Beer cooler : have I charged it correctly?

[acnerd]

Like a lot of other engineers, I was taught that a full site glass means the unit is charged correctly. I have learnt so much on this forum in the last two weeks about superheat and subcooling and doing it scientifically rather than by feel. So I would like to explain a job I did today on a callout and give you all the figures I measured, and then leave it to the experienced boys to comment on it. CONSTRUCTIVE CRITICISM ONLY PLEASE!!

An NRS Beer Master that runs on R134a had lost its charge due to a capillary wearing through against the discharge pipe. Cut a long story short, replaced the LP switch, put in a new drier, pressure tested to 200 PSI (@55'C SLT), evacuated to 4 Torr and started charging it. The evaporator uses an internally equalised TEV. Condenser fan runs at 1 speed. Cellar setpoint 13'C.

Initially I put in 2 KG and watched the site glass as it started to fill up. Before it was completely full, I stopped and allowed the system to run for 10 minutes to stabilize.
Readings: 20 degrees on condenser
31 degrees off condenser
23'C on evaporator
13'C off evaporator
36 PSI suction pressure
14'C SST
115 PSI liquid pressure
33.5'C SLT
So that gives me 9K superheat and 1.5K subcooling. My understanding of that is that it is still short of refrigerant. I also realise that we should only really be taking these measurements when room is close to design temperature. This will take hours to get there due to all the warm beer kegs though.

I continued to add small quantities of refrigerant in 50g intervals and monitored SH and SC values, which did not change much.

I then blocked up the condenser coil to simulate warmer days and pushed the values up as follows:

20 degrees on condenser
41 degrees off condenser
19'C on evaporator
11'C off evaporator
40 PSI suction pressure
16'C SST
160 PSI liquid pressure
45'C SLT
Obviously the unit has been running for awhile, so the room temperature is dropping slowly.
So now my SH is 9K and my SC is 1.5K. No change????

I have stopped charging at 2.5KG and the best SC I got was 2K and the SH stays at 9K. The cellar temp down to 18'C and off evaporator at 11'C.

So the big question, was I on the right track? My concern is that if I keep charging it while the room is still about 5 to 7'C from design temperature, that I am making a mistake. I was aiming for a SC value of between 4.5K and 8.5K. The SH would be 60%-70% of the evaporator TD.

Thanks in advance for your answers! :o

[Gary]

First let's start with abbreviations and definitions:

SST = Saturated Suction Temp
SLT = Suction line temp
SCT = Saturated Condensing Temp
LLT = Liquid Line Temp

TD is the difference between air on temp and saturation temp.
dT is the difference between air on temp and air off temp.

[Gary]

20 degrees on condenser
31 degrees off condenser

Cond dT = 31-20= 11K dT

23'C on evaporator
13'C off evaporator

Evap dT = 23-13 = 10K dT

36 PSI suction pressure = 5C SST

TD = 23-5 = 18K TD

14'C SLT

Expected SH = 18*.6 to 18*7 = 10.8K to 12.6K

SH = 14-5 = 9K SH

115 PSI liquid pressure = 35C SCT

33.5'C LLT

35-33.5 = 1.5 SC

High SH + Low SC = System undercharged.

[Gary]

20 degrees on condenser
41 degrees off condenser

Cond dT = 41-20= 21K dT

19'C on evaporator
11'C off evaporator

Evap dT = 19-11 = 8K dT

40 PSI suction pressure = 7C SST

TD = 19-7 = 12K TD

16'C SLT

Expected SH = 12*.6 to 12*.7 = 7.2K to 8.4K

SH = 16-7 = 9K SH

160 PSI liquid pressure = 46C SCT

45'C LLT

46-45 = 1K SC

21K cond dT = Insufficient cond airflow

High SH + Low SC = System undercharged.

[Gary]

Unblock the condenser (blocking it was not necessary) and add more refrigerant.

[Magoo]

The TEV could be an MOP , restricting load on compressor at high space temperature.

[Gary]

The TEV could be an MOP , restricting load on compressor at high space temperature.

I think you are probably right.

