The other day I was on a site visit on a packaged roof top unit to try to assist a contractor who was having problems with some nuisance lock outs. Of course we went through all of the usual checks of gas pressures, air flow, temperature rise, etc. We went to see if his incoming power was balanced and if it was 208 or 230 volts. As we were checking across each pair, we were reading 234 volts approximately. We then checked each leg to ground and found one of the power legs was a “wild or stinger” leg. This normally is not a problem because all the pairs were 234 volts.
What we did find is that the wild leg was attached to the L1 terminal on the contactor. L1 also powers the transformer and sends the line voltage to the inducer terminal on the ignition control. We figured this may be one of the problems causing his nuisance lock outs. On a 3 phase unit, all of our controls come off either L1 or L3 on the contactor. If you have a wild leg, that should always go to the L2 terminal on the contactor because all that feeds is 1 leg of the compressor and 1 leg of a 3-phase blower motor. It does not go to any controls or single phase components.
So we switched the L1 wild leg to L2, and of course, checked rotation of the motor and compressor, found them running backwards, and reversed L1 & L3 and now everything was rotating correctly and the wild leg only powered the compressor and motor.
I could end this story here but we found one other thing.
Now I have been at this trade for a very long time and I know better than to stick a screw driver into a control panel while the unit is powered, but guess what I did? Yep — I stuck a screw driver into the powered control panel and, sure enough, touched a live terminal and then touched the screw driver to the cabinet. Needless to say, the screw driver welded itself to the cabinet (and I ruined a perfectly good screw driver in the process) but what didn’t happen is what this story is all about. With a dead short to ground — the breaker did not trip!
Now no one was hurt (other than my own ego for doing something I knew better not to do), so I checked with the contractor and he did have a ground wire from the disconnect switch to the ground lug in the unit. However, there was no dedicated ground wire in the conduit. Now, NEC allows conduit to be used as a ground but, in this case, the RTU was a replacement for a 14-year-old unit. Upon investigation, we found most of the compression couplings on the conduit running across the roof were rusted which made for an “incomplete” or poor ground. I told the contractor that he absolutely had to pull a ground wire back to the main panel.
After seeing this, and having only found this out by making a stupid mistake, this led to 2 other possible issues. The first and most important is the issue of SAFETY. With an incomplete ground, the unit was a potential hazard to anyone who worked on it. Given the right conditions, a component with a mild short, and other possible factors, if some one approached the unit and touched the cabinet and provided themselves as a better ground, there could have been very serious consequences.
The other issue with the poor ground is that of flame rectification. Without a good ground, this could have led to some of the nuisance lock outs.
Now, I don’t want you testing for a good ground by destroying a screw driver but I want to make the point that a lot of issues can be caused by both the power coming into the unit, such as the wild leg in the wrong location, and poor grounding of the equipment.
And above all — SAFETY SHOULD ALWAYS BE FIRST. Unless you are testing a live circuit with a meter which is intended for that purpose, any time a unit is powered, tools need to be kept out of the control compartment.
Hopefully, this true story will remind you how important grounding is and how important balanced power and putting a wild leg in it’s correct place can be.