RE: [SI-LIST] : Merits of low dielectric constant

About this list Date view Thread view Subject view Author view

From: George_Tang@Dell.com
Date: Fri Jun 23 2000 - 14:23:36 PDT


I am not a cable vendor, but let me pretend to be one. To make a coaxial
cable for 60 GHz signals, we can do some quick calculations.

To avoid making this cable into a metal waveguide (no higher order modes)
the cutoff frequency of the shield should be greater than 60 GHz. For
starters, let's assume the dielectric constant is 4. 1/2 wavelength of 60
GHz in the medium is about 50 mil. The shield diameter should be less than
50 mil to have the proper cutoff frequency. But using air for dielectric
(Er =1), you can use 100 mil for shield diameter. We'll keep this in mind
for now.

Now let's find the skin loss. To speed up the calculation, I use a
stripline simulation tool and find that for a 50 ohm stripline (trace width
= 20 mil, copper planes spacing = 40 mil, trace centered in between planes,
trace thickness = 1 mil, Er = 4), copper loss is about 1.5 ohms per cm @ 60
GHz. That's pretty significant. The stripline has similar properties as a
coax, and the dimensions I chose are not too far off from an actual high
frequency cable.

To improve on this, it would certainly help to use wider center conductor
for lower copper loss. Wider center conductor gives greater capacitance, so
to keep the 50 ohm impedance, we need to decrease Er. Also, from above, to
keep the same cutoff frequency, we can use a 100 mil diameter shield if Er =
1. A wider shield also allows us to use a wider center conductor to keep
the same ratio (b/a) for 50 ohm coax line. It seems that a low dielectric
constant material does improve the design greatly.

Regards,

George Tang

-----Original Message-----
From: Hassan Ali [mailto:hali@nortelnetworks.com]
Sent: Friday, June 23, 2000 6:37 AM
To: si-list
Subject: [SI-LIST] : Merits of low dielectric constant

I attended a presentation by a high-frequency (1GHz < f < 65GHz) coaxial
cable vendor, and the presenter claimed that their cables use a material
with a very low dielectric constant and therefore are ideal for high-speed
application as they give rise to low capacitive loading. He gave a formula
showing the capacitance (I think per unit length) decreasing as you decrease
dielectric constant. This claim, however, perplexed me as I don't know how a
cable's capacitance per unit length would give rise to a capacitive loading.
All I know from my transmission line classes, a lossless transmission line
with Z0 = sqrt(L/C) would transmit signals exactly the same way regardless
of the value of the p.u.l. capacitance C as long as the ratio L/C is
maintained. Am I missing something here?

Thanks.

Hassan.
 

--
Hassan Ali <hali@nortelnetworks.com>
Equipment & Network Interconnect, Nortel Networks
2 Brewer Hunt Way, Kanata ON, K2K 2B5 Canada
Tel: 613-765-1410 (ESN 395) Fax: 613-765-5512 (ESN 395)

**** 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 - 10:50:43 PST