Finally, I tracked the problem! It was a mistake in BOM that had put the crystal
load capacitors into 0.1uF row! We take these things for granted, especially
since BOM is automatically generated by design software. But this time it was a
visible custom property that was changed to proper value, and the one used to
generate BOM was not. A lesson for the lazy designer for copy/pasting component
symbols...!
The schematic "looked" all ok, but the BOM it generated had wrong values. I will
have to scan through the entire schematic to look for such mistakes, but it is a
lot easier, now that I have got the boards to work!
I was able to get into ISP mode and write and run a test program.
Thanks all for the help.
Girish
--- In l..., "hsutherland2@..." wrote: >
> Hello Girish,
>
> Sorry I missed your original post.
>
> It sounds to me you should focus your investigation on the oscillator. It
should be running even with reset held low (someone correct me if this is
wrong). With an ordinary DMM, I read about 1 volt DC on each crystal pin. My
2138 and 2148 boards read the same.
>
> Not impossible that both your uC's are bad, but that seems unlikely. A
batch of incorrect capacitors or crystals seems more probable.
>
> Or, possibly the parameters of your crystal are such that it won't work
with 33 pF caps. I presume, in this day and age, that you have unmarked surface
mount caps. You could parallel them to see if the high end of the range would
work. Putting them in series would be more of a challenge...
>
> -Hugh
>
> --- In l..., "gmpundlik" wrote:
> >
> > Hello,
> >
> > I am developing a LPC2132/01 based system. I just received the assembled
PCBs and I tried to put in a test program. But the LPC2132 won't go into
ISP mode.
> >
> > Pin P0.31 is left unconnected, and pin P0.14 is used exclusively for ISP
mode select, with a pull-up of 10K. I used LPC Flash Utility v2.2.3, tried baud
rates of 4800, 9600, 19200 and 38400 but none of them could detect the device
id, let alone write anything to its flash.
> >
> > I also tried resetting the controller by shorting the reset capacitor, after
power up. The controller's 3.3V supply is generated by ADP3338 LDO that
does not soft start (have used it earlier with LPC2114), so I guess it is not
the problem with supply voltage ramp.
> >
> > The voltages at all power pins are ok. The battery socket is empty (Vbat
connected to VSS). However, the XTAL2 and XTAL1 clock pins are level at high and
low respectively. Seems that the oscillator is not working. The crystal is 12MHz
with 33pF capacitors.
> >
> > Errata documents do not mention anything relevant to this problem.
> >
> > Is there any issue or a trick to get the LPC2132 working, or could it be
plain dead... on a fresh new board (two of them)!?
> >
> > Thanks,
> >
> > Girish
>
Reply by "hsu...@sbcglobal.net"●August 25, 20092009-08-25
Hello Girish,
Sorry I missed your original post.
It sounds to me you should focus your investigation on the oscillator. It should
be running even with reset held low (someone correct me if this is wrong). With
an ordinary DMM, I read about 1 volt DC on each crystal pin. My 2138 and 2148
boards read the same.
Not impossible that both your uC's are bad, but that seems unlikely. A
batch of incorrect capacitors or crystals seems more probable.
Or, possibly the parameters of your crystal are such that it won't work
with 33 pF caps. I presume, in this day and age, that you have unmarked surface
mount caps. You could parallel them to see if the high end of the range would
work. Putting them in series would be more of a challenge...
-Hugh
--- In l..., "gmpundlik" wrote: >
> Hello,
>
> I am developing a LPC2132/01 based system. I just received the assembled PCBs
and I tried to put in a test program. But the LPC2132 won't go into ISP
mode.
>
> Pin P0.31 is left unconnected, and pin P0.14 is used exclusively for ISP mode
select, with a pull-up of 10K. I used LPC Flash Utility v2.2.3, tried baud rates
of 4800, 9600, 19200 and 38400 but none of them could detect the device id, let
alone write anything to its flash.
>
> I also tried resetting the controller by shorting the reset capacitor, after
power up. The controller's 3.3V supply is generated by ADP3338 LDO that
does not soft start (have used it earlier with LPC2114), so I guess it is not
the problem with supply voltage ramp.
>
> The voltages at all power pins are ok. The battery socket is empty (Vbat
connected to VSS). However, the XTAL2 and XTAL1 clock pins are level at high and
low respectively. Seems that the oscillator is not working. The crystal is 12MHz
with 33pF capacitors.
>
> Errata documents do not mention anything relevant to this problem.
