Had a client in Youngstown get hit with a massive demand spike last month when their HVAC lockout relay failed during a heat wave. System came online all at once instead of staged startup - 340 kW spike that lasted 22 minutes. FirstEnergy's Schedule GS tariff nailed them for $4,200 extra on their July bill. Anyone seen similar failures with Honeywell W7212 relays? This is the second one this year and I'm wondering if there's a pattern here.
FirstEnergy demand spike from HVAC lockout relay failure
Jim, I've seen this exact issue with Georgia Power accounts. The W7212s seem to be failing more frequently in high humidity areas. Had three clients in metro Atlanta get hit this summer - one manufacturing plant saw a 280 kW demand spike that cost them $3,800 under Schedule PL-1. The relay contacts are corroding and staying closed instead of cycling properly. Honeywell issued a service bulletin about it in May but not many contractors know about it yet.
Derek's right about the humidity issue. Duquesne Light territory here in Pittsburgh and we've had similar problems. The key is getting these facilities to install demand limiting controllers as backup. I spec the Energy Sentry units now - they'll shed non-critical loads when demand hits preset thresholds. Cost about $2,500 installed but saves way more than that on the first avoided spike.
We've been pushing staged startup protocols with all our Cleveland commercial accounts after similar issues with FirstEnergy. The problem isn't just the relay failure - it's that most HVAC contractors don't understand demand charges. They wire everything to come online simultaneously for "faster cooling." I've started including demand charge education in every HVAC system audit. Even a 15-minute delay between zones can cut peak demand by 30-40%.
Frank makes a good point about contractor education. Down here with TVA territory, I've found that showing HVAC guys actual demand charge calculations gets their attention. Had one contractor in Knoxville change his entire installation approach after I showed him how his simultaneous startup cost a client $2,900 in one month. Now he stages everything with 10-minute delays and actually markets it as a selling point to customers.
The staging approach is critical but you also need to watch for rebound spikes. Had a Georgia Power Schedule PL-2 account where staged startup worked great until a zone sensor failed. System kept calling for cooling but never satisfied, so all zones eventually came online anyway. Created an even bigger spike than simultaneous startup would have. That's why I always recommend redundant sensors and timeout controls on each zone.
One more thing - make sure to document the relay model numbers and installation dates. I'm building a database of these W7212 failures to present to Honeywell. If we can prove a pattern, they might extend warranties or issue replacements. Have photos of the corroded contacts from three different sites now.