[Magoo]

Hi Gary.
I have been confused by similar problems in the past, system drops into line approaching design. The TEV MOP vavs are a good idea so long as you know they are installed.
magoo

[desA]

@Magoo:

How would you be able to check whether the TEV was MOP, in the field?

[Gary]

MOP or not, the TXV cannot control the flow of liquid if there is no liquid to control.

The SC is much too low for the TXV to function properly.

[Magoo]

Hi desA.
Normally indicated on power head with Danfoss, others are by model number. Very confusing I know.
From memory ,danfoss low temp are -10C medium temp + 10C. Travel with all supplier catalogues is best advise. Danfoss, sporlan, Flicker, alco. And all and sundry.
magoo

[Magoo]

Hi Gary.
Way back on original post, there was a clear sight glass, ie., full flow available to TEV, correct me if I am wrong.

[Gary]

If the sightglass is clear, then his subcooling measurements are wrong.

Hmmmm... in particular the liquid line temps are suspiciously high... way above cond air on temps.

[Magoo]

Hi Gary.
I would suggest that the system is undersized, and that the MOP vav is half protecting compressor under high duty loadings.
The up-side of MOP vavs is if undersized system at high duty then the heat of compression is taxing the condenser, hence minimal sub-cooling of liquid,but keeping compressor within limits.

[desA]

I wonder if the OP could explain a bit more about how the condenser exit temps were taken?

[Gary]

Let's take a wild guess and say he is measuring the condenser outlet instead of the receiver outlet.

And possibly this unit has the receiver under the condenser?

[Gary]

An undermount receiver would explain it. The condenser dumps straight down into the receiver with no liquid seal and no subcooling between the condenser and receiver. If he measured the temp at the condenser outlet, there would be little if any subcooling.

[nike123]

Expected SH = 18*.6 to 18*7 = 10.8K to 12.6K


Hi Gary! Could you explain this 0.6 and 0.7 factors in that expression?

[acnerd]

The TEV could be an MOP , restricting load on compressor at high space temperature.

Could you explain in more detail the function of a MOP TEV and what the abbreviation means? Thanks Magoo.

[acnerd]

I wonder if the OP could explain a bit more about how the condenser exit temps were taken?

The condensing unit is supplied with shut off service valves, identical to a split system. The LLT was taken at the point the gauges are attached on the service valve. The SH measurement was taken from the suction gauge on the return service valve and the SLT at the evaporator where the bulb is mounted. Thanks desA.

[acnerd]

Let's take a wild guess and say he is measuring the condenser outlet instead of the receiver outlet.

I'm not making it that easy for you guys. ;) I definitely measured the liquid temp. after the receiver.

[acnerd]

An undermount receiver would explain it. The condenser dumps straight down into the receiver with no liquid seal and no subcooling between the condenser and receiver. If he measured the temp at the condenser outlet, there would be little if any subcooling.

The receiver is at the same level as the condenser. It's a small upright receiver. Don't quote me on this, but now that I remember, the condenser puts liquid in at the bottom and the liquid is drawn out the top and then through a drier, site glass etc. Thanks for your comments Gary. I will get my abbreviations right in future! :D

[acnerd]

Hi Gary.
I have been confused by similar problems in the past, system drops into line approaching design. magoo

A key bit of knowledge I am missing here is whether I should be trying to adjust SH and SC values when the system is still not at design temperatures. I am worried whether it's a situation of the chicken or the egg?

[acnerd]

Hi Gary.
I would suggest that the system is undersized, and that the MOP vav is half protecting compressor under high duty loadings.
The up-side of MOP vavs is if undersized system at high duty then the heat of compression is taxing the condenser, hence minimal sub-cooling of liquid,but keeping compressor within limits.

It needs to be added here that the cellar is a normal room with a cement floor, not an insulated cold room. It had been high in temp for at least 24 hours. It contains approx. 20 beer kegs, all warm, so I am of the opinion it will take some time to get down to temp? As we all know, a fridge system will be capable of operating with a leak up to a point, where the indoor temps will slowly start to rise .... and then suddenly no cooling.

[acnerd]

Hmmmm... in particular the liquid line temps are suspiciously high... way above cond air on temps.

