Searching \ for '[PIC] Running into programming walls' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/microchip/devprogs.htm?key=programming
Search entire site for: 'Running into programming walls'.

Exact match. Not showing close matches.
PICList Thread
'[PIC] Running into programming walls'
2005\09\10@230718 by PicDude

flavicon
face
Spent a bunch of time today trying to program a few different devices but ran
into some walls.  First up was a 16F818 which programmed successfully only a
few times, intermittently.  Two fresh F819s did the same.  Google and other
resources told me that the F818/9 are funny animals w.r.t. programming, but
trying all the suggestions I could find, it still did not work.  I tried
these with a quickly-hacked-together JDM clone with IC-Prog, and a commercial
ICD2 clone with MPLAB 7.30.  FWIW, both of those programmers can successfully
program 16F872's and 16F628's, so I trust that the programmers work.

So I took a break from that and switched to another chip -- the 16F88.  With
the ICD2 & MPLAB, I was able to successfully program this device, with MCLRE
on (serving as a reset pin).  When I changed MCLRE to off, programming
failed.  Since that, the programmer fails to recognize the 16F88's device ID
anymore and fails programming.  Google found me some info that indicated
newer PICs can get locked up in this way and un-reprogrammable when both
internal-oscillator and MCLR-as-IO was selected, but I had been using the
external RC oscillator config setting throughout all of this.  I had another
fresh 16F88, but it did the same thing -- after changing the config bits in
the code to use MCLR as a reset pin, it won't re-program.

Arrrggghhh!!! I'm stumped, and frustrated.

So my questions -- if some PIC's require Vdd-before-Vpp, why doesn't the ICD2
control Vdd?  Are the 16F819 and 16F88 really Vdd-before-Vpp devices?  I've
never really read-thru any programming specs before, but maybe I'll check one
out.

Most importantly is what I can do at this point, and is there really a
programmer that properly supports the 16F818 and 16F88?  From some research,
the F88 should be easy, but I'm not sure why it locked up.  The F819 appears
to be problematic, but there must be a supported programming HW/SW combo that
works for this, right?

Cheers,
-Neil.





2005\09\11@024835 by Chen Xiao Fan

face
flavicon
face
I think it is a ICD2 firmware bug (or more kind words --"unimplemented
feature"). It can not program a 12F629/675 with internal MCLR and
internal RC oscillator and I have to use a PICkit 1 to program the
12F629. Now I have both PICkit 1 and Promate III and I of course like
Promate III (US$895) slightly better than PICkit 1 (US$35). ;)-

However we use ICD2 to program 16F819 and got no problems at all but we are
using 20M or 10Mhz ceramic resonator. ICD2 should program 16F819 and
16F88 without any problem, at least if the chip is not on a board. Take
note we are using original ICD2.

Where do you get MPLAB 7.30? MPLAB 7.21 is the latest and I will
recommend Version 7.20 against 7.21 if you do not need the new features
of 7.20.

Check your ICD2 clone again to see if it cut corners in some places.
Wisp628 should support both chips and it is very easy to build. Xwisp2
v1.70 just released will support Windows/Linux and OS/2.


Regards,
Xiaofan

{Original Message removed}

2005\09\11@040503 by PicDude

flavicon
face
On Sunday 11 September 2005 01:48 am, Chen Xiao Fan scribbled:
> I think it is a ICD2 firmware bug (or more kind words --"unimplemented
> feature"). It can not program a 12F629/675 with internal MCLR and
> internal RC oscillator and I have to use a PICkit 1 to program the
> 12F629. Now I have both PICkit 1 and Promate III and I of course like
> Promate III (US$895) slightly better than PICkit 1 (US$35). ;)-

I can program 12F629/12F675's easily with a homebuilt Tait parallel programmer
(<$10) and Odyssey on Linux (my preferred environment).

But the important thing here (going back to the 16F88) is that I was using the
*external* oscillator setting.


