Refrigerant R134a. Normal condensing pressures

[mrr]

Hi,

I'd like to ask, what are the normal condensing pressures of R134a? I am working with an experimental plant and I observe (I think) condensing pressures high.

The arriving temperature (ambient temperature) to the condenser is 30 ºC, and now I can see 13.5 barg of discharge pressure and 1.6 barg of aspiration pressure.

In addition, in all the experiments I have done so far the subcooling is about 20K. Isnt it too big? By the way, I have a liquid tank at the outlet of the condenser, so I would expect it can avoid effects related with overcharging. I have heard the condenser was sized for R404A, so it may be oversized for R134A.

Where can be the problem? I can provide more data about the experimental plant.

Regards

[Glenn Moore]

Hi mrr
20k subcooling is very high suggesting overcharging. Most expansion valve manufacturers require only 4k subcooling for the expansion valve to achieve nominal design capacity. So would expect the system is overcharged causing the liquid to back up in the condenser causing high discharge pressures .Remove some of the charge to reduce the degree of subcooling to somewhere in the 5-7K range is more than sufficient

[RANGER1]

Can only add that possibility of non condensables in condenser as well.
Can refrigerant flow down into liquid receiver with no traps, sometimes manufacturer puts trap so last few rows of condenser can hold liquid for subcooling before draining into receiver.

[Tol]

Hi mrr I use a condensing temperature of 45°C for all the refrigerants I work on. In your PT chart will give you a corresponding high side pressure.
I don't know if this will work for you but it should give an idea of your condensing pressure.

[mrr]

Hi again,
Thank you very much for all of your answers. What is happening in my experimental plant is very interesting and intriguing. I think it is worth to go into detail.

Initially, we had (and have yet) a Danfoss AKV-10-6 expansion valve (electronic pulse expansion valve). This EEV was oversized since its controller is always working with a pulse witdh of no more than 30 % (with a cycle of 1 second, the EEV is 300 ms opened and 700 ms closed). In addition, we have refrigerant flowmeter in the liquid line. With the aforementioned EEV the refrigerant measure is not reliable, since, as you can guess, the refrigerant flowrate is pulsant. Thus, we decided to install a "step by step" electronic valve to have a reliable measurement of the refrigerant flowrate. Now, both EEV are installed in parallel, when one is working, the other is closed. Then we can make comparisons.

So here is the thing:

The main componentes of the plant are a bitzer compressor with an inverter. The evaporator is a B15THx20 SWEP heat exchanger, and the condenser is an air battery were a fan with a inverter is installed. The fluid to cool is a glycol solution in a small tank where we have a variable resistance up to 5.4 kW. This system was prepared for R404A and R134a. Finally we put R134a, which means that the components are oversized with respect to R404A.

- The bitzer compressor gives as much as 3.5 kW with R134A (10 ºC of evaporation). Our nominal point was about 0 ºC in the glycol solution, this means an evaporation of about -5 ºC of the refrigerant. The compressor gives about 1-2 kW of power at that point with R134a, depending on the condensing temperature (40-60 ºC).
- With the AKV EEV I use to get between 9 and 12 barg of condensing pressure. This happen when the ambient temperature is between 18-25 ºC. If I increase the ambient temperature (we can control it) to 35 ºC the condensing pressure rises at 16 barg. In all of this situations, the subcooling is always about 20 K.
- As I said, we have a B15THx20 heat exchanger. I can control the flowrate of the glycol solution. I chose about 13 l/min, it seems at that point the heat transfer is optimal.
- The superheat can be controlled. And the AKV EEV is between 20-30 % of pulse width.

- The EEV (step by step) valve was chosen for the aforementioned nominal point (-5 ºC of evaporating temperature, and between 1-2 kW). Software results are about a 30% of aperture. However, when I put in the system the aforementioned conditions, this valve is 100% opened and the superheat cannot be controlled. However, if I increase the ambient temperature about 30 ºC then the condensing pressure also increases and the EEV seems to control the superheat.

