PDA

View Full Version : Something I missed?







busman
13-12-2004, 10:22 AM
I designed my own heat pump for my house. It's been running now for about 2 years now, but there is one question that I have failed to find an answer to now, and thought that someone here, may have an answer.

Here's the problem:
When running in heating, the outdoor (evap) has 7 circuits for the distributor. 3 ton nominal system. R22. Balanced port TXV. According to Sporlan's software, the TXV, distributor, and orifice all work good on paper. I have an un-even frosting pattern on the coil however. I'm wondering now, do I have too many circuits?

Gary
13-12-2004, 04:13 PM
Is the frost uneven between circuits (more frost on some circuits than others)? Or is the frost uneven across the circuits (more frost at beginning of circuits than end of circuits)?

What are the following:

Evap air in temp
Evap air out temp
SST (or low side pressure)
Suction line temp near evap outlet
Suction line temp near compressor inlet

chemi-cool
13-12-2004, 04:20 PM
I will go for poor soldering of the distributor, some of the tubes are partialy blocked.

On another thing, I'd go with Marc, air flow through the coil.

Chemi :)

Gary
13-12-2004, 04:32 PM
We can't see it from here. The only way we can "see" a system online is by the description and the temperature measurements.

Going on the insufficient information given, my guess would be an oversized coil, in which case the evap dT will be very low and the superheat will match the evap TD.

busman
13-12-2004, 06:28 PM
Well, air flow seems to be fine, in fact too much (approx 5000 cfm) with an 825 rpm 1/2hp fan motor 26" 4wing 27 pitch. Oversized coil, yes, old 10 ton a/c condensor coil with 7 circuits, and 1 subcooling loop (not used in heating).

Since it's a heat pump, the outdoor temperature varies much, and affects all the numbers. Seem to notice it most when temperatures get cold outside, and frosting begins on coil.

Several circuits in the middle of the coil is what is frosting up. There is flow through all circuits, all the time. It's not a no-flow situation in some circuits. Solder job looked great on install, looked down inside, and no visible obstructions in the refrigerant path. Upon running, there is a band of frost the appears in the center two or three circuits. It's not at the bottom, (along the ground where ground moisture may be more abundant) or at the top, but dead in the middle.

Distributor is pointing downward to lend to proper distribution.

Moisture and dirt *should* not be present, or any other form of contaimination. Pulled an eight our vacuum on the system at 500 microns when I got it, pressurized with nitrogen. It sat for several years, assembled system purging with nitrogen during brazing process, had another 500 micron vacuum and have suction *filter* and bi-flow liquid filter/drier.

System mass flow, that's something I have not explored. Can you expound on that?

Thanks!

Gary
13-12-2004, 10:20 PM
As I understand it, the outdoor coil is a 10 ton coil with a 3 ton TXV and a 3 ton compressor.

The coil being grossly oversized, likely the superheat at all of the suction ends as well as at the TXV bulb, will be identical to the TD. IOW, the suction line readings will all be at ambient.

Gary
13-12-2004, 11:00 PM
I would agree with you, Gary, if the compressor and the outdoor coil were a match using a 3 Ton TEV. That's not to say what you say is impossible just that under my scenario far more likely.

Provided the TEV has something very close to its factory setting then finding superheat discrepancies between circuits should be a piece of cake. Experimenting with a thermometers position air-side before and after non-frosted and frosted tubes too will provide valuable pointers.

Does the TEV have a MOP function?


Given a 10 ton coil and 5000 CFM of airflow, the 3 ton flow of refrigerant will be completely boiled off long before it gets to the other end of the coil, and the suction cannot rise above ambient because there is no (above ambient) heat source to cause it to do so. Therefore all coil outlet superheats should theoretically match the TD.

Gary
14-12-2004, 01:06 AM
Hopefully Busman will provide us with temperature measurements so we can see what it is doing.

busman
14-12-2004, 04:47 AM
TXV is a Sporlan C-charge head (medium temp refrigeration.)