>
> Is there any issue or a trick to get the LPC2132 working, or could it be plain
dead... on a fresh new board (two of them)!?
>
> Thanks,
>
> Girish
>
Reply by gmpundlik●August 25, 20092009-08-25
P0.31 is connected to nothing. It is a output only pin with internal pullup by
definition, that should not be pulled low.
--- In l..., "Charles R. Grenz" wrote: > gmpundlik wrote:
> > Hello,
> >
> > I further rechecked the schematic and the tested the boards again.
> >
> > The schematic is good, reset circuit has a 10K pullup and 0.1uF capacitor to
the ground. P0.14 has a 10K pullup and jumper to the ground. There are in all 36
pins used for I/O. Pin P0.31 is left open and not used, so it should not pose
any problem. I also changed the crystal from a KDS make that had no datasheet,
to a Vishay crystal of known parameters, 20pF load capacitance so that the 33pF
capacitors matched.
> >
> > Voltages on all power pins are OK. I scoped the serial port signals and
traced with a serial port monitor. The '?' command is being sent, but
there is no response from the microcontroller. The clock shows low on one pin
and high on the other.
> >
> > Pin P1.20 when low is supposed to put pins P16:25 in traceport mode, so I
set it to high by a jumper. Pin P1.26 (enables JTAG when low) is also not used,
left unconnected.
> >
> > I have probably checked for every possible pin that could send the LPC2132
into some other mode and prevent it from responding to ISP.
> >
> > I had used a LPC2114 earlier in this project, that had no problem doing ISP
right at first power up. I am sure LPC2132 is no different. At least, I did not
find any ISP related cases other than bad handling of P0.14 or other hardware
errors.
> >
> > Now, is there anything else I could be missing, in terms of logic levels on
certain pins that I need to maintain on startup?
> >
> > Thanks for any help.
> >
> > Girish
> >
> > What is P0.31 have or connected to?
>
> regards,
> Charles
>
Reply by "Charles R. Grenz"●August 25, 20092009-08-25
gmpundlik wrote: > Hello,
>
> I further rechecked the schematic and the tested the boards again.
>
> The schematic is good, reset circuit has a 10K pullup and 0.1uF capacitor to
the ground. P0.14 has a 10K pullup and jumper to the ground. There are in all 36
pins used for I/O. Pin P0.31 is left open and not used, so it should not pose
any problem. I also changed the crystal from a KDS make that had no datasheet,
to a Vishay crystal of known parameters, 20pF load capacitance so that the 33pF
capacitors matched.
>
> Voltages on all power pins are OK. I scoped the serial port signals and traced
with a serial port monitor. The '?' command is being sent, but there
is no response from the microcontroller. The clock shows low on one pin and high
on the other.
>
> Pin P1.20 when low is supposed to put pins P16:25 in traceport mode, so I set
it to high by a jumper. Pin P1.26 (enables JTAG when low) is also not used, left
unconnected.
>
> I have probably checked for every possible pin that could send the LPC2132
into some other mode and prevent it from responding to ISP.
>
> I had used a LPC2114 earlier in this project, that had no problem doing ISP
right at first power up. I am sure LPC2132 is no different. At least, I did not
find any ISP related cases other than bad handling of P0.14 or other hardware
errors.
>
> Now, is there anything else I could be missing, in terms of logic levels on
certain pins that I need to maintain on startup?
>
> Thanks for any help.
>
> Girish
What is P0.31 have or connected to?
regards,
Charles
Reply by gmpundlik●August 25, 20092009-08-25
Hello,
I further rechecked the schematic and the tested the boards again.
The schematic is good, reset circuit has a 10K pullup and 0.1uF capacitor to the
ground. P0.14 has a 10K pullup and jumper to the ground. There are in all 36
pins used for I/O. Pin P0.31 is left open and not used, so it should not pose
any problem. I also changed the crystal from a KDS make that had no datasheet,
to a Vishay crystal of known parameters, 20pF load capacitance so that the 33pF
capacitors matched.
Voltages on all power pins are OK. I scoped the serial port signals and traced
with a serial port monitor. The '?' command is being sent, but there
is no response from the microcontroller. The clock shows low on one pin and high
on the other.
Pin P1.20 when low is supposed to put pins P16:25 in traceport mode, so I set it
to high by a jumper. Pin P1.26 (enables JTAG when low) is also not used, left
unconnected.
I have probably checked for every possible pin that could send the LPC2132 into
some other mode and prevent it from responding to ISP.
