Changes in 20bcc96: inductive

				Tech/actuator/inductive-load-dat/inductive-load-dat.md
			
          @@ -12,7 +12,11 @@ Almost always, RS232 disconnects are from something called ground bounce, which 

           The problem is worse when the systems all share a ground. If you powered the new board with a battery, not a power supply, this can help to float the ground. Or, if the PC is a laptop, disconnect from the wall and run it on battery.  Try that first if you haven’t. if you haven’t tried, change the cycle time, where maybe the motor is on 5 seconds and then off 5 second. That makes it easier to see when the error happens- at the start or the stop of the motor. If the start, it is going to be a voltage drop on the RS232 chip, and a local capacitor of 5uf or so can make a huge difference. If this happens with the motor turning off, this is almost certainly inductive spike causing a ground bounce. 

          -I usually put some extra capacitance on CPU, comms chips, and any logic chips and make sure there is a good ground between them- either a ground plane, or at least a 20mil trace. Same with power. The caps will handle all short term power changes- but you need the larger power traces to get that power to or from the chip before the cap is no longer able to keep voltage in limits.

          +### extra capacitances 

          +I usually put some extra capacitance on CPU, comms chips, and any logic chips and make sure there is a good ground between them- either a ground plane, or at least a 20mil trace. Same with power. The caps will handle all short term power changes- 

          +

          +### power trace

          +but you need the larger power traces to get that power to or from the chip before the cap is no longer able to keep voltage in limits.

           ### oscilloscope check 

          @@ -25,6 +29,8 @@ Do you have a DSO oscilloscope? You will never see this with a power supply mete

           - add 100uf capacitors to 5V/GND and 3V3/GND

           - use a seperated 12V/2A battery, USB-RS232 board does not keeping reboot, can still can not be opened COM port to use 

           - mosfet control is already optocoupler isolated as the design, and used seperated power supply 

          +- [[DPR1084-dat]] RS232 board only connects with RXD, TXD, GND, no power supply 

          +

...	...	@@ -12,7 +12,11 @@ Almost always, RS232 disconnects are from something called ground bounce, which
12	12
13	13	The problem is worse when the systems all share a ground. If you powered the new board with a battery, not a power supply, this can help to float the ground. Or, if the PC is a laptop, disconnect from the wall and run it on battery. Try that first if you haven’t. if you haven’t tried, change the cycle time, where maybe the motor is on 5 seconds and then off 5 second. That makes it easier to see when the error happens- at the start or the stop of the motor. If the start, it is going to be a voltage drop on the RS232 chip, and a local capacitor of 5uf or so can make a huge difference. If this happens with the motor turning off, this is almost certainly inductive spike causing a ground bounce.
14	14
15		-I usually put some extra capacitance on CPU, comms chips, and any logic chips and make sure there is a good ground between them- either a ground plane, or at least a 20mil trace. Same with power. The caps will handle all short term power changes- but you need the larger power traces to get that power to or from the chip before the cap is no longer able to keep voltage in limits.
	15	+### extra capacitances
	16	+I usually put some extra capacitance on CPU, comms chips, and any logic chips and make sure there is a good ground between them- either a ground plane, or at least a 20mil trace. Same with power. The caps will handle all short term power changes-
	17	+
	18	+### power trace
	19	+but you need the larger power traces to get that power to or from the chip before the cap is no longer able to keep voltage in limits.
16	20
17	21
18	22	### oscilloscope check
...	...	@@ -25,6 +29,8 @@ Do you have a DSO oscilloscope? You will never see this with a power supply mete
25	29	- add 100uf capacitors to 5V/GND and 3V3/GND
26	30	- use a seperated 12V/2A battery, USB-RS232 board does not keeping reboot, can still can not be opened COM port to use
27	31	- mosfet control is already optocoupler isolated as the design, and used seperated power supply
	32	+- [[DPR1084-dat]] RS232 board only connects with RXD, TXD, GND, no power supply
	33	+
28	34
29	35
30	36