No flash gas is present in the liquid line. I have a site glass mounted in the liquid line at mid-point between the air handler, and the outdoor unit. This is outside my kitchen window, so that I can monitor system performance. The only time that bubbles are present, is during restart back to heating after a defrost. That may take 5 minutes for bubbles to clear at that mid-point site glass. There is a second site glass mounted at the inlet of the outdoor TXV. (wont Prof sporlan be happy!) That site glass remains clear almost always because it's at the lowest point in the system. No extra means of subcooling is used in heating mode. It comes directly out of the air handler (second floor) and down into the outdoor unit.

I have experimented with opening up the TXV to decrease superheat. My thinking was attempting to increase the pressure drop across the nozzel and distributor, as opposed to using the TXV pressure drop. This would also create a higher suction pressure too, which should increase the efficiency of the unit in heating mode. Regardless of TXV superheat setting, the results remained unchanged.
Side note: coil is clean. It's been washed out, cleaned out, cleaned with Calgon's Nu-Brite, so there "shoud" not be any thing restricting air flow through any part of the coil.

I will see what other hard data I can gather to see which path to go down next.

I have a second heat pump on the addition to my house. It is a factory built system. It's an ICP (Arcoaire) 12 SEER unit. The homemade system averaged about 15% higher suction pressure (in psia) than that of the factory system in heating mode.

Gary
14-12-2004, 05:29 AM
TXV is a Sporlan C-charge head (medium temp refrigeration.)

No flash gas is present in the liquid line. I have a site glass mounted in the liquid line at mid-point between the air handler, and the outdoor unit. This is outside my kitchen window, so that I can monitor system performance. The only time that bubbles are present, is during restart back to heating after a defrost. That may take 5 minutes for bubbles to clear at that mid-point site glass. There is a second site glass mounted at the inlet of the outdoor TXV. (wont Prof sporlan be happy!) That site glass remains clear almost always because it's at the lowest point in the system. No extra means of subcooling is used in heating mode. It comes directly out of the air handler (second floor) and down into the outdoor unit.

I have experimented with opening up the TXV to decrease superheat. My thinking was attempting to increase the pressure drop across the nozzel and distributor, as opposed to using the TXV pressure drop. This would also create a higher suction pressure too, which should increase the efficiency of the unit in heating mode. Regardless of TXV superheat setting, the results remained unchanged.

With that oversized coil, there is no way you can adjust the TXV superheat. Adjustment is futile.



Side note: coil is clean. It's been washed out, cleaned out, cleaned with Calgon's Nu-Brite, so there "shoud" not be any thing restricting air flow through any part of the coil.

I will see what other hard data I can gather to see which path to go down next.

If you scroll back a page, you will find a list of temperature measurements. Those temperatures will tell the story.



I have a second heat pump on the addition to my house. It is a factory built system. It's an ICP (Arcoaire) 12 SEER unit. The homemade system averaged about 15% higher suction pressure (in psia) than that of the factory system in heating mode.

How much does the electricity for that oversized condenser fan motor cost?

shogun7
14-12-2004, 06:11 AM
With that oversized coil, there is no way you can adjust the TXV superheat. Adjustment is futile.(QUOTE) Gary

With that said I believe you can get some sort of balance by eliminating 3 of the cap tubes and derating the air flow to about 3000 CFM :confused:

Gary
14-12-2004, 06:40 AM
With that oversized coil, there is no way you can adjust the TXV superheat. Adjustment is futile.(QUOTE) Gary

With that said I believe you can get some sort of balance by eliminating 3 of the cap tubes and derating the air flow to about 3000 CFM :confused:


Actually, I had the other approach in mind: Leave the coil as is and drop the airflow down to about 1000 CFM, give or take a couple hundred.

Gary
14-12-2004, 06:55 AM
Morning gentleman. Is that because with the lower than normal TD the coil is more likely to be flooding than anything else, Gary? :)

Almost right. Low TD and superheat to match, but no flooding. With that much heat load it can't flood, even if it is wide open, which it is. It might as well be a cap tube.



You must remember, the TEV being matched to the compressor WILL easily control it's own superheat i.e. act to force an establishement of it's superheat setting by simply closing down until it has dictated the operating TD it requires to see the superheat it desires as superheat, and subcool for that matter, is hugely dependant on final coil TD for its establishment.