> However we use ICD2 to program 16F819 and got no problems at all but we are
> using 20M or 10Mhz ceramic resonator. ICD2 should program 16F819 and
> 16F88 without any problem, at least if the chip is not on a board. Take
> note we are using original ICD2.

The chip is NOT in circuit here either.


> Where do you get MPLAB 7.30? MPLAB 7.21 is the latest and I will
> recommend Version 7.20 against 7.21 if you do not need the new features
> of 7.20.

Oops -- that should've bene 7.20.  IIRC, 7.21 was a development release, which
I usually stay away from.


> Check your ICD2 clone again to see if it cut corners in some places.
> Wisp628 should support both chips and it is very easy to build. Xwisp2
> v1.70 just released will support Windows/Linux and OS/2.

I just got and am still evaluating this clone, which claims to be fully ICD2
compatible.  I went with the ICD2 since it is supposed to support a wide
range of PICs, and I would really like to start using the debugging feature,
even if it means I have to install a Windows OS for now.

Are Microchip's schematics for the ICD2 public anywhere?  I had seen the
original ICD schematics somewhere in the past, but not the ICD2.    BTW, does
your ICD control the Vdd line?  I don't know why, but I find that rather
unusual that it does not.

Cheers,
-Neil.


>
>
> Regards,
> Xiaofan
>
> {Original Message removed}

2005\09\11@042033 by Stef Mientki

flavicon
face
Try it tomorrow again !
I've seen this kind of behaviour with JDM programmer and 16F628,
after reprogramming it for about 10 or 20 times within a few hours,
I had to wait one day to get access to these PICs again.
But I guess I'm about the only one in the whole world who has seen this
fantastic phenomena ;-)
Stef Mientki

PicDude wrote:

{Quote hidden}

2005\09\11@053006 by Chen Xiao Fan

face
flavicon
face
I think I heard those thing among people using JDMs.
JDMs are cheap though. Wisp628 is slightly more expensive
but will still take one day to build. JDM or similar
can be used to bootstrap the Wisp628. Wisp628 is much more
reliable. I do not want to get into wars of programmer
though. So please ignore this if you still prefer to
use JDMs. I have played with PonyProg in year 2002 to
program some EEPROMs and found out it could also program
AVRs and PICs. However I since found it not as good as
those intelligent programmers with MCU inside.

My guess is that Vpp are not really controlled in
most JDM units and it will "soft-destroy" the internal IC
but the chip will recover after sometime. But this
is only a guess. If this is correct, I will guess
Neil's ICD2 clone does not properly regulate Vpp.
The original ICD2 is using a boost converter with
a digital poti to control the output voltage (Vpp)
and the digital poti is not cheap. So most of the
cheaper clones will try to avoid it. Most of
the cheaper clones will probably avoid the boost
converter as well.

Microchip will probably not release ICD2 schematics
to the public but it only takes some time to
draw the schematics if one got the official ICD2.

ICd2 should be able to control the sequence of
Vdd and Vpp. That is why I think it is rather a software
bug. Still I could be wrong.

ICd2 debugging feature is nice to have. This is the thing
Microchip is not doing too well to the open source
communuity. I've raised the question in the Microchip
forum but the moderators choose to ignore.

http://forum.microchip.com/tm.asp?m=110321

I may want to ask their higher level VPs later.
Maybe send an email to the CEO. Just kidding. ;)-
It is not so easy for big companies to support
Linux because of the supporting issues.

Regards,
Xiaofan

{Original Message removed}

2005\09\11@055219 by Wouter van Ooijen

face picon face
> Microchip will probably not release ICD2 schematics
> to the public but it only takes some time to
> draw the schematics if one got the official ICD2.

IIRC it is in the documentation!

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2005\09\11@060456 by Jan-Erik Soderholm

face picon face
HI.
Just so there isn't any missunderstandings...

Chen Xiao Fan wrote :

> My guess is that Vpp are not really controlled in
> most JDM units...

