alarm output I toggled the two outputs 180 degrees out of phase with each other. Bob Ammerman RAm Systems (contract development of high performance, high function, low-level software) -- http://www.piclist.com hint: The PICList is archived three different ways. See http://www.piclist.com/#archives for details.

Bob Ammerman wrote: ...... >> >> Guess you could always write xxxx'xx01 and xxxx'xx10, on regular >> digital pins. > >Sure: But here I only have to write 1 pin and the other 'magically' switches >state. Saves me at least one instruction. Gotta have some use :-) > Yeah, I know what you're talking about. Just make "sure" you comment this new whixxxbang invention, or 6 months down the road you're gonna be looking at this and scratching your head, and wondering who the heck was the guy what wrote that code :). - danM -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! use RemoveMElistservKILLspammitvma.mit.edu?body=SET%20PICList%20DIGEST

Bob Ammerman wrote: >> > Hmmm, which gets me to thinking - ie, Mchp built some damn many >> > "gotchas" into these thingies that piclist can support 1800 >> > "experts" each with a totally different set of experiences >> > <:-))) >> >> Yeah, I guess you're right about the "different set of >> experiences" thing... I mean, I've been on the PICLIST for five >> years and I would've SWORN that there were only about 18 of us >> experts here. >> >> -Andy >> > >Its easy to reconcile this latest difference in perception: > >There are 1800 "experts", but only 18 experts (actually probably a few more >than that). > Bob gets the "A". -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! use listservSTOPspamspam_OUTmitvma.mit.edu?body=SET%20PICList%20DIGEST

Andy Warren wrote: >Dan Michaels <spamBeGonePICLISTSTOPspamEraseMEMITVMA.MIT.EDU> wrote: > >> I was referring to the situation where you are attempting to use an >> analog-configured pin as a digital output, which is what I thought >> Tony was asking about ??????????? > > Dan: > > That IS what he was asking about, but the problem with > analog-in/digital-out pins has nothing to do with "successive" > writes; that's an entirely different issue related to capacitance > and the fact that I/O Writes happen at the end of an instruction > cycle while I/O Reads happen at the beginning of the cycle. > Hmmmm, as I understand it, pins configured as analog but used as digital outputs always "read" as digital "0". Therefore, successive BSF instructions to a sequence of pins will always clear any pins that were previously set, regardless of what is connected to them externally. I thought this was the whole problem Tony was talking about. No? This is a different situation from where the pins are configured as digital, and where they may have caps hanging on them externally, and where you also use a series of BSF/BCF cmds. Here, the caps will hold the pins temporarily. Been there, done both - been burned every which way. [I think I'll shut-up now]. - danM -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! use KILLspamlistservspamBeGonemitvma.mit.edu?body=SET%20PICList%20DIGEST

----- Original Message ----- From: Dan Michaels <EraseMEoricomEraseMEUSWEST.NET> To: <@spam@PICLIST@spam@spam_OUTMITVMA.MIT.EDU> Sent: Tuesday, October 17, 2000 12:45 AM Subject: Re: [PIC]: analog pins > Bob Ammerman wrote: > ...... > >> > >> Guess you could always write xxxx'xx01 and xxxx'xx10, on regular > >> digital pins. > > > >Sure: But here I only have to write 1 pin and the other 'magically' switches {Quote hidden}> >state. Saves me at least one instruction. Gotta have some use :-) > > > > > Yeah, I know what you're talking about. Just make "sure" you > comment this new whixxxbang invention, or 6 months down the road > you're gonna be looking at this and scratching your head, and > wondering who the heck was the guy what wrote that code :). > > - danM Funny, I almost made a remark to that effect in my posting. Apparently great minds think alike, or else fools seldom differ. 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 spamBeGonelistservKILLspammitvma.mit.edu?body=SET%20PICList%20DIGEST > > > > -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! use .....listservspam_OUTmitvma.mit.edu?body=SET%20PICList%20DIGEST

