L1s @500 MHz but how exactly ?

Technical discussions around xCORE processors (e.g. xcore-200 & xcore.ai).
ale500
Respected Member
Posts: 259
Joined: Thu Sep 16, 2010 9:15 am

L1s @500 MHz but how exactly ?

Post by ale500 »

That is the question... should I go for a 25 MHz oscillator and leave the PLL and divider as they are or it is better a 20 MHz oscillator and reprogram the PLL and so on (but that would trigger a reset...) how is it that it should be implemented ?


bearcat
Respected Member
Posts: 283
Joined: Fri Mar 19, 2010 4:49 am

Post by bearcat »

The core frequency can be set in the XN file along with the oscillator frequency. No need to reprogram the PLL in code, unless there is another reason to do that.

Be sure to set the mode 0, 1 pins to match your frequency source.

One note is if the clock rate does not divide evenly, you proabably need to use the following builtins instead for clock blocks:
void configure_clock_rate_at_most(clock clk, unsigned a, unsigned b)
void configure_clock_at_least(clock clk, unsigned char divide)

I would pick my clock frequency according to any other clock needs of your design and cost of parts.
ale500
Respected Member
Posts: 259
Joined: Thu Sep 16, 2010 9:15 am

Post by ale500 »

I ordered some 25 MHz oscillators :). I'm eager to get my hands on the new boards... those L1 sure look tiny :). I hope they are easier to solder than they look!
User avatar
rp181
Respected Member
Posts: 395
Joined: Tue May 18, 2010 12:25 am
Contact:

Post by rp181 »

Look into drag soldering, I solder the chip in under a min. using that method. Works great with a normal tip (not a fine pitch), and works even better with a drag soldering tip (like a hoof tip, with a well to hold solder).
ale500
Respected Member
Posts: 259
Joined: Thu Sep 16, 2010 9:15 am

Post by ale500 »

You mean... apply solder wildly and then use solder-wick ? (That is what I had in mind)
User avatar
Bianco
XCore Expert
Posts: 754
Joined: Thu Dec 10, 2009 6:56 pm
Contact:

Post by Bianco »

I would be careful with drag soldering.. it requires some excersise.
I'd rather spend ten times more time to solder each pin and not screw up like i did when i tried drag soldering. Besides that i'm not sure wether the 128TQFP with 0.4mm pitch is even drag solderable.
ale500
Respected Member
Posts: 259
Joined: Thu Sep 16, 2010 9:15 am

Post by ale500 »

I Have drag-soldered some devices with 0.5 & 0.65 mm pitch in the past. While soldering is a concern to me, solder paste and hot-air (rework station) are also available to me :)
User avatar
leon_heller
XCore Expert
Posts: 546
Joined: Thu Dec 10, 2009 10:41 pm
Location: St. Leonards-on-Sea, E. Sussex, UK.
Contact:

Post by leon_heller »

Drag-soldering using a Metcal system with a mini-hoof cartridge works for me on fine-pitch devices. Plenty of jelly flux is the secret.
User avatar
lilltroll
XCore Expert
Posts: 956
Joined: Fri Dec 11, 2009 3:53 am
Location: Sweden, Eskilstuna

Post by lilltroll »

Have you tried the stencils ?

@ 50 s

c_Qt5CtUlqY

One way that works for singel row QFNs as well is to begin with hot air and fluss and very little paste.
That way it auto-centers and the pads get some connection.
Thereafter you can use the solder iron and some more solder-wire to improve the connection.
If you do not have the touch for drag - you can use the sweep tech.
The solder and tip must keep it's temp at drag soldering.

If the tenn gets to cold you destroy the pcb lanes. This tip has some thermal mass in the video.

5uiroWBkdFY
Probably not the most confused programmer anymore on the XCORE forum.
User avatar
leon_heller
XCore Expert
Posts: 546
Joined: Thu Dec 10, 2009 10:41 pm
Location: St. Leonards-on-Sea, E. Sussex, UK.
Contact:

Post by leon_heller »

That Hakko bevel tip is similar to the Metcal mini-hoof. A standard tip can be used but it's almost impossible to avoid solder bridges.
Post Reply