Searching \ for '[PIC] PIC18LF45K22 with 1.8V - mission impossible?' 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/devices.htm?key=pic
Search entire site for: 'PIC18LF45K22 with 1.8V - mission impossible?'.

Exact match. Not showing close matches.
PICList Thread
'[PIC] PIC18LF45K22 with 1.8V - mission impossible?'
2011\08\10@113434 by Robert Berger

flavicon
face
Hi,

There seems to be a major design flaw with low voltage processors/tools
from Microchip, since when you use Real-ICE with a PIC18LF45K22@1.8V
it's unable to perform a bulk erase (needs at least 2.7V).

Which means you write code - download it on an erased chip, make a code
change, try to reprogram and it does not work, since the real-ice can
not bulk erase the chip at 1.8V. (needs > 2.7V)

I need 1.8V to interface with a main board, so I'm considering to cook
one board with voltage translators, so I can run the PIC with 3.3 V or
so and everything will (hopefully) work.

.... but like this I could use another chip from a different manufacturer
as well ...

My initial reason to use the PIC18LF45K22 was it's capability to work
with 1.8V.

What's your experience with PICs and development at 1.8V and what's your
work around for the problem above.

Please advise,

Robert..."Hardware, n.: The parts of a computer system that can be
kicked." - unknown

My public pgp key is available at:
pgp.mit.edu:11371/pks/lookup?op=get&search=0x90320BF1

2011\08\10@121736 by Jan-Erik Soderholm

face picon face
Note that this has been like this a long time even
with the "5V-parts" that runs down to 2V-something.

They usualy needs at least 4.5V to "bulk erase". They
*can* erase flash at lower voltages, but not *bulk* erase.

Sometimes it is possible to use other erase methods, but
it is normaly only bulk-erase that clears the CONFIG bits.
And then the only way to un-lock a read-protected chip.

Jan-Erik.


Robert Berger wrote 2011-08-10 17:34:
{Quote hidden}

2011\08\10@125241 by alan.b.pearce

face picon face
> There seems to be a major design flaw with low voltage processors/tools
> from Microchip, since when you use Real-ICE with a PIC18LF45K22@1.8V
> it's unable to perform a bulk erase (needs at least 2.7V).

I think you will find it is the processor is the problem.

....
> What's your experience with PICs and development at 1.8V and what's your
> work around for the problem above.
>
> Please advise,

I think some of the newer PIC24 devices with very low core voltages will do what you want. IIRC the just out PIC24E devices have a 1.somethingV core, and I have a feeling some of the slightly less recent ones do too. The 24FJ64GB004 family is shown to operate to 2V when you run Vdd and Vddcore at the same voltage.
-- Scanned by iCritical.

2011\08\10@153400 by Robert Berger

flavicon
face
Hi,

Thanks for the replies and let me elaborate a bit more on my problem.

What I would like to have is a processor, which can operate at 1.8V and
where I can use real-ice to edit/compile/download/debug, edit/compile,...

Although there seem to be methods for the PIC18LF45K22 to erase flash at
1.8V ( ww1.microchip.com/downloads/en/DeviceDoc/41398B.pdf ) it looks
like real-ice always uses bulk erase, which requires >2.7 V.

It's fine for me to reset the configuration with > 2.7V, but I would
like to develop my code with 1.8V.

I would not mind switching the processor, even to something different
from Microchip, but ideally I'm after a development method where I can
operate the chip at 1.8V and can use edit/compile/download/debug without
applying a higher voltage in between with real-ice.

How are you usually developing when you have such constraints?

Develop with a higher voltage and voltage translators to 1.8V and when
everything works go reduce the voltage to 1.8V so the interface to the
other board can be made without voltage translators?

Regards,

Robert







...."Real Time, adj.: Here and now, as opposed to fake time, which only
occurs there and then."

My public pgp key is available at:
pgp.mit.edu:11371/pks/lookup?op=get&search=0x90320BF1

2011\08\11@044508 by Reinhard Daemon

flavicon
face
Jan-Erik Soderholm <jan-erik.soderholm <at> telia.com> writes:

{Quote hidden}

Hello Jan-Erik,

i am working together with Robert on the project.
Just to make it clear:
The configuration bits are once set and are fine.
We do not alter them, we just update the program code and for
doing so *erase* should be enough ("erase before program" is set in the configuration settings of the programmer in MpLab).

We would just need to reprogram the programm code (not ID, EEPROM, CONFIG,....)
@ 1.8 V with RealICE.
But as Robert already stated, we are only able to programm a fresh chip once
(... but we have chosen a *flash* device and not an OTP, hihi).

reinhard

2011\08\11@050252 by Jan-Erik Soderholm

face picon face

Reinhard Daemon wrote 2011-08-11 10:05:
{Quote hidden}

Yes, but I'm quite sure that what they mean with that is a bulk erase.

