Correct pipe-up for a desuperheater

[desA]

The Alfa Laval Technical Manual (4 ed), gives pictures of two different piping arrangements for desuperheaters:

1. Top down



2. Bottom up



Can anyone explain where each arrangment is best used, and why?

How would condensate build-up in the desuperheater be managed, if at all?

[Tesla]

Hi desA
The condensate build up in the second picture would be controlled by the flow rate and temp of water. With a small margin so that the refrigerant would stay outside (to the right) the envelope on an enthalpy chart. Theoreticly any way but with pressure drop any condensation would tend to log in the bottom. So a riser trap double tube could be required at the gas inlet to the desuperheater. In saying all that I just wouldn't pipe a desuperheater with the inlet at the bottom - it could be very dangerous.
Use of the top diagram would be used on a heatpump as a preheat for hot water. It would also be usefull in hot climates or hot days reducing the load on the compressor. Also on a co2 transcritical system. I'm just trying to think where the lower diagram would be used?

[desA]

Thanks Tesla.

Very interesting when it comes to the possibility of condensate build-up in the desuperheater. If height were available, I'd suggest to mount the desuperheater higher than the condenser, with condensate fall into the condenser inlet - then it would be free-flowing. In some situations this is not so simple. To reduce height, the desuperheater can also be laid horizontally, allowing condensate to drain (simple system).

[desA]

In saying all that I just wouldn't pipe a desuperheater with the inlet at the bottom - it could be very dangerous.

Can you expand on why it would be dangerous? Water steam-out, or condensate logging??

[Magoo]

Hi DesA.
De-superheating is not condensing, the discharge gas remains in the vapour stage. Generally relates to between 15 and 20 % of THR. Variable conditions can really screw things up. So I generally have heat recovery water loop pump speed controlled relative to exit vapour temperature.
magoo

[nike123]

♦ A separate desuperheater is installed, figure 2. A BPHE
is excellent for this application. Here a small water flow
is heated to 80 °C and the remaining latent heat is
cooled in either a water or air cooled condenser
In a PHE the pressure drop is normally, large enough to
allow for upward vapour flow. The inevitable condensation
means that droplets (liquid refrigerant and oil) have to
move upward and a certain minimum pressure drop is
necessary, in the order of 1 - 3 kPa/m flow length.

page 101..........

[desA]

Hi DesA.
De-superheating is not condensing, the discharge gas remains in the vapour stage. Generally relates to between 15 and 20 % of THR.

This is fine for a steady condensing temperature. Things begin to get interesting when the Tc,sat value changes over a heating cycle, though.

Variable conditions can really screw things up. So I generally have heat recovery water loop pump speed controlled relative to exit vapour temperature.
magoo

Fair-enough. Problem again is what to do when you have a heating cycle with changing Tc,sat? Surely, if you're wanting to use control, you'd need more than one temperature to be certain that no condensation takes place?

[desA]

:
♦ A separate desuperheater is installed, figure 2. A BPHE
is excellent for this application. Here a small water flow
is heated to 80 °C and the remaining latent heat is
cooled in either a water or air cooled condenser
In a PHE the pressure drop is normally, large enough to
allow for upward vapour flow. The inevitable condensation
means that droplets (liquid refrigerant and oil) have to
move upward and a certain minimum pressure drop is
necessary, in the order of 1 - 3 kPa/m flow length.

page 101..........

Thanks nike - that is a good reference. Had noticed that as well.

This will place some restriction on the system performance, in terms of additional pressure drop. This will have to be closely watched, I would think.

A very interesting juggling act.

[nike123]

Fair-enough. Problem again is what to do when you have a heating cycle with changing Tc,sat? Surely, if you're wanting to use control, you'd need more than one temperature to be certain that no condensation takes place?

If you use differential temperature between pipe at exit of desuperheater and outside air temperature (or condenser water in temperature for water cooled condenser), I presume you could control water flow thru desuperheater with variable speed pump and some 4-20mA exit on PLC controller.

[Magoo]

Hi DesA,
The primary mover is heat recovery that is driven by discharge gas temperature, so either have a variable volume water control driven by temp., or better would be pressure control for straight line control, but would be more expensive to manage component wise.Like transducers and PLCs.
Perhaps the good old thermo bulb water reg.. vav would be applicable for smaller units. They can be pretty course in regulating though. The pressure driven ones are softer control.

[Tesla]

It could be dangerous if/when liquid oil/refrigerant logs in the bottom of desuperheater - when the logged liquid is rapidly heated and pressurised on compressor startup hydrostatic pressures could event, rupturing some part of the system around that point. Condensate (some) is inevetable due to oil seperation and refrigerant durring off cycle. Perhaps a float level control used as a drain. For control I think both pressure and temp need to be measured to calculate enthalpy. This may also be an application where P, I and D are used to control heat transfer.

[desA]

Hi DesA,
The primary mover is heat recovery that is driven by discharge gas temperature, so either have a variable volume water control driven by temp., or better would be pressure control for straight line control, but would be more expensive to manage component wise.Like transducers and PLCs.
Perhaps the good old thermo bulb water reg.. vav would be applicable for smaller units. They can be pretty course in regulating though. The pressure driven ones are softer control.

Thanks very much, Magoo. Applied wisdom, as usual.

