HGBP with unloaders

[cool-hvacer]

Hi all...

New to the site, but think it is great!! I am a Mech Eng with limited HVAC/R but have a glycol chiller system that I have been asked to help with. So any comments would be mucho appreciated!

I have a few fundamental questions with respect to hot gas bypass / compressor unloading... Our system has both (2 stages of unloading on the compressor and a HGBP system). All of them are hooked up to a thermostat that activates them sequentially, in 3 deg increments.

I have gone through the control schematics, and it appears that the HGBP gets activated first, then the two stages of compressor unloading, and then the Liquid Line solenoid (pumpdown to low pressure cut out). Does this order sound right?!?!

Also, the HGBP valve that is currently on the system is rated for the full capacity of the chiller. Since the capacity reduction for the chiller is suppost to be 100-67-33-0, should the HGBP valve be specified for only 1/3 of the total capacity?!?!

Thanks in advance!!

[cool-hvacer]

Hee, hee....

No takers?!?!

Well... I just reread my original post, and I guess the question is not so clear....

Should the compressor be unloaded first and then HGBP or should it be HGBP and then compressor unloading?

Or does it matter?!?!

[Brian_UK]

Hey, give us a chance to finish work, eat food and sit at the computer ;)

Don't be so impatient.

Personally I would expect the compressor to unload first and then the HGBP come into play.

[US Iceman]

Unload the compressor all the way down to the last step of capacity control, then allow the hot gas bypass to come on.

The other way is a complete waste of energy and the compressor may not unload, because the hot gas may keep the suction pressure up.

[cool-hvacer]

I just came across an article (by Zentner) that suggests you have to size the valve for the entire capacity, even if you offload at the compressor first.

Does this make sense??

The arguement was that when the solenoid kicks, the valve will already be opened the appropriate amount?

Still - does this preclude the idea of using a valve sized only for the remaining load?!!?

Anyone with experience on this?!?!

TX!

[US Iceman]

...that suggests you have to size the valve for the entire capacity, even if you offload at the compressor first.


Think about this.

If the hot gas valve is allowed to energize before the compressor unloads, then the hot gas will allow gas into the evaporator (or suction depending on configuration) as soon as the load starts to decrease. The addition of hot gas for the bypass will keep the suction pressure high close to normal (if adjusted correctly), which means the compressor will not unload (if suction pressure controls the unloader mechanism).

So... what you have is a full hot gas bypass which keeps the compressor fully loaded all of the time (irrespective of the actual cooling load).

Will it work? Sure.

Is it cost efficient? No way!

[cool-hvacer]

Oops... OK - let me clarify a bit more....

The system we currently have is actuated by a thermostat on the glycol lines - this thermostat offloads in three stages. Two compressor unloads and one HGBP.

I think we are both in agreement that it is currently wired incorrectly. HGBP unloads first, then 2 compressor stages. So I am going to change the thermostat contacts to do the HGBP last.

What the Zentner article mentions is that your HGBP valve is modulated by downstream pressure, and probably sits behind a solenoid valve. So - even though the HGBP valve may be open, the solenoid will not kick on until you reach the final stage of unloading.

This is where I get a "little fuzzy / lost":

The article suggests that you need a valve big enough to handle the entire tonnage, and even though, you will not be passing the tonnage, the pressures will be correct.

I don't understand why a valve for 1/3 (or the last of the capacity) will not be adequate?!!? And it seems to me that when that solenoid clicks on for the 3rd stage of unload, a huge blast of hot gas is going to go into the evaporator....

:rolleyes:

[US Iceman]

OK, so if the HGBV is sized for 100% of the load, but the hot gas solenoid valve does not energize until the load drops to about 25-33%, then the hot gas solenoid and HGBV will be operating at about 25-33% of their total full load capacity. For the HGBV, this will have poor control capability.

If the compressors have one step of capacity control and you have two compressors and one HGBV then you have something like this:

Both compressors fully loaded = 100% capacity

Both compressors; one fully loaded & one half loaded = 75% capacity

Both compressors; both half loaded = 50% capacity

Both compressors; one off & one half loaded = 25% capacity

Both compressors; one off & one half loaded w/ HGBV = 0 to 25% capacity

Or something like that. This is just an exmaple of how it could work.

