Rolling of Titanium Tube-Tubesheet connection Air cooled Cooler

Here is a tip, visit the web site below and download the designand fabrication manual (pages 25 and 26);

Metengr,

Thanks, that is helpful. Information is practical, I will read it carefully.
Do you have some personal expierence with this topic?

Thanks Euroweld

I do and the experience I have says not to groove tubesheets for Ti tubes. This is because they are hard and the expansion into the groove can cause cracking at the groove. Since we never grooved Ti tubesheets (or Ti clad tubesheets) I can't tell you if cracking occurs or not.

On the other hand, your tubes are quite a bit thicker than what I have have experience with. My experience is more in the range of 0,5 mm thick so I don't know what the effectg would be on tubes that thick.

rmw

You will need to build a mock up and test rolling conditions and the resulting pull out strengths.
You want to use the least expansion that gives good joint strength.
I have seen people used a group of 6 shallow grooves or even just a roughened hole ID in the tubesheet.
You will need to use 5 pin rolls and good torque control (do not roll to size, this will not give consistent sealing).

How will the unit be shipped after fabrication?
You will probably need to leak test and then do some re-rolling after delivery.


= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

rmw and EdStainless,
I agree with your response. Do you know of any publications that discuss the potential for cracking of thin wall tubing when expanding into tubesheets with square edged grooves? To me, it seems more like "good engineering practice" and just plain common sense to avoid this combination, but current standards and specifications do not prohibit this arrangment, even though it may be evident in the heat exchanger industry.

Cracking is not my concern, I am more worried about under-rolling and over-rolling.
This is a thin tubesheet for such thick tubes. For tubesheets like this I am more accustomed to seeing tubes with walls no more than 0.7mm.


= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

EdStainless,

I agree that cracking is not a concern for such thick tubes (2.11 mm, 0.083", 14 BWG) for the air cooler case described in this thread. When you had mentioned that you had seen the use of 6 shallow grooves it raised my question of groove design regarding the thinner wall tubes like 0.70 mm, 0.028", 22 BWG and thinner walls. Note that titanium tubes as thin as 0.40 mm, 0.016", 27 BWG are being installed, with potential for even thinner tubes of 0.30 mm, 0.013", 30 BWG. These thin walled tubes are mainly used for applications such as steam surface condensers. Either no grooves or shallow multi v grooves are typically used for these condenser tubesheet holes. The shallow grooves provide a higher pull out loading.

rmw mentioned his experience would be to not groove tubesheets with titanium tubes, and his experience has been with the thinner wall tubes of 0.5 mm, 0.020", 25 BWG. rmw reference to grooves may be the TEMA type square edged grooves.

Although this information may not apply to the original question of this thread, it may be known by a few manufacturers or users that use of grooves of the squared edged type should be avoided when expanding thin wall tubes because of the potential for cracking at the groove area. This would apply to both mechanical roller expansion and hyrdraulic expansion. When using the hydraulic expansion process, grooved holes are required to effect a seal. Common groove is a wider squared edged type, TEMA notes as 1/4" wide.

I believe the potential for cracking of thin walled tubes in square edged grooved tubesheet holes exists. Many manufacturers, like rmw's company, do not use grooves to avoid cracking of the tube. Some manufacturers will use the shallow grooves as you describe when pull out loading dictates.

Since there are few articles written on this matter, I ask the question does anyone have experience with thin wall tube cracking at grooves or know of any publications that discuss the potential for cracking of thin wall tubing when expanding into tubesheets with square edged grooves? Comments welcomed.

EuroWeld,

For your case, you might want to contact the tube supplier for their recommendations on tube expanding. If you take a look at Valtimet Technical Standards (see link) under the Tube To Tubesheet Attachment section, it recommends for mechanical expansion a tube wall reduction range of 5-12%. I believe the % wall reduction might be on the lower end of the range when expanding into like materials. Also note the caution when using the hydraulic expansion method with thick wall tubes.

You did not mention if the tubes will be welded to the tubesheets. If welded, you may be aware that titanium requires special procedures and a high level of cleanliness during the welding process. The tubes would first be expanded into the tubesheets then followed by welding. Avoid any lubricants on the rollers as they may contaminate the weld. Some manufacturers have used a dry roller approach or a safe refrigerant fluid as a lubricant. Suggest you discuss the welding techniques with professionals who do this work often.

