Lpc2000

Hello,

I'm trying to get a new PCB up and running using an LPC2214, but 
cannot get the chip to talk to the JTAG port. I'm fairly new to both 
the processor, and ARM in general, so sorry if this is a really dumb 
problem.

I'm using a Segger J-Link (IAR rebadged) and Rowley compiler.

I have connected to and run test code on the Olimex LPC2214 
development board, so I know the J-Link is OK.

I have checked and double checked the JTAG connection against the 
Olimex schematic.

I have the BSL (P0.14) connected to a pullup. Boot0 and Boot1 are 
open circuit (high through internal pullups). RTCK is pulled low by 
10K. 

The processor oscillator is running.

I've scoped the levels on the JTAG interface with the J-Link 
unplugged, and they are identical on my and the Olimex board, with 
the exception of TCK, which is at 0V on my board and approx 0.4V on 
the Olimex board

Whenever I try to connect from the Rowley IDE I get "Too few devices 
on the JTAG bus. Selected Device :0, No. of Devices 0". I guess this 
means the JLink simply can't see the processor for whatever reason.

The board and chip are new, and have not have any code downloaded 
into them. Do I need to download code using the Flash Loader utility 
from the Phillips website, or should I be able to connect straight 
away through the J-Link? 

Does everything else sound OK, or could I have got a duff processor?


Thanks

Simon Ansley

Hi

I had a simmilar Problem when I built my Jtag clone for the AVR.
The Problem was I forgot to power up the processor.
I your case somethin simmilar must be the problem, since the
Jtag link is recogniced.
Triple check all your power connections to the processor.

Cheers
Michael

>From: "Simon Ansley" <simon@...>
>Reply-To: lpc2000@yahoogroups.com
>To: lpc2000@yahoogroups.com
>Subject: [lpc2000] Newbie problem : Connecting LPC2214 to Segger J-Link
>Date: Sat, 15 Oct 2005 17:39:35 -0000
>
>Hello,
>
>I'm trying to get a new PCB up and running using an LPC2214, but
>cannot get the chip to talk to the JTAG port. I'm fairly new to both
>the processor, and ARM in general, so sorry if this is a really dumb
>problem.
>
>I'm using a Segger J-Link (IAR rebadged) and Rowley compiler.
>
>I have connected to and run test code on the Olimex LPC2214
>development board, so I know the J-Link is OK.
>
>I have checked and double checked the JTAG connection against the
>Olimex schematic.
>
>I have the BSL (P0.14) connected to a pullup. Boot0 and Boot1 are
>open circuit (high through internal pullups). RTCK is pulled low by
>10K.
>
>The processor oscillator is running.
>
>I've scoped the levels on the JTAG interface with the J-Link
>unplugged, and they are identical on my and the Olimex board, with
>the exception of TCK, which is at 0V on my board and approx 0.4V on
>the Olimex board
>
>Whenever I try to connect from the Rowley IDE I get "Too few devices
>on the JTAG bus. Selected Device :0, No. of Devices 0". I guess this
>means the JLink simply can't see the processor for whatever reason.
>
>The board and chip are new, and have not have any code downloaded
>into them. Do I need to download code using the Flash Loader utility
>from the Phillips website, or should I be able to connect straight
>away through the J-Link?
>
>Does everything else sound OK, or could I have got a duff processor?
>
>
>Thanks
>
>Simon Ansley
>
>
>

Simon Ansley wrote:

