[SI-LIST] : A/D and D/A Converters PWR/GND Connection

Dr. Edward P. Sayre ([email protected])
Tue, 20 Oct 1998 20:08:25 -0400

Jay and the SI list:

I agree completly with Jay regarding the use of solid grounds. The problem
with single point grounds on a system or shelf basis is that there are
always parasitic paths that at high frequencies act like another ground
path {and often there is an unintended ground or grounds as well}. Thus,
there is the real possibility of having system wide high frequency ground
loops while they are absent to low frequencies and DC. This happens often
in -48VDC powered systems.

It is possible to maintain the semblence of a single point ground on logic
cards by tying the AGND and DGND planes at one point (usually near the
connector) or creating motes and other physically isolating structures. In
the latter case, if you have a bus or other large number of lines leaving
the area, then the small connecting copper strip between the isolated area
and the "noisy ground" may turn out to be pretty wide and the isolating
feature of the solution is lost. And, don't forget, don't pass the traces
beneath the missing copper forming the isolating area or you will pay in
EMI emissions.

Specialized analog chips, especially those which span chassis ground to
logic ground such as those common in telephony designs are typically
galvanically isolated with transformers. This is an effective DC and LF
isolation mehtodology, but does very little isolation for high frequencies.
They can, however, with sophisticated design techniques become part of a
filter network to form a stop band to noise outside the band of interest.

Analog power supplies often cause thermal problems since the the drop is
entirely dissipated as heat and must be placed in accordance with the
principles of good thermal mamagement in the system. However, as Jay
says, they will fix certain noise problems, (provided that the
input/outputs are well bypassed, etc.).

Lastly, it is helpful also to include a CM choke in series with most
regulators, both analog and switching, to block logic card noise from
propagating through the power system distribution wiring.

Hope this helps,

ed sayre
57 PM 10/20/98 -0400, you wrote:
>To whomever sent this item (name wasn't listed),
>I would recommend using the most solid Ground possible. You may want to use
>"islands" for the analog and digital regions, connecting them together
back at
>the power supply point ("single point grounding"). That way you don't have
>digital noise causing problems in the analog circuitry. However, if you also
>have high-speed signals routed across the plane boundary you can have other
>problems (like EMI) due to the lack of a continuous return path. As for
>on a previous design we had good results by isolating the analog power from
>digital power using high frequency ferrite beads and ceramic capacitors
(on the
>analog side of the bead) for A/D converters. For more sensitive circuitry, a
>separate ANALOG (no, not a switcher) was necessary to provide clean power to
>the analog circuitry. It depends on what you're doing with the analog power,
>but I would not think that would be necessary for an A/D converter. Your
>mileage may vary.
>For what it's worth, Analog Devices in their applications seminars and notes
>recommends common connections for both power and Ground for A/D converters.
>Their reasoning is that they provide separate pins on the devices so the
>package parasitics (mainly inductance) don't share noise in getting to the
>outside world.
>Jay Diepenbrock
>Advisory Engineer
>Interconnect Technology & Qualification
>IBM Global Procurement, B8UA/061, RTP, NC
>Phone: 919-543-8804 Fax: 919-543-3642
>Email: [email protected]
>**** To unsubscribe from si-list: send e-mail to
[email protected] In the BODY of message put: UNSUBSCRIBE
si-list, for more help, put HELP. si-list archives are accessible at
http://www.qsl.net/wb6tpu/si-list ****

| ------------------------------------- |
| "High Performance Engineering & Design" |
| Dr. Ed Sayre e-mail: [email protected] |
| NESA, Inc. http://www.nesa.com/ |
| 636 Great Road Tel +1.978.897-8787 |
| Stow, MA 01775 USA Fax +1.978.897-5359 |

**** To unsubscribe from si-list: send e-mail to [email protected] In the BODY of message put: UNSUBSCRIBE si-list, for more help, put HELP. si-list archives are accessible at http://www.qsl.net/wb6tpu/si-list ****