Since you mentioned the Wisp628 in the paragraph before,

Note that the Wisp628 doesn't do that either "as-is".
It needs a few extra components (as outlined on Wouters
Wisp628 page), and then it simple shorts Vdd to GND
a short period, which in itself puts some constraints
on the power supply (no PSU with "soft-start-", not
to many uF's on-baord). A 7805 based PSU with a few
10's of uF has usualy worked for me.

I've been re-selling Wisp628's in Sweden for a couple
of months, and I've had *three* customers locking themselfs
out from 12Fxxx since then. All of them using the blink-a-led
test program (using intosc and int-MCLR) from Wouters page...)
And since these are all PIC-beginners, well, it's not the
best start they could get... :-)


Jan-Erik.



2005\09\11@064333 by Xiaofan Chen

face picon face
No it is not!

On 9/11/05, Wouter van Ooijen <spam_OUTwouterTakeThisOuTspamvoti.nl> wrote:
>
> > Microchip will probably not release ICD2 schematics
> > to the public but it only takes some time to
> > draw the schematics if one got the official ICD2.
>
> IIRC it is in the documentation!
>
> Wouter van Ooijen
>
> -- -------------------------------------------
> Van Ooijen Technische Informatica: http://www.voti.nl <http://www.voti.nl>
> consultancy, development, PICmicro products
> docent Hogeschool van Utrecht: http://www.voti.nl/hvu <http://www.voti.nl/hvu>
>
>
>

2005\09\11@064823 by Xiaofan Chen

face picon face
I am talking about **Vpp**. What I mean is the Vpp voltage is rather constant for Wisp628 yet those JDMs will probably have big variations.
As for the Vpp/Vpp sequence, it is not probably not so easy for ICSP programmers if no Vdd control is possible. However ICD2 does not
even properly program the standalone 12F629 with internal MCLR and RC oscillator configuration.
Regards,
Xiaofan

On 9/11/05, Jan-Erik Soderholm <.....jan-erik.soderholmKILLspamspam@spam@telia.com> wrote:
{Quote hidden}

>

2005\09\11@064925 by Xiaofan Chen

face picon face
Sorry I hit the send button too fast.
"As for the Vpp/Vpp sequence" should be "As for Vpp/Vdd sequence control"

On 9/11/05, Xiaofan Chen <xiaofancspamKILLspamgmail.com> wrote:
{Quote hidden}

> > --

2005\09\11@090617 by Gerhard Fiedler

picon face
Chen Xiao Fan wrote:

>> does your ICD control the Vdd line?  I don't know why, but I find that
>> rather unusual that it does not.

> ICd2 should be able to control the sequence of Vdd and Vpp.

I'm not sure how it would do that if set to external Vdd. I regularly
program boards with the ICD2 where the PIC is powered from the board. The
ICD2 simply can't control Vdd in this setting.


FWIW, we never had problems with programming the 16LF88 (external -MCLR)
with the ICD2 in-circuit. We did have problems using a clone sold here in
Brazil in this application, though.

Gerhard

2005\09\11@092548 by olin piclist

face picon face
PicDude wrote:
> So I took a break from that and switched to another chip -- the
> 16F88.  With the ICD2 & MPLAB, I was able to successfully program
> this device, with MCLRE on (serving as a reset pin).  When I changed
> MCLRE to off, programming failed.  Since that, the programmer fails
> to recognize the 16F88's device ID anymore and fails programming.
> Google found me some info that indicated newer PICs can get locked up
> in this way and un-reprogrammable when both internal-oscillator and
> MCLR-as-IO was selected, but I had been using the external RC
> oscillator config setting throughout all of this.

Google is only partially right.  If you programmer doesn't pay attention to
Vdd and Vpp order and sometimes Vpp rise time, then this can happen.  I just
checked the 16F88 and 16F818/819 programming specs, and both of them require
Vdd before Vpp and a fast rise time on Vpp.

It is always possible to completely erase and reprogram any properly
functioning flash PIC regardless of the state it was previously programmed
in.  There are usually two problems when this isn't working: The programmer
doesn't perform the correct sequence or the external circuit is interfering
in the case of in-circuit programming.

