Random question: replace chip on Startkit?
-
asid61
- Active Member
- Posts: 36
- Joined: Sat May 31, 2014 6:39 am
Random question: replace chip on Startkit?
The Startkit runs a USB series XMOS chip. Is it possible to use a heatgun or oven to remove the existing chip and replace it with the same model of chip? This would be to allow for easy access to the USB portions of the XMOS by getting rid of the restrictions on tile 1.
-
sethu_jangala
- XCore Expert
- Posts: 589
- Joined: Wed Feb 29, 2012 10:03 am
startKIT comes with jtag interface included in the chip with out any need to use an external debug adapter. Code for Jtag interface is programmed into the OTP of the startKIT and uses one tile to run all the related functionality. So, if you replace the device with other device(whose OTP is not programmed), the device will not work as expected.
Sethu.
Sethu.
-
Ross
- XCore Expert
- Posts: 966
- Joined: Thu Dec 10, 2009 9:20 pm
- Location: Bristol, UK
You would have to wire up the jtag to an external xtag (pads on the back). Also the second tile will be set to boot from the xtag that used to be in the other tile.. would be a bit of effort to get things going (im not sure if it is possible - you would need to check access to the various lines on the start kit layout)
-
pstnotpd
- XCore Addict
- Posts: 161
- Joined: Sun Jun 12, 2011 11:47 am
Considering the pricing of the silicon v.s. that of the startkit itself I really wonder if it is worthwhile.
I was actually thinking about this time ago after watching some re-balling vids. Imho the problem with freeing the 2nd tile is not as much the contents of the OTP, but the fact that the internal fuse is blown such that it can only boot from there and not from the TP1 header.
As I understand it the OTP contains a bootloader which actually loads and runs the configuration through Xrun and this can be "intercepted" to basically do the dynamic loading and SPI trick.
It does however also block any attempt to set up a static configuration through the JTAG chain and do a multi-startkit configuration using an xn. I've yet to explore declaring the 2nd startkit a device further, but those attemps have not given any result yet.
In my project a startkit with OTP in place but the fuse not blown would probably work best for static configuration, but in the mean time I started thinking about another way involving re-programming the switches dynamically.
I was actually thinking about this time ago after watching some re-balling vids. Imho the problem with freeing the 2nd tile is not as much the contents of the OTP, but the fact that the internal fuse is blown such that it can only boot from there and not from the TP1 header.
As I understand it the OTP contains a bootloader which actually loads and runs the configuration through Xrun and this can be "intercepted" to basically do the dynamic loading and SPI trick.
It does however also block any attempt to set up a static configuration through the JTAG chain and do a multi-startkit configuration using an xn. I've yet to explore declaring the 2nd startkit a device further, but those attemps have not given any result yet.
In my project a startkit with OTP in place but the fuse not blown would probably work best for static configuration, but in the mean time I started thinking about another way involving re-programming the switches dynamically.
-
segher
- XCore Expert
- Posts: 844
- Joined: Sun Jul 11, 2010 1:31 am
There is no separate fuse, the "boot from OTP" bit is part
of the OTP itself.
The OTP isn't locked.
Actually figuring out how to change the contents to do what
you want it to do will probably require a stack of throw-away
boards though ;-)
of the OTP itself.
The OTP isn't locked.
Actually figuring out how to change the contents to do what
you want it to do will probably require a stack of throw-away
boards though ;-)