Post by thread:[PIC]: Problems with timing

'[PIC]: Problems with timing'

Jinx wrote:

> MOQ for SMT 10F is a 3000-lot reel. Not negotiable, will not break

> a reel, already asked. And it's the same price as the DIP - NZ$0.91

What about samples directly from Microchip?  Samples for the 10F206 have

been flowing well for at least a month now.

*****************************************************************

Embed Inc, embedded system specialists in Littleton Massachusetts

(978) 742-9014, http://www.embedinc.com

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Joe Jansen wrote:

> Also, before posting I made sure that there were no errors,

> warnings, or even messages from mpasm when I compiled inside the IDE

> (MPLAB 6.60), so contrary to suggestion, compilation was not an issue.

Then you must have changed the code before sending it to the list.  The code

you posted have everything starting in column 1, which is a bad idea, makes

it hard to read, and will produce lots of warnings from MPASM.

*****************************************************************

Embed Inc, embedded system specialists in Littleton Massachusetts

(978) 742-9014, http://www.embedinc.com

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Maybe it is a mail reader issue then.  Using gmail, and looking back

in the thread, the only things I see in column 1 are the #include, the

#defines, and the jump labels.

Agree that using no whitespace is a PITA to read, and can see how it

would look sloppy if seen that way.  All I can offer is that it was/is

indented in the code, and the reader I am using shows it that way in

the thread.  Apologies if the cut and paste into the gmail compose

window somehow messed that up...\

--Joe

On Sun, 5 Sep 2004 10:52:12 -0400, Olin Lathrop

<spam_OUTolin_piclistTakeThisOuTembedinc.com> wrote:

{Quote hidden}> Joe Jansen wrote:

> > Also, before posting I made sure that there were no errors,

> > warnings, or even messages from mpasm when I compiled inside the IDE

> > (MPLAB 6.60), so contrary to suggestion, compilation was not an issue.

> 

> Then you must have changed the code before sending it to the list.  The code

> you posted have everything starting in column 1, which is a bad idea, makes

> it hard to read, and will produce lots of warnings from MPASM.

> 

> 

> 

> 

> *****************************************************************

> Embed Inc, embedded system specialists in Littleton Massachusetts

> (978) 742-9014, http://www.embedinc.com

> _______________________________________________

> http://www.piclist.com

> View/change your membership options at

> http://mailman.mit.edu/mailman/listinfo/piclist

>

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Joe Jansen wrote:

> Maybe it is a mail reader issue then.  Using gmail, and looking back

> in the thread, the only things I see in column 1 are the #include, the

> #defines, and the jump labels.

>

> Agree that using no whitespace is a PITA to read, and can see how it

> would look sloppy if seen that way.  All I can offer is that it was/is

> indented in the code, and the reader I am using shows it that way in

> the thread.  Apologies if the cut and paste into the gmail compose

> window somehow messed that up...\

Maybe you used tabs instead of hard spaces?  Handling of TABs varies so

widely that there is no guarantee what a file will look like on someone

else's system.  I just checked in the archives, and it seems you indeed used

TAB characters instead of hard spaces.  It looks a bit messy there too, this

time because sometimes the parameters are way off to the right, but maybe

that's how the tabs in Internet Explorer happened to be set on my system.

It certainly looks different than how Outlook Express displayed your

message, although I've never explicitly set tab stops in either.

When are people finally going to learn DON'T PUT TABS IN SOURCE FILES!?

By the way, I've got a program that fixes up the formatting of MPASM files.

It aligns labels, opcodes, operands, and comments in specific columns and

always writes hard spaces for whitespace whether the original file contained

TABs or not.  It's called ASPIC_FIX and is available for free as part of the

PIC development environment at http://www.embedinc.com/pic.

*****************************************************************

Embed Inc, embedded system specialists in Littleton Massachusetts

(978) 742-9014, http://www.embedinc.com

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

> What about samples directly from Microchip?  Samples for the

> 10F206 have been flowing well for at least a month now

I'm interested in the 10F, but my work diary is fit to burst as it is.