> Are the 16F819 and 16F88 really Vdd-before-Vpp
> devices?

Yes.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

2005\09\11@093611 by Xiaofan Chen

face picon face
Oh yes you are right. If the PIC is powered from the board, then one has
to control the Vdd from the board (switch off the power) so it may not
be possible at all. This is true for any programmers. I think Wouter
has a good explanation last time. For example, PICkit 1 is able to
program the 12F629 with internal RC oscillator and internal MCLR
because it also control the target power. It is a programmer cum
target board.

However ICD2 does not even program a standalone 12F629! We need to
check the hardware ICD2 in order to see if it is really a hardware
limitation or a firmware bug for the original ICD2. For some cheap
clones, this may really be a hardware limitation.

Regards,
Xiaofan

>From ICD2 help file:

ICDWarn0033: You have selected Internal MCLR and Internal Oscillator
in your configuration settings. If your code makes use of port pins
that correspond to Clock and Data pins in programming mode, you may
not be able to reprogram your device. See on-line help for this
warning for more information. (OK/Cancel)

When Internal MCLR is used with MPLAB ICD 2 for programming, both Vpp
and Vdd are powered together, and then Vpp is pulled high to Vihh to
enter programming mode. This means that your code will be running
before Vpp goes to Vihh. If that code makes use of port pins that
correspond to Clock and Data pins in programming mode, there is a
chance their values may not be 0, as necessary to enter programming
mode. Therefore, the device could not be reprogrammed.

Click OK to continue programming or click Cancel to cancel the
programming operation.

Work-Around

When External MCLR is used, this is not a problem, as Vpp can go
directly to Vihh. Also, if External Oscillator is used, the external
oscillator can be kept from running, thus keeping the code from
running, until Vpp is at Vihh.




On 9/11/05, Gerhard Fiedler <EraseMElistsspam_OUTspamTakeThisOuTconnectionbrazil.com> wrote:
{Quote hidden}

2005\09\11@120942 by PicDude

flavicon
face
I had hacked together the JDM as an experiment.  I should be able to program
both chips on the ICD2 though.

Cheers,
-Neil.

On Sunday 11 September 2005 03:20 am, Stef Mientki scribbled:
> Try it tomorrow again !
> I've seen this kind of behaviour with JDM programmer and 16F628,
> after reprogramming it for about 10 or 20 times within a few hours,
> I had to wait one day to get access to these PICs again.
> But I guess I'm about the only one in the whole world who has seen this
> fantastic phenomena ;-)
> Stef Mientki


2005\09\11@125051 by PicDude

flavicon
face
On Sunday 11 September 2005 04:30 am, Chen Xiao Fan scribbled:
> I think I heard those thing among people using JDMs.
> JDMs are cheap though. Wisp628 is slightly more expensive
> but will still take one day to build. JDM or similar
> can be used to bootstrap the Wisp628. Wisp628 is much more
> reliable. I do not want to get into wars of programmer
> though. So please ignore this if you still prefer to
> use JDMs. I have played with PonyProg in year 2002 to
> program some EEPROMs and found out it could also program
> AVRs and PICs. However I since found it not as good as
> those intelligent programmers with MCU inside.

No, I don't prefer to use the JDM -- it was a quick hack-together for testing.  
I prefer to purchase a decent supported programmer, but did not want to spend
$160 on it (for the original ICD2).  And when I find one that works properly,
I intend to pick up a second one, as I always like to have a backup device.  
I had thought of getting Olin's easyprog ($59 each if I build it myself), but
it did not list the 16F818/819 as supported devices, and I prefer Linux.  I
tried the ICD2 clone because it was very low cost, and I am planning to
experiment with LPLAB under Linux to use it.

I did find Rob's webpage with the updated Linux Wisp software, but Wouter's
page still does not list the F818/F819 as supported devices.  It's not that
new, but there must be a reason that these programmers have avoided these
devices. ???


