PDA

View Full Version : Open flash intercooler question







Den123
06-04-2017, 08:47 AM
Hello everybody.
Maybe someone can help me. Now I am calculate a horizontal open-type flash intercooler for ammonia. I would like to clarify that whether the high-stage compressor will not be flooded. Maybe there are specialists with real experience with this vessel type.
Here are the conditions for selection, drawing in attachment.
Two stage cycle with open flash horizontal intercooler with one piston compressor for two stages and two mecanical float regulators.
Capacity: 164 kW, ammonia,
Evaporation temperature: minus 35 C,
Temperature intercooler: plus 10C,
Condensation temperature: plus 35 C,
Vapor speed 0.25 m/s.

mbc
06-04-2017, 02:09 PM
HI
Donot for get to put 1 or 2 holls in top of pipe insid flash tank on the your intery gas from first stage to cool in flash intercooler

Den123
22-04-2017, 09:09 AM
Thanks, it s anti-siphon hole?

Josip
22-04-2017, 09:54 AM
Hi, Den123 :)


welcome to RE forums...


Hello everybody.
Maybe someone can help me. Now I am calculate a horizontal open-type flash intercooler for ammonia. I would like to clarify that whether the high-stage compressor will not be flooded. Maybe there are specialists with real experience with this vessel type.
Here are the conditions for selection, drawing in attachment.
Two stage cycle with open flash horizontal intercooler with one piston compressor for two stages and two mecanical float regulators.
Capacity: 164 kW, ammonia,
Evaporation temperature: minus 35 C,
Temperature intercooler: plus 10C,
Condensation temperature: plus 35 C,
Vapor speed 0.25 m/s.

Your intercooler is OK, but I have some suggestion ...

open flash intercooler we can use with f r e o n 22 (and similar refrigerants) as the pressure within flash vessel is still enough high to push the liquid refrigerant to expand down to the evaporator pressure...

with ammonia this pressure within flash vessel is close to evaporating pressure ... for that reason the better solution is closed circuit system ...

Also liquid make up is better to be at bottom to inject ammonia into liquid ....

Of course do not forget that hole on gas injection pipe coming from the first stage ... must close to the top of the vessel ... you can also install outside (3/$") pipe as a jumper between gas inlet pipe and vessel ... in this case you are sure that hole is not forgotten ;)

Hope this will give you some idea ...

Best regards, Josip :)

Den123
22-04-2017, 11:02 PM
Thanks, i use two mechanical float regulators for that punp system: higt side regulator between condenser and flash intercooler (+35 to - 10) and middle side regulator between intercooler and circuit receiver (-10 to -35). Inside of intercooler i make two holes 10 mm dia. First (anti-siphon hole) on top of first stage pipe inlet inside of vessel, second on bottom of suction to hight compressor stage elbow for condense drain.

Josip
23-04-2017, 05:15 AM
Hi, Den123:)



Thanks, i use two mechanical float regulators for that punp system: higt side regulator between condenser and flash intercooler (+35 to - 10) and middle side regulator between intercooler and circuit receiver (-10 to -35). Inside of intercooler i make two holes 10 mm dia. First (anti-siphon hole) on top of first stage pipe inlet inside of vessel, second on bottom of suction to hight compressor stage elbow for condense drain.


Maybe you can upload your scheme ...much better to understand ....

circuit receiver (-10 to -35) ? ... is that suction separator for -35*C system:confused:



Best regards, Josip :)

RANGER1
23-04-2017, 06:49 AM
No sign of oil drain off bottom of vessel, it would be mandatory as well as oil collection vessel to avoid any oil buildup.
Also liquid level make up & float level control column a. It close.
You do not want level to e effected by incoming high pressure liquid.
No relief valve stub as well.

Tycho
26-04-2017, 08:40 PM
The open flash intercoolers I have dealt with have been a pain to get to remain stable, however these have been on systems with 1-4 screw compressors. On one of the systems we had to replace the open flash with a closed one because it was impossible to maintain a steady level with the varying load. this was one where the LP receiver asked for liquid from the intercooler, and then the intercooler asked for liquid from the HP receiver.