In both cases (with the AKV (pulsed) and the "step by step" EEV) the subcooling is high, about 20 K. I have a receiver after the condenser, my understanding is that a tank receiver guarantees liquid feeding of refrigerant to the EEV and compensates for possible excess of refrigerant (correct me if I am wrong). In addition, the higher subcooling, the greater capacity of the EEV. What can be the reasons the "step by step" EEV cannot control the superheat unless I rise the condensing pressure? (I selected this EEV taking into consideration a small delta P between aspiration and discharging pressures). Should I have selected the EEV for the evaporator capacity instead of the compressor capacity?

We are going to change the step by step EEV for a bigger one. However, I still do not know the exact cause of the problem and I'd like to know it. If you have any question or need any clarification please let me know and I will edit the post. Waiting for hints!

Regards,

[Glenn Moore]

Hi met
What Bitzer compressor model are you using?
Which controller are you using for the AKV valve ?
What stepper model valve and controller are you using?
Please send this info then I can check valve selections and controller parameters that need setting

[mrr]

Hi Glenn,

Thanks. Here is the info:

- Compressor: Bitzer 2JES-07(Y)

- AKV 10-6: http://www.ra.danfoss.com/TechnicalI....02_AKV_VS.pdf
The driver is an OSAKA controller which works fine.

- The stepper model valve is a E2V11: http://www.carel.com/documents/10191...c4?version=1.0. The controller is a EVD Evolution http://www.carel.com/product/evd-evolution. I am sure the controller is not the problem. I have set manual operation and I set 100 % aperture and the stepper EEV does not control the superheat.

I am 100% sure that the controllers are not the problem. There must be another cause.

I think I should put this post also in "troubleshooting".

Best,

[PaulZ]

Hi mrr
If you are always getting 20k sub-cooling regardless of the ambient I would have to agree with Ranger's statement about non-condensables and Glenn's about overcharge as possible causes, you would need to check both and eliminate them as causes.
Paul

[mrr]

Hi PaulZ,

Can it be that the condenser is oversized? If not, can you explain why? How could I eliminate non-condensables?

[PaulZ]

Hi mrr
If the condenser was oversized the discharge pressure would be lower. You have said the ambient air is 30 oC and you have 13.5 barg discharge. You have said this is an air cooled condenser so I would expect the discharge to be about 10 to 11 barg at that ambient.
Then you raised the ambient to 35 oC and ended up with 16 barg discharge, should be about 12 to 13 barg. This does not indicate over condensing but does indicate non-condensables.
Regards
Paul
I suggest you reclaim all the refrigerant evacuate the system to 500 microns and the recharge with new refrigerant.

[Glenn Moore]

Hi
The AKV valve has a capacity of approx 8kw, the E2V11 has a capacity of approx 3 kW , and the compressor has a capacity of approx 2 kW . So the AKV valve needs a no 4 orifice fitted to match the system capacity and the E2V11 needs replacing with a bigger valve. I suspect you have a high pressure drop through the plate HX which will deranged the EEVs capacity but by increasing the condensing pressure you increase the delta p across the valve which increases its capacity. This scenario often happens with these EEVs as they react differently to conventional tevs. Change the valve and the AKV orifice and re evacuate and charge the system by measuring the subcooling and stop charging at about 8k sub cooling that is sufficient then the condensing pressures should be as Paul suggests. When selecting EEVs always allow for some pull down load as all to often the selection of the valve is to small and the valve will go 100% and only starts to control once the pull down load has gone and the valve and compressor duties match at the design condition but for long periods the system runs totally inefficiently due to the componentry mismatch at the high load condition

[mrr]

Hi Glenn and PaulZ,

I think you both were right, thank you for your suggestions!. New refrigerant was charged and vacuum was done. I dont really like how the refrigerant was charged since there was not an accurate measurement of the mass refrigerant charged. I would have liked to do the according to Glenn suggestion (8 k of subcooling).

Anyway, now I get coherent discharge pressures and the new stepper EEV seems to control the superheat. However, I'd like to find why is happening the following:

- When working with the stepper EEV, If I have a low ambient temperature (let's say 9 ºC) and I put the condenser turbine to work 100 % I can decrease the condensing pressure to 5 barg. In this cases the superheat increases, the valve go to 100% and I observe the refrigerant flowrate oscillating in the liquid line (I have a flowmeter which measures the volumetric flow of the refrigerant, it is placed before the expansion valve, in the liquid line). It is like the refrigerant flow is not constant. Is this normal? What can be the cause of this? I expect the flowrate to be more or less constant for a constant aperture of the EEV.