I had used a LPC2114 earlier in this project, that had no problem doing ISP
right at first power up. I am sure LPC2132 is no different. At least, I did not
find any ISP related cases other than bad handling of P0.14 or other hardware
errors.
Now, is there anything else I could be missing, in terms of logic levels on
certain pins that I need to maintain on startup?
Thanks for any help.
Girish
--- In l..., "gmpundlik" wrote: >
> Hello,
>
> I am developing a LPC2132/01 based system. I just received the assembled PCBs
and I tried to put in a test program. But the LPC2132 won't go into ISP
mode.
>
> Pin P0.31 is left unconnected, and pin P0.14 is used exclusively for ISP mode
select, with a pull-up of 10K. I used LPC Flash Utility v2.2.3, tried baud rates
of 4800, 9600, 19200 and 38400 but none of them could detect the device id, let
alone write anything to its flash.
>
> I also tried resetting the controller by shorting the reset capacitor, after
power up. The controller's 3.3V supply is generated by ADP3338 LDO that
does not soft start (have used it earlier with LPC2114), so I guess it is not
the problem with supply voltage ramp.
>
> The voltages at all power pins are ok. The battery socket is empty (Vbat
connected to VSS). However, the XTAL2 and XTAL1 clock pins are level at high and
low respectively. Seems that the oscillator is not working. The crystal is 12MHz
with 33pF capacitors.
>
> Errata documents do not mention anything relevant to this problem.
>
> Is there any issue or a trick to get the LPC2132 working, or could it be plain
dead... on a fresh new board (two of them)!?
>
> Thanks,
>
> Girish
>
Reply by gmpundlik●August 22, 20092009-08-22
Sorry I did not explain the complete situation. P0.14 is pulled up by 10K
resistor, and grounded though a jumper. So when I attempt to flash the program,
I put the jumper in place, and rest of the times it is pulled high without the
jumper.
I think I know what can be the problem. The pullup resistor on P0.14 was tied to
5V instead of 3.3V. There was a issue with the power supply circuit in the
begining, the first stage 15V switching regulator catch diode was connected in
reverse (bad footprint used by the PCB designer), and I had to replace the
regulator. The subsequent power stages 5V and 3.3V were OK, therefore I guess
they were safe. But, if somehow the 5V rail could have been affected, it could
have damaged the P0.14 pin. Rest of the microcontroller may well be ok, but
there is no way it can respond to ISP requests now.
I am building another board with the diode put correctly, so I will know by next
week.
Can a reversed catch diode of a switching regulator inflict such damage on rest
of the circuit?
Thanks,
Girish
--- In l..., Prerak Sheth wrote: >
> To have a flash programming enabled, you need to get p0.14 to low
> level exclusively.
> If it is pulled up and you are not grounding to program, then it would
never
> recognize programming mode.
> This may sound too simplistic, but I have faced this number of times my
self
> Regards,
> Prerak
>
> On Thu, Aug 20, 2009 at 10:36 PM, gmpundlik wrote:
>
> >
> >
> > Hello,
> >
> > I am developing a LPC2132/01 based system. I just received the assembled
> > PCBs and I tried to put in a test program. But the LPC2132 won't go
into ISP
> > mode.
> >
> > Pin P0.31 is left unconnected, and pin P0.14 is used exclusively for ISP
> > mode select, with a pull-up of 10K. I used LPC Flash Utility v2.2.3,
tried
> > baud rates of 4800, 9600, 19200 and 38400 but none of them could detect
the
> > device id, let alone write anything to its flash.
> >
> > I also tried resetting the controller by shorting the reset capacitor,
> > after power up. The controller's 3.3V supply is generated by ADP3338
LDO
> > that does not soft start (have used it earlier with LPC2114), so I guess
it
> > is not the problem with supply voltage ramp.
> >
> > The voltages at all power pins are ok. The battery socket is empty (Vbat
> > connected to VSS). However, the XTAL2 and XTAL1 clock pins are level at
high
> > and low respectively. Seems that the oscillator is not working. The
crystal
> > is 12MHz with 33pF capacitors.
> >
> > Errata documents do not mention anything relevant to this problem.
> >
> > Is there any issue or a trick to get the LPC2132 working, or could it be
> > plain dead... on a fresh new board (two of them)!?
> >
> > Thanks,
> >
> > Girish
> >
> >
> >
>
>
Reply by Prerak Sheth●August 22, 20092009-08-22
To have a flash programming enabled, you need to get p0.14 to low
level exclusively.