I just want to know the relative superheats of the two circuit types, frosting and non frosting, before we move on. Never assume anything, a possible assumption here is that you have even air flow.

I'm all for getting lots of temperature measurements. :)

busman
14-12-2004, 05:53 PM
TXV is externally equalized (a must for distributor setups.) I don't recall the exact design temperature that I used. I attempted to try numbers at many different ambients to account for all the different temperatures that the unit would be exposed to.

I don't "think" oil logging is a problem with this setup. It's the same tubes over and over, and frosting begins just after defrost, almost too soon for it too begin to log any oil.

I have a set of gauges afixed to the unit. This morning, with an ambient of 36F, it was running 48psig (24F SST). No superheat reading available yet. I'm home taking care of kids today (sick wife), so that reading is going to be a little be difficult.

Will post more info soon. Thanks!

chemi-cool
14-12-2004, 06:49 PM
Hi busman.

At 36F, frost on the outdoor coil is normal.
I dont know what control your defrost, but you can use a thermostat for controlling the amount of frost.

At these low ambients, you will have more defrost cycles per hour.

Chemi :)

Gary
14-12-2004, 08:56 PM
I have to say that it is fun watching you convert temperatures for a change, Marc. :)

As I see it, the factory superheat setting is more like 7F. Since the saturated mixture is boiling off long before it reaches the bulb, the TXV is "seeing" 12F superheat (equal to TD), and is running wide open in a futile attempt to bring down the superheat.

I would love to do some experimenting here, but we need temperature readings so that we can do before and after comparisons.

chemi-cool
14-12-2004, 09:03 PM
One more thing that comes to mind, busman.

Is the outdoor fan sucking the air from the coil or pushing it through?

If it pushes the air through the coil or if it sucks it but it is close to the coil then there is almost no air movement at the centre of the coil :cool:

Chemi :)

busman
14-12-2004, 09:16 PM
I used a balanced port TXV in this application, because I thought it would do a better job of controling the refrigerant in this wild application.

I will go out and get some pics, and parts list of this setup soon.

busman
14-12-2004, 09:26 PM
Hi busman.

At 36F, frost on the outdoor coil is normal.
I dont know what control your defrost, but you can use a thermostat for controlling the amount of frost.

At these low ambients, you will have more defrost cycles per hour.

Chemi :)

Frosting I expect. It has a demand defrost control on it to handle defrost when necessary, as opposed to time/temperature/thermostat conventional system. This employs two thermistors that compare ambient with coil temperature, and activates defrost when needed.

The issue is the imbalance of frosting on the coil between circuits.
I also have put on a time delay relay that runs the outdoor fan for an extra 2 minutes after compressor shut down to allow warmer air in milder temps to help melt off frost, and prevent the need for more call for defrosts.

shogun7
15-12-2004, 03:14 AM
[QUOTE=busman]Well, air flow seems to be fine, in fact too much (approx 5000 cfm) with an 825 rpm 1/2hp fan motor 26" 4wing 27 pitch. Oversized coil, yes, old 10 ton a/c condensor coil with 7 circuits, and 1 subcooling loop (not used in heating).

How does one use a 3 ton valve to feed a 10 ton heat exchanger and not starve the coil even if its wide open?
Seems to me you will never get the proper mass flow to feed all the circuits properly (evenly?) thats why I suggested cutting out some circuits and air flow. He's pumping 5 to 7 tons of air through the coil with 5000 CFM in addition to starving it :confused:

busman
15-12-2004, 04:48 AM
Well, it's a 10 ton condensor coil, but if you go by coil face area, it's about close to some of your more garden variety residential systems. I think the way that got 10 tons out of this stinker was moving a huge amount of air, and not really worried about SEER rating.

I was looking at the feeding characteristics tonight, (but was unable to spend enough time to take temps :rolleyes: ) The distributor tubes are off of opposite sides of the distributor, so it kinda rules out blockage on one side of the distributor idea.