>Hello,
>
>I'm trying to get a new PCB up and running using an LPC2214, but 
>cannot get the chip to talk to the JTAG port. I'm fairly new to both 
>the processor, and ARM in general, so sorry if this is a really dumb 
>problem.
>
>I'm using a Segger J-Link (IAR rebadged) and Rowley compiler.
>
>I have connected to and run test code on the Olimex LPC2214 
>development board, so I know the J-Link is OK.
>
>I have checked and double checked the JTAG connection against the 
>Olimex schematic.
>
>I have the BSL (P0.14) connected to a pullup. Boot0 and Boot1 are 
>open circuit (high through internal pullups). RTCK is pulled low by 
>10K. 
>
>The processor oscillator is running.
>
>I've scoped the levels on the JTAG interface with the J-Link 
>unplugged, and they are identical on my and the Olimex board, with 
>the exception of TCK, which is at 0V on my board and approx 0.4V on 
>the Olimex board
>
>Whenever I try to connect from the Rowley IDE I get "Too few devices 
>on the JTAG bus. Selected Device :0, No. of Devices 0". I guess this 
>means the JLink simply can't see the processor for whatever reason.
>
>The board and chip are new, and have not have any code downloaded 
>into them. Do I need to download code using the Flash Loader utility 
>from the Phillips website, or should I be able to connect straight 
>away through the J-Link? 
>
>Does everything else sound OK, or could I have got a duff processor?
>
>
>  
>
I just brought up a new design using LPC2106 and LPC2138 processors.  
The first thing that I did was to see if they were "alive".  I connected 
to them via the UART0, gave them the '?' and echo'ed the "Synchronized" 
back to them. 

Once I had done that, I could assume that Clock operation was correct, 
Power Supplies were ok, Reset Circuit was okay, ...  IOW, the processors 
had the proper conditions to run with.

Next step was to attempt to connect via the Abatron BDI2000 JTAG, that 
worked as expected.  But, I was not about to subject the BDI2000 to an 
unknown board until  I had some level of confidence in the board.

I would scope the pins to see if the JTAG signals get to the processor 
pins, and see if the signals are clean, without glitches which would 
suggest a short.

I know that the BDI2000 will not attempt to connect to the board unless 
it first sees a VCC supply present on the JTAG connector.  Did you 
"edit" the JTAG connector pins and removed anything?


Hope this helps?

TomW



-- 
Tom Walsh - WN3L - Embedded Systems Consultant
http://openhardware.net, http://cyberiansoftware.com
"Windows? No thanks, I have work to do..."
----------------------------------------------------

Tom,

> >
> I just brought up a new design using LPC2106 and LPC2138 processors.
> The first thing that I did was to see if they were "alive".  I 
connected
> to them via the UART0, gave them the '?' and echo'ed 
the "Synchronized"
> back to them.
>

Many thanks, I've found the relevant bit in the manual, made up a 
cable and
tried that. Glad I'd hung a 232 converter on that port! The processor 
acts
as described.

> Once I had done that, I could assume that Clock operation was 
correct,
> Power Supplies were ok, Reset Circuit was okay, ...  IOW, the 
processors
> had the proper conditions to run with.
>

OK, now I've done that, I'll try the JTAG again. As I said before, 
I've
compared the signals (without the JTAG i/face connected) with my 
Olimex
board, and they appear identical except for TCK. Guess I need to find 
out
why that's different.

>
> I know that the BDI2000 will not attempt to connect to the board 
unless
> it first sees a VCC supply present on the JTAG connector.  Did you
> "edit" the JTAG connector pins and removed anything?
>

I've got the VCC present on pins 1 and 2, Gnd on all of the rest of 
the even
numbered pins.
The connector is wired as on the Olimex schematic. I haven't found 
anything
obvious like shorts to GND, so it may be something more subtle


Thanks again for the help, even a little progress has made me feel a 
bit
more confident.

Simon Ansley

Simon Ansley wrote:

>I've got the VCC present on pins 1 and 2, Gnd on all of the rest of 
>the even
>numbered pins.
>The connector is wired as on the Olimex schematic. I haven't found 
>anything
>obvious like shorts to GND, so it may be something more subtle
>  
>

Usually, I use two X-acto knife blades as probes to my Ohm-meter, then 
carefully buzz out the connections.  It is easy to have a surface mount 
pad not soldered. 

Usually it is something simple, open  / shorted connection on a new board.

TomW

-- 
Tom Walsh - WN3L - Embedded Systems Consultant
http://openhardware.net, http://cyberiansoftware.com
"Windows? No thanks, I have work to do..."
----------------------------------------------------

Sorry, can't help you just confirm your problem. 

I have tried the same with the Segger J-Link (IAR rebadged) and the
IAR (Olimex) LPC2148 eval board. I get exactly the same errors as you do.

My guess is that the IAR rebadged segger unit, has different firmware
in it so it will only work with the IAR compiler :-(. This is just my
guess though.

Let me know if you get anywhere, have you tried talking to Rowley
Associates?