For the moment I'm quite content to know what they do, how they

do it, and that they're out there if I need them. Maybe sometime

in the future an altruist, list or local, will have small lots of SMT for

sale at a reasonable cost

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Why are you pushing the 10F so hard?  Except for price and size, there is no

reason to use a 10F over teh 12F.  Many people here don't have the equipment

to program a SOT-23 chip.  I haven't even seen the what my picstart needs to

do it.

When you're buying one or five chips, the price differnece between a 10F and

12F629 isn't even worth considering, especially since Microchip doesn't sell

in small quantities.

The size isn't an advantage here either.  Fabrication labs could save money

by not needing through hole components, and they can build a higher board

density, but that doesn't apply here.  A DIP-8 package is small enough to be

dwarfed by whatever is around it in the home-made project.  Why go to the

trouble of dealing with the SOT-23 chip (programming and soldering) when you

won't save any space in the project?

From: "Olin Lathrop" <.....olin_piclistKILLspam@spam@embedinc.com>

Sent: Sunday, September 05, 2004 7:36 AM

> What about samples directly from Microchip?  Samples for the 10F206 have

> been flowing well for at least a month now.

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

pic@canadaspeaks.com wrote:

> Why are you pushing the 10F so hard?

I'm not "pushing" anything.  I only noted that a 10F would be a better fit

than a 12C508A for the task described by the OP.  My remaining responses

were in reply to other people saying they were having trouble getting parts.

I've had parts since before Masters, and I'm pretty sure the 10F206 pipeline

has been fully turned on.  The problem is probably due to a lag on the part

of distributers, which is why I recommended contacting Microchip for

samples.

> Except for price and size,

Right, except for the minor detail of price and size.  "... but other than

that, Mrs. Lincoln, how did you like the play?"

> Many people here don't

> have the equipment to program a SOT-23 chip.

So far I'm programming them in circuit.  For prototypes I put standard RJ-12

programming connectors on the boards.  In production, there will be some

bare pads that the test fixture will connect to with pogo pins.

> I haven't even seen the what my picstart needs to do it.

There is a SOT-23 ZIF socket and DIP adapter available, but I don't know if

that is supported thru the Picstart+ yet.  I think you need a PicKit1 for

now, but I'm not totally sure about that.

> When you're buying one or five chips, the price differnece between a

> 10F and 12F629 isn't even worth considering,

True, but I don't recall the OP saying what his volumes were.

> The size isn't an advantage here either.  Fabrication labs could save

> money by not needing through hole components, and they can build a

> higher board density, but that doesn't apply here.

Why do you say that?  Maybe I missed something the OP said?

> A DIP-8 package

> is small enough to be dwarfed by whatever is around it in the

> home-made project.

Well, maybe some homemade projects.

*****************************************************************

Embed Inc, embedded system specialists in Littleton Massachusetts

(978) 742-9014, http://www.embedinc.com

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

On Sep 5, 2004, at 4:39 PM, Olin Lathrop wrote:

>

> I'm not "pushing" anything.  I only noted that a 10F would be a better 

> fit

> than a 12C508A for the task described by the OP.

I'm actually moderately excited by the 12F508 and company.  Sounds like 

a good

fit in a lot of places, cheaper and easier to work with than a 12C508.  

Ditto

for the 16F54, 16F57, etc.  An impressively clever marketing move, 

IMO...

BillW

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

On Sun, 2004-09-05 at 19:39, Olin Lathrop wrote:

> > A DIP-8 package

> > is small enough to be dwarfed by whatever is around it in the

> > home-made project.

> 

> Well, maybe some homemade projects.

And, as a hobbyist, let me add that there are DEFINATELY some apps where

the 10F would be a perfect fit for a hobbyist as well. Anything that

needs to be tiny, i.e. something as simple as an RFID type system. TTYL

-----------------------------

Herbert's PIC Stuff:

http://repatch.dyndns.org:8383/pic_stuff/

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

>Try asking for the SOT-23 package.  That's the "real"

>package for the 10F series anyway.  The 10F206 has

>been released in SOT-23 package.  The DIP packages

>are mostly meant for debugging, so there are a lot

>fewer of them around and I don't think they are taken

>very seriously by Microchip.

And to have a handful of DIP ones for sticking in a proto board, put a

SOT-23 onto one of those SO-series to DIP converter boards.

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

> The DIP packages.......I don't think they are taken

> very seriously by Microchip.

