RC snubber design

Discussion in 'General Electronics Chat' started by JimG, Jul 24, 2011.

  1. JimG

    Thread Starter Member

    Dec 7, 2009
    53
    0
    I have an Arduino-based PID controller on the boiler heater in my espresso machine. The Arduino is powered from mains (120VAC, 60Hz) through a 9V unregulated wall wart.

    Part of my PID system includes a 16 x 2 character LCD display (generic Hitachi-compatible controller). The LCD is powered by the Arduino's 5VDC on-board regulator.

    Here is the problem that I hope someone can help me solve: about 1 in 10 times when the "brew" button on the espresso machine is flipped, the LCD display becomes filled with garbled non-characters. From that point forward, the LCD remains garbled until the PID system is powered on/off. The Arduino continues to work fine even after the LCD has gone south.

    When the brew switch is flipped, two things happen in the machine: a solenoid valve is opened and a vibratory pump starts to run. The problem seems to be more prevalent when the brew switch is turned off, so I figure the collapsing field in one or both of these inductive loads is creating a nasty transient which is messing up the LCD.

    So I want to try putting an RC snubber across the brew switch on the espresso maker, but would like some advice on sizing the R and C. I have a scope, but am not sure how to set up triggering to actually see the transient.

    Jim
     
  2. ErnieM

    AAC Fanatic!

    Apr 24, 2011
    7,388
    1,605
    I would also look at these "inductive loads" as they will kickback large transient voltages. I've seen 400V off a relay. Without a schematic I could not say where to probe, but generally one side of the solenoid is fixed, the other side is switched. Probe the switch side, and trigger on levels larger then the fixed side.

    A snubber is generally used when you want to preserve the energy in the core, otherwise just a diode is required.
     
  3. #12

    Expert

    Nov 30, 2010
    16,284
    6,797
    1/2 LI^2 = 1/2 CV^2

    Measure the inductance. Calculate the highest current that could flow when the voltage is at its peak. Now you can figure energy as 1/2 L times (peak I squared).

    Decide on the peak voltage you will allow to happen. Use energy = 1/2 C times (voltage squared). That will find the capacitor size.

    Then you add a resistor so that the switch will not close the circuit with the capacitor charged and allow a (theoreticaly) infinite current. If you can read the current limit on the switch, the allowable current from the capacitor is the labeled current minus the load current if the switch is flipped at the peak voltage. The load being inductive means the instantaneous start current is zero.

    Got it?
     
  4. JimG

    Thread Starter Member

    Dec 7, 2009
    53
    0
    Thank you for the replies. Following is a link to the espresso machine schematic:
    http://www.frcndigital.com/coffee/Silvia/SilviaWorks.html

    The solenoid (3 way valve) is actuated only by the brew switch, while the pump is actuated by either brew or hot water switches.

    The pump is always connected to mains neutral (once the main power switch is on). The solenoid is completely isolated until the DPST brew switch is flipped.

    Where should I probe, and how should I set up triggering based on the schematic? I have two channels on my old Scopemeter and can trigger on either.

    I have a couple of decent Fluke DMMs (289, 189), but neither measures inductance. I understood the instructions for sizing C then R, but since i can't measure L, I might have to approach this experimentally.

    I might not have made it clear that everything in the machine is 120VAC, so I don't think I can use diodes?

    Jim
     
  5. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    You could use TVS diodes, but not standard diodes. You'd want the TVS diodes rated for greater than the peak-peak AC voltage; for 120VAC RMS, the peak voltage is ~169.7v - so you'd want TVS diodes rated for 190v or more.
     
  6. JimG

    Thread Starter Member

    Dec 7, 2009
    53
    0
  7. #12

    Expert

    Nov 30, 2010
    16,284
    6,797
    Yes. or maybe a little lower voltage, like 190 as wookie said. This is a good method, transient voltage squashers.

    Finding the inductance is easy in this case. The solenoid is operated by 120 volts, so just measure the current and calculate the inductance. If you ignore the resistance part of the equasion, you err on the safe side.
     
  8. JimG

    Thread Starter Member

    Dec 7, 2009
    53
    0
    The next one down in the series has a min. breakdown voltage of 171, so I thought that might be cutting it kinda close.

    Makes sense! I'm not sure I can do the same thing with the pump, though. It's spec sheet says it has an internal rectifier, half wave I think.

    Thanks again.

    Jim
     
Loading...