Hot oil leaks

[axl2005]

Hello Wise Fellows

I am using a reactor heated/cooled with a external dimpled jacket, to perform the reaction we need to heat it until reaction is finished after that we agitate and cool very fast our product.From 400 F to 80 F in 2 min. Our heating/cooling media is mineral oil. We have been experienced oil leaks very often. Normally we weld this part of the jacket, but we know that we are putting more stress in the wall metal. (We believe that is because of stress metal during cooling and heating cycle).
Do you know any material that we can use instead of weld it?

Material Stainless Steel 316

Thanks again for your support

 

[moltenmetal]

Are you asking how to seal a threaded joint against severe temperature cycling like that?

There's a product called "Expando" which is the only thing I'd even consider for this service. It's a graphite-bearing cementitioius pipethread sealant. A real bear to use and to disassemble afterward, but when used properly it is VERY effective at preventing hot oil leakage in threaded fittings.

But why not just weld it?

 

[TBP]

moltenmetal - the first time I went looking for "X-Pando", I spelled like you did, and got nothing related on my google search

 

[axl2005]

I can weld this reactor but we waste a lot of time making this. Normal downtime is about 48-72 hours. We expect to solve this problem in a few hours.

rgds

 

[Compositepro]

Metal structures can experience tremendous stresses when going though such rapid temperature change. Unless your reactor was specially designed to do so it is likely that changing the temperature so rapidly is destroying the equipment. Unless the reactor is very small you are also not likely too be cooling the batch as fast as you describe but only the jacket fluid temperature. The problem is parts of the reactor cool faster than others and that results in stress due to dimensional changes. Equipment must be designed appropriately to handle these conditions.

 

[unclesyd]

As stated above you are cooling the SS reactor too fast. SS especially doesn't like to heated or cooled very rapidly due to it's high thermal expansion. Any heatsink in the reactor will cause extremely high stresses in that area with resultant cracking due to differential temperatures. This can happen even on the surface of the metal being heated or cooled.

At what temperature are you introducing the cooling media?

 

[moltenmetal]

TBP: we've got a case of it in cans and it keeps forever (as long as you don't open the cans!) so we won't have to Google that for the next ten years or so. And I was too lazy to go back on the shelf and look up the tradename spelling!

 

[TBP]

We've got about a quart can that will last us for years - as long as we keep it sealed. We normally don't like to use it, because it's sometimes a one-way trip It also isn't very well suited for maintenance work, because production people want things back in service 15 minutes ago, and won't let it cure for the required 24 hours.