The opinion from the purchasing dept at Arrow is that they'd be

surprised if 10F DIP saw the light of day. The point behind the

10F is its size and MC's goal to produce the smallest micro. The

12 series has 8-pin DIPs covered and the 10F is (presently) a lot

less sophisticated than the 12

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Joe,

On Sat, 4 Sep 2004 21:19:48 -0400, Joe Jansen wrote:

> Also will be adding the cap on the power / gnd.  Shoudn't be an issue

> in this case, since I have the power supply with a big filter cap, but

> one never knows.

Well actually there are basically two reasons for capacitors across power supply lines - the one you have 

deals with one, which leaves the other!  :-)

A "big filter cap" looks after things coming in from outside, and helps to smooth out dips in the supply 

voltage due to varying loads in your circuit, but it tends to be some way from the processor (generally next 

to the voltage regulator) and is usually an electrolytic of a good number of uF.

The other thing that needs doing is "decoupling", which is to "short out" noise generated by the processor 

itself, and that needs to be close to the processor (ideally soldered direct to the pins of the processor 

itself but this is rarely practical) and is typically a 0.1uF ceramic.  Without this all sorts of strange 

things can happen, which is why Bob (I think) suggested it.  They used to make IC sockets that had decoupling 

caps built-in across the power supply pins, but these seem to have disappeared and wouldn't have had the right 

pins connected for a PIC anyway, but they were a Good Thing!

Now some purists may disgree completely with my description above, in that it may be over-simplistic (or 

perhaps actually wrong!), but it works for me and I like simple explanations where they work!  :-)

Cheers,

Howard Winter

St.Albans, England

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

Hi, this may seem like a dumb question but are you saying the one cap

have to be connected to power and the other one to ground or actually

one capacitor connected between power and ground?  Sorry..

On Sat, 11 Sep 2004 11:17:26 +0100 (BST), Howard Winter

<hdrwKILLspamh2org.demon.co.uk> wrote:

{Quote hidden}> Joe,

> 

> On Sat, 4 Sep 2004 21:19:48 -0400, Joe Jansen wrote:

> 

> > Also will be adding the cap on the power / gnd.  Shoudn't be an issue

> > in this case, since I have the power supply with a big filter cap, but

> > one never knows.

> 

> Well actually there are basically two reasons for capacitors across power supply lines - the one you have

> deals with one, which leaves the other!  :-)

> 

> A "big filter cap" looks after things coming in from outside, and helps to smooth out dips in the supply

> voltage due to varying loads in your circuit, but it tends to be some way from the processor (generally next

> to the voltage regulator) and is usually an electrolytic of a good number of uF.

> 

> The other thing that needs doing is "decoupling", which is to "short out" noise generated by the processor

> itself, and that needs to be close to the processor (ideally soldered direct to the pins of the processor

> itself but this is rarely practical) and is typically a 0.1uF ceramic.  Without this all sorts of strange

> things can happen, which is why Bob (I think) suggested it.  They used to make IC sockets that had decoupling

> caps built-in across the power supply pins, but these seem to have disappeared and wouldn't have had the right

> pins connected for a PIC anyway, but they were a Good Thing!

> 

> Now some purists may disgree completely with my description above, in that it may be over-simplistic (or

> perhaps actually wrong!), but it works for me and I like simple explanations where they work!  :-)

> 

> Cheers,

> 

> 

> Howard Winter

> St.Albans, England

> 

> 

> 

> 

> _______________________________________________

> http://www.piclist.com

> View/change your membership options at

> http://mailman.mit.edu/mailman/listinfo/piclist

>

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist

On Sat, 11 Sep 2004 23:56:21 +1000, Wan Zulhelmi Wan 

Ahmad Kamar wrote:

> Hi, this may seem like a dumb question but are you 

saying the one cap

> have to be connected to power and the other one to 

ground or actually

> one capacitor connected between power and ground?  

Sorry..

Both the "big filter" and "decoupling" capacitors are 

connected between +V and 0V, but are in different places 

physically (the former close to the power-input, the 

latter as close to the PIC as possible).

Cheers,

Howard Winter

St.Albans, England

_______________________________________________

http://www.piclist.com

View/change your membership options at

http://mailman.mit.edu/mailman/listinfo/piclist