Si Si Siev errrr senior.
They were never giving you the full story then. The equipment would be shipped hot back to Waltz Mill where it would be maintained between outage seasons. A system was not dedicated to a single plant but carried between plants. With minor retooling it would be used on 44's 51's and any other generator that you find out there. It was interesting using them in CE designs though.
The cooling issue was solved using a combination of venturi coolers with heat pipes for some components and a tent with a spot cooler for other parts. I don't recall all of the details about the control system but I do remember the erratic behavior of the ROSA I systems. What you are saying might help to make sense as to why they switched to brushless DC motors and multiple resolvers per axis.
Did you ever make it out to Waltz Mill to work at the mock-ups? I probably was there 8 to 10 times a year for durations of at least a week.
I quit working there in 1997 and went to work for their competition in 2000. I moved from Upstate NY to Lynchburg VA in 2001. Ultimately I left the industry in 2008 after the passing of my first wife and now serve as a lowly IT Manager. Westinghouse and Waltz Mill now seem like a lifetime ago.
That would be ROSA III and it was a behemoth. Sure it was portable in the same sense that the Navy labels anything with handles portable. The main motor control unit (Universal Amp Box) did indeed require that you take it down to modules to get it in the building. Nothing related to ROSA III would be removed from a radiological area after first use due to needing local spot coolers in containment with it. Very few things are ever intended to be carried in then removed from a commercial plant. BTW they went with 60HZ and hooked local 480v instead of power from outside. The only downfall to the system other than being heavy as hell was the fact that if you changed out an actuator or resolver it was going home to be recalibrated before being used again. The process was not doable in the field.
Considering you never actually saw the inside of containment it is no surprise that you have spotty recollection about the particulars of plant design. The problems that you describe would be mostly talking about "chiller" plants. Whoever decided packing the containment building with ice to reduce the size requirements of containment needs to be shot. A few of these plants are jungle gyms.
ROSA III went the way of the betamax as it was too expensive to maintain and much cheaper equipment could do 90% of what it could do. It was intended to be a "one robot does all" maintenance tool. It didn't succeed on that count and the costs doomed it.
If you are referring to the ROSA arm, sir your 80 lb estimate is low and if you are referring to the UAB found in ROSA III suffice to say we took it apart to get it into the building . Tube diameters were between 3/4 and one inch but the wall thickness was no where near that.
The reason plants had radiation monitoring equipment going into the plant was related to a Westinghouse engineer becoming contaminated in the north test cell at Waltz Mill and dragging that from Waltz Mill all the way to a site where it was discovered on the way out not in. So, the monitors where there to protect the plant from having to explain "unexplainable" contaminations.
The "day labor" comment was essentially true however the lifetime part of that is dead wrong. back then (until 1992) exposure limits were 3 R per quarter and 5 R per year (Federal limits, wholebody, form 4) with a lifetime limit of 1(N-18)R or one rem per your age minus 18. So assuming that you were grabbed off of a bar stool a few days before a change in quarter at the ripe old age of 23 (1(23-18)=5) and jumped into the steam generator for the typical 3-5 minutes it would take to accumulate 3 rem. A few days later when the quarter changed they threw you back in for a little less than 2 rem more. You have worked a week and hit 5 rem total committed dose. Next year you could get another rem according to the law. In 1992 the law was changed (10cfr20) and the quarterly limit went away along with the lifetime limit.
However, what utility in its right mind is going to change its story from "it isn't safe for you to be exposed to more than 1 rem for every year you have been over 18" to "sure come get all you like up to 5 rem per year. No worries." That is a lawsuit waiting to happen. I am sure even now people with high lifetime exposures are either refused access to the RCA or maybe even refused unescorted access to avoid being the cause of the person receiving more than the old 1N-18 rule.
It was a little bigger than a football actually. My arm is a little bigger than a football and it has been in that hole. I was on site for the first 5 months of that outage and was never so glad in my life to be away from some place.
Take a look at DC Cook or Indian Pt. for that matter. The NRC will shut down a plant. If you want to call a spade a spade, the pushing match over inspections vs profitability have a predictable swing with predictable consequences. Let the bean-counters convince you that they can run without inspections and pretty soon you will start having Davis Besse like events. Let the Nimbys win the safety at all cost argument and pretty soon you have 140+ day outages again like we saw as recently as the 90's that drive the price per KW/H way up. The NRC is just the poor bunch of engineers caught in the middle of this political infighting.
Thats my opinion after working in the industry for 20 years anyways.
Tell that to my Nikon 1. These little mirrorless cameras are the unsung heroes these days. The convenience of a point and shoot with the speedy focus of a DSLR. They have manual settings to boot (albeit cumbersome in certain situations). Need a slightly different focal length, no problem they are interchangeable just like their bigger cousins and even can use the bigger SLR lenses with an adapter. Since it is mirrorless it has an ungodly frame rate. Its high ISO performance looks alot like film.
Don't get me wrong, it hasn't replaced my D300 for action photography or the D7000 for low light work. It did move the D70 to the closet probably never to return and will spend more time with me than the D40.
People shouldn't get all worked up over what she said about cell phone cameras and $200 cameras being good enough. What she was saying is that for the intended presentation (online @ 72dpi) almost anything will work. Go to print a 10X14 @ 300dpi with your phone, I dare you. That little sensor has real problems with noise so that even if you had the pixels to print that large, they would be very ugly pixels indeed.
Actually the first company to jump into the mass transit AVL field is from NZ. Check your facts next time.
Given this is post accident, I doubt they go anywhere without a teletector or similar. I am also sure they carry a more mundane rate meter such as an RO-2 or similar.
You want to slow it down and run it by me again how a GM tube is a better rate meter than an ion chamber? You are kidding, right? If you were to say a GM is better at detecting then I would agree. Quantifying, not so much. I would take a semi-proportional range device for the purpose any day over a GM.
Small quibble, RO-2 survey meters are the most common that I have seen in the commercial plants in the US. The top range of an RO-2 is 5 R/hr. The 6150 Teletector w/ ratemeter option is the most common high radiation survey instrument that I have seen and it is limited to 1 SV which is 100 R shy of the needed range in this case. Other than that I agree wholeheartedly.
I suspect you haven't really worked with the technology much. Correctly configured KVM runs a Win200X server just fine. Hint - use LVM for storage instead of a container file if I/O is high on the machine. Also use virtio drivers for best performance. Most people don't go for LVM because of the added complexity moving VMs from one host to another.
I believe that the OP is.
I spoke of Proxmox earlier and I still think it would work for you. Most of the solutions (Proxmox included) will use ISCSI if available. Freenas could fit that bill nicely. Storage replication is a nice bonus to taking this route. NFS is also usable for virtual machine disks. LVM can be used also provided you are willing to setup replication where you need HA capabilities (ala DRBD).
Another happy Proxmox user here. I have been using it in production here for the last 2 years. Prior to that I had tried most of the hypervisors out there. Xen was glitchy in the networking end of things. Xen also seemed to have disk I/O performance issues. Plain jane KVM worked well but suffered on the management side of things.
Proxmox's real strength is the combination of openvz (as close to the metal as you get in VM) and the flexibility to fully virtualize any OS you care to (KVM). Add to that the fact that Proxmox handles HA and it is getting to the point that it handles distributed file systems (ceph and sheepdog). General maintenance is a breeze.
I wish that I could speak to how well it scales out but I haven't had to setup a large cluster with 100's of VM's . I imagine it does well though.
I am interested in others experience with large installs.