<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"> <HTML><HEAD> <META http-equiv=Content-Type content="text/html; charset=iso-8859-1"> <META content="MSHTML 6.00.2900.2180" name=GENERATOR> <STYLE></STYLE> </HEAD> <BODY bgColor=#ffffff> <DIV> Hello List,<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o:p></o:p> <o:p> </o:p> Last week piano technician John Ross from <?xml:namespace prefix = st1 ns = "urn:schemas-microsoft-com:office:smarttags" /><st1:place><st1:City>Windsor</st1:City>, <st1:State>Nova Scotia</st1:State></st1:place> provided us with very useful information that has helped us make progress on solving a minor, but persistent quality problem associated with the H5 Humidistat in our Piano Life Saver System.  The problem and our current approach to eliminating it are detailed in the following discussion.<o:p></o:p> <o:p> </o:p> The challenge with a small percentage of the H5's is that upon installation, the yellow low water warning light works fine.  However, sometime after installation, the yellow light fails and will not turn on when the water in the Humidifier is low.  Usually, this is discovered by the piano owner who calls the technician stating one of the following two things, either  “It has been a long time and the yellow light has not come on to tell me to add water.”  Or “the red light is on and the yellow light has never come on”.  The technician makes a service call and finds the tank nearly dry in the first case and completely dry in the second.<o:p></o:p> <o:p> </o:p> John encountered this problem, and happened to have two H5 Humidistats in his shop that he could use for the replacement.  First, he decided to test them in his shop and found that one of these failed within the three hours time.  While this frequency of occurrence far exceeds the rate we have seen in the field, the coincidence was very fortunate as the information broke a technical impasse we were experiencing in finding the root cause of the problem.<o:p></o:p> <o:p> </o:p> John and I talked on the phone at length about the issue.  He has had some experience in electronics prior to becoming a piano technician.  He thought there might be some sort of thermal effect underlying the problem.  He suggested that proximity to a dehumidifier, as in a vertical installation, might be providing heat to cause the problem.  I explained that we had seen it in grand pianos on occasion, where the Dehumidifiers are much farther from the Humidistat.  The idea of temperature being involved was new and kept me thinking.<o:p></o:p> <o:p> </o:p> We set up tests where we sealed the vent holes on the Humidistats and operated them overnight.  Sealing the vent holes allowed for excessive heat build up inside the Humidistat.  The vent holes normally operate to dissipate heat from the transformer on the circuit board inside the Humidistat.  (The transformer was added to the H5 to meet the increasingly stringent Underwriter’s Laboratory  safety standards for our product.)  However, the inside of the Humidistat does get slightly warmer during operation.  With the vent holes sealed, the inside of the Humidistat reaches a much higher temperature than during normal operation.  Using this procedure we were able to duplicate the failure in a small number of Humidistats in our in-house inventory.<o:p></o:p> <o:p> </o:p> I met with our design engineer, Keith Howell, and our senior electronics technician, Randy Kaufman on Friday.  Under microscopic examination we had previously observed hairline cracks in a capacitor in the yellow light circuit in units returned to us from the field.  However, the cracks were found in returns where we could not duplicate the problem as well as in units where we could.  At our meeting on Friday we agreed that the warming of the case during the first few hours of operation was enough to widen such a crack and cause the circuit to fail.  While this doesn’t tell us what is causing the crack it gives us an excellent new approach to investigate this more thoroughly.<o:p></o:p> <o:p> </o:p> Our next steps are as follows.  We are testing all Humidistats under the extreme condition of having the vent holes sealed during overnight operation.  Failures are removed from production and will be closely analyzed for why they failed.  Gayle, our ten-year old son, Robbie and I spent the weekend testing our entire inventory.  Product shipped from our factory as of Monday, April 28th will be free of this defect.<o:p></o:p> <o:p> </o:p> We will begin testing bare circuit boards while hot to see if we can reject them prior to assembly.  If they fail under these conditions, it will show that the cracking of the capacitor is being done by our circuit board supplier.  If this is so we will audit their factory to determine where in their process the problem is caused.<o:p></o:p> <o:p> </o:p> If it does not appear on the bare boards, we will look even more closely at our process to see if there is anything we are doing to cause a physical shock to the circuit board that could create a hairline crack in the capacitor of concern.  We have studied our process closely and do not see anything of concern.  However, the recent thermal hypothesis and the use of heat in our testing will finally allow us to make a definitive determination and move forward.<o:p></o:p> <o:p> </o:p> John has suggested that technicians test Humidistats in their shops prior to installing the System.  This seems like a prudent idea until old inventory works its way through the distribution system.  Simply place the Humidistat in an upright position (vents facing upward as in normal installation).  Plug in the two-light panel and let the Humidistat run for a couple hours.  Contact us at 800-438-1524 if you encounter a problem and we will quickly furnish a replacement.<o:p></o:p> <o:p> </o:p> I am grateful for John’s input to the situation.  It really has made a difference.  Thank you, John.<o:p></o:p> <o:p> </o:p> If any of you have encountered this situation in the field, I apologize for the frustration you must have experienced.  Also if you have a return that you have not sent back to us, please do.  It will help us solve this permanently at its root cause, rather than eliminating the defect through testing at the end of our assembly process.<o:p></o:p> <o:p> </o:p> I welcome your comments, concerns and suggestions.  Please contact me off the list at <A href="mailto:roger@dampp-chaser.com">roger@dampp-chaser.com</A>.<o:p></o:p> <o:p> </o:p> Sincerely,<o:p></o:p> <o:p> </o:p> Roger Wheelock, VP<o:p></o:p> Dampp-Chaser Corporation<o:p></o:p></DIV></BODY></HTML>