Thanks for pointing me to the Simocode - when I looked on Siemens website the fanciest device targeted at LV motors that I could find was the ESP200.
My dilemma as a project engineer is to implement the most advanced technology I can that can be easily maintained and troubleshot with the existing plant maintenance organization, which consists of a controls group (10 engineers with 2/4 yr degrees), instrumentation group (12 instrument techs, 1 shop supervisor, 1 engineer), and electrical group (14 electricians, 1 shop supervisor, 2 engineers). The intended service life of our installations is 50 years - we're not reconfiguring, reprogramming ,etc. as often as the manufacturing world. The maintenance folks' preference is to have each subsystem (controls, instrumentation, electrical) segregated from the others and interconnected in ways that make it easy to troubleshoot them separately. For example, any time a PLC, DCS, or instrument controls a motor we install an interposing relay cabinet in the substation, which allows electricians to troubleshoot the starters back to the relay, and allows controls/instrument techs to troubleshoot their circuits up to the relay. Aside from minimizing cross-training, safety is given as another reason for this because it keeps instrument techs and controls engineers out of 480V buckets and 5kV starters.
I just evaluated MV protective relays and decided to switch from Multilin to SEL given the situation above. The SEL automatically records a lot of data (that Multilin can only capture part of with a laptop manually connected) that can be used for failure analysis, and has the flexibility to program front panel LEDs and display messages so that connecting a laptop to a relay to troubleshoot will be rarely required - everything the electricians need to know they can learn from the front panel, the schematic (including a logic drawing), and with their multimeter. The tradeoff is SEL relays are much more complex to set up initially, but that I can manage easily on the project side - the complexity doesn't negatively affect the maintenance department.
I'm aware of AB's Intellicenter package with one HMI for the MCC that is pre-networked and pre-configured. I'm working on setting up a demo with AB so I can explore the functionality of the HMI. I've seen these in an operating facility and talked to happy customers but I haven't yet explored the HMI myself. The issue I'm anticipating is the amount of work it takes to program the start/stop logic into the device logix controller in each E3+ - right now when we commission an MCC we preload a standard settings file into the MM300 that only requires slight tweaks by electricians in the field to accomodate various motor control schemes. If the E3+ requires a laptop and is programmed in function block or something that will be a tough sell to the maintenance dept. The AB 857 and 825 are kind of odd relays - I'm assuming they exist for the customer that has gone to AB lock, stock, and barrel that wants a relay that integrates with the other AB products very easily on Devicenet, Controlnet, or Ethernet/IP. I would be surprised if anyone outside this group would buy them - most other 'casual' MV motor relay users would go to Multilin for ease of use. The 'serious' MV motor relay users tend to end up with Basler or SEL for functionality reasons, and because they are large enough to support them with internal resources. AB sells very few of these relays from what I hear - the majority of AB MV starters ship with Multilin 469 or 369 relays.
All of the above is the opposite of what Siemens, AB, and Schneider are pushing - integration of all control systems, drives, motor starters, etc. onto one network or set of networks that work together and one controller platform. AB and Siemens in particular only seem to design their products to work (or work very well) with their other products. We're already standardized on Honeywell DCS, Triconix Trident safety PLCs, and Modicon PLCs so it doesn't make much sense to introduce S7 or Logix 5000 into the mix just to be a glorified protocol converter or interface to Profibus, Devicenet, Ethernet/IP, etc. Yes these protocols are open, but they're not nearly as open as the prime vendors would have us think. The more programmable platforms we have, the less knowledge we have about any one of them, and the harder they're going to be to troubleshoot in 10 years when it stops working. One of my goals is to have a minimum number of programmable devices to deal with, which is done by eliminating gateways and protocol converters. As much of a pain as Modbus and Modbus TCP are, they are truly open and with a day or two of work, different vendors equipment can talk together reliably after an interface is built.
To some extent, our maintenance situation is understandable simply given the scale of the refinery - 5 Honeywell TDC 3000 DCS systems, 35 Modicon Concept and Unity PLCs, 6-8 Troconix Trident safety PLCs, 14 major double-ended substations, 25+ smaller substations, 1300 LV motors, 175 MV motors, 70 MW typical power consumption. I have no idea how many instruments we have, but it's probably a lot more than the number of instruments/sensors in typical manufacturing plants.
I know some process plants are embracing the traditional AB/Siemens approach to automation by using area maintenance teams with a more interdisciplinary focus instead of groups of discipline specialists that cover the whole plant. This works if the plant is willing to develop and designate subject matter experts for various areas that are resources for the area maintenance teams, but this is tough to implement well in a large facility with diverse equipment.
This got a bit off track, but I thought some background on my plant environment was in order since there are only a few process industry folks on here that know the different constraints that we face.
