Issue link: https://iconnect007.uberflip.com/i/994883
50 DESIGN007 MAGAZINE I JUNE 2018 removed prior to testing. Most materials used in the PCB industry have relatively low thermal conductivity and copper has extremely good thermal conductivity. A common range of ther- mal conductivity for high-frequency materials is from 0.2 W/m∙K to 0.5 W/m∙K and copper has about 400 W/ m∙K for thermal conductiv- ity. When a material is tested for thermal con- ductivity and if copper remains on the sample being evaluated, the thermal conductivity test method will report a much higher value than if the sample has the copper removed prior to testing. Since most thermal modeling software requires thermal conductivity values of the raw substrate, it is more appropriate for the test procedure to exclude the influence of the copper during this test. Another consideration for data sheet com- parisons is peel strength and several items can be an issue for fair assessments. One issue that can be a significant concern is the use of one copper type to generate lower insertion loss and a different copper type used to report the peel strength value of the material. It is well known that a copper type which has a smooth surface (at the copper-substrate inter - face) will have lower insertion loss due to lower conductor loss. However, the smoother copper will typically cause the peel strength to be lower. Some material suppliers will often use smoother copper to demonstrate better inser- tion loss performance, while their data sheet will have peel strength values generated from using a copper with a rougher surface. This is usually noted on the data sheet as a footnote next to the peel strength value, which cau - tions the reader that the peel strength test- ing was done with a different copper than the copper used for other test methods on the data sheet. DESIGN007 John Coonrod is technical marketing manager at Rogers Corporation. increase in frequency will cause the Dk to decrease slightly when tested within the micro- wave range of frequencies or within the lower millimeter-wave range of frequencies. Again, it is possible to test the same material in two different tests, which are using the same test method and get two different answers for Dk and both answers are correct. That can be true due to the frequency-dependence of materials and when using the same test method but at different frequencies, the Dk value should be different. There are also similar accuracy concerns for comparing data sheets as it relates to dissipa- tion factor (Df). The main concern regarding Df is typically the test methods being com- pared at different frequencies. Df is frequency- dependent and with an increase in frequency, the Df should increase. Sometimes data sheets will show the Df values at 2 GHz or 2.5 GHz when targeting applications in that frequency range and then other data sheets will report Df for the material when tested at 10 GHz. The Df should be higher at 10 GHz than 2 GHz, so the designer needs to make sure the frequency and test method is the same when comparing Df values for different materials. Additionally, there are other scenarios to con- sider for different properties and one example would be the thermal conductivity of a mate- rial. There are different test methods for deter- mining the thermal conductivity of the material and when using the same test method, there are different sample preparations which can alter the results of this property. One notewor- thy difference for testing thermal conductivity and using the same test method is whether the material is copper clad or if the copper was The main concern regarding Df is typically the test methods being compared at different frequencies.