Metal Core as a separate choice
Metal Core PCBs (MCPCB) was in version 0.6 and 0.7 thought to be specified by the materials. This does not seem logical anymore with the new material setup discussed in another thread. I suggest that Metal Core is specified as a separate element:
"metal_core": {
"present": true/false,
"material": "aluminum"/"copper" #(others can be added later),
"thickness_minimum": X #(in micrometers),
"thickness_maximum": X #(in micrometers),
}
Agree?
On the stackups, the core can be set in the same manner, but without the "present"-tag, as this is given if present in the stackup.
I did deliberately not add a pure "thickness" tag, as setting a min and max to the same number would yield the same result.
-
As a follow-up question: Should "heavy_copper" be a choice in addition to "aluminum" and "copper"?
-
Andreas Lydersen It must also be possible to include the "metal core" (or what I might rather call "thermal substrate") in the stackup description. Although the thermal layer is usually at the bottom of the stackup, it could be in the center of the stackup.
Although the material is likely either aluminum or copper, it may be desirable to specify the grade of metal and/or the thermal conductivity.
As to specifying max and min but not thickness: The paradigm should be consistent throughout CircuitData. One disadvantage to the approach here is that it leads you to assume that the nominal value is the average of max/min. Which prevents a specification such as 2.0 +0.1/-0.05 mm.
-
Bruce McKibben Would you be OK with using "thermal substrate" as an alias, so that it can be search for but is not the name of the element? I see that most manufacturers state "metal core" as their capability.
What kind of values are "grade" and "thermal_conductivity"? Is there a special UoM that can be used for them, assuming they are numbers?
I totally agree on the fact that min/max should be consistent throughout the language, and will look into it. I'm doing the same thing on the proposed material change as we speak (actually "as we chat"). The last thing you mention should be taken care of by the proposal from Athanasios Kallinis in his post about tolerances https://www.circuitdata.org/t/m2zck0/what-about-tolerances
-
Andreas Lydersen I am ok with the alias. Note: Bergquist refers to their products as thermal substrates (trademarked Thermal Clad).
I don't actually know a lot about metal-based board materials. But I have seen different aluminum grades mentioned, such as Al 5052 and Al 6061. For example, see http://www.bergquistcompany.com/thermal_substrates/base.htm
Thermal conductivity is in units: watts per meter-kelvin (W/m·K)
-
I'm a bit uncertain as to if adding grade and thermal_conductivity is a bit over the top at this stage? Do you see it as a need when specifying a board Bruce McKibben ? Are there any other opinions out there? Athanasios Kallinis
-
Andreas Lydersen I am unqualified to say much about how customers wish to specify their metal-based boards. I can only report what I have seen in material descriptions. Thermal conductivity is a parameter we added to our system a couple of years ago.
