Refrigerant Delta T Across Evaporator

[FieldGen3]

I saw a journeyman take the temperature rise of the refrigerant across the evaporator of a walk in cooler. First temp was taken between the outlet of the txv and the distributor, second temp was taken at the outlet of the evaporator near the sensing bulb. He thought he was checking superheat, I know this is incorrect, but I've learned to stop arguing and just look it up later. My question, is there a benefit to this information in troubleshooting? Will adjusting the txv spring change this delta t? He thought it would but was satisfied with the reading he got so no adjustment was made. Thanks for any input.

[icemeister]

Welcome to HVAC-Talk!

What he was doing...incorrectly...is called the two-temperature method of measuring evaporator superheat.

The Heatcraft I&O Manual (See p.34) the proper way is to measure the first temperature at the suction line where the TXV bulb is attached and the second temperature at the outlet of one of the distributor tubes (where it enters the coil).

Theoretically, this will give you a fairly decent superheat reading because the inlet temperature should be approximately equal to the saturation temperature of the refrigerant in the evap, but it fails to take into account a couple of variables which can throw it off.

First is there's always some pressure drop through the coil itself, so the reading you get at the inlet will actually be a bit warmer than the average saturation temperature in the coil.

Then, if you have a refrigerant blend there's the effect of the glide which for some flavors may be 10ºF or more through the coil.

For these reason, I prefer to take the second reading at a return bend at about the midpoint of a coil circuit. It helps to account for both the coil pressure drop and the refrigerant glide. The best way of course is to use the temperature/pressure method for the most accurate superheat readings.

What your journeyman was doing by measuring the second temperature at the distributor didn't account for the pressure drop in the distributor, which is typically around 25 psi, and also the pressure drop in the distributor tubes, which can be 10 psi or so. With those factors in play, I can't imagine how he considered his superheat to be OK.

[Tommy knocker]

Journeyman installer?

[FieldGen3]

Thanks for the explination. It's starting to make a lot more sense.

[Phase Loss]

if you have a refrigerant blend there's the effect of the glide which for some flavors may be 10ºF or more through the coil.

Doesn't the 2 temperature method bypass all that glide business.

Temperature is Temperature

The glide only comes into play once you gauge up.

[jpsmith1cm]

Doesn't the 2 temperature method bypass all that glide business.

Temperature is Temperature

The glide only comes into play once you gauge up.

Let's look at 407a.

With a 40# evaporator, your bubble point (evaporator inlet) is going to be 8 degrees. Your dew point (outlet) is going to be about 18 degrees.

So, even if you're running a 0 degree superheat (measured by the pressure-temperature method), it will appear that you're running a 10 degree SH by using the two temperature method.

[Phase Loss]

So if I had a 40# suction.

I have 8* liquid/vapor at the beginning of the coil

I have 18* vapor at the end of my coil

Now I need an additional 6* superheat for my cooler

I now have a 24* suction line

what a crappy refrigerant.

[icemeister]

So if I had a 40# suction.

I have 8* liquid/vapor at the beginning of the coil

I have 18* vapor at the end of my coil

Now I need an additional 6* superheat for my cooler

I now have a 24* suction line

what a crappy refrigerant.

The issue isn't that it's a crappy refrigerant, it's just that the two-temperature method is a crappy way to determine superheat with a blend.

If you use the temperature/pressure method based on the dewpoint temperature, there's no need to be concerned about the glide. It's really no different than it is with a single-component refrigerant.

Where you need to be aware of glide is if you need to know what the evaporator temperature actually is...such as when you're initially setting up EPRs on a rack system. If the case specs call for a specific evap temp it's easy with a single component gas, but with a blend you need to determine the mean evap temperature, which according to DuPont, is the temperature within about 65%-75% of the total glide. here's a link explaining it (See p.2):

DuPont Suva Temperature Glide PDF

Basically it's saying that for a refrigerant with a 10ºF glide, the mean evap temp will be approximately 7ºF above the bubble point...or 3ºF below the dewpoint temperature, so in JP's example it would be 18ºF-3ºF or 15ºF.

[jpsmith1cm]

The issue isn't that it's a crappy refrigerant, it's just that the two-temperature method is a crappy way to determine superheat with a blend.

If you use the temperature/pressure method based on the dewpoint temperature, there's no need to be concerned about the glide. It's really no different than it is with a single-component refrigerant.