Oh, sorry, it was asked before and I forgot to answer. It's not the original fan motor. I replaced it with a slower RPM motor. The original was a 1 hp 1075 rpm motor. Now I have an 825 rpm 1/2hp motor, with original fan blade. I could get away with a 1/3 hp, but when it comes to the coil frosting up, the motor begins to lag down too much. So, electrical draw is much more appealing. Original 5 Amps, now just under 2.

As promised here is the stat sheet on the parts:

Copeland ZR34K1PFV Scroll Compressor
Sporlan BFBVE body with C charge power head
Sporlan ASC7 Aux Side connector
Sporlan 1113-7-1/4 distributor
Sporlan #3 nozzle for distributor

Now, here's something that I forgot all about, until I started looking at this thing again tonight. Sporlan did not offer a 3/16 connection for distributor that I needed. I needed 3/16 according to their program to gain enough pressure drop for a proper operating distributor. So, I used a 1/4 distributor. I used 1/4" tubing stubs, exactly cut, fitted the 3/16's into the 1/4" tubing, at exact insertations, (so all would be equal). The other part that I noticed, was I had to use an ASC7 (7/8) connector to fit the distributor. So, I had to sweat a reducer between the TXV and the ASC.

peterweston
15-12-2004, 11:51 AM
:D Does the unit ever cylce? Not sure on the temp as i am in Australia and indoor heating not used that often.

busman
15-12-2004, 12:31 PM
So have I dropped off my internal coil velocity too far you think?

busman
15-12-2004, 12:47 PM
:D Does the unit ever cylce? Not sure on the temp as i am in Australia and indoor heating not used that often.

Average wintertime ambient here can range from 25-35F (sorry guys, I'm stuck on the dumb American system of measurements) over night, which is not too bad. However, last year, was a real bear for this area. Overnight lows were down to about +5F for about 3 weeks straight. And this heat pump was still heating. Yes, the unit does cycle very well.

Gary
15-12-2004, 03:07 PM
When are you going to give us some temperature readings, so we can see what this thing is doing?

Evap air in temp
Evap air out temp
SST (or low side pressure)
Suction line temp near evap outlet
Suction line temp near compressor inlet

A measurement is worth a thousand calculations.

Gary
15-12-2004, 03:32 PM
Oh, sorry, it was asked before and I forgot to answer. It's not the original fan motor. I replaced it with a slower RPM motor. The original was a 1 hp 1075 rpm motor. Now I have an 825 rpm 1/2hp motor, with original fan blade. I could get away with a 1/3 hp, but when it comes to the coil frosting up, the motor begins to lag down too much. So, electrical draw is much more appealing. Original 5 Amps, now just under 2.


What makes you think you are getting 5000 CFM of airflow?

busman
15-12-2004, 11:12 PM
Okay, here are the winning numbers:

34F Ambient Temp
46#psig Suction Pressure
(22F) SSTemp
32F Suction Line Temp at compressor
32F outlet on a circuit that has known to frost (not during measurement)
Discharge air temperature 29F.

So, looks like about a 5TD on the coil, running 10F Superheat.

Gary
15-12-2004, 11:41 PM
Okay, here are the winning numbers:

34F Ambient Temp
46#psig Suction Pressure
(22F) SSTemp
32F Suction Line Temp at compressor
32F outlet on a circuit that has known to frost (not during measurement)
Discharge air temperature 29F.

So, looks like about a 5TD on the coil, running 10F Superheat.

34 - 29 = 5F dT
34 - 22 = 12F TD
29 - 22 = 7F approach
32 - 22 = 10F superheat

So, looks like about a 12F TD on the coil, running 10F superheat.

I'm surprised that the superheat is 2F below the TD. I would question the instrumentation. But I'm thinking Marc may not agree... LOL

Gary
15-12-2004, 11:46 PM
Here's an experiment you may want to try:

Remove the fan blade and flip it over, so that it "backhands" the air. This will greatly reduce the airflow. Then take a new set of readings.

With a 5F dT, you are not getting 5000 CFM, or anywhere in that neighborhood. By downsizing the motor and RPM you have brought the CFM down almost into the acceptable range.