Of our 26 Titanium heat enchanters I know the the first 200 have grooves, eith two or thre deo= depending on TS thicken.these are shallow rectangular grooves cut with a standard grooving too. All the tubes were rolled and welded.. Their approach is using an expanding arbor type expandingmandrel for very precise dimensional accuracy on expansion.. This approach has very distinct advantages over rolling or the normal hydraulic setting. Enclosed is a sketch of how the tu tube setting is accomplished.
To checkout their website just type "A-M-H.com in the address line.

Hello unclesyd,
I am not aware of any shell and tube or steam surface condenser manufacturer who use the process you mention. Also, I have not seen this type expanding process specified or approved by the refinery, chemical or power industry customers who buy shell and tube heat exchangers and power plant surface condensers. Are there any technicle articles written that desribe this type expanding process specific to the the shell and tube and condenser industry?

The groove you describe might be similar to the standard TEMA groove of 1/8" wide by 1/64" (0.015") depth. Are the titanium tubes you mention for the heat exchangers thin wall 22 BWG and thinner, or a heavier wall tube? I know that certain standards like API limit wall thickness to a heavier wall than 22 BWG. My concern will continue to be with the expanding of thinner wall tubes (22-30 BWG) into square edged grooves and the potential for cracking regardless of the expanding process. Cracking has certainly been experienced with expanding into over size holes, expanding beyond the backside of the tubesheet and excessive expanding torque and pressure.

I' the deatails Monday,
I haven't asked about any standards or approvals as ll our work is in house or on site, nothingfor the ousiide and based on some recent work done supposedly by compenet people I have have to try it.As Imentioned in another post this tooling is likethe toaster waiting on sliced bread. It's taken a long timefor me to find this a it's been out there for a number of years.
I realize your concern hope we cna find out moreinformation.

I wasn't able to get the exchanger information so I'll start early in the morning. the technician in the lab told me they are plugging, Pop-A-Plug, a tube in the older one. By his description they now have a vibration problem probably due to a loose baffle. I don't like the new approach as they will not pull or scope a failed tube, They say that their reliability program will tell them when to replace or retube an exchanger.

Here is a detail of one of the smallest of the Titanium HX's. With the thinner tube sheet these have only two grooves, while anything thicker has 3 grooves with the same weld detail. The largest on site are 70"' diameter with 2404 1" 18 ga tubes with a 2 3/4" tube sheet with 3 grooves. These ae steam calanderia. 2 of these clanderia have been in service since the 60's.
Our service is HNO3/Organic Oxidation.


I hope this picture works. You may have to print it.

UncleSyd,

Wow, when I think of Ti tubes, I usually think in terms of 22-23-24 ga. wall thicknesses.

rmw

Just remember that some of these bundles were made when fabricating Ti was bear, a big bad one, The 70" one was the biggest in the chemical industry for about a moth until DuPont built a bigger one. In fact the tubes for the first ones came from DuPont. It was a time lag until the tube making process was able to produce at thinner wall tube. In reality you call up and got what they had, not necessarily what you desired. All during this time the military was eating up all the Ti, When the military slacked off the Hawaiian heat pump project got it all.
The heaver wall tubes have proved a blessing in disguise as now they are seeing a little corrosion on the reboilers.

UncleSyd,
I could not open the picture you attached, besides, you mention 18 BWG tubes, and I am talking about the thinner tube walls like rmw mentions, including 25 and 27 ga now being used for power plant condenser applications.

One reason for the thicker tubes is that the process steam is three phase with a lot of turbulence. Any thinner tubes wouldn't last due to vibration. Vibration around this process is so bad that even though the process lines are Ti you have to switch to SS for the pump connections, Ti couples to the pumps fail in a very short time for fatigue. The stainless is usually about 6 diameters from the pumps.

UncleSyd and all,
See attached sketch of the shallow multi v groove design for thin wall tubes, 22 thru 27 BWG, that is best when rolling these thinner tubes. The TEMA type square edged grooves and similar may cause tube cracking or splitting of these thin walls. This design may be similar to the shallow grooves mentioned by EdStainless.