The biggest concern I have for your design is that it is horizontal and not very large, which will mean that you would need a very low and stable liquid level to prevent the level from surging, and also that you plan to use the intercooler to subcool liquid before it is introduced to the -35 part of the system.
The problem/challenge I can see you will be having is how to maintain a stable liquid level in a horizontal intercooler

I am interested in knowing how you would be controlling the liquid level over the three stages.
HP receiver -> intercooler -> LP receiver.

If it had been me, I would have had a low level float on the LP Receiver that called on liquid from the Intercooler, and then the intercooler would call on liquid from the HP receiver.
this would still place a very varying load on the intercooler, and when calculating the valves between each stage you would have to take into consideration that from the HP receiver to the intercooler the difference is +35 to +10C, so you would need to calculate the capacity of the valve in that temp/pressure range, to let through enough liquid to maintain a stable level in the intercooler. And also when feeding the LP drum from the intercooler, you need to calculate the size of the valve to have the capacity to feed from +10 to -35, and this valve has to correspond with the valve that is feeding the liquid from the HP receiver to the intermediate receiver to make sure that you are not feeding to little liquid into the intermediate receiver.
because if you are underfeeding the receiver, the gas from the first stage doesn't have enough liquid head to condense and then the intermediate suction pressure will rise, and the intermediate suction temperature will rise, and the intermediate discharge temperature will rise proportionately to the suction temperature.

I have been on systems like this, and they are near impossible to get to work stable, at least my experience is that they are notoriously unstable, and once you have gotten it dialed in at 100% load the compressor decreases capacity and everything is unstable again.

you didn't say what the system will be doing, is it production with varying load, or storage with a very stable load, so I can only give you my personal opinion, and my personal opinion is that if you absolutely have to/want to subcool the liquid from the HP recieiver, then I have to go with Josip and say that you should go for a closed intercooler.
I'm guessing it is a productionline system with varying load, since you want an intercooler though :)

I have serviced a system with a gram piston compressor that must have been at approximately the cooling capacity your system has.
The compressor is servicing 4 vertical platefreezers with 30 blocks, 75mm width, with a freezing time of 2.5 hours.

This system has a standing intercooler, I can't vouch for the internals as I have only seen the schematics of the system, but I imagine it to be something like this:
14766

it's been a while since I have been there, but for the life of me, I can not remember if the liquid from the HP receiver went through the -10 intermediate stage before it went to the -35 stage, or if it went straight from the HP receiver to the -35 stage.
I am pretty sure it went straight from the HP receiver to the -35 stage, and the intercooler was fed with liquid from the same HP receiver.



As it's been a while since I've been there, and it's not far from where I am, I'll drop by them tomorrow and have a look and take some pictures for you:)


If you have height restrictions and have to have a horizontal intercooler, I would have to add a few changes :) (no offence, just my opinion)

I would lengthen the Liquid inlet pipe, so that liquid is introduced under the liquid level, to prevent too much disturbance of the liquid, if liquid is sprayed in above the liquid level it would cause a lot of splashing and liquid droplets in the gas stream that might be picked up by the suction to the 2nd stage.
I would also remove the snorkel on the liquid outlet pipe, to prevent oil from gathering in the intermediate vessel, and let the oil settle in the oil separator below the LP receiver.
Like in this drawing: ("drain holes" marked as have been mentioned earlier in the thread :))
14764


Another altenative, if you don't want oil to go to the LP receiver is to add an oil sump in the intercooler, but then I would also redesign the vapor inlet pipe, to make sure that any/all oil would have to move past the trap before it reaches the liquid outlet.
in this instance I would have kept the snorkel on the liquid outlet.
This however would mean that the operators would have to be diligent in draining the oil.
14765