Dan Michaels <TakeThisOuTPICLIST.....TakeThisOuTMITVMA.MIT.EDU> wrote: > Hmmmm, as I understand it, pins configured as analog but used as > digital outputs always "read" as digital "0". Therefore, > successive BSF instructions to a sequence of pins will always clear > any pins that were previously set, regardless of what is connected > to them externally. I thought this was the whole problem Tony was > talking about. No? Dan: What you say is true, and it IS the problem that Tony was having. However, you wrote an earlier message saying that my explanation -- which was almost word-for-word identical to what you just wrote -- was "not exactly correct". I guess I just don't see where our explanations differ, as we seem to be in violent agreement. > [I think I'll shut-up now]. Me, too. I think the explanation's clear now, and besides, I have a plane to catch in the morning. -Andy === Andrew Warren - TakeThisOuTfastfwdKILLspamspamix.netcom.com === Fast Forward Engineering - San Diego, California === http://www.geocities.com/SiliconValley/2499 -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! use .....listservRemoveMEmitvma.mit.edu?body=SET%20PICList%20DIGEST

>Now that gets me to thinking: Here is a neat trick to get a 'push-pull' >output from two PORT A pins: >1: define them as analog (ADCON1) >2: TRIS them as outputs. >Now, when you BSF either of them the other will automatically be cleared. >This would have been useful in a system I built that drove a piezo >transducer directly off two PIC pins. This also sounds like a real crazy scheme to confuse anyone trying to reverse engineer your code. It would have them going round in circles trying to work out why you are connecting a piezo sounder across analog inputs, and be looking for the interrupt routine that is listening into them.... but then why is the code doing things to the digital outputs????? -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! use RemoveMElistservspamBeGonemitvma.mit.edu?bodyT%20PICList%20DIGEST

> I had RA3 controlling the SCL line, and analog enabled with RA0, RA1 and > RA3 set as analog. (ADCON1 set as 04h) RA3 was set as an output, not as > an input. > > This did not work. I turned off the A2D and it did work. > > I've now swapped the SCL pin to RC0 and all works OK with A2D enabled. > > Anyone know why? The A/D setting overrides the digital useage of a port on the 16 series. You have limited selection of A/D inputs and voltage reference inputs in ADCON1. The real gotcha is that the A/D wakes up enabled on the earlier 16 series parts. ***************************************************************** Olin Lathrop, embedded systems consultant in Devens Massachusetts (978) 772-3129, spamBeGoneolin@spam@spam_OUTcognivis.com, http://www.cognivis.com -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! use TakeThisOuTlistservspammitvma.mit.edu?body=SET%20PICList%20DIGEST

> > I though if an anlog pin was set as an output you could use it as > > an output pin capable of driving whatever. > > That's true, Tony. No, it's not. Some I/O modules will control the pin regardless of the TRIS setting, or otherwise "take over" the operation of the pin. The A/D is a good example of this. ***************************************************************** Olin Lathrop, embedded systems consultant in Devens Massachusetts (978) 772-3129, olinEraseMEcognivis.com, http://www.cognivis.com -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! use RemoveMElistservEraseMEspam_OUTmitvma.mit.edu?body=SET%20PICList%20DIGEST

----- Original Message ----- From: Olin Lathrop <@spam@olin_piclistRemoveMEEraseMECOGNIVIS.COM> To: <EraseMEPICLIST@spam@MITVMA.MIT.EDU> Sent: Tuesday, October 17, 2000 8:25 AM Subject: Re: [PIC]: analog pins > > > I though if an anlog pin was set as an output you could use it as > > > an output pin capable of driving whatever. > > > > That's true, Tony. > > No, it's not. Some I/O modules will control the pin regardless of the TRIS > setting, or otherwise "take over" the operation of the pin. The A/D is a > good example of this. The datasheets I've read have said that you must TRIS analogs as inputs if you want to read the analog voltage. They say that if you TRIS the pin as an output, the A/D can still happen, but that it will convert the pin voltage as driven by the digital output. I stand ready to be corrected. 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@listservspam_OUT.....mitvma.mit.edu?body=SET%20PICList%20DIGEST

