diff --git a/FilterDerivative.cpp b/FilterDerivative.cpp
index 6446267..5e101bf 100644
--- a/FilterDerivative.cpp
+++ b/FilterDerivative.cpp
@@ -2,7 +2,7 @@
 #include "Arduino.h"
 
 float FilterDerivative::input( float inVal ) {
-  long thisUS = micros();
+  unsigned long thisUS = micros();
   float dt = 1e-6*float(thisUS - LastUS);   // cast to float here, for math
   LastUS = thisUS;                          // update this now
   
diff --git a/FilterDerivative.h b/FilterDerivative.h
index 30aa2f1..c49cc96 100644
--- a/FilterDerivative.h
+++ b/FilterDerivative.h
@@ -3,7 +3,7 @@
 
 // returns the derivative
 struct FilterDerivative {
-  long LastUS;
+  unsigned long LastUS;
   float LastInput;
   
   float Derivative;
diff --git a/FilterOnePole.cpp b/FilterOnePole.cpp
index 8c42eb7..7dae08f 100644
--- a/FilterOnePole.cpp
+++ b/FilterOnePole.cpp
@@ -17,7 +17,7 @@ void FilterOnePole::setFilter( FILTER_TYPE ft, float fc, float initialValue ) {
 }
 
 float FilterOnePole::input( float inVal ) {
-  long time = micros();
+  unsigned long time = micros();
   ElapsedUS = float(time - LastUS);   // cast to float here, for math
   LastUS = time;                      // update this now
 
diff --git a/FilterOnePole.h b/FilterOnePole.h
index 8035759..c0ab393 100755
--- a/FilterOnePole.h
+++ b/FilterOnePole.h
@@ -25,11 +25,11 @@ struct FilterOnePole {
 
   float X;      // most recent input value
 
-  // elapsed times are kept in long, and will wrap every
-  // 35 mins, 47 seconds ... however, the wrap does not matter,
+  // elapsed times are kept in unsigned long, and will wrap every
+  // 71 mins, 35 seconds ... however, the wrap does not matter,
   // because the delta will still be correct (always positive and small)
   float ElapsedUS;   // time since last update
-  long LastUS;       // last time measured
+  unsigned long LastUS;  // last time measured
 
   FilterOnePole( FILTER_TYPE ft=LOWPASS, float fc=1.0, float initialValue=0 );
   
diff --git a/FilterTwoPole.cpp b/FilterTwoPole.cpp
index 7a8c6fa..3d1ed1d 100755
--- a/FilterTwoPole.cpp
+++ b/FilterTwoPole.cpp
@@ -120,7 +120,7 @@ void FilterTwoPole::setAsFilter( OSCILLATOR_TYPE ft, float frequency3db, float i
 float FilterTwoPole::input( float drive ) {
   Fprev = drive;                      // needed when using filter as a highpass
 
-  long now = micros();                      // get current time
+  unsigned long now = micros();             // get current time
   float dt = 1e-6*float(now - LastTimeUS);  // find dt
   LastTimeUS = now;                         // save the last time
   
diff --git a/FilterTwoPole.h b/FilterTwoPole.h
index 9b63f1c..4d0384e 100755
--- a/FilterTwoPole.h
+++ b/FilterTwoPole.h
@@ -41,7 +41,7 @@ struct FilterTwoPole {
 
   bool IsHighpass;  // false for normal output, true will make a lowpass into a highpass
 
-  long LastTimeUS;  // last time measured
+  unsigned long LastTimeUS;  // last time measured
 
   FilterTwoPole( float frequency0 = 1, float qualityFactor = 1, float xInit = 0);