Lpc2000

Hello,

I've got a design where 2 LPC2129 will be used. They will 
communicate with eachother. My first guess is that the clocksignals 
for both controllers should be the same to prevent phase-shift or 
small frequency differences when using two crystals.

I believe that it is possible to clock the microcontroller by an 
external clock, so I would like to clock both controllers by one 
external source.

Has anyone experience with this, or done this before? Are there 
other issues that require attention when using two controllers in a 
system? 

I've got one controller that will measure continuously, send the 
result to the other controller (sender is master), the other 
controller then has to manipulate this data a bit and send it to 
either COM or CAN, and control an amplifier or something.

Regards,
Laurens

laurensswaans wrote:

>Hello,
>
>I've got a design where 2 LPC2129 will be used. They will 
>communicate with eachother. My first guess is that the clocksignals 
>for both controllers should be the same to prevent phase-shift or 
>small frequency differences when using two crystals.
>
>I believe that it is possible to clock the microcontroller by an 
>external clock, so I would like to clock both controllers by one 
>external source.
>
>Has anyone experience with this, or done this before? Are there 
>other issues that require attention when using two controllers in a 
>system? 
>
>I've got one controller that will measure continuously, send the 
>result to the other controller (sender is master), the other 
>controller then has to manipulate this data a bit and send it to 
>either COM or CAN, and control an amplifier or something.
>
>  
>
Pretty much what I'm doing, I have two processors on my board: LPC2106 
doing outside world communications through multiple serial interfaces, 
LPC2138 doing data collection and running an SD card via SPI.

I run both from a 14.745mHz oscillator, not a crystal, then couple the 
processors to the clock signal via 100p cap into the X1 input of the 
PLL.  They work fine.

I don't think you need to be that concerned with phasing the clocks, 
SPI, I2C or plain RS232 is pretty tolerant.

TomW


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

Thanks Tom,
   
  It's good to hear that someone has already succesfully used this idea. I've almost finished the PCB lay-out. Looks pretty need!
   
  Regards, Laurens

Tom Walsh <tom@...> wrote:
  laurensswaans wrote:

>Hello,
>
>I've got a design where 2 LPC2129 will be used. They will 
>communicate with eachother. My first guess is that the clocksignals 
>for both controllers should be the same to prevent phase-shift or 
>small frequency differences when using two crystals.
>
>I believe that it is possible to clock the microcontroller by an 
>external clock, so I would like to clock both controllers by one 
>external source.
>
>Has anyone experience with this, or done this before? Are there 
>other issues that require attention when using two controllers in a 
>system? 
>
>I've got one controller that will measure continuously, send the 
>result to the other controller (sender is master), the other 
>controller then has to manipulate this data a bit and send it to 
>either COM or CAN, and control an amplifier or something.
>
>  
>
Pretty much what I'm doing, I have two processors on my board: LPC2106 
doing outside world communications through multiple serial interfaces, 
LPC2138 doing data collection and running an SD card via SPI.

I run both from a 14.745mHz oscillator, not a crystal, then couple the 
processors to the clock signal via 100p cap into the X1 input of the 
PLL.  They work fine.

I don't think you need to be that concerned with phasing the clocks, 
SPI, I2C or plain RS232 is pretty tolerant.

TomW


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




  SPONSORED LINKS 
        Microprocessor   Microcontrollers   Pic microcontrollers     8051 microprocessor 
    
---------------------------------
  YAHOO! GROUPS LINKS 

    
    Visit your group "lpc2000" on the web.
    
    To unsubscribe from this group, send an email to:
 lpc2000-unsubscribe@yahoogroups.com
    
    Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service. 