Gary
15-12-2004, 11:51 PM
Just to clarify:

When you measure the change in temperature of a single thing, such as a stream of air, the change is a delta-T or dT.

When you compare the temperatures of two different things, such as the ambient temperature and the refrigerant temperature, the difference is a TD.

Gary
16-12-2004, 12:13 AM
Instruments? Well, saturated suction back up in the evap is probably even a little higher while the actual temperature within the suction line might even be a little lower :)

Otherwise, none of those number surprise me. And my concerns about TEV superheat control conflicting with with the TD ceiling are very real indeed. As ambients drop I expect the compressor capacity loss affecting a reduction in TD to overtake the TEV's superheat setting adjustment by dP/dT ratio.

But none of these figures will tell us whether we have poor air, refrigerant or turbulence distribution.

There was no difference between the coil outlet superheat and the compressor inlet superheat. How could there be any significant difference on other circuit outlet superheats?

Gary
16-12-2004, 12:29 AM
Given a substantial reduction in airflow (flipping the blade), this should bring the TXV within its full control range, and circuit outlet superheat differences would then be much more pronounced.

shogun7
16-12-2004, 03:05 AM
Dummy, it's only a 10 Ton coil at it's nominal TD. Otherwise, it is just a coil. Until you state the index TD it is just a coil and not yet a coil with a capacity rating.

If it's a 10 Ton coil with a 15K TD then it's a 3.3 Ton coil at a 5K TD except then with poorer tube turbulence and heat transfer. Possibly even with poorer heat transfer in some circuits over other circuits.
I know that stupid! Did I say it was doing 10 tons at it's existing conditions? NO! I was referring to the fact that it's surface area was capable of doing 10 tons at std conditions. and it's not a 3ton coil at 5K it's a 10 ton coil doing 3 tons of refrigeration!!!:eek: so with all your bulsh** your looking either stupid dumb or fuffering from constipation of the cerebral :rolleyes: And don't call me names...Dummy!

shogun7
16-12-2004, 03:17 AM
Agreed, but only if it isn't a distribution problem caused by the excessive TEV closure by the SH/TD conflict I'm talking about :)
Fliping the blade may increase TD but I predict a greater frost problem will develop as it will not make much difference to suction temp as ambient falls ;) This coil is just too big for the application to be efficient but it seem to work for busman in a fashon :confused:

busman
16-12-2004, 03:34 AM
One other thing I noticed. I compared the two running systems tonight when I took the readings. The 12 SEER factory system was running at 40# of suction pressure. I like the higher suction pressure of the homemade one better.

Counterflow - okay, I know what counterflow means, but counterflow what? Counterflow of air, counterflowing refrigerant, or counterflowing of what?

shogun7
16-12-2004, 04:04 AM
One other thing I noticed. I compared the two running systems tonight when I took the readings. The 12 SEER factory system was running at 40# of suction pressure. I like the higher suction pressure of the homemade one better.

Counterflow - okay, I know what counterflow means, but counterflow what? Counterflow of air, counterflowing refrigerant, or counterflowing of what?

When you install the coil for counterflow operation,you make the suction connection on the entering air side of the coil. This maintains a temperature difference between the coil surface temperature and the leaving refrigerant temperature. It provides one of the control means for the thermostatic expansion valve. As this temperature difference becomes small, suction header location becomes more important. If you neglect the minor effect of pressure drop in an evaporator, the temperature of the refrigerant is constant. The importance of counterflow operation from a capacity standpoint in a direct expansion coil is far less than in a water coil. In a water coil, there is a major change in the temperature of the water passing through the coil. :)

Gary
16-12-2004, 06:11 AM
One other thing I noticed. I compared the two running systems tonight when I took the readings. The 12 SEER factory system was running at 40# of suction pressure. I like the higher suction pressure of the homemade one better.


Flip the fan blade and see what happens. Try it, you'll like it.

I would expect little if any difference in suction pressure. What I expect to change is the superheat.