What could work is some kind of boot-loader. They never do a bulk-erase
(since that would erase the boot-loader itself) but the page-mode
erases that usualy/sometime works at lower voltages.

So if re-programming at low voltages is a definite need, a boot-loader
could be a solution.

>
> We would just need to reprogram the programm code (not ID, EEPROM, CONFIG,...)
> @ 1.8 V with RealICE.
> But as Robert already stated, we are only able to programm a fresh chip once

Yes, since that is "bulk erased" from factory. It doesn't matter
if *your* bulk erase fails... :-)


> (... but we have chosen a *flash* device and not an OTP, hihi).
>
> reinhard
>
>

2011\08\11@051923 by Xiaofan Chen

face picon face
On Thu, Aug 11, 2011 at 4:05 PM, Reinhard Daemon <spam_OUTr.daemonTakeThisOuTspamtugraz.at> wrote:
> i am working together with Robert on the project.
> Just to make it clear:
> The configuration bits are once set and are fine.
> We do not alter them, we just update the program code and for
> doing so *erase* should be enough ("erase before program" is set in
> the configuration settings of the programmer in MpLab).
>
> We would just need to reprogram the programm code (not ID, EEPROM, CONFIG,...)
> @ 1.8 V with RealICE.
> But as Robert already stated, we are only able to programm a fresh chip once
> (... but we have chosen a *flash* device and not an OTP, hihi).

What if you enable LVP? That might work for things like PICkit 2.
http://ww1.microchip.com/downloads/en/DeviceDoc/PICkit%202%20Readme%20v2-61-00%20%28a%29.txt

Microchip PM3 also supports LVP but that is rather expensive.

The help inside MPLAB seems to suggest that Microchip MPLAB
ICD2, ICD 3, Real ICE and PICkit 3 do not support LVP inside MPLAB
and they recommend to use PM3.

You can try support.microchip.com as well to confirm this.

-- Xiaofa

2011\08\11@052658 by Xiaofan Chen

face picon face
On Thu, Aug 11, 2011 at 5:19 PM, Xiaofan Chen <.....xiaofancKILLspamspam@spam@gmail.com> wrote:
> On Thu, Aug 11, 2011 at 4:05 PM, Reinhard Daemon <r.daemonspamKILLspamtugraz.at> wrote:
>> i am working together with Robert on the project.
>> Just to make it clear:
>> The configuration bits are once set and are fine.
>> We do not alter them, we just update the program code and for
>> doing so *erase* should be enough ("erase before program" is set in
>> the configuration settings of the programmer in MpLab).
>>
>> We would just need to reprogram the programm code (not ID, EEPROM, CONFIG,...)
>> @ 1.8 V with RealICE.
>> But as Robert already stated, we are only able to programm a fresh chip once
>> (... but we have chosen a *flash* device and not an OTP, hihi).

I remember that MPLAB ICD 3 and Real ICE should be able to support
this use case. When the bulk erase is not possible, they should
be able to try row erase. Please go to Microchip Support and
raise a ticket.
http://support.microchip.com

Or hopefully someone in Microchip forum can answer the
question. I do not have a Real ICE myself.

{Quote hidden}

-- Xiaofa

2011\08\11@052735 by RussellMc

face picon face
Is it possible to place the processor in a stable tate such that it's
not outputing any high signals, then raise it's Vdd t0 >= 2.7V. bulk
erase it and then lower voltage, all without other drama? (Probably
not, but maybe worth the asking).

- If the PIC is internally configures to expct 1.8V there 'may be issues'

If an I/O "blips" to > 1.8V during erase summat else may suffer.

Removable adaptor board?


 Russell



On 11 August 2011 07:33, Robert Berger
<.....gmaneKILLspamspam.....reliableembeddedsystems.com> wrote:
{Quote hidden}

>

2011\08\11@080530 by Robert Berger

flavicon
face
Hi,

On 08/11/2011 12:02 PM, Jan-Erik Soderholm wrote:
>
> What could work is some kind of boot-loader. They never do a bulk-erase
> (since that would erase the boot-loader itself) but the page-mode
> erases that usualy/sometime works at lower voltages.
>
> So if re-programming at low voltages is a definite need, a boot-loader
> could be a solution.

A boot loader will not help much, since the edit/compile/__debug__ cycle
will break. Real-ice will not know where the symbols are are we load the
PIC18 via a bootloader.