> The datasheets I've read have said that you must TRIS analogs as inputs if > you want to read the analog voltage. They say that if you TRIS the pin as an > output, the A/D can still happen, but that it will convert the pin voltage > as driven by the digital output. I stand ready to be corrected. There may be more than one A/D implementation. I remember running into an A/D gotcha on one of my first PIC projects, which might have been a 16C77 or 16C924. I wasn't using analog in, so I ignored the A/D module. What I didn't realize was that the A/D wakes up enabled, and that it was taking over the analog input pins. I was trying to use some of them as digital outputs, but I couldn't get them to do anything even though TRIS and the port register were set correctly. When I finally figured out to disable the A/D module, everything worked the way I intended it. I've also noticed that the A/D is supposed to wake up off now on newer PICs like the 16F87x. Perhaps these A/Ds also have slightly different operation. I sorta remember that you can individually disable the A/D function on select pins by using TRIS on the 17C series. ***************************************************************** Olin Lathrop, embedded systems consultant in Devens Massachusetts (978) 772-3129, spamBeGoneolinEraseMEcognivis.com, http://www.cognivis.com -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! use listservspamBeGonemitvma.mit.edu?body=SET%20PICList%20DIGEST

>> Bob Ammerman wrote: ........... >> >> Yeah, I know what you're talking about. Just make "sure" you >> comment this new whixxxbang invention, or 6 months down the road >> you're gonna be looking at this and scratching your head, and >> wondering who the heck was the guy what wrote that code :). >> >> - danM > >Funny, I almost made a remark to that effect in my posting. Apparently great >minds think alike, or else fools seldom differ. > Or else we were both listening to the recent thread regarding hazards of writing too-clever code and leaving it uncommented. [ie, I know the guy who wrote this code must have been "real" bright, cause I sure can't figure it out :)]. - danM -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! use RemoveMElistserv@spam@spamBeGonemitvma.mit.edu?body=SET%20PICList%20DIGEST

At 09:16 AM 10/17/2000 -0400, you wrote: {Quote hidden}>----- Original Message ----- >From: Olin Lathrop <.....olin_piclist@spam@EraseMECOGNIVIS.COM> >To: <.....PICLISTRemoveMEMITVMA.MIT.EDU> >Sent: Tuesday, October 17, 2000 8:25 AM >Subject: Re: [PIC]: analog pins > > >> > > I though if an anlog pin was set as an output you could use it as >> > > an output pin capable of driving whatever. >> > >> > That's true, Tony. >> >> No, it's not. Some I/O modules will control the pin regardless of the >TRIS >> setting, or otherwise "take over" the operation of the pin. The A/D is a >> good example of this. > >The datasheets I've read have said that you must TRIS analogs as inputs if >you want to read the analog voltage. They say that if you TRIS the pin as an >output, the A/D can still happen, but that it will convert the pin voltage >as driven by the digital output. I stand ready to be corrected. > The following is straight out of my 16C7x d/s. All one has to do is figure out Mchp-engineer-speak. [it's kinda like reading "Brave New World" sometime, or was that "1984"?] Now, as I recall, everything worked as stated here, except for: "Pins configured as digital inputs, will convert an analog input." =========================== The ADCON1, TRISA, and TRISE registers control the operation of the A/D port pins. The port pins that are desired as analog inputs must have their corresponding TRIS bits set (input). If the TRIS bit is cleared (output), the digital output level (VOH or VOL) will be converted. The A/D operation is independent of the state of the CHS2:CHS0 bits and the TRIS bits. Note 1: When reading the port register, all pins configured as analog input channels will read as cleared (a low level). Pins configured as digital inputs, will convert an analog input. Analog levels on a digitally configured input will not affect the conversion accuracy. Note 2: Analog levels on any pin that is defined as a digital input (including the AN7:AN0 pins), may cause the input buffer to consume current that is out of the devices specification. -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! use .....listservSTOPspam@spam@mitvma.mit.edu?body=SET%20PICList%20DIGEST

Olin Lathrop <PICLISTEraseME@spam@MITVMA.MIT.EDU> wrote: > > > I though if an anlog pin was set as an output you could use > > > it as an output pin capable of driving whatever. > > > > That's true, Tony. > > No, it's not. Yes, dammit, it IS. > Some I/O modules will control the pin regardless of the TRIS > setting, or otherwise "take over" the operation of the pin. That's true, too. > The A/D is a good example of this. And that's verifiably false. Read the datasheet, Olin, or just... oh, I don't know... actually TRY it. -Andy === Andrew Warren - RemoveMEfastfwdspamBeGoneix.netcom.com === Fast Forward Engineering - San Diego, California === http://www.geocities.com/SiliconValley/2499 -- http://www.piclist.com#nomail Going offline? Don't AutoReply us! use spamBeGonelistservKILLspam@spam@mitvma.mit.edu?body=SET%20PICList%20DIGEST