Re: [SI-LIST] : ATA source termination

About this list Date view Thread view Subject view Author view

From: Chuck Hill (chuckh@altaeng.com)
Date: Fri Apr 28 2000 - 11:31:48 PDT


Dennis,

Perhaps I can clarify this. On the source (driving) end, it is necessary
to provide some series resistance to dampen ringing that occurs after the
driver has switched. The on resistance of the device and the series
resistor combine to provide this damping. The ATA spec calls for this
resistive damping to be 1.0 times the characteristic impedance of the
line--my opinion is the ideal value is slightly less than that since the
ringing can still be well controlled and risetime slightly faster (LC model
instead of distributed transmission line, we are in an in-between frequency
range). Typical drivers have 10-20 ohms of ON resistance so series
resistors should be 33 to 47ohms. Higher resistance values make it harder
to achieve good DC margin on the low level (0.8v).

The 82 ohms apply to those lines which are input only. Even CMOS inputs
have some series resistance along with the capacitance. This combined
resistance on the input should be near the line impedance.

Remember, this is not a precision system. There is a lot of tolerance
allowed for. When adjusting components, look for good edge speed and not
too much ringing and crosstalk and reflection noise which gets closer to
switching thresholds.

Chuck Hill, consultant

At 11:00 AM 4/28/00 -0600, Rehm, Dennis wrote:
>I am coming up to speed on the ATA spec and have found a point where the
>spec seems inconsistent.
>
>Regarding the source termination resistors at a device, I quote from D.2.2.6
>
>"Ideal termination at each connector is when the impedance seen looking back
>toward the source matches the cable impedance in the forward direction. For
>devices, this means that the sum of driver output impedance and termination
>resistance match the cable impedance(typically 80 to 85 ohms), minus 5 to 10
>% to allow for attenuation due to the capacitive loading of other devices on
>the cable."
>
>However in Table 6 of section 4.2.2.2 the device termination impedance
>specified varies from 22 to 82 ohms. I assume that 82 represents matching to
>the cable impedance which would actually disagree with D.2.2.6 above as you
>should match to 5 -10 % less. To be consistent with D.2.2.6 this table
>should have all entries be the same at about 70 to 75 ohms. But I guess I am
>missing something.
>
>
>
>Dennis
>
>
>-----Original Message-----
>From: Sean Murray [mailto:seandm@mmspec.com]
>Sent: Friday, April 28, 2000 8:58 AM
>To: 'si-list@silab.eng.sun.com'
>Subject: [SI-LIST] : RF boards
>Importance: High
>
>
>Anyone have any feelings on what subtle differences when fabricating a
>board can do to an RF board. Such as going from FR-4 to Getek; 50/200
>gold/nickle vs. 30/100 gold/nickle. Is there any books out there that
>deal with strictly the layout of RF boards??
>
>Regards,
>Sean Murray
>Sales Manager
>
>M&M Specialties Inc.
>1236 W. Southern Ave. #106
>Tempe, AZ. 85282-4518
>Phone: (480) 858-0393 x105
>Cell: (602) 684-7859
>Fax: (480) 858-1882
>Email: seandm@mmspec.com
>WWW.mmspec.com
>
>
>
>**** To unsubscribe from si-list or si-list-digest: send e-mail to
>majordomo@silab.eng.sun.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
>****
>
>
>**** To unsubscribe from si-list or si-list-digest: send e-mail to
>majordomo@silab.eng.sun.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
>****
>

**** To unsubscribe from si-list or si-list-digest: send e-mail to
majordomo@silab.eng.sun.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
****


About this list Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Wed Nov 22 2000 - 10:50:09 PST