Heat transfer calculation 1

[RomanKatz]

Hello friends,

I have a drum heating stove (hot air, known Q and speed). I use it do heat viscous materials. I am trying to write a heat transfer formula to use in excel that can correlate between heating temp and heating time of the stove to material temp that is inside the drum. My task is to calculate the heating time and temp so that the heated material would flow down the pump.

can someone provide any guidelines? like is it convection to the top of the drum, and then conduction via drum and conduction to the material? any thoughts would be appreciated.

Thanks.

Best Regards
Roman Katz

 

[Compositepro]

The main heat transfer mechanism will be natural convection on the inside wall of the drum and natural or forced convection on the outside wall. The viscous material on the inside wall will rise to the top and reach oven temperature relatively quickly, and material at the center will slowly drop to the bottom. After a couple hours the material at the center may not have risen in temperature hardly at all, while the material at the top is at oven temperature. Once the viscosity of the material has dropped sufficiently to allow good convective flow, the mixing will make the temperature uniform.

So the process is quite complex and almost impossible to calculate without computer analysis (CFD) and extensive material data. For this reason drum heating is almost always done "overnight". The main thing to realize is that it is very important to have effective heating applied to the bottom of the drum. Do not place the drum on the insulated floor of the oven. Heat applied to the top half of the drum is useful to prevent heat loss, but contributes nothing to heating the bottom half.

The speed at which heat travels through a material due to conduction is proportional to the "thermal diffusivity", which is the ratio of thermal conductivity to heat capacity. Viscous organic materials have low thermal conductivity and high heat capacity, resulting in exceptionally low thermal diffusivity.

 

[LittleInch]

"known Q and speed" - Is this a fan oven / stove?

What are the dimensions we're talking about here?

If you have a large volume to surface area ratio then it will different to a smaller volume to surface area ratio.

How viscous is "viscous" - do you have any data on this as it will make a vast difference.

It's probably pretty easy to heat the outside of your viscous mass, but then quite slow to heat the middle core - think baked Alaska here.



Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Latexman]

LI,

I expect it's what we call a hot box. They are very common in chemical plants. Benko is one of the big guys in the supply of these. We have at least one in every plant. Mainly, we heat viscous surfactants to speed up pumping rates. We have two sizes, an 8 drum/2 tote or a 16 drum/4 tote (double decker). Compositepro is exactly right, most instructions get simplified to the operators to start preparing for the next days production, and to heat the material overnight. We even standardise on the temperature setpoint and fan speed to simplify the operation to minimise confusion. It works most of the time. Let's see what the OP says. I'm curious too.

Good Luck,
Latexman

 

[danschwind]

I agree with Compositepro and Latexman. I've dealt with a "custom" hot box project some years ago and no one expects that the content of the drums to be heated fast. It is usually overnight (~8h).

If you need to heat it faster, I reckon that the use of a "drum heating jacket" will speed up the process compared to air as you can operate with a higher temperature. Need to be sure that the drum and the product can handle the higher temperature though.

Daniel
Rio de Janeiro - Brazil

 

[IRstuff]

For a viscous material, it's all about the transient heating of that material and its thermal conductivity and convection coefficient.

Ostensibly, one could speed up the heating of the drum with internal thermally conductive rods to get the external heat to the interior faster; or possibly have a drum that looks like a Bundt cake

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[LittleInch]

Ah. Learn something new.

Missed the bit about the "drum heating stove".

Sounds like a few temperature probes and a data recorder would be needed.

Far too many variables otherwise.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[pierreick]

Hi,
This is typically a Hot Box , normally we introduce the material the day /night before and use it the next morning .That is my experience .
Another possibility is to use electrical blanket/mantel designed for drums or IBC .
I don't think anybody will perform calculation , it's based on experience or empirical.
There is no way to measure the temperature inside the drums which are sealed during the operation .
I have seen people using hot bath to melt the product .
my view.
Pierre

 

[RomanKatz]

Hello again,

Thank you all for your replies. It is a hot box, or a fan oven for 4 drums. The viscous material is for lubricants production. It is what gives the lubricant "W", meaning high viscosity at winter, low during summer. I used 2 drum heating belts 2.5 KW each for like 4 hours on one drum during winter time and yet something remained at the bottom. Even with outside insulation blanket.

This hot oven works, just got started with it during night and heated 4 drums of this polymer and pumped one drum in just 30 min, every drop of it.
Regarding the viscosity at room temp is hard to determine since it has to be diluted but from similar materials I assume its 1000nds CST.

I understand that calculation is not going to happen, so any suggestions how can I optimize the hot box time for this material or others?

Only try and see?

Thanks.

Best Regards
Roman Katz

 

[LittleInch]

Well "try and see" also goes under the name of "experimentation"....

You could instrumentise a special drum with sealed sensors to measure temperature at various locations or try the CFD route.

But at the end of the day it's whether it works or not that makes the difference.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[RomanKatz]

Yes, You are correct. I'll try that.

Best Regards
Roman Katz

 

[Latexman]

any suggestions how can I optimize the hot box time for this material or others?

Is the hot box capacity affecting the lubricants production? I.e. is the tail wagging the dog? If no, be conservative with heating time to keep it that way. If yes, maybe more hot box capacity is needed.

Our most common problem is people forgetting to prepare the correct raw materials (RMs), and last minute production schedule changes and having to wait on RMs to heat up. When the production schedule changes, the first thing we do is load the hot box as required and get it going! The hot box planning and loading responsibilities must be crystal clear and understood by all.

Good Luck,
Latexman

 

[SuperSalad]

VI improvers or viscosity modifier polymers, like you are referring, to are notoriously difficult to get full drains from containers like a drum even at relatively high temperatures. Unless you are going to upend the drum and let it drip drain for an hour, all the ones I have experience with are simply too tacky and thick to get them all out without opening the drum and scraping the bottom and side wall.

It definitely depends on the particular viscosity modifier you are using, but a majority of the ones I've used are thick enough that I'd expect conduction would a bigger heat transfer mechanism than convection. There will be some convection, but if you're trying to build a functional preliminary model to use for production purposes, you'd be better off treating it like a heat sink (all conduction/"worst case"), then modifying it using real world measurements to fine tune the model.

That all being said, heat helps, yes, but even that won't get you to a 100% drain of a drum. This stuff just sticks

Andrew H.

http://www.MotoTribology.com

 

[IRstuff]

One calculation that would definitely be worth the effort is the Biot number, which can give you an idea how poorly the bulk thermal conductivity is, relative to the surface area. Lubricants are poor thermal conductors, and as such, they need a higher surface area to volume than other materials to achieve faster thermal changes. A Bundt cake/toroidal shape potentially can decrease the heating time. Alternately, some thermally conductive studs or other thermal shorts inside the drums would greatly reduce the distance heat has to travel in the lubricant itself.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[Latexman]

A special shaped heating vessel, though surely more effective heating-wise, would bring up contamination issues between viscosity modifiers and the final products. Hot boxes aren't expensive, and you are putting the pure RM where it goes, so I would empirically figure out how much time is good enough, and use a hot box. KISS

Good Luck,
Latexman

 

[pierreick]

Hi ,
Consider this resource with industrial application.

https://nzifst.org.nz/resources/unitoperations/httrtheory.htm

Pierre