[Gozzy]

I fully agree with Iceman. All that HGBP is doing is putting a false load on that evap during low ambient for constant volume. Those compressors have to fully unloaded before that HGBP starts pulling that false load.

[cool-hvacer]

Is it just me or is this system "whacked" :rolleyes:

The unloading is controlled by a thermostat. 4 stages - 2.5 degree increments. The thermostat is actually hooked up to the glycol "return"?!?!

As I mentioned earlier, the current sequence is:
1st stage - HGBP solenoid valve
2nd stage - compressor to 67%
3rd stage - compressor to 33%
4th stage - L.L. solenoid closes-pumpdown/shutdown.

I will change it to the following sequence:
1st stage - compressor to 67%
2nd stage - compressor to 33%
3rd stage - HGBP solenoid valve
4th stage - L.L solenoid closes-pumpdown/shutdown.

Now the HGBP valve itself needs to be adjusted manually for what pressure it will begin to actuate. So it seems a little crazy that someone could go and turn the thermostat up and your pressures for HGBP would not work anymore?!?!

OK - Rant is over :confused:

So the questions I have are?

1) Is having the thermostat on the glycol return the "reasonable " thing to do?!?

2) I suspect the design is such that if all the cooling loads are removed, the system will move down through its 4 stages, pumpdown, and go into low pressure cutout. Do I need to have less than 1/3 of the total load capacity for the HGBP for it to ever reach this state?

3) Back to the HGBP valve question :confused::confused::confused: - will having a 9 ton valve that actuates over 6psi give me 3 tons of bypass in 2 psi pressure drop?

4) If I do rip out the current 9 ton valve and replace it with a 3 ton (or less) are there other things I need to investigate first?!?!

Sorry if the questions seem too basic!

[Grizzly]

cool-havacer
I have a basic question for you.
Do the Solenoids "energise" to load the compressor.
Or Is it as I suspect they energise to unload the compressor.
Now lets get really basic.
The purpose of a hot gas bypass is to supply sufficient "load" to enable the compressor to run at lower load conditions than it would normally do!
So it comes into effect at the minimum load stage.
This allows the compressor to cover a larger range of operation.



now the HGBP valve itself needs to be adjusted manually for what pressure it will begin to actuate. So it seems a little crazy that someone could go and turn the thermostat up and your pressures for HGBP would not work anymore?!?!
Now the HGBP valve itself needs to be adjusted manually for what pressure it will begin to actuate. So it seems a little crazy that someone could go and turn the thermostat up and your pressures for HGBP would not work anymore?!?!


Now it seems crazy to you. But when you first posted you also did not understand!
So why is it wrong to assume someone else "got it wrong"!
We all do from time to time!
Grizzly

[US Iceman]

I suspect the design is such that if all the cooling loads are removed, the system will move down through its 4 stages, pumpdown, and go into low pressure cutout. Do I need to have less than 1/3 of the total load capacity for the HGBP for it to ever reach this state?


If the cooling loads are removed and the HGBV is the last step of capacity control, the system will never pump down and shut off. Why? Because the hot gas valve will continue to stay on until the load returns and increase, which will force the compressors to load back up.

If the load disappears while the HGBV is initialized then you might allow a 30-60 minute run time on a timer. If the load does not return in that time frame, then shut off the last compressor. Why? There is no sense in wasting energy waiting for the loads to return.

[cool-hvacer]

For Grizzly:

1st: Solenoids energize to unload compressor.

2nd: Assume someone sets the thermostat and the HGBP valve to actuate at a specific low temp (lets say 10F for fun). Unloading happens as expected
@19F first stage unloads compressor to 67%
@16F second stage unloads compressor to 33%
@13F third stage open HGBP solenoid. Now the actual HGBP valve will not open until pressure drops to whatever you have set it for. (In this case probably 7.5 to 13.5psig corresponding to 2 to 12.5 F temperatures).

So yes - it does seem crazy that someone could walk up to the front of the unit, and turn the thermostat dial temperature up 30 deg, and the system would now behave like:
@39F first stage unloads
@36F second stage unloads
@33F HGBP solenoid opens
BUT.... glycol temp continues to drop all the way down to 12.5F before HGBP begins to work.

If by saying this is incorrect, I have offended anyone, just "lightly;)" cuff me across the back of my head, but please explain to me why this OK.

For Iceman:

Ok - but seeing that the unit does not have a timer, and does have a 4th stage (L.L. shut off and pumpdown mode), do you think the original design intention is for it to actually shut down as opposed to run in HGBP mode for a period of time.

