Propose better unit-handling in equation.

pierre-haessig · Nov 2, 2022 · 55070ff · 55070ff
1 parent f325ed3
commit 55070ff
Showing 1 changed file with 2 additions and 1 deletion.
diff --git a/Modelica/Blocks/Sources.mo b/Modelica/Blocks/Sources.mo
@@ -941,13 +941,14 @@ The Real output y is a trapezoid signal:
   equation
     y = if time < startTime then wMin else
       if time < (startTime + max(duration,eps)) then
-        10^(log10(wMin) + (log10(wMax) - log10(wMin))*min(1, (time-startTime)/max(duration,eps)))
+        wMin * (wMax/wMin)^min(1, (time-startTime)/max(duration,eps))
       else
         wMax;
      annotation (defaultComponentName="logSweep",
        Documentation(info="<html>
 <p>The output <code>y</code> performs a logarithmic frequency sweep.
 The logarithm of frequency <code>w</code> performs a linear ramp from <code>log10(wMin)</code> to <code>log10(wMax)</code>.
+It uses <code>wMin*(wMax/wMin)^x</code> instead of the equivalent <code>10^(log10(wMin)+log10(wMax)-log10(wMin)*x)</code> to help with unit-checking.
 The output is the decimal power of this logarithmic ramp.
 </p>
 <p>For <code>time &lt; startTime</code> the output is equal to <code>wMin</code>.</p>