|
Advice Contact Forum SiteMap Sponsors Home PC Hardware from Made-in-China.com |
. | |
|
Prices: |
LOSTCIRCUITS |
||
| ATA Not-So-FAQs Or: Why Ribbon Cables are unsuitable for RF transmission of data |
![]() |
|
|
(Article by snn47, March 17, 2003) -.. ..-. ..... .--. -.-- | ||
Mechanical and RF-properties of the cable
Measuring and calculating the RF-properties is only easy in systems that have identical impedance from source to sink which includes all transmission components in between.
IDE, however, is not a precisely defined system; neither the interface of the controller and HDD nor the flat-ribbon-cable can be considered constant under all conditions. The specifications define some of the physical dimensions and leave too much tolerance just in the specifications of the electrical properties of the flat-ribbon-cable not even to mention the interface of the controller and HDD.
| 80 wire | 40 wire | ||
| Conductor | AWG | 30 | ~28 |
| Ground-signal-ground | |||
| Single ended impedance | [Ohm] | 70-90 | 110 |
| Capacitance | [pF/ft] [pF/m] | 13-22 42-72 | |
| Inductance | [mH] | 0.08-0.16 | |
| Propagation delay | [nsec/ft] [nsec/m] | 1.35-1.65 4.43-5.41 |

| Dimension | Millimeter | Inches |
| A1 | 50.800 | 2.000 |
| A2 | 50.165 | 1.975 |
| A3 | 0.635 | 0.025 |
| A4 | 0.6858 | 0.027 |
| A5 | 0.3175 | 0.0125 |
| T1 | 0.127 | 0.005 |
| T2 | 0.0406 | 0.0016 |
| T3 | 0.0508 | 0.002 |
| T4 | 0.102 | 0.004 |
The simplified calculation of the cable properties seems to be easy and straightforward, measuring them is definitely not, especially if it's the propertiesof the entrie system and not just the flat-ribbon-cable alone.
Formulas & Abbreviations
Impedance of a two wire line well above ground plane can be calculated with Z = [276 log (2D / d) / K] / eR
and the universal formulas for known values of L and C are:
| Z= | ____ \/ (L/C) | and | v= | 1 \/ (L/C) |
This is just the ideal approach, without any resistance in the cable or the isolation, nor any of the more complex RF-effects that may change with increased frequency like frequency like skin-effect or dielectric changes and losses. Now let's look at (hopefully correct) data from real IDE-cable from data sheets.
next page: => Real Cables =>