Lpc2000

--- In lpc2000@yahoogroups.com, "tkreyche" <tkreyche@w...> wrote:
> 
> I'm laying out a board to test the LPC2138 and looked at circuit 
> diagrams for various 2106 and 2138 boards, to get the JTAG right.
> 
> Some of the diagrams (see Leon's in the file section) use a 74x244 
to 
> buffer the JTAG signals and optionally a transistor on the reset 
> line. The Keil 2138 board only uses a few pull-up/pull-down 
> resistors, no buffer/transistor (there are a few minor differences 
> between 2106/2138 on activating JTAG).
> 
> On the IAR 2106 board I figured they did it just to isolate signals 
> between the primary/secondary JTAG ports. 
> 
> I don't see any compelling reason to use the buffer unless you 
prefer 
> to zap the buffer with static rather than the LPC. Or maybe if you 
> have a questionable JTAG-parallel/usb converter.
> 
> Anybody (Leon?) have any insight?

Signal integrity is important with interfaces like JTAG - clean sharp 
pulse edges are required for proper operation, although the 
frequencies are low. A buffer might not be essential in every case, 
but the cost is minimal, so it's a good idea to use one. 
Manufacturers' JTAG interfaces for other devices like FPGAs are 
always buffered, although I have heard of people getting away without 
them by keeping the leads short.

Leon