LP Rotor Disk Assembly

You do not specify tolerances, you just say "mounted on to rotor with negative interference".

Simple answer, heat disc and or cool shaft, fit disc, check shaft for straightness, balance assembly, overspeed spin test, check shaft for straightness, check balance again.

If you want to be really sure take the assembly up to about 50 deg C above maximum steam temperature and check that there is no deformation or change of shape due to the interference fit induced stresses.

By the way, who designed the interference fit? If you get it wrong it either moves or the overspeed spin test gets exciting.

... not enough details have been provided... . Ok, the diameter of the disc's bore is 0,55 mm less than rotor's diameter. All bores have from 430 up to 450 mm diameter.

The diameters were measured after disassembly of the discs, so the diemensions are the same like they were before. Additionally, not only negative allowance will keep discs on the rotor as each disc has a key.

The way how to mount disc on the rotor is almost ready as I wrote, but my intention is to get another point of view how to do it. I understand that it would be necessary to give more detail how we are going to do, so let me clarify what is my proposal:

The disc is heated in horizontal position. The disc temperature which makes the bore is bigger enough, to perform the assembly process, is calculated, controled during warming and finally the internal diameter of disc's bore will be checked to be sure if is more than 0,50 than rotor's diameter. If it is ok, the rotor hanging in vertical position will enter the hot disc and push it down to be sure if the disc is on its right place. The force will be visible as the disc is lying on some kind of "spring bed". We can control the force as hanging rotor can be move down or up using a crane.

Then after of couple of quarters when the temperature difference between disc and rotor will decrease up to ~ 60 Celsius deg, the rotor with disc will be lifted up and move away to turn the rotor upside down to assembly next disc. (It is double flow rotor) In the same time the secound disc will be prepared and whole process will be repeted.

The gland bushes, oil guard bushes and couplings are assembled on to the rotor when it is in vertical position.

I think more less you can have some picture what is my idea of disc assembly. I have skipped a lot of details but main message should be clear, now.

The discs prior to assembly will be static balanced and the last overhaul activity is rotor low speed blancing.

We do not have possiblity to take the complet rotor to some chamber with higher temp. but what your wrote will be done in room temp.

Anyway, thank you for your remarks

Looks like you have an ANSI B4.1 Class FN 4 Interference fit. 21 thou interference for 17" shaft roughly - seems OK to me. You may want to consider cooling the shaft as well (dry ice should be good enough) - gives you more time to fit the wheel if things don't go altogether according to paln.

Otherwise your procedure seems fine to me. I always like to high speed spin test if at all possible as it tends to relieve stresses before balancing that could be relieved with the machine in service. For the same reaon I like to heat the rotor to slightly above running temperature for a few hours. Not necessary to spin it during the process, just some heating blankets and rotate every now and then to avoid any possibility of rotor sag.

I do not think that designer took care of specs you mentioned, especially when it is russian design, but it seems that ANSI is similar to other technical documentation.

Certainly, shaft cooling is good proposal, I will try to discuss it with my colleagues.

Opening this thread I wanted to know if somebody had similar case and went trough assembling discs in diferent way, for example: mounting them on fixed in horizontal position, shaft

Anyway your remaks "Guest" are worth reading.

We are going to launch assembly process in the end of this month, so if I find/invent new idea I will use it.

Thank you so far