And this is where I am asking for you guys with the experience.... What do you normally see....? What do you suggest is the correct way to do it....? Should I modify our equipment to do it that way...? If I do (modify it), anything else to worry about?

Thanks for all the good replies!!

Oh ya - does the thermostat bulb on the glycol return (as opposed to supply) seem normal?

[US Iceman]

Ok - but seeing that the unit does not have a timer, and does have a 4th stage (L.L. shut off and pumpdown mode), do you think the original design intention is for it to actually shut down as opposed to run in HGBP mode for a period of time.

And this is where I am asking for you guys with the experience.... What do you normally see....? What do you suggest is the correct way to do it....? Should I modify our equipment to do it that way...? If I do (modify it), anything else to worry about?


Who is to say what's normal? I have never seen a chiller with hot gas as the first stage of capacity control unless the compressors have no capacity reduction capability themselves.

As the load decreases and the compressor capacity reduction lowers the total system capacity you will reach a point where the hot gas is called for (otherwise the suction pressure will drop as the load decreases). When the last compressor unloads (one is already off) and the suction pressure decreases because of low load you need hot gas to keep the suction pressure (and the evaporating pressure) to remain constant.

As the load continues to decrease the HGBV will open more to maintain the suction pressure. See what's happening? The suction pressure will never decrease, therefore the system will not pump down unless the liquid line solenoid is de-energized. If the return temperature hits the thermostat set point with no load, then the system may pumpdown and then shutoff.



Oh ya - does the thermostat bulb on the glycol return (as opposed to supply) seem normal?


I do not have a problem with the bulb being in the return line as long as you adjust the thermostat accordingly for the desired supply temperature.

Remember, the supply temperature desired is what the chiller will do at full load. For example, if you have 19°F return temperature at full load then the supply temperature will be something less, however, is it 10° less, 9° less, or what? At that point you need to make sure the glycol mixture has the proper concentration to ensure no freeze-ups!

Are you having fun yet?;)

PS: What is the refrigerant?

[cool-hvacer]

Refrigerant is r134a
Glycol is at 40% - freeze point -10F

The following is my "graphical" attempt to figure out what is going on as the load drops....

Note the system is at high altitude, so there are several "Pressure" columns. I just use the psia to avoid confusion.

I put in both 9 and 3 ton HGBP valves at started them opening at 27.2 psia (fully open at 21.2psia).

I guessed at what was going on in between 9 and 6 tons, as well as between 6 and 3 tons. I am really lost as to what happens below 3 tons.....

Hope the graphic shows up.....

And Oh ya - I have been having fun with this for quite a while now!!:D

[US Iceman]

You mentioned 2 HGBV's. Is there one for each compressor, or what's going on with that?

[cool-hvacer]

Nope - single compressor.... One HGBP valve....

The original valve is for 9 tons (total capacity of the system) but if I switch it to the final stage of unloading, I was considering one for only 3 tons.

I am still a little fuzzy as to which one to get.:confused: If I am not mistaken, the 9ton one will bypass faster for a small psi drop. I.E. If the 9 tons valve actuates over (say) 6psi, then won't it pass 3 tons over 2 psi?!?! My understanding is that a lower psi drop is neccessary to prevent evaporator freezing.

But the other question (that I have alluded to a few times) is do I really want to bypass the full capacity, or should I just let the temperature drop till the system hits the forth stage, and pumps down/shuts down? Maybe by selecting a 2 ton HGBP so there is always still a ton of cooling to keep the temperature coming down...

:D

"Iceman" - I do appreciate you staying with me on this one!!

[US Iceman]

This is like boxing in the dark because I do not know exactly what you have.

If you have two compressors; each with one step of capacity control; are the compressor circuits independent or are the two compressors on a common circuit?

The problem is oil return and equalization between the two compressors.

Can you describe this more please?

[cool-hvacer]

System has one compressor, and one HGBP valve.

I put two HGBP valves on the chart because I am trying to pick which one to use, 3ton or 9ton.

[US Iceman]

Ok then, let's see if I have this right. One compressor with two steps of capacity control and one HGBV. Sorry for being a little slow I'm trying to work on other stuff also.