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

  


		
---------------------------------
How much free photo storage do you get? Store your holiday snaps for FREE with Yahoo! Photos. Get Yahoo! Photos

[Non-text portions of this message have been removed]

Tom,
   
  Could you please tell me what kind of oscillator circuit you use? I'm quite confused about what the LPC2129 datasheet specifies.
  What I get from it is that I need an oscillator that has a minimum rise/fall-time of 5ns, a minimum RMS-voltage of 200mV, but the input pin (XTAL1) can not withstand voltages exceeding V18 (which is 1.8V???). All oscillators I see are 5V or rarely 3.3V... What do you use??? Do you use a 5V oscillator and attenuate the signal?
   
  Thanks in advance,
   
  Laurens
  

Tom Walsh <tom@...> wrote:
  laurensswaans wrote:

>Hello,
>
>I've got a design where 2 LPC2129 will be used. They will 
>communicate with eachother. My first guess is that the clocksignals 
>for both controllers should be the same to prevent phase-shift or 
>small frequency differences when using two crystals.
>
>I believe that it is possible to clock the microcontroller by an 
>external clock, so I would like to clock both controllers by one 
>external source.
>
>Has anyone experience with this, or done this before? Are there 
>other issues that require attention when using two controllers in a 
>system? 
>
>I've got one controller that will measure continuously, send the 
>result to the other controller (sender is master), the other 
>controller then has to manipulate this data a bit and send it to 
>either COM or CAN, and control an amplifier or something.
>
>  
>
Pretty much what I'm doing, I have two processors on my board: LPC2106 
doing outside world communications through multiple serial interfaces, 
LPC2138 doing data collection and running an SD card via SPI.

I run both from a 14.745mHz oscillator, not a crystal, then couple the 
processors to the clock signal via 100p cap into the X1 input of the 
PLL.  They work fine.

I don't think you need to be that concerned with phasing the clocks, 
SPI, I2C or plain RS232 is pretty tolerant.

TomW


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




  SPONSORED LINKS 
        Microprocessor   Microcontrollers   Pic microcontrollers     8051 microprocessor 
    
---------------------------------
  YAHOO! GROUPS LINKS 

    
    Visit your group "lpc2000" on the web.
    
    To unsubscribe from this group, send an email to:
 lpc2000-unsubscribe@yahoogroups.com
    
    Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service. 

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

  


		
---------------------------------
Yahoo! Messenger  NEW - crystal clear PC to PC calling worldwide with voicemail 
		
---------------------------------
How much free photo storage do you get? Store your holiday snaps for FREE with Yahoo! Photos. Get Yahoo! Photos

[Non-text portions of this message have been removed]

Laurens Swaans wrote:

>Tom,
>   
>  Could you please tell me what kind of oscillator circuit you use? I'm quite confused about what the LPC2129 datasheet specifies.
>  What I get from it is that I need an oscillator that has a minimum rise/fall-time of 5ns, a minimum RMS-voltage of 200mV, but the input pin (XTAL1) can not withstand voltages exceeding V18 (which is 1.8V???). All oscillators I see are 5V or rarely 3.3V... What do you use??? Do you use a 5V oscillator and attenuate the signal?
>   
>  
>
I use a 3.3v osc and attenuate that.  It is digikey P/N: 535-9255-1-ND

Along with that osc, I also have 1)3.9K, 2)2.2K, and 2)100p caps.  The 
two 2.2K resistors are each locate near one of the CPUs.  Those 2.2K 
resistors go to GND on one side and the osc feed for the other.  The osc 
feed is the 3.9K resistor in series with the output of the osc output.  
This forms an attenuator for the osc signal (which is HUGE!).  Then, the 
100p caps are used to connect each CPUs' X1 input to the attenuated 
signal, see diagram:


OSC--------3.9K---.-------------------.
                  |                   |
  X1----100p------|    X1----100p-----|
                  |                   |
                 2.2K                2.2k
                  |                   |
                  |                   |
                 GND                 GND

Hope this comes out, heh.

Anyhow, I have a 400mv signal @ each X1 input.  The DC voltage of the X1 
input is around 1.2volts, well within spec.

Regards,

TomW

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