You've got me wondering Gary. They have never maintained this unit and it was in a right state. Condenser blocked, evaporator half blocked. I blew the coils out from the other side with nitrogen and then did a chemical coil clean on the condenser coils, and then rinsed it out. With a 10 to 20 degree delta-T, maybe it's not clean enough?

[acnerd]

The condenser dumps straight down into the receiver with no liquid seal and no subcooling between the condenser and receiver. If he measured the temp at the condenser outlet, there would be little if any subcooling.

I've been thinking about this, and may change my mind on what I said earlier. (Should have looked harder at the fridge circuit :() I think the liquid receiver has only one access point in the top, where the liquid line is tee'd into it. It's a pump down system, so maybe they only use it as a storage vessel on off-cycle. It would give symptoms then of no liquid seal?

[Gary]

Was the sightglass full?

Was the LLT measured near the sightglass?

[Gary]

Hi Gary! Could you explain this 0.6 and 0.7 factors in that expression?

In another thread, Magoo presented the theory that during the pulldown the SH should be 60-70% of the TD.

I haven't tested this, but it makes very good sense. The TD is an indicator of heat load and as the heat load decreases the TXV should close proportionately, resulting in proportionate SH.

For some reason the SH in this system is staying fixed at 9K as the TD drops, and we are wondering if this might be as a result of a MOP charge in the TXV bulb.

[Gary]

You've got me wondering Gary. They have never maintained this unit and it was in a right state. Condenser blocked, evaporator half blocked. I blew the coils out from the other side with nitrogen and then did a chemical coil clean on the condenser coils, and then rinsed it out. With a 10 to 20 degree delta-T, maybe it's not clean enough?

The condenser did not show indications of insufficient airflow until you "blocked up the condenser coil to simulate warmer days ". Blocking up the condenser is not necessary and throws off the measurements.

[nike123]

The condenser did not show indications of insufficient airflow until you "blocked up the condenser coil to simulate warmer days ". Blocking up the condenser is not necessary and throws off the measurements.

I would say that only when room temperature is at/near design temperature that condenser air flow should be partially blocked ( if outdoor temp is too low) to get condensation near design temperature and fine adjust subcooling by charge.
Do you agree with that?

[nike123]

In another thread, Magoo presented the theory that during the pulldown the SH should be 60-70% of the TD.

I haven't tested this, but it makes very good sense. The TD is an indicator of heat load and as the heat load decreases the TXV should close proportionately, resulting in proportionate SH.

For some reason the SH in this system is staying fixed at 9K as the TD drops, and we are wondering if this might be as a result of a MOP charge in the TXV bulb.

I remember that thread and I taught that that could be the case, but I wanted to be sure. Thanks!

[Gary]

I would say that only when room temperature is at/near design temperature that condenser air flow should be partially blocked ( if outdoor temp is too low) to get condensation near design temperature and fine adjust subcooling by charge.
Do you agree with that?

At very low ambient temps the airflow should be blocked off to simulate warmer ambient, but one would think that the system would already have a fan control for that purpose.

Subcooling is limited by ambient temp. For example, if the SCT is 60F/15.5C and the ambient is 50F/10C, then it is not possible to achieve 15F/8.5K SC, because the TD is only 10F/5.5K.

I'm thinking a proportionate rule could be devised for SC, similar to Magoo's SH rule, i.e. SC = 60-70% of cond TD. This should be tested to get the right proportions.

[Brian_UK]

Could you explain in more detail the function of a MOP TEV and what the abbreviation means? Thanks Magoo.Have a read of this Sporlan document about TEVs, an explanation of MOP starts on page 6.

http://sporlan.jandrewschoen.com/10-9.htm

Also click on the Literature section, there is a lot of useful information there.:)

[Magoo]

Acnerd
MOP stands for maximum operating pressure, TEV limits liquid feed rate. Check Brian_UK's connect to Sporlan

[acnerd]

Was the sightglass full?

Was the LLT measured near the sightglass?

Yes it was full. The sightglass is 200mm after the liquid receiver and I measured at the liquid line shut-off service valve which is a further 300mm away from the sightglass. It's then approx 3m run to the TEV.