At full load we have; the compressor fully loaded and no hot gas bypass. As the load decreases, the first step f compressor capacity is reduced (hot has still off). As the load continues to decrease the last step of compressor capacity is reduced (hot has still off). Now, as the load continues to decrease the suciton pressure will start to drop a bit and you want to energize the HGBV. The HGBV no allows gas to flow into what I am assuming is the evaporator. This false loads the evaporator and provides sufficient pressure to maintain the suction pressure.

Now, at this point the load is zero, but the compressor capacity is 33%. This means the HGBV should have a capacity to provide about 3 tons (or 33% of 9 tons, which is what I think I saw in your posts).

The HGBV must provide an equivalent amount of gas equal to 3 Tons at the normal operating conditions which for the HGBV are: dicharge pressure minus the normal evaporating pressure. So what you have now is: valve flow capacity ~3 Tons with a specific delta-P.

The HGBV has a sensing line that is connected to the suction line, which makes this valve a downstream pressure regulator. As the downstream pressure (which is the suction pressure) decreases, the valve strokes open to maintain this pressure. At 33% load the HGBV may not pass any vapor. At 0% load, the HGBV is fiully open delivering the full 3 Ton capacity of the valve.

If the load stays at zero percent, the system will continue to run at 33% of it's full load capacity but no actual cooling is taking place. this is where the energy waste comes into the picture.

Let's see where this goes. It's getting late here.

[Gary]

Maybe I am reading this incorrectly, but it seems to me that the HGBV has nothing to do with capacity control and everything to do with freeze protection. It maintains a minimum evaporating pressure. It is energized whenever the liquid line solenoid is energized and de-energized for pumpdown.

[Gary]

Why haven't we seen the control schematics?

[Grizzly]

Sorry for being thick Guys but I do tend to agree with the statement that a hot gas bypass is not a form of capacity control.
The ones I have come across activate on low load conditions and are only fitted to recips?
My assumption is they then allow sufficient gas flow through the compressor .
Allowing it to run smoothly at low load but more importantly.
The higher gas flow maintains movement of oil through the system ensuring sufficient oil return back to the Compressor.
Just a thought but surely the valve needs to be capable of handling total system capacity.
If any of the cap loading coils fail then there cylinders are going to run loaded!
Coolhavacer.
Would it not be possible for us to know what make Compressor we are talking about?
We must surely have someone on the forum who can then advise of the designed control strategy.
Grizzly

[US Iceman]

This is going to get interesting I see.

Hot gas bypass is a form of load control, if you want to put a face on it. As the load decreases, and with the compressor at minimum capacity, the suction pressure would decrease with the load. Under extreme conditions of long periods of operation at this condition the chiller could conceivably freeze up. Also, during this time the outlet glycol temperature would continue to decrease and the low gas velocity (low mass flow) could cause oil return issues and motor cooling problems on semi-hermetic or hermetic compressors.

Therefore, in an effort to reduce these issues the use of a HGBV provides the opposite of the above:


Stable evaporating pressures (and hence temperatures)
relatively constant glycol supply temperature during low load operation. This helps to stabilize the controls on the loop side.
stable gas flow for oil return and motor cooling
and, it eliminates or reduces short cycling during low loads.

[Grizzly]

This is going to get interesting I see.

Hot gas bypass is a form of load control, if you want to put a face on it. As the load decreases, and with the compressor at minimum capacity, the suction pressure would decrease with the load. Under extreme conditions of long periods of operation at this condition the chiller could conceivably freeze up. Also, during this time the outlet glycol temperature would continue to decrease and the low gas velocity (low mass flow) could cause oil return issues and motor cooling problems on semi-hermetic or hermetic compressors.

Therefore, in an effort to reduce these issues the use of a HGBV provides the opposite of the above:


Stable evaporating pressures (and hence temperatures)
relatively constant glycol supply temperature during low load operation. This helps to stabilize the controls on the loop side.
stable gas flow for oil return and motor cooling
and, it eliminates or reduces short cycling during low loads.


Thanks Iceman.
Grizzly

[US Iceman]

I should also submit my assumptions for what I have been saying.

I am assuming the hot gas is injected downstream of the TXV (but before the evaporator inlet). This allows the TXV to control the gas superheat and return gas temperature constant for motor cooling. The evaporator does not notice the difference.

If the hot gas is injected into the suction line (between the evaporator suction connection and the compressor suction valve, you could use some liquid injection for desuperheating. Otherwise, the compressor motors could still overheat because of high return gas temperatures.