Cheers

Martin

--- In lpc2000@yahoogroups.com, "Simon Ansley" <simon@a...> wrote:

>
> Hello,
> 
> I'm trying to get a new PCB up and running using an LPC2214, but 
> cannot get the chip to talk to the JTAG port. I'm fairly new to both 
> the processor, and ARM in general, so sorry if this is a really dumb 
> problem.
> 
> I'm using a Segger J-Link (IAR rebadged) and Rowley compiler.
> 
> I have connected to and run test code on the Olimex LPC2214 
> development board, so I know the J-Link is OK.
> 
> I have checked and double checked the JTAG connection against the 
> Olimex schematic.
> 
> I have the BSL (P0.14) connected to a pullup. Boot0 and Boot1 are 
> open circuit (high through internal pullups). RTCK is pulled low by 
> 10K. 
> 
> The processor oscillator is running.
> 
> I've scoped the levels on the JTAG interface with the J-Link 
> unplugged, and they are identical on my and the Olimex board, with 
> the exception of TCK, which is at 0V on my board and approx 0.4V on 
> the Olimex board
> 
> Whenever I try to connect from the Rowley IDE I get "Too few devices 
> on the JTAG bus. Selected Device :0, No. of Devices 0". I guess this 
> means the JLink simply can't see the processor for whatever reason.
> 
> The board and chip are new, and have not have any code downloaded 
> into them. Do I need to download code using the Flash Loader utility 
> from the Phillips website, or should I be able to connect straight 
> away through the J-Link? 
> 
> Does everything else sound OK, or could I have got a duff processor?
> 
> 
> Thanks
> 
> Simon Ansley
>

Hi,

You'll need to copy the latest JlinkARM.dll (the one that matches your 
firmware) from the Segger distribution into the CrossWorks bin 
directory. For the next release we'll access the JLinkARM.dll in the 
Segger distribution to avoid problems with .dll and firmware 
incompatibilities.

Regards
Michael

>Sorry, can't help you just confirm your problem. 
>
>I have tried the same with the Segger J-Link (IAR rebadged) and the
>IAR (Olimex) LPC2148 eval board. I get exactly the same errors as you do.
>
>My guess is that the IAR rebadged segger unit, has different firmware
>in it so it will only work with the IAR compiler :-(. This is just my
>guess though.
>
>Let me know if you get anywhere, have you tried talking to Rowley
>Associates?
>
>Cheers
>
>Martin
>
>--- In lpc2000@yahoogroups.com, "Simon Ansley" <simon@a...> wrote:
>  
>
>>Hello,
>>
>>I'm trying to get a new PCB up and running using an LPC2214, but 
>>cannot get the chip to talk to the JTAG port. I'm fairly new to both 
>>the processor, and ARM in general, so sorry if this is a really dumb 
>>problem.
>>
>>I'm using a Segger J-Link (IAR rebadged) and Rowley compiler.
>>
>>I have connected to and run test code on the Olimex LPC2214 
>>development board, so I know the J-Link is OK.
>>
>>I have checked and double checked the JTAG connection against the 
>>Olimex schematic.
>>
>>I have the BSL (P0.14) connected to a pullup. Boot0 and Boot1 are 
>>open circuit (high through internal pullups). RTCK is pulled low by 
>>10K. 
>>
>>The processor oscillator is running.
>>
>>I've scoped the levels on the JTAG interface with the J-Link 
>>unplugged, and they are identical on my and the Olimex board, with 
>>the exception of TCK, which is at 0V on my board and approx 0.4V on 
>>the Olimex board
>>
>>Whenever I try to connect from the Rowley IDE I get "Too few devices 
>>on the JTAG bus. Selected Device :0, No. of Devices 0". I guess this 
>>means the JLink simply can't see the processor for whatever reason.
>>
>>The board and chip are new, and have not have any code downloaded 
>>into them. Do I need to download code using the Flash Loader utility 
>>from the Phillips website, or should I be able to connect straight 
>>away through the J-Link? 
>>
>>Does everything else sound OK, or could I have got a duff processor?
>>
>>
>>Thanks
>>
>>Simon Ansley
>>
>>    
>>
>
>
>
>
>
>
>
> 
>Yahoo! Groups Links
>
>
>
> 
>
>
>
>
>  
>