- Glenn, can I ask why do you say I should change the E2V11? As you said, it has a capacity of about 3 kW, and the compressor a capacity of 1-2 kW for 0,-5 ºC of evaporating temperature. So, why to change?

- I am trying to find the problem here, and I do not know if it is related mainly with the evaporator or the condenser. If the evaporator is bigger (in capacity) than the compressor, can be this the problem?

Regards,

[RANGER1]

For best expansion valve control condensing pressure should be controlled very accurately, say with variable speed fan on condenser.
When selecting expansion valve capacity criteria is shown with different pressure drops across valve.
The lower pressure drop , equals lower capacity of valve & higher bigger capacity.
When you select a valve you should have a design condensing pressure in mind to suit all conditions.
In your case select example 45
deg C condensing & control that at all times.
This can be adjusted as long as expansion valve is sized accordingly.
Mic compressor has varying condensing pressures, it will have varying capacities as well due to efficiencies, so you might have to think about that as well if you can allow that in testing.
Example lower condensing pressure, better compressor performance, lower suction pressures.
If you want constant results, select design condensing pressure & control that with condenser fan etc
Of course lower condensing pressure more efficiency & lower power consumption, but depends on what you are doing & trying to achieve.

[mrr]

Hi Ranger,

Thank you for your answer. The thing is that I considered what you are saying when selecting the stepper EEV. My EEV provides about 2.9 kW for the lowest pressure drop, which should be enough since my thermal load (a controlled resistance) is within the range of [1,2] kW. Still, I am trying to understand why the refrigerant flowrate fluctuates when the pressure drop is very low and then the superheat is no longer under control, it is like the EEV is not 'properly feeded'.
I am wondering if an oversized evaporator of an oversized condenser has something to do with this, i.e, my evaporator has a capacity of 8 kW and my compressor has a max capacity of 3.5 kW. I am sure people with experience know why this happens.

Regards,

[Glenn Moore]

Hi mmr
I understand you are using a plate HX which will have some kind of distributor nozzle inside the plate . this distributor will cause a pressure drop between the EEV and the plate . This P/D can loss you capacity and I have seen this many time with AKV and Stepper valves. Both AKV and Stepper valves capacity charts give fully open duties only , ie there is no more capacity left in the valve, apart from increasing the PD across the valve by raising the condensing pressure. (Not an ideal thing to do as energy consumption increases). TEV's capacity charts change with the evaporator and condensing condition.
So to fully determine what is going on we need to know the pressure drop from the expansion line after the EEV and the suction line. So ideally we need a pressure tapping just before the EEV , one just after the EEV and one on the suction line to see what the evaporator is actually doing. With Ternary blended refrigerants if you get to high a PD through the evaporator it can cause a major loss in duty.
On many distributors on coolers I have had to drill out the orifice or venturi to reduce its PD to allow correct operation of the expansion device. I have experienced PD's of more than 6.5Bar through a distributor which had a much to small orifice fitted .
AKV valves should never be used with Plate HX as during the off cycle you can fully evacuate the plate during the OFF portion of the PWM cycle, causing poor duties and often short cycling on the LP control. You can use 2 AKV's in parallel using Danfoss controllers in a Master/ slave set up where the 2 AKV's fire in a halfway rectified mode , this works very well although expensive.
If you can get these Pressure readings, as above, it may be helpful to understand whats going on.
I have been troubleshooting expansion valves and evaporators over 30 years on many sites and there is always an answer somewhere, its a very interesting subject , which with these new refrigerants you can get an expansion and a re expansion effect.

[RANGER1]

Glenn,
I would have thought 5 bar discharge would be to low for any system design with any type of Tx
valve & in the situation where Mrr has variable speed condenser fan.
From my experience very constant discharge pressure gives best superheat control, so you would have to control condenser fan on some sort of optimal pressure to suit Tx valves, power consumption etc c
Very unusual in the real world to have condenser & evaporator way oversize like this one, so compressor suction could be raised as well by slowing compressor if possible to control plant with some balance.