Regards,

Robert

...."Getting an education was a lot like having a sexual disease. It made
you unsuitable to a lot of jobs and you had the urge to pass it on."- A
weirdo quoting PTerry

My public pgp key is available at:
pgp.mit.edu:11371/pks/lookup?op=get&search=0x90320BF1

2011\08\11@081007 by Robert Berger

flavicon
face
On 08/11/2011 12:19 PM, Xiaofan Chen wrote:
>
> What if you enable LVP? That might work for things like PICkit 2.
> ww1.microchip.com/downloads/en/DeviceDoc/PICkit%202%20Readme%20v2-61-00%20%28a%29.txt
>
LVP is enabled by the configuration, but this does not seem to help
since real-ice wants to perform a bulk erase.


> Microchip PM3 also supports LVP but that is rather expensive.

We are actually not after a solution to programm the devices, we can do
that (once), what's not working is the edit/compile/debug cycle.

Regards,

Robert


...."My employer doesn't even agree with me about C indentation style." -
Used as a disclaimer

My public pgp key is available at:
pgp.mit.edu:11371/pks/lookup?op=get&search=0x90320BF1

2011\08\11@081514 by Robert Berger

flavicon
face
On 08/11/2011 12:26 PM, Xiaofan Chen wrote:
>
> I remember that MPLAB ICD 3 and Real ICE should be able to support
> this use case. When the bulk erase is not possible, they should
> be able to try row erase. Please go to Microchip Support and
> raise a ticket.
> http://support.microchip.com

A ticket has been raised, but I'm not sure you can access it:
http://www.microchip.com/support/IndividualTicket.aspx?TicketNo=211790&p=0

And this is more or less what Microchip support says:
"The Bulk Erase requirement is a limitation of the chip and NOT of the
Programmer.

The Programming specification requires a Voltage of min. 2.7V for the
Bulk Erase."

>
>>
>> The help inside MPLAB seems to suggest that Microchip MPLAB
>> ICD2, ICD 3, Real ICE and PICkit 3 do not support LVP inside MPLAB
>> and they recommend to use PM3.

PM3 will not help, since it's just for programming and not for debugging
and we are fine with programming (once).

Regards,

Robert


....If it's there, and you can see it, it's real. - If it's not there,
and you can see it, it's virtual.- If it's there, and you can't see it,
it's transparent.- If it's not there, and you can't see it, you erased
it. (from some mailing list)

My public pgp key is available at:
pgp.mit.edu:11371/pks/lookup?op=get&search=0x90320BF1

2011\08\11@082008 by Robert Berger

flavicon
face
Hi,

On 08/11/2011 12:26 PM, RussellMc wrote:
> Is it possible to place the processor in a stable tate such that it's
> not outputing any high signals, then raise it's Vdd t0 >= 2.7V. bulk
> erase it and then lower voltage, all without other drama? (Probably
> not, but maybe worth the asking).

Not really e.g. when you hit a breakpoint while debugging.

>
> - If the PIC is internally configures to expct 1.8V there 'may be issues'
>
>  If an I/O "blips" to > 1.8V during erase summat else may suffer.

I'll blow my main board;)

>
> Removable adaptor board?

Well that's only for people who write bug free code (I would like to
debug my code and frequently update it until it works)

Looks like the best solution will be voltage translators, so the PIC can
be run at 3.3V where everything is supposed to work and the voltage
translators translate to 1.8V so to main board can not get killed.

Once the software is stable we can connect the PIC and the main board
directly.

Regards,

Robert

{Quote hidden}

>> -

2011\08\11@082616 by Reinhard Daemon

flavicon
face
Robert Berger <gmane <at> reliableembeddedsystems.com> writes:

> LVP is enabled by the configuration, but this does not seem to help
> since real-ice wants to perform a bulk erase.
>
Robert, you are right, i have also tried with the LVP configuration bit enabled, but it was in vain.
The oscilloscope told me that PICKIT3 and REALICE do not "look" for the LVP bit,
but they do always use the high programming voltage (around 9 V) at the MCLR# pin.
Mihai, a person from the microchip support team, has just emailed to me:

QUOTATION-BEGIN
For the other issue with the REAL ICE not being able to program but once a
PIC18LF45K22 at 1.8 V, I have searched our tickets to see if one was opened, and I see one of my colleagues took an issue like this for a customer from Greece,
so I am assuming this might be your colleague.
I looked on the issue and I have also tried to program with REAL ICE this device at 1.8 V, sometimes it worked, sometime did not,
at 1.9-2 V it worked all the
time. But I will let my colleague to handle it.
QUOTATION-END

in my case, on your board, i am NEVER able to programm the chip a second time at 1.8 V ..., but i am always able to refresh it with PICKIT3 (supplying 3 V to the target board). I strongly recommend to buy a PICKIT3 plus the adapter set.