{Quote hidden}

Correct -- this ICD2 uses an offset 78L12 to generate the Vpp.


> ...
> ICd2 should be able to control the sequence of
> Vdd and Vpp. That is why I think it is rather a software
> bug. Still I could be wrong.
> ...

To clarify then, your original ICD2 does have the ability to control Vdd,
right?


Cheers,
-Neil.


2005\09\11@130215 by Jims Email
picon face
PICDude,

You might want to consider the BASICMICRO ISPPro.  It'd a good programmer,
good support, low cost, and small size.   An excellent buy.

                                           Regards,

                                           Jim
{Original Message removed}

2005\09\11@130946 by PicDude

flavicon
face
I've not been able to find this anywhere.

Cheers,
-Neil.

On Sunday 11 September 2005 04:52 am, Wouter van Ooijen scribbled:
{Quote hidden}

2005\09\11@131204 by PicDude

flavicon
face
On Sunday 11 September 2005 05:04 am, Jan-Erik Soderholm scribbled:
> Note that the Wisp628 doesn't do that either "as-is".
> It needs a few extra components (as outlined on Wouters
> Wisp628 page), and then it simple shorts Vdd to GND
> a short period, which in itself puts some constraints
> on the power supply (no PSU with "soft-start-", not
> to many uF's on-baord). A 7805 based PSU with a few
> 10's of uF has usualy worked for me.

I was just looking at that and I'm not a fan of shorting out the PS.  Surely
there must be a cleaner way to do this, by intercepting the Vdd line to the
target?

Cheers,
-Neil.


2005\09\11@131426 by PicDude

flavicon
face
On Sunday 11 September 2005 05:48 am, Xiaofan Chen scribbled:
> I am talking about **Vpp**. What I mean is the Vpp voltage is rather
> constant
> for Wisp628 yet those JDMs will probably have big variations.

When I was experimenting with this yesterday, I saw 11.8V on this line when on
the laptop, and 11.9 from a desktop PC.  The ICD2, btw, shows 12.7V on this
line.

Cheers,
-Neil.


2005\09\11@131710 by PicDude

flavicon
face
On Sunday 11 September 2005 08:06 am, Gerhard Fiedler scribbled:
> FWIW, we never had problems with programming the 16LF88 (external -MCLR)
> with the ICD2 in-circuit. We did have problems using a clone sold here in
> Brazil in this application, though.
>
> Gerhard

Have you any idea what the differences were?  Was it that the original ICD2
has Vdd control while the  clone did not?  Or was there a difference with the
Vpp levels?

Cheers
-Neil.


2005\09\11@132619 by Jan-Erik Soderholm

face picon face
PicDude wrote :

> On Sunday 11 September 2005 05:04 am, Jan-Erik Soderholm scribbled:

> > Note that the Wisp628 doesn't do that either "as-is".
> > It needs a few extra components (as outlined on Wouters
> > Wisp628 page), and then it simple shorts Vdd to GND
> > a short period, which in itself puts some constraints
> > on the power supply (no PSU with "soft-start-", not
> > to many uF's on-baord). A 7805 based PSU with a few
> > 10's of uF has usualy worked for me.
>
> I was just looking at that and I'm not a fan of shorting out
> the PS.  Surely there must be a cleaner way to do this, by
> intercepting the Vdd line to the target?

Probably, but then it's not something you simply can
put in between the Wisp628 and the target as a "dongle"
on the cable, AFAIK...

Jan-Erik.



2005\09\11@132857 by PicDude

flavicon
face
On Sunday 11 September 2005 08:25 am, Olin Lathrop scribbled:
> Google is only partially right.  If you programmer doesn't pay attention to
> Vdd and Vpp order and sometimes Vpp rise time, then this can happen.  I
> just checked the 16F88 and 16F818/819 programming specs, and both of them
> require Vdd before Vpp and a fast rise time on Vpp.

But does Vdd-before-Vpp also imply a max time between these 2 events?  Or can
Vdd be on for many seconds before Vpp comes on?