[mrr]

Hi again,

Thank you. Glenn, the plate exchanger is a SWEP B15Tx20 http://www.swep.net/products/b15t/. The fluid to cool is glycol in a small tank, no more than 20 l, with a resistance heater.

I think there is no big pressure drop between the EEV and the heat exchanger, the EEVs are in parallel and placed no more than 0.5 m from the heat exchanger.

The AKV works well, although it is oversized as I said, since its controller gives no more than 30% of duty cycle, its behaviour is the same independently of the condensing pressure, i.e., it can easily "flood" the evaporator and the superheat decreases. That does not happen with the stepper EEV in some conditions. When this happens, the refrigerant flowrate fluctuates and the condensing pressure is relatively small. I am still wondering: 1- why this effect of fluctuating flowrate and 2- is it caused because the stepper EEV is small due to an oversized condenser or evaporator?. As I said, when the pressure drop is enough the refrigerant flowrate seems to be constant, and the EEV was chosen knowing the minimum possible pressure drop. So, there must be something else.

Do you advice any software for calculating the evaporator and condensder capacities? I am wondering if the problem is the evaporator capacity in comparison with the EEV capacity. I do not have data of the condenser since it is was built specifically for the experimental plant.

I can control the condensing pressure well, but what I wanted is that my system could "flood" the evaporator as the AKV can do.

Regards,

[RANGER1]

http://energyrent.no/ViewFile.aspx?ItemID=576

http://www.swep.net/refrigerant-hand...rant-handbook/

If you were to select a plate heat exchanger, you would first come up with a design relating to heat load,flow, type of refrigerant, temperatures of product etc then send it off to in your case SWEP.
They would then select a PHE to suit your needs & give a few options with data sheet for design conditions.
It may also give pressure drops across orifices in each circuit, if not ask for it.
Why not make an inquiry to SWEP with your heat exchanger or a new one, hopefully they will help.
Then you can size components to suit

Once you get away from design it may then becomes more difficult to understand if & why problems occur.

Why is 5bar sct to 1.6bar sst is needed, it's not real world to me if you want guaranteed performance, against R&D, which SWEP would have already done.
SWEP indicate in link 0.5 to 2 bar pressure drop in distributor on PHE

[Glenn Moore]

Hi mrr
The AKV can flood under all conditions as its approx. 3x the compressor duty, whereas the Stepper valve is almost the same capacity as the compressor. As Ranger has said condensing pressures like 5 Bar are not realistic and for R134A its typically in the range of 8-11Bar . When you lower the condensing pressure the AKV still has sufficient capacity to do the job whereas the Stepper valve even though fully open cannot fill the evaporator.
Some time ago a Professor working for Danfoss carried out many experiments with TEVs and evaporators under Lab conditions and found what we call the MSS of the evaporator. The MSS can be 1k superheat or 15K superheat or anywhere in between .The MSS or Minimum Stable Superheat all evaporators have an MSS point somewhere, you have to find it by experimentation. In the Danfoss Superheat controller EKC 315A, EKC 316, EKD 316 etc all have 2 modes of superheat control. Factory setting (1)is normally MSS control, this setting can be changed to (2) UPLOAD. I always set to UPLOAD and then find the optimum MSS at minimum load and at MAXIMUM load this then controls the valve under all conditions and the valve controls the evaporator to the best effect across all duties.
During this R&D project it was found that even though the s/heat was stable the refrigerant in the evaporator had a flow and ebb movement similar to the sea washing up and down onto the beach. This effect could not fully be explained but the flow through the evaporator could be seen to be disturbed as it rose around a bend , some of the refrigerant carried on around the band and some of the refrigerant would fall back down the bend giving unstable flow. Gravity could also have an effect as also how good the evaporator plates are fitted together or how the return bends on an evaporator are brazed can all have an effect on the flow , this seems to be amplified with low P/ds. From what Ranger has said about the P/D through the Swep plate, if the P/D is 2 Bar then the Stepper would have nothing left to allow a decent flow to satisfy the evaporator.

