[TCML] + 30mA = dead garage door opener

[Joe Mastroianni <joe@xxxxxxxxxxxxxx>]

Blew up my garage door opener last eve.  I'm still not comfortable with rolling the coil out into the driveway.  I put a vent fan in the roof of the garage, in fact, to evacuate the O3 & NOx, so I wouldn't have to coil in public.   I went as far as unplugging the opener whenever I ran my tests.

Alas - even unplugged, the opener couldn't take it. 

I had expected the radio would overload or maybe the front end filters would fry.  Not a chance.  I am subscribing to the "ain't much RF from a coil" theories proffered on this site.  

I am still running the 4.5" x 22" secondary with double toroids - a 3x12 at 23 inches and a 4.5x18 at 28 inches as measured from the bottom of the secondary.  All of my prior coupling problems were solved on the last go around with help from Bart  and JavaTC. (things go infinitely easier when you have the expert at hand )   I am now running the secondary with the base in the same plane as the primary - no altitude adjustment necessary, now that I've carved 2.5 turns out of the inside of the primary.

The sequence of events is this - I acquired a new 15/60 NST.   Previously I had been running a 15/60 in parallel with a 15/30.   I swapped out the 15/30 for the new 15/60, figuring I'd ramp up the power in increments so I could observe the effects with my SRSG based coil. 

Now effectively running 15/120,  I did not increase the primary cap tank.  I plan to do that this weekend.  Slowly increasing the variac I noticed two effects:

1) The SRSG timing was ever so slightly off now.  Where before it was running perfectly with no phase adjustment, I had to kick in about 3-5 degrees (my estimate based on prior measurements with this Freau phaser arrangement) to get all electrodes to fire reliably.  Undoubtedly the added amperage filled those caps faster,  and possibly inductance of the additional NST changed the phase a bit.  

2) No safety gaps fired until  I got to about 2/3 power, but I had not readjusted them with the new NST, so there's no surprise here.  

3) The streamers were far more "energetic", and what I mean by that is that where before a streamer would form and stick around for some fraction of a second before evaporating and being replaced by another - they stick-around-time of the streamers was shorter.    Overall streamer length was about the same at 2/3 power as full power before, and this is a qualitative measure, but it seemed within the realm of what might be expected with a potential 33% increase in rated input current.  Of course, the actual increase in current depends on the loading of the NSTs. Tilbury has some curves in his book suggesting a very linear inverse relationship between NST output V and I with a dV/dI of somewhere in the neighborhood of -0.3, (actual equation:  Vo(kV) = -0.3*Io(mA) + 0.9798 ) suggesting that at max rated current the output V would be under 10% of rated output.  

I did take care to unplug the opener, and I had in fact even wrapped the power cord around an RF choke.  Alas, this precaution was a completely inept attempt. The streamers energized the garage door spring, which is connected directly to the gantry holding the garage door opener motor + controller + drive belt.   Some currents must have flowed through that gantry into the housing of the opener itself.  The effect is that the door will now only open.  The opener only closes the door on occasion (during which time it appears to functional normally).   The radio control is unaffected, which leads me to believe there is no RF-induced damage.  It's strictly a streamer hit.

Oddly, the streamers also hit the wiring for the garage door IR sensor.  Those seem to function normally as well during my tests.

My opener is a Chamberlain model 3280.  It appears from the code it is throwing that either the RPM sensor or the logic board itself is fried.  I have ordered replacements for both, as the components are significantly cheaper than the whole motor (e.g. ~ $80 + shipping). 

When the replacement parts arrive I will dismantle the unit and see if there is any obvious spark damage visible.  My greatest fear is that what has happened is that one of the limit switches has been somehow compromised and that the parts I replace have nothing to do with the failure - and that the problem can be solved by simply bending a contact back into place.   But I dont have the patience to dismantle the opener in the absence of any repair parts.  So that experiment will wait till later this week.



Cheers, 
Joe_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla