TAC32 Power loss

I had an overtime service call last Friday on a ThyssenKrupp TAC32 hydro - customer said the unit was down. I believe this is the only TAC32 we have on service. It does have a user interface tool (UIT) in the controller. It is pretty user friendly. The fault log on the UIT had several door faults and a couple other random faults: 1089 (Car gate closed but DCL not active), 1040 (door watchdog closing failure), 1049 (open door protection failure), 1104 (gate and lock error), 1005 (DOL was detected while HW closed),1006 (Hoistway interlock failed to close), 1007 (safety string opend during a run and caused an emergency stop), 1019 (normal limits and car position do not agree, selector lost, car position not at top or bottom landing, the I/O DL, NTST, NTSB do not agree), 1021 (a safety node is offline, inspection indicates an invalid I/O condition, the status of an I/O indicates a combination of active and inactive inputs that are not allowed).

This is 3 stops. At first if I put a car call in for 3 from the 1st floor or a call for 1 from the 3rd floor, the car call would drop out during the run. The elevator would level into 2 and stop with the doors close. If I hit 2 the doors would open or I could put a call into 1 or 3 and the elevator would take off and complete a normal run. I also noticed the floor stops were off by about 3/4".

I put the car on inspection and got on the cartop. I ran fine on cartop inspection. I got off and put it back to normal. Then I started getting door problems. The doors kept cycling like it was not seeing the correct sequence of gate lock, door lock, and DCL making up. I put the car back on inspection. I used the manual door open and door close buttons and the gate switch, door lock, and door close limit appeared to be lined up correctly.

Eventually, I just relearned the doors, adjusted the floor stops, and did a jack resync. It ran without a single fault. I cycled the power just to see if maybe there was a memory issue. The changes I made held. I stopped by the job Monday when I was in the neighborhood hitting other maintenance stops. There was not a single fault logged from the weekend and it was running great.

Friday afternoon, we did have a bunch of thunderstorms hit the area. I asked around. Someone told me that if a TAC32 loses power during a run, these problems can happen.

I was just wondering if anybody else has experienced something similar to this on a TAC32?

This also has 2 stage telescoping jacks. I do have a couple other Dover jobs (DMC controllers) that have twin post telescoping jacks. I have heard that guys use Catepillar oil additive to help relieve stick-slip problems on noisy jacks. Does the catepillar oil help?

The Cat oil does help but only if you put it right in the jack. Gorman packing also helps a lot for stick slip in leveling. If you want to see an immediate improvement, increase level speed and smooth slowdown. As far as your main concern, I have never seen anything really buggy like that on a TAC32 but we don’t really deal with lightning here. A lot of the faults you list could be caused by the car CAN channel going down. It’s a good idea to check the CAN plug on the CWIL, make sure wires aren’t on the insulation and I’ve seen things in the COP pull out or dislodge that plug when it opens. What makes the least sense to me is the door faults and the cycling you saw that was fixed by only doing a new door scan. You didn’t adjust the limits at all?

No, I did not move the limits at all. Just did a door relearn and save.

One or two months ago, a bearing in the idler arm on the door operator went bad. A new idler arm and door belts were put on. I did not do the replacement. Another mechanic did. I checked with him. He did move the DCL a little bit and did a door learn and saved it.

If i get an callbacks, I will check the CAN plug and double check all the the plugins in the COP.

Thank you for the feedback.

The door faults are probably the DCL6 point being set properly. Let me explain. You mentioned the door reopening like it didn't see DCL and a 1089 fault. As it was explained to me, TKE CPU generic software only allows a few milliseconds after making gates and lock up to see a door close limit, but the DCL and DCL6 both qualify for purposes of monitoring for the 1089 fault condition. So if DCL6 IO is at proper time before gate and interlock, you should never see that 1089. DCL6 is not an actual switch, it is an output from door card measured in motor revolutions away from learned DCL point. DCL6 is set to 0 by default and needs to be set up on door card. Its typically 12-13 on a linear operator and 3.5-4 on a harmonic operator. This can be very intermittent. So if you aren't seeing in now doesn't mean it won't return.