If we look at the Condenser , if it is over sized ,by controlling the Fans with a speed controller then the oversize becomes less important as the condenser is there to simply condense the gas to a liquid under a stable fan speed control ie stable discharge pressure.
If we look at the evaporator + the EEV valves . The only thing the EEVs are interested in is to control the evaporators superheat at the set point from the controller. The controller for these valves work best with a pressure transmitter and a temperature sensor measuring the actual real evaporator superheat. So when selecting the correct EEV for the system select it to be between 50 to 70% open at design conditions . You could use the compressor capacity details to select the valve as the compressor has a fixed capacity unless it has an invertor fitted. The valve will simply fill the evaporator under the demand from its controller . If the valve is sized correctly it will fill the evaporator and control the s/heat as set point.

EEVs can and do work with very low P/Ds tests done have been as low as 2 Bar across the valve , although some R&D work I did with York on their ENERGY PAK Chiller we had the unit run throughout the Winter months and the P/D across the valve went as low as 0.8 Bar (The valve was a PHTQ85-4 ). The chiller ran without problems although the valve was fully open all this time as its capacity with this low P/D was not sufficient to control the superheat set point of 4K , it could only achieve about 10K but it still worked although the suction was on the low side.

But I agree with Ranger if you are doing some experimental work it would be best to have a fully matched system rather than the mis match hybrid, you are working with. We can only assume what is actually going on.

I have a 34 Channel chart recorder that I used on site for trouble shooting and R&D work. It taught me a lot about evaporators and their behavior under differing conditions . We found on multiple evaporator systems that some evaporators go to sleep when close to set point and one becomes the master, with TEVs you could never see this, but with electronics you can see so much provided you understand what and why your seeing it.

[mrr]

Dear Gleen,

Thank you for your detailed answer and for your time. Also thanks to Ranger and PaulZ I did not answer before because I have been doing some tests.

As I said, I selected the stepper valve to be less than 50 % at design conditions.Even considering very low pressures the selected stepper valve can manage the compressor capacity.

There is something else I realized, even with the AKV when I have relatively low pressures in the condenser I can "hear" a sound in the liquid line, near the flowmeter. This sounds like a pre-evaporation or a lack of feeding to the expansion valve. So there must be something else, the problem is not the EEV. I have manually set the EEV to 100% and I can see of the superheat decreases to 0 K.

To me, it seems the lowering the condensing pressure creates a sort of obstruction in the liquid line. I am wondering if it is due the actual refrigerant charge or something related with the MSS curve you mentioned, though superheat seems to be well controlled in 7 K.

In addition, I do not really know if having the heat plate exchanger in a co-current setup has something to do with this phenomenon.

Regards.

[Glenn Moore]

Hi mmr
As I mentioned before, using an AKV pulsed valve on a plate HX is not to be recommended due to the fact that you virtually empty the plate of refrigerant during the valves off cycle part of its PWM . But also with an AKV it is necessary to keep the speed of the liquid in the liquid line down to 1 m/sec to prevent noise in the liquid line during the opening and closing of the AKV. So the liquid line itself has to act like a small liquid receiver before the AKV. It seems that liquid line noise occurs as the AKV opens the liquid line is quickly drained by the AKV valve opening the line to the near empty plate hx, so the gas is sucked through the AKV orifice rather than forced through by the higher liquid pressure. This sudden pressure drop in the liquid line causes the liquid to evaporate and bubble, then as the AKV closes the liquid line refills with liquid and the bubbles burst causing the noise in the line. The solution is to increase the liquid line pipe size before the AKV valve to slow the liquid speed down to 1 m/sec or lower.
I have heard this noise a few times with AKV's and its normally just the liquid line pipe is simply to small not for the job simply for the correct operation of the AKV valve.

[mrr]

Dear Gleen,

Thank you. But bear in mind that the commented effect occurs for both the stepper valve and the AKV (pulsed) valve (remember I can choose which EEV to use).

This effect happens when I have low condensing pressures that can be reached if I have low ambient temperatures. It is interesting that even the sound changes. In addition, we have a liquid receiver installed before the EEV. Thus, I think there must be something else that causes that noise.

On the other hand, I am relieved to see that the commented effect happens regardless witch EEV I use, so the Carel EEV (stepper valve) seems to be properly chosen since it regulates the superheat.

Regards