Talk:Regenerative heat exchanger

Factual accuracy and cleanup

Originally placed in the article by User:R Stillwater; moved here by User:Jafet 04:11, 27 October 2008 (UTC)[reply]

There is an error in the diagram showing the counter flow heat exchanger. The bottom diagram shows flow going from cold to hot (left to right), but it should be reversed going from hot to cold. The hot fluid heats the cold and is cooled.

Also there needs to be diagrams for a fixed matrix and rotary regenerators. Neither are mentioned in the article on heat exchangers.

When not just plain wrong, this article is confusing. The article on heat exchangers describes a regenerative heat exchanger as either a shell and tube type or flat plat exchanger, both of which would isolate the two fluid paths. This article seems to indicate that the two flows are not kept strictly separate by talking about leakage from one to the other. What? The early article does not mention anything periodic about the operation of the regenerator, something this article implies is essential to its function. If there is cyclic nature to the operation, what triggers the cycles and why does that improve efficiency?

I was left with an image of a heat exchanger where the energy exchange was between the an insulated metal exchanger that was heated by one flow for period. That flow stopped, and the fluid to be heated flowed through the heat exchanger taking up the heat, and this cycle was repeated. In a power plant with continuous operation, regenerators would need to be constructed in pairs, one heating while the other cooled. The fact that I am not sure this is correct, is because the article is so poorly written.

I wanted to understand whether a regenerator was just a particular type of counter flow heat exchanger, or something different. This article seems to imply a difference: the cyclic operation and the not entirely separate flow paths. But since I couldn't understand these differences, it didn't help me.

Editing of this page needs to also involve an improvement of the section on regenerative heat pumps listed under heat exchangers.

In reply to above (I either forgot how to indent, or never knew):

I believe the diagram is correct. The whole point of a counterflow heat exchanger is to maintain as small a temperature difference as possible between the two flows (which go in opposite directions), so that the flows into/out of each end enter/exit at the same temperature.

Aside: It is something I often think about. I came here (actually first to the "heat exchanger" article) to find out more about how efficient the process is. I am amazed to find that it is ideally 100% efficient. Which of course it has to be when you think about it, but that just didn't seem right.

I agree the rest of the article needs diagrams, and appears to be the result of joining together two related but otherwise completely independent descriptions of something(s). --Adx (talk) 12:37, 24 November 2008 (UTC)[reply]

The diagram is correct, although it could use more prominence on the flow arrows to make them more visually obvious than the temperatures. As it is, the tendency is to interpret the very obvious colour change as being related to "flow", when actually it's the opposite.

(I've also taken the liberty of indenting your reply, hope that's OK) Andy Dingley (talk) 12:52, 24 November 2008 (UTC)[reply]

Pulsed or continuous

Don't Regenerative heat exchangers have two periods ?? a hot one, and a cold one. During one period, there is only one gas flow, and the heat exchange is done between the gas and the solid matrix. This article seems to assume that the two flows occur at the same time, wich is plain wrong.

It's not "plain wrong", it's an alternative (and generally preferred) way of doing it.

Obviously you can't have the heating and the cooling flows together. So they are separated by either separating them in time (pulsed batch flow, such as a regenerator piston in a reciprocating Stirling engine) or separated in space (continuous flow) by moving the two two separate flows through different parts of the regenerator. This is conveniently done by making the regenerator as a honeycombed barrel and slowly rotating it. Flow is kept axial, along narrow axial passages, and doesn't mix between the two sides, one of which carries the hot flow and the other the cold. These barrel regenerators are popular with gas turbines, where weight isn't a problem but fuel efficiency is. Andy Dingley (talk) 15:34, 2 April 2009 (UTC)[reply]

Answer to precedent message

Ok sorry for "plain wrong", but as far as i know, a regenerator is not a heat exchanger, because the heat is transfered from the hot to the cold fluid thru a temporary heat storage (the regenerator matrix). Rummel has proposed in 1930 a model of "pseudo recuperator', in which the regenerator in the periodic steady state is modeled as a pseudo recuperative heat exchanger (recuperator). However, this is an approximation, and it as been controversial, because its not clear when this model is accurate.

So i think this article should first make a clear difference between regenerator and heat exchanger, and then talk about the fact that it can be seen as a heat exchanger in certain cases.

Also, rotary regenerator are not heat exchangers, even if the principle looks like heat exchangers. The heat is still transfered by a temporary material.

To explain it "with hands", we could say that it is the same difference between a pipe and a bottle. You can transport water with both, but the physic is not really the same. The pipe is steady, while the bottle must be filled and then empty.

Regards —Preceding unsigned comment added by 132.168.64.223 (talk) 14:02, 21 April 2009 (UTC)[reply]

I'll try to clean this up

I didn't intend to work on this but when I tried to link it to hot blast I noticed it was in error. I have some knowledge of regenerative heat exchangers on boilers from working in a steam plant.Phmoreno (talk) 23:07, 27 January 2013 (UTC)[reply]

Re:Suggested simplification needed

In response to the suggestion to include in the discussion some simplification, it would help to have some flow diagrams (including both parallel and series regenerators), which is beyond my skills. — Preceding unsigned comment added by 38.121.180.9 (talk) 18:35, 7 October 2015 (UTC)[reply]