From: Andy Peters (firstname.lastname@example.org)
Date: Wed Apr 05 2000 - 11:48:25 PDT
Thanks to all who've responded. Here's what I intend to do. I found a TI
part (CDCF2509) with built-in source terminations and nine outputs. The
oscillator will feed the reference input. The five "1Y" outputs will feed
an FPGA and its four SDRAMs (two of the SDRAMs are on top of the board, the
other two are on the bottom). One of the "2Y" outputs will feed a third
FPGA's clock input. A second "2Y" output will feed a second CDCF2509. The
five "1Y" outputs of the second PLL buffer will feed the second FPGA and its
four SDRAMs. The "2Y" outputs of the second PLL buffer will be disabled.
The board has a clocks-only layer, so we don't think we'll have too much
trouble trying to keep the clock line lengths all equal.
The only "gotcha" here is that the VME interface control chip, a Cypress
(@#$%!) part, requires an 80 MHz clock, but that part has a built-in
power-on reset circuit that enables it to configure itself. That
configuration fails if the clock is not running, and the ~1 ms PLL buffer
startup time is too long. It seems as if the only solution is to run two
traces from the oscillator - one to the first buffer and the second to the
VME chip. I don't think it's a problem keeping those traces equally short,
and the oscillator data sheet says it can drive +/- 16mA.
Again, much thanks to the gurus.
Sr Electrical Engineer
National Optical Astronomy Observatories
950 N Cherry Ave
Tucson, AZ 85719
520 318 8191
"Money is property; it is not speech."
-- Justice John Paul Stevens
**** To unsubscribe from si-list or si-list-digest: send e-mail to email@example.com. In the BODY of message put: UNSUBSCRIBE si-list or UNSUBSCRIBE si-list-digest, for more help, put HELP.
si-list archives are accessible at http://www.qsl.net/wb6tpu
This archive was generated by hypermail 2b29 : Thu Apr 20 2000 - 11:36:03 PDT