> It is always possible to completely erase and reprogram any properly
> functioning flash PIC regardless of the state it was previously programmed
> in.  There are usually two problems when this isn't working: The programmer
> doesn't perform the correct sequence or the external circuit is interfering
> in the case of in-circuit programming.

Exactamundo, and I'm would expect MPLAB to perform the correct sequence, which
is why I'm trying to figure out the specifics of the various hardware
options.  But I've run into much information that indicates problems and
"finnicky" behaviour with programming the 16F818/9.

Cheers,
-Neil.


2005\09\11@134737 by Wouter van Ooijen

face picon face
> I was just looking at that and I'm not a fan of shorting out
> the PS.  Surely
> there must be a cleaner way to do this, by intercepting the
> Vdd line to the
> target?

When your target is 7805/78L05 powered I see no problem.

Intercepting the power an switching it off (and shorting any remaining
charge!) might be cleaner, but is not as easy to add to a target
circuit.

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2005\09\11@134738 by Wouter van Ooijen

face picon face
> I've not been able to find this anywhere.

I recall having seen a full circuit diagram somewhere, but it was
probably on a reverse-engineering webpage.

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2005\09\11@134739 by Wouter van Ooijen

face picon face
> I did find Rob's webpage with the updated Linux Wisp
> software, but Wouter's
> page still does not list the F818/F819 as supported devices.  
> It's not that
> new, but there must be a reason that these programmers have
> avoided these
> devices. ???

sheer lazyness (in updating the webpage). wisp628 supports the 818/819,
either with the xwisp on my website or if not I'll send you the latest
xwisp.py which surely does.

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2005\09\11@141233 by olin piclist

face picon face
PicDude wrote:
> I did find Rob's webpage with the updated Linux Wisp software, but
> Wouter's page still does not list the F818/F819 as supported devices.
> It's not that new, but there must be a reason that these programmers
> have avoided these devices. ???

There are a lot of devices to support, so you start with the most popular
ones and work down.  There has been no previous request for 16F818/819
support, but several for dsPICs and 16F88.  Therefore those will get done
first.  I'm not even sure I have samples of the 16F818/819 waiting for me to
implement programmer support.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

2005\09\11@141622 by olin piclist

face picon face
PicDude wrote:
> But does Vdd-before-Vpp also imply a max time between these 2 events?
> Or can Vdd be on for many seconds before Vpp comes on?

Usually not, but in this case there was a note about that.  There was also
the usual note about raising MCLR from 0 to Vpp quickly.  I'm not totally
sure the Vdd to Vpp time restriction applies if MCLR is raised quickly.  You
could look at the spec and decide for yourself.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

2005\09\11@153657 by PicDude

flavicon
face
On Sunday 11 September 2005 12:47 pm, Wouter van Ooijen scribbled:
> > I was just looking at that and I'm not a fan of shorting out
> > the PS.  Surely
> > there must be a cleaner way to do this, by intercepting the
> > Vdd line to the
> > target?
>
> When your target is 7805/78L05 powered I see no problem.
>
> Intercepting the power an switching it off (and shorting any remaining
> charge!) might be cleaner, but is not as easy to add to a target
> circuit.
>
> Wouter van Ooijen

I'm thinking out-of-circuit programming, where if Vdd is supplied by the
programmer, it should be simple to cut off/block.

Cheers,
-Neil.


2005\09\11@153957 by PicDude

flavicon
face
On Sunday 11 September 2005 12:47 pm, Wouter van Ooijen scribbled:
> sheer lazyness (in updating the webpage). wisp628 supports the 818/819,
> either with the xwisp on my website or if not I'll send you the latest
> xwisp.py which surely does.
>
> Wouter van Ooijen


Don't know if you remember, but a year or 2 ago I built a Wisp628 and had some
issues, which IIRC was mostly due to the software not being fully mature at
the time and also w.r.t. Python version conflicts.  I was looking at Rob's
Wisp SW, and think I'll try that.  But I have to find the old Wisp, and
wondering which of many boxes it might be in.

