When the PCB design is completed, the laminate layer is determined, and the PCB board has been put into fabrication, the simulation evaluation found that the high-speed signal channel margin may be unsafe. Is there still a chance to improve?
Let’s first introduce the loss classification of PCB. There are three kinds of losses: conductor loss, dielectric loss and radiation loss.
You may hear more about the first two. In fact, the loss of PCB is mainly the first two. As for the radiation loss, it is also related to the dielectric constant DK. In addition, the radiation loss will only exist in the microstrip line. If it is properly designed, it can be reduced to a relatively low level, accounting for a very small proportion in the total loss.
Among them, the dielectric loss is mainly caused by the polarization phenomenon of the dipole. As shown in the figure below, the higher the applied voltage frequency, the greater the current, the greater the number of dipoles swinging in the material, the greater the amount of dipole movement under the action of the electric field, and the greater the volume resistivity of the medium. The power loss in the system is higher. In order to describe the material properties that measure the law of dipole motion, the concept of DF came into being.
Let me talk about another principle of conductor loss. First of all, we must know that there is an important concept in the high-speed theory called the skin effect. At higher frequencies, the current will run along the surface of the conductor, that is, at high frequencies, the size of our resistance depends on the current flowing through. The size of the cross-section, the smaller the cross-section through which the current flows, the greater the resistance, so the conductor loss also gradually increases with the increase in frequency.
To sum up the loss factors mentioned above, the sheet material mainly determines the dielectric loss, that is, the size of the sheet DF that we often say has the greatest impact on the loss, and it is for this reason that we distinguish between different grades of sheets . In addition, the line width and copper thickness of the trace affect the conductor loss.
The above mentioned is very consistent with the scenario we summarized. The board is fixed, the dielectric loss is basically fixed, the stacking and design are fixed, the wiring structure is fixed, and the conductor loss is basically fixed.
So if we still want to improve the loss in this case, we must see if there are any factors that can affect our loss.
In addition to the above factors, we also introduced the influence of copper foil roughness. The surface of the copper foil is relatively rough (in order to increase the adhesion between the copper foil and PP), so the roughness of the copper foil needs to be considered at high speed, and the roughness of the copper foil will also affect the loss of the trace.
In fact, this loss can also be included in the conductor loss. The principle is probably like this. Due to the skin effect, the current will be transmitted on the copper teeth. When passing through the protruding copper teeth, the current transmission path becomes longer than the smooth copper surface, so it will further increase the DC and AC resistance at the same time, so as to increase the conductor loss.
We are familiar with several types of copper foils with different roughness grades, including ordinary STD copper foil, RTF reversal copper foil and HVLP ultra-low profile copper foil. Of course, there are now HVLP2 and HVLP3 copper that continue to be optimized on the basis of HVLP copper foil.
But the reality is like this. Many friends know that copper foil with different roughness will affect the loss, but they don’t know how much influence it has. Can there be quantitative data to give it.
In fact, everyone thinks it is not easy. The not easy point is that the loss is composed of several large parts, which mainly include the sheet DF, the trace width and copper thickness, the reference layer thickness and the copper foil we are talking about now. Roughness.
If you want to extract the effects of different types of copper foil roughness separately, then you must ensure that the factors are consistent in order to simply extract them. The specific point is to ensure that the board is the same, the wiring structure is the same, and the thickness of the wiring reference is also the same. Only when the roughness of the copper foil is different can the influence of the copper foil roughness be known separately. Do you think it is possible to do it?
Of course, since Mr. High-speed asked, it means that Mr.High-speed must have done it! Mr. High-speed also specially made a test board for this purpose, in order to compare the difference between the popular RTF and HVLP copper foils. Yes, it is simply the difference brought about by the difference in copper foil!
Mr. High-speed makes HVLP and RTF copper foils on the upper and lower parts of the same laminate. Because the upper and lower parts are symmetrical, the wiring structure and the thickness of the dielectric are exactly the same. If the same laminate is used, it means that the board is the same, and it is the same time. Processing will not produce processing errors, because you can successfully do this by walking the same length of traces on the trace layers of the designated RTF and HVLP copper foils!
Just like this test board, we take the RTF and HVLP copper foil traces on the L5 and L12 layers respectively, and the trace length is the same (the trace length is 3 inches after de-embedding), so in addition to the copper foil roughness, Other factors can be guaranteed to be the same. In this way, the two pairs of traces are tested, and the effect of two different copper foil roughness on loss can be obtained by pure comparison.
We immediately give the loss difference tested, as shown below: From the loss results, RTF copper foil is indeed worse than HVLP copper foil (everyone knows this). From the quantitative analysis, the effect of pure copper foil roughness on loss has reached the level of 15%-20%, which can be regarded as a big influencing factor.
This article is considered to be a relatively long article. First of all, we theoretically analyze several major factors that cause PCB loss. Then we shared with you a test method of how to separately extract the influence of different copper foil roughness on the loss, so as to quantitatively get the influence of different roughness on the wiring loss.