AlmostRetired
Chemical
- Feb 1, 2016
- 19
I’ve been a reader of Eng-Tips for a long time, but this is my first post. As my handle suggests, I’m an old stager Chem Eng just doing a bit of consulting work here in Australia as I transition to retirement. I think most of your regular posters here are from the USA, so it’s worth noting I worked in the states (Ohio) for several years 1995-2001.
I’m interested in hearing what experience anyone has with bolt-on/clamp-on dimple plate heating jackets on tanks, and any suggestions on how to improve performance.
My client recently purchased a second-hand mixing tank, about 12,000 litres capacity. They needed to heat this tank to around 70 C, with 10 bar steam their usual heating medium. My suggestion was to retrofit a dimple-plate jacket, and I even recommended a local supplier (mea culpa ….) – who at this stage will remain nameless, suffice to say they are well-reputed here, probably the Aussie equivalent of Mueller in the US for similar add-on or immersible heat transfer equipment.
Cutting to the chase … the client has called me again because, on commissioning, the heat transfer performance of this jacket was woeful. On their other tanks, all with either integral jackets or coils, the heat transfer rate is typically 100-150 kW (U = 200-300 W/m2.C or more), whereas this one, with admittedly smaller jacket area per unit volume, was only giving 15-30 kW at U = 15-30 W/m2.C.
We did all the obvious stuff – made sure we bled out all air, checked that the steam trap was working properly and of adequate capacity – and came to the conclusion that (lack of) heat transfer from jacket to tank was the only possible reason for the problem.
This particular jacket is formed in one piece to fit the profile of the tank, and is connected only by the two flanged ends of the one-piece jacket being bolted together - No other clamping bands, no heat transfer cement, no welds ….
Figuring that the issue had to be (lack of) heat transfer from jacket wall to vessel wall, almost certainly due to differential thermal expansion between the two creating an air gap, we stripped off the insulation to find (a) most of the jacket closure bolts were loose – I mean not even finger tight (they had been confirmed spanner-tight when cold), and (b) a gap of typically 2-3 mm between the jacket inner wall and the vessel.
Eureka, as the man said – differential expansion had created a substantial air gap between the two, and hence our poor heat transfer.
Obviously, doing up the bolts tight again while hot should help, but we also need to be cautious because doing them up too tight when hot might lead to the jacket crushing the (lighter-gauge) tank when it cools down. Basically, I can only say this was a very poor form of affixing the jacket to the tank.
Discussions with the coil vendor (who unfortunately got paid months ago) were fruitless – only at the end did he concede the generic statement “lack of heat transfer cement can reduce performance by up to 75%” – but with no admission as to why such a lack in the original design was his responsibility. OK, my client approved his drawing – but it only showed dimensions and nozzle orientation, no mention of the method of fixing.
Meanwhile we have sourced a suitable heat transfer cement from another supplier, it’s not expensive even for a 4mm thick application, but it is going to be a pain to retro-add – we’re gonna have to strip off all the insulation, take the jacket off the tank to apply the cement, then re-install all when done. Possibly we have a case against the original supplier, but meanwhile we will lose several days production while the tank is out of service doing the HT cement retro.
So, I’d like to know … has anyone else had a similar experiences with add-on tank jackets and an air gap, and if so how did you solve the poor heat transfer problem? Any ideas beyond those I’ve suggested above?
If we can’t get satisfactory performance from the jacket, my thought is to add some steam-heated tubular baffles in the tank – they would be physically easier to install than a coil – any comments on that?
I’m interested in hearing what experience anyone has with bolt-on/clamp-on dimple plate heating jackets on tanks, and any suggestions on how to improve performance.
My client recently purchased a second-hand mixing tank, about 12,000 litres capacity. They needed to heat this tank to around 70 C, with 10 bar steam their usual heating medium. My suggestion was to retrofit a dimple-plate jacket, and I even recommended a local supplier (mea culpa ….) – who at this stage will remain nameless, suffice to say they are well-reputed here, probably the Aussie equivalent of Mueller in the US for similar add-on or immersible heat transfer equipment.
Cutting to the chase … the client has called me again because, on commissioning, the heat transfer performance of this jacket was woeful. On their other tanks, all with either integral jackets or coils, the heat transfer rate is typically 100-150 kW (U = 200-300 W/m2.C or more), whereas this one, with admittedly smaller jacket area per unit volume, was only giving 15-30 kW at U = 15-30 W/m2.C.
We did all the obvious stuff – made sure we bled out all air, checked that the steam trap was working properly and of adequate capacity – and came to the conclusion that (lack of) heat transfer from jacket to tank was the only possible reason for the problem.
This particular jacket is formed in one piece to fit the profile of the tank, and is connected only by the two flanged ends of the one-piece jacket being bolted together - No other clamping bands, no heat transfer cement, no welds ….
Figuring that the issue had to be (lack of) heat transfer from jacket wall to vessel wall, almost certainly due to differential thermal expansion between the two creating an air gap, we stripped off the insulation to find (a) most of the jacket closure bolts were loose – I mean not even finger tight (they had been confirmed spanner-tight when cold), and (b) a gap of typically 2-3 mm between the jacket inner wall and the vessel.
Eureka, as the man said – differential expansion had created a substantial air gap between the two, and hence our poor heat transfer.
Obviously, doing up the bolts tight again while hot should help, but we also need to be cautious because doing them up too tight when hot might lead to the jacket crushing the (lighter-gauge) tank when it cools down. Basically, I can only say this was a very poor form of affixing the jacket to the tank.
Discussions with the coil vendor (who unfortunately got paid months ago) were fruitless – only at the end did he concede the generic statement “lack of heat transfer cement can reduce performance by up to 75%” – but with no admission as to why such a lack in the original design was his responsibility. OK, my client approved his drawing – but it only showed dimensions and nozzle orientation, no mention of the method of fixing.
Meanwhile we have sourced a suitable heat transfer cement from another supplier, it’s not expensive even for a 4mm thick application, but it is going to be a pain to retro-add – we’re gonna have to strip off all the insulation, take the jacket off the tank to apply the cement, then re-install all when done. Possibly we have a case against the original supplier, but meanwhile we will lose several days production while the tank is out of service doing the HT cement retro.
So, I’d like to know … has anyone else had a similar experiences with add-on tank jackets and an air gap, and if so how did you solve the poor heat transfer problem? Any ideas beyond those I’ve suggested above?
If we can’t get satisfactory performance from the jacket, my thought is to add some steam-heated tubular baffles in the tank – they would be physically easier to install than a coil – any comments on that?