Where you need to be aware of glide is if you need to know what the evaporator temperature actually is...such as when you're initially setting up EPRs on a rack system. If the case specs call for a specific evap temp it's easy with a single component gas, but with a blend you need to determine the mean evap temperature, which according to DuPont, is the temperature within about 65%-75% of the total glide. here's a link explaining it (See p.2):

DuPont Suva Temperature Glide PDF

Basically it's saying that for a refrigerant with a 10ºF glide, the mean evap temp will be approximately 7ºF above the bubble point...or 3ºF below the dewpoint temperature, so in JP's example it would be 18ºF-3ºF or 15ºF.

This whole glide thing bit me on my ever-increasing backside a while back.

Had a Hussmann sushi case (ASPSM? I forget the exact model) that wasn't holding temp. R407a rack converted from R-22.

EPR was running about 40# or 18 degrees, right? Not quite.

The effective evaporator temp was, as icemeister said, somewhere in the middle. I tend to simply average them. This made the actual evap temperature closer to 13 degrees.

Too cold evaporator caused both excess frost buildup and required extended time to defrost completely.

To add to the problem, other techs were LOWERING the evap pressure, exacerbating the problem.

I wound up sitting and thinking on what was actually happening, bringing the evap pressure UP to 48# which gives an AVERAGE of 20 degrees evap temp.

Problem SOLVED.

[Phase Loss]

This is embarrassing I never gave much care for glide

Just hooked up my digi's and it did the work for me never really looked at a PT chart with vapor/liquid pressures.

makes me think back a few years ago, back to a chain of stores my old company used to service.

Each case used a case controller that used a pulsing metering device that based superheat off of coil inlet/outlet temperature difference.

Each store was R-404A and worked well.

Then they put in a 407C store...and it was oil fail central.

[Capz]

Welcome to HVAC-Talk!


For these reason, I prefer to take the second reading at a return bend at about the midpoint of a coil circuit. It helps to account for both the coil pressure drop and the refrigerant glide. The best way of course is to use the temperature/pressure method for the most accurate superheat readings.

icemeister, what model thermometer do you use and how do you secure it in order to get an accurate temp reading on an evaporator tube? Thanks

[icemeister]

icemeister, what model thermometer do you use and how do you secure it in order to get an accurate temp reading on an evaporator tube? Thanks

I use a Fluke 52-2 with a Fluke 80PK-8 pipe clamp for the coil outlet or suction line. For the return bends it get more problematic, but I have found a plain type K thermocouple wire probe secured with a very small hose clamp covered with some cork tape works OK. A good surface probe might be nice too...but I don't have one.

But.... I really don't recommend the two temp method and haven't used it much at all. If the superheat reading is that important, it's better to just burn in a pressure tap and be sure.

[Capz]

I also use the Fluke 52-2 with the K thermocouple wire probe and some foam or cork tape. I was just curious to know if anyone had a better method for temperature accuracy

[GT Jets]

This is embarrassing I never gave much care for glide

Just hooked up my digi's and it did the work for me never really looked at a PT chart with vapor/liquid pressures.

makes me think back a few years ago, back to a chain of stores my old company used to service.

Each case used a case controller that used a pulsing metering device that based superheat off of coil inlet/outlet temperature difference.

Each store was R-404A and worked well.

Then they put in a 407C store...and it was oil fail central.

This actually proves a point I tried to make a few times and got slammed for it.... Glad you knew the manifold set was smarter than you were....

Not being condescending, just appreciating you know the difference.

GT

[Phase Loss]

I knew it read dew/bubble point...so I never cared to look into it


technology made me lazy & stupid there.

[itsiceman]

I also use the Fluke 52-2 with the K thermocouple wire probe and some foam or cork tape. I was just curious to know if anyone had a better method for temperature accuracy

Cooper 4011 fits great on the little tubes well as suction lines or return bends.
Much easier/better than the beaded K-Type IMO

[itsiceman]

Phase Loss This might help it sink in. This was on a unit overfeeding. In the pic it shows bubble, dew for the pressure and suction temp.
It's easier for me to grasp glide when seen this way vs pt charts and graphs but maybe somebody has a better way of explaining glide. HTH

[Phase Loss]

The impossible negative superheat scenario. DUN DUN DUUUUNNNNNNNNNN

[bsullins]

I also use the Fluke 52-2 with the K thermocouple wire probe and some foam or cork tape. I was just curious to know if anyone had a better method for temperature accuracy

Try the fluke 80pk-3a touch probe. I have several of the fluke temp clamps, but like the touch probe best.