[acnerd]

Have a read of this Sporlan document about TEVs, an explanation of MOP starts on page 6.

http://sporlan.jandrewschoen.com/10-9.htm

Also click on the Literature section, there is a lot of useful information there.:)

Thanks, interesting reading. They say MOP TEV's are generally used in a/c applications to prevent compressor loadings in high demand times where the evaporator is warm and the TEV would be open wide. Obviously I would have to check the beer cooler again, but in my understanding of what the MOP TEV does, it would be wasted on this system? It would have a 2 degree dead band, run between say 12 to 14'C and the difference in evaporator pressure between on and off cycle would only be a few PSI? At design temperature, the TEV wouldn't be opening to maximum, would it?

[Gary]

Yes it was full. The sightglass is 200mm after the liquid receiver and I measured at the liquid line shut-off service valve which is a further 300mm away from the sightglass. It's then approx 3m run to the TEV.

I don't see how it would be possible to have a clear sightglass with only 1-2K SC. And the liquid line temp measurement is way above ambient.

I'm thinking your temp measurement is way off. What type of instrument are you using to measure the temps? Has it been calibrated?

[acnerd]

I don't see how it would be possible to have a clear sightglass with only 1-2K SC. And the liquid line temp measurement is way above ambient.

I'm thinking your temp measurement is way off. What type of instrument are you using to measure the temps? Has it been calibrated?

Ok, went to the van and got it out. It's a digital thermometer using K-type thermocouples. I make use of a surface touch probe which is about 3 months old. The thermometer was calibrated in Jan 2009.

This is a link for it..... if it works.

http://uk.rs-online.com/web/search/searchBrowseAction.html?method=searchProducts&searchTerm=206-3738&x=0&y=0

I am going back tomorrow to carry out final adjustments on the LP switch now the room is down to setpoint temperature. I will take the measurements again and report back.

No-one has answered my question yet : can I get accurate SC and SH measurements when the room is not at design temperature?????? With a lot of heat to remove, would the SH not be higher than normal, and the SC lower than normal???

[Gary]

No-one has answered my question yet : can I get accurate SC and SH measurements when the room is not at design temperature?????? With a lot of heat to remove, would the SH not be higher than normal, and the SC lower than normal???

Yes... yes... and no. The accuracy of your measurements depends upon the accuracy of your instruments, the superheat should be higher than it would be at design temp and the subcooling should be normal.

[acnerd]

The accuracy of your measurements depends upon the accuracy of your instruments,

Peter_1 put the following statement to me another thread:

"Therefore I show students the measuring fault of their meters. They know the evaporating temperature of the system. They have to measure on several bends (the first ones) of an evaporator. They never measure evaporating temperature but some degrees higher (+/-3 to 5°C higher) This is partially due to the thermal resistance of the copper. This is then their measuring fault and this value can also be used at the outlet of the evaporator when measuring SH."

I never knew I had to make allowances for the material I was measuring? Is that where I am going wrong then?

[Gary]

Peter_1 put the following statement to me another thread:

"Therefore I show students the measuring fault of their meters. They know the evaporating temperature of the system. They have to measure on several bends (the first ones) of an evaporator. They never measure evaporating temperature but some degrees higher (+/-3 to 5°C higher) This is partially due to the thermal resistance of the copper. This is then their measuring fault and this value can also be used at the outlet of the evaporator when measuring SH."

I never knew I had to make allowances for the material I was measuring? Is that where I am going wrong then?

If you are determining evaporating temperature by pressure measurement rather than bend temperature, then this fault does not apply and no allowances need to be made.

[acnerd]

Thanks for your help Gary, I'll post the final measured values tomorrow night when I get home.

[acnerd]

Ok here are the final measurements:

Room setpoint: 13'C
Evaporator air on: 14'C
Evaporator air off: 9.5'C
Suction pressure: 30 PSI
SLT: 8'C to 12.5'C (TEV hunting?)
Condenser air on: 22.5'C
Condenser air off: 33.5'C
Liquid pressure: 110 PSI
LLT: 33'C to 35'C (why???)

