This site has helped me out so just giving back a little.
I have been working on some 30gtn's that have the usual issues. The T05X alarms that haunt the 30gtn, oil failure issues, etc etc,
For the compressor Feedback alarms, I had read about relay problems, pressure control snubbers etc.
I have had one genuine relay contact failure. I found a honeywell style relay that is just the ticket for the 30gtn, that is sold by carrier as a totaline part. Its got 2 sets of contacts, one is silver pilot duty and the other is power duty .
I have had a machine that gave a persistent T051 feedback error. Judging from all the jumpers, relays, pressure controls and what not lying in the bottom of the machine, I am not the first guy on this chiller and its problem. I tried hardwiring the molex plug on that circuit, running my own wires, hanging the relays instead of hard mounting, different pressure controls with snubber isolators, the whole nine yards. A couple of times, I got a few days out of it with no alarms, but the alarm eventually came back. I eventually resigned myself to ordering Carrier's feedback board kit, but I found out my MBB did not support this option, so I decided to try a few things before investing 4 or 5K more into this machine.
I put a datalogger across the contacts of the compressor safeties. The datalogger did not show the contacts breaking , but a slight oscillation in the voltage across the safeties. the control relay never does really break, but it causes the coil strength of the relay to slightly fluctuate. Its enough to cause the MBB to see mA 5 dc signal as failed.
My fluke was showing the same minor fluctuation across the safeties, being that its like .2v fluctuation, its not the typical contact break voltage you see like 0 then 24 then 0 etc. This is with the orifice snubbers in place.
So the work around for this is to replace the safeties with manual reset safeties. Jump the 24v loop in the cabinet, and put the safeties in the 120 volt out line from the feedback relay that goes out to the contactor for that compressor. I've done it twice now and works like a charm, and you don't have to spend a ton on the carrier kit.
Next problem, oil failures, washed out oil pumps and broken valve plates.
After trying everything from tech support and the service manual, look at the condensing temperature thermistor. Flow across an orifice depends on pressure drop. If you sporadically get the odd pressure spike on the high side, it overfeeds, then the chiller sends the EXV into a tizzy trying to correct for it. The 30GTN needs to have really stable condensing temperature to work. The condensing thermistor has to be super tight on a clean surface(and on the correct U bend). A dab of heat transfer compound helps to. The standing static temperature check does not work very well on these chillers. You have to watch the thermistor performance as the temperatures are actually changing. I have had 2 thermistors pass the standing temp check no problem, but watch your condensing pressure on the CCN compared to guage, it will never be exact as its deduced from PT tables read from the return bend, but if they stray more than 25 psi from actual guage, its a thermistor problem or the charge is out of whack.
A note about the network service tool 5. I bought it and it came with no instructions, but it can be figured out. Its a thousand times better than trying to troubleshoot off the CCN display, or navigator. The baud rate and refresh rate is a little slow, but you can at least see trends developing among groups of sensors.
About condenser coils, never install fan cycling as a cheap workaround of the motormaster option. The condenser coils can't handle it. I had one machine someone did this on, and the condenser kept popping, finally got MM installed, and its been great since. Also these chillers like to run really low headpressure compared to standard r22 machines, if you are popping circiut breakers, try to get the head down or look for a fan motor control problem. You can set the condensing temperature for the rest of the fans in the CCN, I find that 105 gives the most stable operation.
Feel free to add.