Also as a last tiny question for all of you, usually on the single two stage piston compressor systems I have been on, the intercooler works at -10 not +10 (pressure converted to temperature, because it's easier than dealing with psi, psig etc etc :)) and I see that if the intermediate suction pressure rises from -10 to around 0C, the intermediate discharge temperature rises drastically, the compressor usually shuts down because of high intermediate discharge temperature if the intermediate suction temperature rises above 0C (gram/york/sabroe compressors)

RANGER1
26-04-2017, 09:28 PM
Tycho,
I'm assuming he is using Witt floats or similar for liquid level control.
Agree all flow rates have to balance to be stable which is the beauty of the Witt float system.
One on condenser drain leg or liquid receiver, one on intercooler, then excess into low side accumulator, like a weir system.

Tycho
26-04-2017, 10:12 PM
Tycho,
I'm assuming he is using Witt floats or similar for liquid level control.
Agree all flow rates have to balance to be stable which is the beauty of the Witt float system.
One on condenser drain leg or liquid receiver, one on intercooler, then excess into low side accumulator, like a weir system.

I'm assuming he is using Danfoss SV4 floats because they are also very stable, and also pretty much same in function as the Witt, though with a different design, and also because I have mostly worked with danfoss components ;)

Since he didn't say anything about the system other than the design of the intercooler vessel itself, we are both just guessing here :)

so my concern for the liquid flow was that many Engineers, and I mean engineers as in office engineers, like to have the liquid level control on the HP receiver, so when the level rises on the HP rec, they want to dump it into the LP side.
However, with an intermediate vessel, if they dump the HP liquid into the intermediate vessel, the level will rise faster than it is able to drain.

Which is why I like systems of this design to be of the kind where the LP vessel "asks" for liquid and then starts draining the intermediate vessel, and then the intermediate vessel "asks" the HP vessel for more liquid.

So in reality there should only be 1 level controller on the intermediate vessel, and that should be the one that feeds it liquid when the level is too low.

so he could do with 2 level controllers, one that feeds the LP drum from the intercooler when the level in the LP drum is getting low, and one on the intercooler that feeds the intercooler liquid from the HP drum when the intercooler has a low level.

But then again, this is a small 164Kw system with a single 2 stage piston compressor, so it will be a compact system, and based on the size of the intermediate vessel in the drawing, and my experience with similar systems, I personally have doubts that a horizontal open flash container of this size will be a good solution.

600 by 1200 mm approx by looking at the drawing, there is going to be pandemonium inside that vessel when the system is running.

again, this is all my opinion :)

RANGER1
27-04-2017, 09:21 PM
Tycho,
Not that it matters, but drawing indicates 790mm inside diameter & if to scale about 3000mm long:D

Den123
02-05-2017, 12:40 AM
Many thanks for all answers. I made a mistake in my first post, intercooler temperature is minus 10 C. This refrigeration system is make for two cold storage (1000 tonn of fish in every cold store), than load is stable. I make a new drawing for vessel. 1476814769

Tycho
04-05-2017, 07:35 PM
Many thanks for all answers. I made a mistake in my first post, intercooler temperature is minus 10 C. This refrigeration system is make for two cold storage (1000 tonn of fish in every cold store), than load is stable. I make a new drawing for vessel. 1476814769


Please remember that I'm not an Engineer, just a service technician :)

I have spent some time to re-educate myself, to make sure I am on track, and I came across this gem: http://www.pfri.uniri.hr/~bernecic/literatura/PPO_BS_BPUS/Multy_stage_ref_comp.pdf
Where it is explained in detail and in words that even a service technician can understand :)

the design I have seen most when it comes to two stage piston compressors is the one shown in Fig.126(a) on page 11 of the PDF, including the water intercooler. the water intercooler has been a pipe jacket welded on the outside of the 1st stage discharge, where water flows counter flow to the discharge.