So the SH varies between 4K and 8.5K and the SC between 1K and -1K.
Sight glass full during operation!

If you work out the evaporator TD based on the above figures, then the SH should be between 1K and 4K ..... a bit low?

So something is not right in the design of this system. True or False?

[acnerd]

If you work out the evaporator TD based on the above figures, then the SH should be between 1K and 4K ..... a bit low?

So something is not right in the design of this system. True or False?

Sorry worked the evaporator TD out incorrectly:
(I used the SLT instead of the SST value)
TD = air on evap - SST
= 14'C - 4'C
= 10'C

So SH is 0.6 x 10'C to 0.7 x 10'C
is 6K to 7K

So the SH is pretty close. It's just the low SC values!

[Gary]

Something is very wrong with the SC measurements. Negative SC is not possible. Either the LLT measurement is wrong or the high side gauge measurement is wrong.

[Gary]

Attach your high side gauge to a container of refrigerant and a temperature probe to the side of the container. The temperature should correspond to the pressure on a P/T chart. If they do not correspond then either the gauge is out of calibration or the temp probe is out of calibration.

[Gary]

At this point, I am thinking that the gauge reading must be off (although it could still be the temp probe) and the system overcharged. Overcharge can cause the SH to hunt.

[acnerd]

Attach your high side gauge to a container of refrigerant and a temperature probe to the side of the container. The temperature should correspond to the pressure on a P/T chart. If they do not correspond then either the gauge is out of calibration or the temp probe is out of calibration.

OK, here's what I found:

R134a: Hp and LP both measure 70 PSI, bottle is 22'C

P/T chart shows: approx 74 PSI at 22'C

R22: Hp and Lp both measure 120 PSI, bottle is 22'C

P/T chart shows: 124 PSI at 22'C

I read in a fridge magazine that most engineers don't have their gauges adjusted to suit atmospheric pressure and it can throw SH and SC values out by a few degrees Kelvin. I have no idea if my gauges are perfectly correct or what the allowable error is. But as it stands, the gauge manifold is reading 4 PSI low or the temp probe is reading nearly 1 degree high!

[Gary]

You might try putting your temp probe in icewater (stir the ice) to see if it shows 0'C.

[acnerd]

You might try putting your temp probe in icewater (stir the ice) to see if it shows 0'C.

Good idea! :)

[acnerd]

I put ice in water and let it stand for 10 minutes, then put more ice in and stirred. I used the liquid sensor, surface touch sensor and air sensor probes. All measured 2,5'C approx. If the ice water was 0'C, wouldn't it be frozen? I put a standard glass mercury thermometer in and it measured 1'C. I don't think any of the above is accurate enough for a definitive answer.........

[Gary]

Let's see how the new numbers work out:

old numbers:

Room setpoint: 13'C
Evaporator air on: 14'C
Evaporator air off: 9.5'C
Suction pressure: 30 PSI
SLT: 8'C to 12.5'C (TEV hunting?)
Condenser air on: 22.5'C
Condenser air off: 33.5'C
Liquid pressure: 110 PSI
LLT: 33'C to 35'C (why???)

new numbers (factoring in temp and pressure estimated errors):

Room setpoint: 13'C
Evaporator air on: 11.5'C
Evaporator air off: 7'C
Suction pressure: 23.2 PSI/-3C
SLT: 5.5'C to 10'C (TEV hunting?)
Condenser air on: 20'C
Condenser air off: 31'C
Liquid pressure: 103.2 PSI/32C
LLT: 30.5'C to 32.5'C (why???)

SC = 32-30.5 = 1.5K... to 32-32.5 = -0.5K
SH = 5.5--3 = 8.5K... to 10--3 = 13K

I would be tempted to say that the SC is low because the system is undercharged... but the sightglass is clear.

Seems we are still missing something.

I am running out of ideas... anyone have a fresh supply?

[Djorn]

As far as I know, any amount of SC, as long as you have some, will give solid liquid, but may still be undercharged for all “conditions”.
Is the probe use to measure LLT insulated? I have seen many instances where temp is measured in the airstream of the cond. Outlet, and the probe is picking up some of that heat. I usually put some “foam tape” around the probe to make sure what I’m measuring is just the LLT.
Does the unit have a CPR valve, a heat exchanger or any other device installed?