[desA]

It could be dangerous if/when liquid oil/refrigerant logs in the bottom of desuperheater - when the logged liquid is rapidly heated and pressurised on compressor startup hydrostatic pressures could event, rupturing some part of the system around that point. Condensate (some) is inevetable due to oil seperation and refrigerant durring off cycle. Perhaps a float level control used as a drain. For control I think both pressure and temp need to be measured to calculate enthalpy. This may also be an application where P, I and D are used to control heat transfer.

Surely this would occur in a condenser just as easily. There would always be some amount of liquid in the condenser, after the system stops at the end of a heating cycle. Why don't we see these explosions at this point?

[Tesla]

Simply because we don't (commonly) pipe the inlet at the bottom of the condenser. In fact in Aus and NZ I believe the standards clearly states there should be a free draining path for any condenser. Is this piping configuration found in practise? I am interested to hear. I wrote "rupturing" not exploding - an explosion implies a flamable liquid. For memory there was a post of a simmilar event on this forum early last year or late year before. Also in the field I have (when I was young and dumber) experienced hydrostatic pressure build up when I excessivly over charged a system which had a blocked TEV.

[desA]

Simply because we don't (commonly) pipe the inlet at the bottom of the condenser. In fact in Aus and NZ I believe the standards clearly states there should be a free draining path for any condenser.

Ok, you're referring to the bottom in pipe arrangement. Fair comment in that situation.

I'd be interested in seeing what the AUS/NZ law has to say about a free-draining path for a condenser. To me, this makes perfect sense, anyway, to be honest.

Is this piping configuration found in practise?

If you're referring to a free-draining condenser path, I'd have to say that this would probably be fairly rare in many heat-pumps. At worst, the condenser discharge runs along the base of the unit around condenser discharge level, through a filter-drier, then off to the TXV - often with some lift component.

Actually, for the HWHP types, with integrated storage tanks & heat-pump sat on top of the tank, it is physically impossible to have a free-draining condenser. In that case, the sub-cooled liquid is lifted some meters up to the heat-pump.

I am interested to hear. I wrote "rupturing" not exploding - an explosion implies a flamable liquid. For memory there was a post of a simmilar event on this forum early last year or late year before. Also in the field I have (when I was young and dumber) experienced hydrostatic pressure build up when I excessivly over charged a system which had a blocked TEV.

Ok, fair comment. Rupturing.

[Tesla]

I wonder desA
could this hydrostatic effect be utilised safely in a HWHP system to transfeer the pressure to temperature advantagiously for a more efficient energy transfeer? I think we may now have the control technology to achieve this - what are your thoughts on this? I was taught on my first heat pump that a lot of extra refrigrant charge (head pressure) was required to achieve the desired temp in the condenser.

[Tesla]

Allow me to ellaborate desA
With new advances recently in mechanical technologies such as bearingless compressors, titainium *(silverplated) heatexchangers, and transcritical processes (prof clevlend's) we can achieve and control much greater pressures/temperatures in a vapour compression system. To add improvement we would add some Schoubuger fluid flow technology with a little sacred geometry (3D) for heatexchanger design ideas... Then we can move on to plasma technologies (the fifth state - that we know of) but more being considered

[desA]

I wonder desA
could this hydrostatic effect be utilised safely in a HWHP system to transfeer the pressure to temperature advantagiously for a more efficient energy transfeer? I think we may now have the control technology to achieve this - what are your thoughts on this?

Can I ask you to perhaps re-phrase your question?

I was taught on my first heat pump that a lot of extra refrigrant charge (head pressure) was required to achieve the desired temp in the condenser.

When you say ' achieve the desired temperature', are you referring to Tc,sat?

What you'll find is that Tc,sat is fairly closely linked to the amount of liquid back-up in a plate condenser. If the charge is low-enough to run the SC ~ 0K, the Tc,sat value will drop off fairly quickly. Addition of a small amount of charge, over the point at which the liquid just begins to sub-cool, will raise Tc,sat a fair few K.

Oddly-enough, I'm presently quantifying just how sensitive this SC 0K point is, as part of my experimental program. So far, it does seem to be pretty much a go -no go affair. Very interesting. I'll work up the range to see where the Tc,sat sensitivity begins to drop off.

Now, past experience has shown that tube-in-tube condensers do not seem to be so sensitive, but, I'd expect thats mainly because they almost always run some SC, due to their long thermal contact lengths.

I would be careful to add too much refrigerant, since Tc,sat heading upwards, would be lowering COP, unless absolutely necessary.

The final positioning of Tc,sat will also depend very strongly on the water heating trajectory within the condenser itself. A high TD, tends to lower the Tc,sat. It would not make sense to add charge to try & overcome that - better to balance the system on SH & SC, in my view. I currently have systems where COP,av over the whole heating cycle (T ~ 30 - 67'C), is around 5.25 - under Asian conditions - circulating water flow - far higher for once through.

[Tesla]

Ok desA
Do you think the hydrdostatic phase is worth while investigating for the heat transfeer process on a HWHP? Where the pressure avalanches at it's rate in increase.

[desA]

^ I think that I may be able to answer that a bit more fully over the next 2 months, or so - from detailed test results.

Practically though, if the condenser outlet temp is dropping off, this ends up cooling the incoming water flow, where you actually want to be heating it. This is counter-productive for a heat-pump, but does provide some stability for the TXV, in terms of liquid cooling.

[Gary]

In the bottom up configuration, the potential problems (on both refrigerant and water sides) are not insurmountable, but then the question is... why bother?

Is there some inherent advantage to bottom feed?

And if so, why isn't the desuperheater in figure 1 bottom fed?