If I understand your last drawing correctly, the level control for the intercooler will be like this:
LP side asks for liquid from the intercooler, and then when the intercooler gets low level, it will ask for liquid from the HP receiver?


Hope this helps :)

Den123
06-05-2017, 04:29 AM
Thanks, here is the system 14787 Regulating is make by two mechanical float regulators. No solenoids, step motor ets.

RANGER1
06-05-2017, 08:30 AM
Den, Have seen this design on a number of jobs, plant is critically charged to a certain degree, as any excess liquid always ends up in low side accumulator.
low side accumulator has to be able to handle any excess liquid when on defrost, pump cavitation fault etc.From PID it shows only 2 air coolers, so defrost "could" be an issue to generate enough hot gas to defrost, usually about 70% on 30% defrost.
Have seen a plant like this & they incorporated a heat sink for defrost,basically coil in a tank of water generating a load.Depending on climate it is in something would have to be incorporated to assist if I'm reading drawing correctly. There is a pressure regulator in discharge line to divert some or all hot gas generated, hopefully it would be enough.
Good luck hope all goes well

Tycho
08-05-2017, 07:26 PM
What Ranger1 said, plus, I have some questions :)
I have looked at your PID, don't know if it's just me, but the one I downloaded was of bad quality.
14788

1: why the hotgas connection to the LP drum and intercooler drum? I like the possibillity to "drain" the HP gas into the LP drum, but I don't get the connection to the intercooler drum.

2: I notice that you have connected the hotgas for defrost to the same pipe as the liquid inlet.
I have a pet peeve with this because most fan evaporators have the "octopus" connection, on ships (where I work mostly) evaporators and plate freezers have nozzles on the liquid inlet, and these nozzles are where dirt or slow moving oil will accumulate.
Therefore if the hotgas is coming in the same way as the liquid, it will take the path of least resistance.

So in my opinion, the flow should be reversed when running hotgas.
Hotgas should come in on the suction side and be forced through the entire coil from the oposite side of the liquid inlet, and the hotgas overflow should be on the inlet side of the evaporator.
This way, the pressure can build up all across the evaporator, and help force debris or oil that is stuck in the evaporator.

3: what is this? Airpurger?

4/5: Is this an equalizing line between the Condenser and intermediate cooler across the float valve? or just a service line where the valve can be opened to dump the liquid from the condenser into the intermediate vessel?
I'm asking because I don't understand. If it had been me, I would have had a pilot receiver along with the float valve out of the condenser, because I "think" the float valve will be open all the time and let the intercooler get filled with liquid. Probably I'm just worried because I am used to systems with more consumers where the HP receiver will fluctuate more :)

6: what is this heat exchanger for?

Tycho
09-05-2017, 12:17 AM
Not criticizing here, just trying to learn and see different ways of doing things :)

RANGER1
09-05-2017, 09:30 AM
Not criticizing here, just trying to learn and see different ways of doing things :)

Tycho, my take on things,
Liquid receiver is low temperature accumulator, no need for anything between condenser/intercooler.
Heat exchanger for glycol underfloor heating.
Air purger yes.
Hot gas defrost agree, but can be done either way.

Den123
11-05-2017, 03:56 AM
Ok, i try to explain :rolleyes:. I dont known how can i attache a pdf file with good quality, only image. Below my answers.
1. Yes, this is service hot gas line to circuit reciver and intercooler for empty vessel.
2. For defrosting we use both ways for hot gas: with connect to liquid line and out to suction line for DSI plate freezers or with hot gas connection to suction line vith out from liquid line, and i agree with you than the best desicion its change of medium direction during defrost, it must be hot gas inlet to suction line and defrost outlet from liquid side of evaporator. We remake it this way, thanks.
3. Yes, this is Hansen airpurger.
4. This is purge line with solenoid valve to purger.
5. This is hight side float regulator.
6. The plate heat exchanger ,four-condenser with partial ammonia gas condense, approx 38 Kw for glicool heating to plant self need. It must be switch off by close inlet and outlet valves.