With calibration, as long as water and ice are in the glass, the temp will be approx. 00 C.

[acnerd]

It's a goof idea to shield the probe from the hot exhaust air from the condenser. I made sure of that.

The fridge unit has only the basic parts, no regulating valves whatsoever.

The fact that the system is running and cooling the product with a full sight-glass is enough for the customer.

Unfortunately for me I'm left with the niggling worry in the back of my head: is it me, the gauges or the sensors?? :mad:

Thanks for all the help guys. :cool:

[Gary]

Unfortunately for me I'm left with the niggling worry in the back of my head: is it me, the gauges or the sensors?? :mad:


I would calibrate the sensor to the ice water, then use the sensor on the refrigerant bottle to calibrate the gauges.

[acnerd]

Thanks Gary. I will look closely at the digital thermometer and see if I can do it myself.

[tbirdtbird]

The system does seem to work. The gauges and temp probe don't seem to be too far out. (water can be 0*C or ice can be 0*C...its all about latent heat. Ice water s/b 0*C as prev. stated). Gary has said in the past that the sight glass doesn't tell the whole story necessarily. I think Djorn has a point...wherever the temp probe is s/b very well insulated. Also liquid pressure seems low. I am not convinced there is a design issue.

[Gary]

Aside from the sightglass, this system says borderline charge. I would add a little and see if the LLT drops.

[RefrigNoob]

I would be tempted to say that the SC is low because the system is undercharged... but the sightglass is clear.

Seems we are still missing something.

I am running out of ideas... anyone have a fresh supply?

Could the orientation, flow direction and size of the sightglass have an influence? Similar to the filterdrier posts between yourself and desa?

Also, could the sightglass be clear because there is subcooling, but not enough to ensure a clear sightglass under various conditions. Isn't that why we charge to 5-10 deg SC? To ensure a clear sightglass under various conditions?

[Djorn]

Does the system have a receiver? If the unit has a receiver, than there should not be too much harm in adding some more gas, and after all, you can always take it out if you have too!
With Subcooling, I always visualize that if you back up a little liquid in the condenser, per se, you give more “time” to remove sensible heat from the subcooled liquid and thereby increase Subcooling.
The need for decent Subcooling helps with pressure drops in the piping. Pressure equals temperature and pipe risers and long runs and all kinds of things can rob you of sensible temperature and a solid column of liquid at the TEV.
If a device, like a sight glass perhaps, can change the pressure and you don’t have much Subcooling, that could cause bubbles in the sight glass I suppose.

[chillyblue]

Hi mate

Read this thread,
http://www.refrigeration-engineer.com/forums/showthread.php?t=11200
I had the same issues as you.
I'm still not convinced that having 7 deg C of subcooling does not cause excessive discharge pressures.
I guess i'm wrong but do cond unit maunfacturers actually design there systems big enough to cope with high amounts of subcooling.
I charged a system the other day with way more refrigerant than i would have normally charged, just to get subcooling to around 7 deg C.
The system was working well way before reaching 7 degC of subcooling.

Gary, why do you have to take the subcooling reading at the reciever outlet rather than condenser outlet.
I would guess the reicever this guy is using will be sitting in front of the condenser fan so air off the condenser will be heating the reciever surface raising it's temperature and lower the amount of subcooling.

[Gary]

Gary, why do you have to take the subcooling reading at the reciever outlet rather than condenser outlet.
I would guess the reicever this guy is using will be sitting in front of the condenser fan so air off the condenser will be heating the reciever surface raising it's temperature and lower the amount of subcooling.

This is a valid point... and something I had not considered. It is not possible for the liquid line to be cooler than its surroundings.

Therefore, the liquid line should not be measured in an area that is subjected to condenser leaving air, but should be measured further downstream where the line is subjected to cooler temperatures.

For the same reason, the sightglass should also be further downstream.

Good observation, Chillyblue :cool:

[tonyelian]

hi
it seems ther some thing missing here size of the room compressor &fan coil capacity so we can go from there