diff --git a/src/classes/PMsrHandler.cpp b/src/classes/PMsrHandler.cpp
index a13ab2b6..9a6755e4 100644
--- a/src/classes/PMsrHandler.cpp
+++ b/src/classes/PMsrHandler.cpp
@@ -838,6 +838,8 @@ Int_t PMsrHandler::WriteMsrLogFile(const Bool_t messages)
             fout.width(8);
             fout.precision(NeededPrecision(fRuns[runNo].GetFitRange(j)));
             fout << left << fixed << fRuns[runNo].GetFitRange(j);
+            if (j==0)
+              fout << " ";
           }
           fout << endl;
         } else if (sstr.BeginsWith("packing")) {
@@ -941,10 +943,17 @@ Int_t PMsrHandler::WriteMsrLogFile(const Bool_t messages)
           fout << endl;
         } else if (sstr.BeginsWith("range")) {
           fout << "range    ";
-          fout.precision(2);
-          fout << fPlots[plotNo].fTmin[0] << "   " << fPlots[plotNo].fTmax[0];
+          fout.precision(NeededPrecision(fPlots[plotNo].fTmin[0]));
+          fout << fPlots[plotNo].fTmin[0];
+          fout << "   ";
+          fout.precision(NeededPrecision(fPlots[plotNo].fTmax[0]));
+          fout << fPlots[plotNo].fTmax[0];
           if (fPlots[plotNo].fYmin.size() > 0) {
-            fout << "   " << fPlots[plotNo].fYmin[0] << "   " << fPlots[plotNo].fYmax[0];
+            fout << "   ";
+            fout.precision(NeededPrecision(fPlots[plotNo].fYmin[0]));
+            fout << fPlots[plotNo].fYmin[0] << "   ";
+            fout.precision(NeededPrecision(fPlots[plotNo].fYmax[0]));
+            fout << fPlots[plotNo].fYmax[0];
           }
           fout << endl;
         } else {
@@ -4668,7 +4677,7 @@ void PMsrHandler::CheckMaxLikelihood()
 //--------------------------------------------------------------------------
 /**
  * <p>Calculates the needed precision of Double_t values for WriteMsrLogFile and WriteMsrFile of the fit range.
- * If a precision of > 4 decimal places is needed, a warning is placed and a value of 4 is returned.
+ * If a precision of > 6 decimal places is needed, a warning is placed and a value of 6 is returned.
  *
  * \param dval value for which the precision has to be estimated
  *
@@ -4679,20 +4688,24 @@ UInt_t PMsrHandler::NeededPrecision(Double_t dval)
 {
    UInt_t prec=0;
 
-   Int_t ival = static_cast<Int_t>(round(dval*1.0e4));
+   Int_t ival = static_cast<Int_t>(round(dval*1.0e6));
 
    if ((ival % 10) != 0)
-     prec = 4;
+     prec = 6;
    else if ((ival % 100) != 0)
-     prec = 3;
+     prec = 5;
    else if ((ival % 1000) != 0)
-     prec = 2;
+     prec = 4;
    else if ((ival % 10000) != 0)
+     prec = 3;
+   else if ((ival % 100000) != 0)
+     prec = 2;
+   else if ((ival % 1000000) != 0)
      prec = 1;
-   else if (static_cast<Double_t>(ival)-dval*1.0e4 > 0.1) {
+   else if (static_cast<Double_t>(ival)-dval*1.0e6 > 0.1) {
      prec = 4;
      cerr << endl << ">> PMsrHandler::NeededPrecision(): **WARNING** fit range value = " << dval << ", is asking for a rediculous resolution.";
-     cerr << endl << ">>   Will limit the precision to 4 decimal places, only." << endl;
+     cerr << endl << ">>   Will limit the precision to 6 decimal places, only." << endl;
    }
 
    return prec;