[Gary]

At full load we have; the compressor fully loaded and no hot gas bypass. As the load decreases, the first step f compressor capacity is reduced (hot has still off). As the load continues to decrease the last step of compressor capacity is reduced (hot has still off). Now, as the load continues to decrease the suciton pressure will start to drop a bit and you want to energize the HGBV. The HGBV no allows gas to flow into what I am assuming is the evaporator. This false loads the evaporator and provides sufficient pressure to maintain the suction pressure.


Let's imagine a low water flow condition (plugged strainer, 3 phase pump running in reverse, etc.). The return water is warm, but the chiller barrel will freeze because the HGBV is not activated because the return water is warm.

[Gary]

As I see it, the hot gas solenoid should be energized when the liquid line solenoid is energized.

The HGBV should be adjusted low enough that it does not bypass at any time during normal operation, but high enough that it keeps the barrel from freezing. IOW, the hot gas is strictly for freeze protection.

The two solenoids are wired in parallel to allow pumpdown.

[US Iceman]

IOW, the hot gas is strictly for freeze protection.


That could be, although I have never seen this done except on Red Water chillers where the hot gas bypass starts to open if the water temperature starts to dip below 32°F. And...when this usually happens it is during low load conditions.

[cool-hvacer]

Ok - maybe these will help.... A couple of documents. I will put a link on, so as not to fill up RE will big files.

irtfweb.ifa.hawaii.edu/Facility/mechanical/

There are schematics for the control, original specs, compressor specs, and thermostat details.

The compressor is a Carlyle O6EM475 (semi-hermetic reciprocating)

The thermostat (Johnson Controls A36) unloads in 4 stages at roughly 2.5 degree increments. It has is sensing bulb on the glycol return line.

The thermostat is labelled on the control schematics as 1TC1, 1TC2, 1TC3, and 1TC4.

1TC4 actualtes first (on drop in glycol temp) and opens up the HGBP solenoid valve.

1TC3 actuates next to energize compressor unloading stage 2

1TC2 actuates next to energize compressor unloading stage 1

1TC1 actuates last (in schematic it should be shown as normally closed). It closes liquid line solenoid, HGBP solenoid, and pumpsdown until low pressure cutout.

Currently the HGBP Valve is a Sporlan ADRHE-6 with a capacity of approx 9 tons.

I guess the questions we are trying to tackle are:

1) Do I change the controls to compressor unloading before HGBP or leave them as is?

2) Do I keep the same HGBP valve (at 9 tons) or switch it to one for 3 tons.

3) At what Temp do I set the HGBP to start actuating

I mentioned it before but again, system is
R134
Design suction temp 2 deg F
Design cond temp 74 deg F
Design capacity of 9 tons at:
entering glycol temp 10 deg F
leaving glycol temp 15 deg F

FYI I have the glycol proportioned for -10 deg F freeze

Hope this helps:eek:

[cool-hvacer]

I guess the HGBP vs compressor unloading is kind of a "belt and suspenders" type question. If you have one do you need the other?

Or how do you get them to work nicely with each other?

:D

[Gary]

I guess the HGBP vs compressor unloading is kind of a "belt and suspenders" type question. If you have one do you need the other?

Or how do you get them to work nicely with each other?

:D

By adding an operating relay.

In schematic1, 1TC1 is shown as closing on drop in temp. This is a mistake in the drawing. It should be shown as closing on rise in temp. (like FZT is shown)

[Gary]

1TC2, 1TC3 & 1TC4 close on drop in temp, so let's start with warm water.

We start the chiller. 1TC1 closes on rise in temp, energizing the operating relay, which enables all of the controls, and the warm chiller starts out fully loaded.

As the water temp drops 1TC4 closes, energizing the hot gas solenoid. If the HGBV were actually bypassing at this point, the temp would not drop any further. The current water temp would be maintained, the chiller would run 24/7... and the unloaders would serve no purpose.

So... we can assume that the pressure setting on the HGBV must be low enough to allow the unloaders to come into play. IOW, the system is still fully loaded.

The temp continues to drop, closing 1TC3 to energize an unloader.

The temp continues to drop, closing 1TC2 to energize the second unloader.

At this point, although the hot gas solenoid is energized, there is still no hot gas bypass.