Reinhard



2011\08\11@090322 by Xiaofan Chen

face picon face
On Thu, Aug 11, 2011 at 8:25 PM, Reinhard Daemon <r.daemonspamspam_OUTtugraz.at> wrote:
> Robert Berger <gmane <at> reliableembeddedsystems.com> writes:
>
>> LVP is enabled by the configuration, but this does not seem to help
>> since real-ice wants to perform a bulk erase.
>>
>
> Robert, you are right,
> i have also tried with the LVP configuration bit enabled,
> but it was in vain.

As I mentioned before, you can use the US$35 Microchip PICkit 2.
>What if you enable LVP? That might work for things like PICkit 2.
> http://ww1.microchip.com/downloads/en/DeviceDoc/PICkit%202%20Readme%20v2-61-00%20%28a%29.txt


-- Xiaofa

2011\08\11@091000 by Xiaofan Chen

face picon face
On Thu, Aug 11, 2011 at 9:03 PM, Xiaofan Chen <@spam@xiaofancKILLspamspamgmail.com> wrote:
> On Thu, Aug 11, 2011 at 8:25 PM, Reinhard Daemon <KILLspamr.daemonKILLspamspamtugraz.at> wrote:
>> Robert Berger <gmane <at> reliableembeddedsystems.com> writes:
>>
>>> LVP is enabled by the configuration, but this does not seem to help
>>> since real-ice wants to perform a bulk erase.
>>>
>>
>> Robert, you are right,
>> i have also tried with the LVP configuration bit enabled,
>> but it was in vain.
>
> As I mentioned before, you can use the US$35 Microchip PICkit 2.
>>What if you enable LVP? That might work for things like PICkit 2.
>> http://ww1.microchip.com/downloads/en/DeviceDoc/PICkit%202%20Readme%20v2-61-00%20%28a%29.txt
>

The other possibility is Olin's LProg. But PIC18F45K22 is not listed.
http://www.embedinc.com/products/lprog/index.htm



-- Xiaofa

2011\08\12@020126 by cdb

flavicon
face
::looks like the best solution will be voltage translators, so the PIC can
::be run at 3.3V where everything is supposed to work and the voltage
::translators translate to 1.8V so to main board can not get killed.

If you're prepared to buy a new programmer - Mikroelectronica have recently brought one out that works with 1v8 -> 5v parts (US$99.00)


Colin


--
cdb, RemoveMEcolinTakeThisOuTspambtech-online.co.uk on 12/08/2011
Web presence: http://www.btech-online.co.uk   Hosted by:  http://www.justhost.com.au
 This email is to be considered private if addressed to a named  individual or HR department, and public if addressed to a blog,  forum or news article.

2011\08\12@052842 by William \Chops\ Westfield

face picon face

On Aug 10, 2011, at 12:33 PM, Robert Berger wrote:

> I would not mind switching the processor, even to something different
> from Microchip, but ideally I'm after a development method where I can
> operate the chip at 1.8V and can use edit/compile/download/debug  
> without
> applying a higher voltage in between with real-ice.

IIRC, low voltage operation is one of the reasons that TI has been  working with FRAM to (eventually) replace flash in their  microcontrollers...

BillW

2011\08\12@200635 by Barry Gershenfeld

picon face
I believe I have seen this behavior but can't confirm it.  Using lesser
(5-volt) PICs, when my supply voltage is too low--falling below 4.5v--MPLAB
will "hang" at the "Erasing" step for a very long time (10 seconds is a long
time when you're accustomed to half a second).  It then proceeds to program
the part at normal speed, and everything seems find.  This is with an ICD2.
I believe it switches to the "slow way" of erasing.

The above may not be helpful in this case, but I've also been able to do
that bootload-then-debug thing.  This works when the bootloaded program is
an address-exact replica of the one compiled.  After programming the part, I
compile the program again (to produce the proper symbol files).  I keep it
from programming the part.  Enable the programmer, and read in the entire
image (bootloader+program).  Now change over to debugger, and program the
part (this adds the debug monitor).  I tell it to run.  It complains.  I say
go ahead anyway, and it works, as long as I let it run past where the
bootloader is.  This works with 16F/18F parts.  Not sure about the others.
Also, I use C.  Don't know a reason that assembly code wouldn't also work,
but I haven't tried it

More... (looser matching)
- Last day of these posts
- In 2011 , 2012 only
- Today
- New search...