If it is pulled up and you are not grounding to program, then it would never
recognize programming mode.
This may sound too simplistic, but I have faced this number of times my self
Regards,
Prerak
On Thu, Aug 20, 2009 at 10:36 PM, gmpundlik wrote:
> Hello,
>
> I am developing a LPC2132/01 based system. I just received the assembled
> PCBs and I tried to put in a test program. But the LPC2132 won't go into
ISP
> mode.
>
> Pin P0.31 is left unconnected, and pin P0.14 is used exclusively for ISP
> mode select, with a pull-up of 10K. I used LPC Flash Utility v2.2.3, tried
> baud rates of 4800, 9600, 19200 and 38400 but none of them could detect the
> device id, let alone write anything to its flash.
>
> I also tried resetting the controller by shorting the reset capacitor,
> after power up. The controller's 3.3V supply is generated by ADP3338
LDO
> that does not soft start (have used it earlier with LPC2114), so I guess it
> is not the problem with supply voltage ramp.
>
> The voltages at all power pins are ok. The battery socket is empty (Vbat
> connected to VSS). However, the XTAL2 and XTAL1 clock pins are level at
high
> and low respectively. Seems that the oscillator is not working. The crystal
> is 12MHz with 33pF capacitors.
>
> Errata documents do not mention anything relevant to this problem.
>
> Is there any issue or a trick to get the LPC2132 working, or could it be
> plain dead... on a fresh new board (two of them)!?
>
> Thanks,
>
> Girish
>
>
>
Reply by gmpundlik●August 21, 20092009-08-21
I checked all the pin usage again, and found that I am using pin P1.20 as input
for a sensor, whose value is indeterminate at startup. This pin when low at
startup, makes pins P1.16:25 as trace port. And at this point, no sensors are
connected and P1.20 is being pulled low.
Apparently, this trace port affects functions of only certain pins (of course,
that is not desired), and is connected directly to the ARM core. The
documentation is not clear whether it affects ISP mode. I could switch this
input to a nearby pin that is defined as output in my application, but before I
perform some surgery on the board, does anybody have any experience with using
the traceport and ISP together on the same board?
Thanks,
Girish
--- In l..., "gmpundlik" wrote: >
> Hello,
>
> I am developing a LPC2132/01 based system. I just received the assembled PCBs
and I tried to put in a test program. But the LPC2132 won't go into ISP
mode.
>
> Pin P0.31 is left unconnected, and pin P0.14 is used exclusively for ISP mode
select, with a pull-up of 10K. I used LPC Flash Utility v2.2.3, tried baud rates
of 4800, 9600, 19200 and 38400 but none of them could detect the device id, let
alone write anything to its flash.
>
> I also tried resetting the controller by shorting the reset capacitor, after
power up. The controller's 3.3V supply is generated by ADP3338 LDO that
does not soft start (have used it earlier with LPC2114), so I guess it is not
the problem with supply voltage ramp.
>
> The voltages at all power pins are ok. The battery socket is empty (Vbat
connected to VSS). However, the XTAL2 and XTAL1 clock pins are level at high and
low respectively. Seems that the oscillator is not working. The crystal is 12MHz
with 33pF capacitors.
>
> Errata documents do not mention anything relevant to this problem.
>
> Is there any issue or a trick to get the LPC2132 working, or could it be plain
dead... on a fresh new board (two of them)!?
>
> Thanks,
>
> Girish
>
Reply by gmpundlik●August 21, 20092009-08-21
Yes, tried that too, with both FlashMagic and NXP utility. The host keeps
sending 0x3F ('?') and doesn't get anything back.
--- In l..., "Bruce Paterson" wrote: > >Is there any issue or a trick to get the LPC2132
working, or could it
> be >plain dead... on a fresh new board (two of them)!?
>
> I don't know the problem you are having, but to assist, you can bring
up
> a terminal session on the ISP port and type ????? at various baud rates.
> If ISP is working it will reply with "SYNCHRONISED". No different to
> FlashMagic, but might save some debugging time.
>
> Thanks,
>
> Girish
>
>
Reply by Bruce Paterson●August 20, 20092009-08-20
>Is there any issue or a trick to get the LPC2132
working, or could it be >plain dead... on a fresh new board (two of them)!?
I don't know the problem you are having, but to assist, you can bring up
a terminal session on the ISP port and type ????? at various baud rates.
If ISP is working it will reply with "SYNCHRONISED". No different to
FlashMagic, but might save some debugging time.