[Gary]

At this point we need to make a decision. Do we want to cycle the system?... or hold a steady minimum temp and run 24/7?

If the HGBV is set for a temp above 1TC1, the system will run 24/7.

If the HGBV is adjusted for a temp lower than 1TC1, the system will cycle and the HGBV is strictly for freeze prevention.

And either way, why not place a jumper across 1TC4 and let the operating relay energize the hot gas solenoid? It seems that the designer of the system does not want to have the hot gas solenoid energized above the 1TC4 setting. Why?

[Gary]

What is the chilled glycol being used for?

[Gary]

I guess the questions we are trying to tackle are:

1) Do I change the controls to compressor unloading before HGBP or leave them as is?

Leave them as is. If anything, we would want the hot gas solenoid to be activated at an even higher temperature... or be activated with the liquid line solenoid. Keep in mind that activating the hot gas solenoid does not activate the HGBV. The HGBV is activated by its pressure setting.


2) Do I keep the same HGBP valve (at 9 tons) or switch it to one for 3 tons.

If all goes wrong, we want the HGBV to be able to overcome the total compressor capacity and keep the barrel from freezing.


3) At what Temp do I set the HGBP to start actuating

This is the key question... and the answer is: It depends upon what you want to accomplish.

If you want the system to run 24/7 at a minmum temperature then you will need to precisely adjust the valve to maintain a temperature between 1TC1 and 1TC2... and if you change the thermostat setting you need to change the HGBV setting.

If you want the system to cycle off and on, then you are using the hot gas for freeze protection and you want to keep the HGBV setting just above the freezestat setting.

[Gary]

1TC1 actuates last (in schematic it should be shown as normally closed).

As an interesting aside, we might consider what we mean by "normally open" and "normally closed". By "normally" we mean non-energized (no electricity, no flow, no pressure, no heat, etc.).

As a rule, wiring diagrams should be drawn with all controls in the normal position.

In schematic1, we see that the freezestat (FZT) is drawn correctly. It opens on a drop in temperature (crossbar under the contacts) and its contacts are open (crossbar not touching the line at the right end), indicating the normal (no heat) position.

The flow switch (FS) is also drawn correctly, opening on drop in flow (crossbar under the contacts) and shown in the normal (no flow) position (open contacts).

In schematic2, although the symbols are drawn correctly HDT, HPS, 1TC2, 1TC3 & 1TC4 should all be showing closed contacts (crossbar touching the line), indicating their "normal" non-energized position.

[cool-hvacer]

Hey Thanks guys!!!

Hmmm...... (smell the smoke?? I am thinking hard now:)) Not ignoring - just digesting!

Like Gary says, I guess it comes down to one thing - do I want this thing to cycle or not when unloaded?

The glycol is used in two AHU's to cool a large volume of space to approx 25 deg F. We only need to cool thevolume for half the day, so there could be a big energy savings if we allowed it to shut down.

Any gut feelings as to whether this amount of startup/shutdown would be hard on the equipment vs the energy costs?

And again guys!! Thanks for all the help!

[Gary]

My gut says shut it down.

[rbsguzman]

Hi all...

New to the site, but think it is great!! I am a Mech Eng with limited HVAC/R but have a glycol chiller system that I have been asked to help with. So any comments would be mucho appreciated!

I have a few fundamental questions with respect to hot gas bypass / compressor unloading... Our system has both (2 stages of unloading on the compressor and a HGBP system). All of them are hooked up to a thermostat that activates them sequentially, in 3 deg increments.

I have gone through the control schematics, and it appears that the HGBP gets activated first, then the two stages of compressor unloading, and then the Liquid Line solenoid (pumpdown to low pressure cut out). Does this order sound right?!?!

Also, the HGBP valve that is currently on the system is rated for the full capacity of the chiller. Since the capacity reduction for the chiller is suppost to be 100-67-33-0, should the HGBP valve be specified for only 1/3 of the total capacity?!?!

Thanks in advance!!
The hot gas by pass is always used for the last step of unloading when compressors totally unloaded cannot match the smallest load required by any cooling apparatus.Special care should be taken when employing this type of method due to it can overheat compressors winding if there is no a good de-superheating valve injecting liquid, and that is when is used by-passing directly into the compressor. I guess the safest way to by-pass hot gas would be into the evaporator, 2 inches downstream the TXV and therefore there is no for a de-superheating valves