Cheers,
-Neil.


2005\09\11@154348 by PicDude

flavicon
face
On Sunday 11 September 2005 01:12 pm, Olin Lathrop scribbled:
> There are a lot of devices to support, so you start with the most popular
> ones and work down.  There has been no previous request for 16F818/819
> support, but several for dsPICs and 16F88.  Therefore those will get done
> first.  I'm not even sure I have samples of the 16F818/819 waiting for me
> to implement programmer support.


I'd be quite thrilled if I could get the 16F88 working for now, since it will
serve the same purpose for me as the F818/9 -- pretty much a 18-pin PIC with
internal oscillator and A/D.  I'm not really using any other peripherals, and
code size is only ~400 words.  On the PCB, I can easily jumper an MCLR
resistor and keep that as an MCLR pin.

Cheers,
-Neil.


2005\09\11@160422 by Wouter van Ooijen

face picon face
> I'm thinking out-of-circuit programming, where if Vdd is
> supplied by the
> programmer, it should be simple to cut off/block.

Wisp628 is specifically designed as an in-circuit programmer.

But if you want ex-circuit programming with a Wisp628, just add the
dongle circuit, and supply power using separate 78L05 's for the progger
and the target.

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2005\09\11@160422 by Wouter van Ooijen

face picon face
> Don't know if you remember, but a year or 2 ago I built a
> Wisp628 and had some
> issues,

Sorry, I get many Wisp628 issues, questions, etc (you'd be amazed what
people can do wrong!) . And my memory for names is not exactly famous :(

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: http://www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: http://www.voti.nl/hvu


2005\09\11@165112 by Gerhard Fiedler

picon face
PicDude wrote:

>> FWIW, we never had problems with programming the 16LF88 (external -MCLR)
>> with the ICD2 in-circuit. We did have problems using a clone sold here in
>> Brazil in this application, though.
>
> Have you any idea what the differences were?  Was it that the original ICD2
> has Vdd control while the  clone did not?  Or was there a difference with the
> Vpp levels?

No, sorry, I don't know what is different. The clone worked intermittently,
and the ICD2 worked without any problems ever. (This was an ICSP
application.) BTW, this is the clone I'm talking about:
http://www.angelfire.com/linux/kitpic/

I'm pretty sure that the ICD2 does not generally use Vdd control, because
many of my boards I can program both with the ICD2 providing Vdd and the
board providing Vdd. Doesn't seem to make a difference.

Gerhard

2005\09\12@022715 by Chen Xiao Fan

face
flavicon
face
I understand that Wisp628 is specifically designed as an
in-circuit programmer. Still maybe you want to control the
Vpp-target and Vdd-target from Wisp648/Wisp2550
(not yet done, I know) for off-line (ex-circuit) programming.

"To supply power using separate 78L05 for the programmer and
the target" may not be the right answer since people still need
to control the sequence of the target power and the target Vpp.

PICkit 1 and PICkit 2 handles this slightly better when the chip
is off-line since it can control the sequence of Vdd and Vpp.
For Wisp628, the Vdd is always there since Vpp is generated from
Vdd. It would be better to add a line to control the Vdd sequence.
One more transistor and several resistors and a slightly change of
the firmware are needed. Take note PICkit 2 (as well as BFMP)
is also designed to be an ICSP programmer.

If the chip is on board and the target is already power up, then
there is no elegant solutions for any programmers. One have to
power down the chip first. Wouter's idea to short the power
supply may be okay for some cases but not universally applicable.

Regards,
Xiaofan

{Original Message removed}

2005\09\12@105033 by alan smith

picon face
For what its worth, I just finished a project last month that I started with an F818...moved to F819 for more code space, never a problem programming them in circuit using the ICD2.

Chip was powered from the target.

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around
http://mail.yahoo.com

More... (looser matching)
- Last day of these posts
- In 2005 , 2006 only
- Today
- New search...