Hi all,
I have two questions about a tubesheet I want to model. See the image I uploaded for an explanation of the situation. The tubesheet is used as a (blind)flange too and has an extra head that is screwed onto the tubesheet.
The image represents a tubesheet with U-bend tubes, but I have a same sort of construction with an heat exchanger with one fixed and one floating tubesheet which both have the same kind of head opposite to the tubes screwed on the tubesheet.

My first question is how I can model such a configuration in PV-Elite? For the stationary tubesheet it looks like configuration e. For a floating tubesheet there is not such a configuration in ASME section UHX. Is this even possible?

For the configuration e I've uploaded two PVI files. One with one flange and one with two flanges. I'm having trouble getting the channel on it correctly. How should I model it?

Thanks in advance for your replies.

Corne,

I think the flanged channel bolted to the back of the tubesheet is a configuration that the code rules do not address. Is it feasible to reduce the outside diameter of the tubesheet below the bolt pcd and have another wn flange on the channel side to clamp the tubesheet? This is an easier configuration to model.

ps The two .PVI files do not appear to be available for download and cause an error when attempted.

Stan, thanks for your reply. Don't know why the PVI files don't work. Strange. I'll re-upload them tomorrow at work.

I have to calculate the tubesheets this way. These are re-builts for existing heat-exchangers. Our client however has lost all information about the old exchanger except some rough sketches but he wants the new ones exactly the same.
I actually have to calculate them according to the Dutch RToD, but as these have less different situations for calculating tubesheets in comparison to ASME I already turned to ASME for the calculations.

I wanted to use 'configuration e' cause the channel is there and the pressure differences will be the same as in the 'configuration e' case. Only my channel will be more flexible. So adding a very flexible channel, and not looking at the channel results itself must give me a good impression of the needed tubesheet thickness I guess. The channel can be evaluated as a seperate pressure part.

I have troubles with adding the channel model correctly though. If anyone has any ideas or comments, feel free to respond. The files will be back online tomorrow (I hope).

I've tried to upload the two PVI files to my private webspace, which didn't work either.

They are now uploaded there as 1 zip file: zip file. I hope this is downloadable for all.

What am I doing wrong in these files when a want to model 'configuration e' with a channel?

Thanks in advance for your answers.

corne,

I have looked at the two pvi files but am not sure where the difficulty lies. The files run but warn of overstress at hydro. Can you be more specific.

The stresses are the least of my concerns atm. First I want to get the model right.

As you see the first model has a standard tubesheet in between two flanges. As I've picked tubesheet 'configuration e' with a flange on 1 side and a channel on the other one I want to get rid of 1 of the 2 flanges in my model.
That should be model 2. I've modelled a piece of straight pipe to reflect the channel, but I have a feeling that's not the way to go.
So how do I correctly model 'configuration e' with a channel on 1 side of my tubesheet and a flange on the other?

corne,

If you are using 'configuration e' then the tubesheet is integral with channel and the way you have modelled model 2 would match that configuration but,

The sketch you included on the first post suggests that what you have is a tubesheet gasketed both sides which would be 'configuration d'.

Model 1 looks closer to the actual layout, although element 20-30 would be channel side (and not shell side as modelled).

The tubesheet is not gasketed on both sides. It's used as a flange too.
Construction:
- shell
- flange
- gasket
- tubesheet

The tubesheet has a bolted channel, but this channel is smaller than the actual shell. That's why I find it not OK to model it as a 2-sided gasketed tubesheet. There will be bending stresses due to bolt forces too.
I want the in practice bolted channel modelled as a normal channel on the outside of the tubesheet (configuration e). Maybe a little less stiff cause of the difference in connection. I can't seem to find a correct way to model the channel though. I can model the tubesheet using the tubesheet input screen and choose configuration e there, but how do I set the size of the channel (which is the measurement for the area where the difference in pressure is).

Can anybody enlighten me on how to enter this in PVElite?