Overall heat trasnfer coefficient

[Covboy69]

We've got an odd design of heat exchanger reactor (Its sort of multi pass cross flow as a best description in Heat exchanger terms.)

When we calculate values for U we get different values for whenever we run the heat exchanger at different stream 1 flowrate, stream 2 flowrates, stream 1 inlet temperature and stream 2 inlet temperature. So giving the sales guy a U value to use in the sales literature is proving difficult.

Is it normal to get U values that vary (Considerably) depending on the conditions or is something odd going on physically inside the thing or during calculations?

Hope someone can be of some help!

 

[Covboy69]

Okay so apologies for poor spelling!

 

[Latexman]

Heat exchanger reactor? Is the reaction limiting the design or the heat transfer? Is there a catalyst or is the reaction spontaneous? Is the vendor aware that his eqwuipment must satidfy both the reaction and heat transfer requirements, so they can get the right resources to work on it? If various cases must be met for operational success and knowing that this heat exchanger reactor is not typical, I would give the vendor data on all the cases and let them size it for the worst case.

Good luck,
Latexman

 

[Covboy69]

The vendors tend to want to know U values as if its purely a heat exchanger (so Hot oil in one side, cold water etc in on the other side and measure the temperature changes and flowrates and then calculate U) I was just wondering if it was normal to have U values that vary so greatly from condition to condition?

 

[Latexman]

Most vendors I've dealt with can calculate U values themselves, and I give them a performance spec (flows, temps., physical properties, but no U). There have been occasions where I have had the design done in-house and given the vendor a mechanical spec and drawing (shell size and thickness, tube size and thickness, tube count, baffle details, etc.). I don't think I'd work with a vendor that could not calculate the U!

If your flows and properties span from laminar to turbulent flow, you can have a wide range of U's.

Good luck,
Latexman

 

[Covboy69]

I think we may indeed be in the laminar to turbulent regime so that may be the problem/cause....Caculating the reynolds number for this beast is a challenge in itself...

A lot of our vendors are lab based chemists (Its only a small unit...) whom seem a little bemused by the whole heat transfer coefficient thing (No offence to any heat trasnfer savvy chemicsts reading!) so seem to want to know a U value and the higher the better...

Thanks!

 

[rmw]

As I read the OP, I understand that you are varying each stream's flow rate and/or temperature. As the U value is a function of a lots of different factors most or all of which are affected by flow rate and temperature I am not surprised that you are getting different U values. I really don't want to go into it much deeper than that other than to remind that temperatures changes change viscosity's, pressure drops, etc all of which affect the performance of heat transfer surface. Changing flow changes a whole 'nother set of factors.

You need to pick a design condition representative of whatever you establish as a benchmark with respect to throughput, output, or whatever it is that your sales guys are trying to sell and calculate and use the U value for that specific point.

Any variance from that point is a whole different world.

rmw

 

[Covboy69]

Many thanks for that answer rmw! Its what I was hoping for and was what I was assuming for a while! The T range inside this thing can be so large that the properties of say water would change hugely as it passed through it...

Many thanks for everyones Input!