Gary
16-12-2004, 04:38 PM
If the low TD is forcing the TEV to take evasive action and close then an increased TD by air flow reduction would allow the valve to open some more.


Being a medium temp valve, let's imagine that the factory superheat setting is about 7F. The bulb is sensing 10F superheat and is opening as a result. However the saturated mixture is boiling off before reaching the bulb, so the coil stays right there at 10F superheat.

Given less airflow, I would expect the saturated mixture to then reach the bulb, allowing for 7F superheat. IOW, it would allow the valve to close some more, not open some more.

BTW, I'm thinking the suction line is in the (lower temp) discharge air stream, which would account for lower than ambient suction line temp. This would explain why the superheat does not currently match the TD.

busman
17-12-2004, 03:02 AM
I think that the TXV is throttling as it should be. Last year, I do remember experiementing opening up the TXV more, to reduce superheat.

Now, if I could figure out how to post the pics from the morning. I took some pics of the frosting pattern. I was not as pronouced this morning, as I have seen it before, but you get a general idea of what's happening. Maybe this something I'm just going to have to live with, or not, I don't know. I was unable to take the specific circuit readings this morning, because I had a relay die on it this morning. My switchover relay to allow my hydronic heat to take over Aux heat failed, and would not allow any Aux heat to come on (hydronic or electric backup) Needless to say, the temp was plumiting when I got up this morning when it went into defrost.

Hopefully, conditions will come around again here pretty soon, for good dewpoint & temp's for making another good batch of frost, so that I can get some good readings.

I'm a little leary of some of my readings with this thermocouple thermometer. Whenever the instrument gets cold itself, the temperature likes to drift. So I must keep the instrument warm while taking readings in my coat pocket.

Two pics are: (1) the in the dark photo, soon after defrost, and the (2) photo is just as defrost begins.

busman
17-12-2004, 03:14 AM
Other pics to check out:
First, is just a general pic overhead of the unit.
Second, is the inside of the unit, and a pic of the distributor.

The unit is a TTA120 Trane.

Gary
17-12-2004, 01:43 PM
It looks like each circuit does four complete loops, two of which are frosted. All of the refrigerant is boiled off half way through. It will be interesting to see what it looks like with the blade flipped over.

shogun7
18-12-2004, 01:05 AM
Indeed, our opinions don't differ on what makes TEV's open or close, right now they simply differ on which scenario this particular TEV rests, whether it will open or close :)
Let see Gary says 7*f SH Soooo I would suspect if Busman flipps the the fan and it goes to oh say 5*f it will stay closed ...However if it goes to OH say 10*f it should go open. Do you agree marc?? :confused:

shogun7
18-12-2004, 02:09 AM
So Gary, we know that the larger the coil surface area of the evaporator, the closer the coil temperature will be to the entering air temperature. (low TD)This happens because of a higher operating coil temperature with a larger coil and the A/C unit would be running higher evaporating (suction) pressures and temperatures. So if we cut the air flow too low it seems to me that the frosting problem will be exiserbated. What say you? :rolleyes:

Gary
18-12-2004, 05:56 PM
So Gary, we know that the larger the coil surface area of the evaporator, the closer the coil temperature will be to the entering air temperature. (low TD)This happens because of a higher operating coil temperature with a larger coil and the A/C unit would be running higher evaporating (suction) pressures and temperatures. So if we cut the air flow too low it seems to me that the frosting problem will be exiserbated. What say you? :rolleyes:

The TXV is running wide open, metering as much refrigerant as its orifice will allow. The refrigerant is entirely vaporized by the time it gets through two of the four loops in each circuit.

With less airflow, the liquid/vapor mixture will make it further through the circuit before it is entirely vaporized. Given enough airflow reduction, liquid/vapor mixture will make it to the TXV sensing bulb and the TXV will begin to control the flow.

At that point, all four loops in each circuit will be frosted. IOW, the frost should be evenly distributed over the entire coil.

busman
18-12-2004, 09:35 PM
Well, something else I would like to throw in the mix here of discussion, is the amount of refrigerant charge. If I slow down the fan speed, [in my mind] it will require more refrigerant to run in colder temperatures. This thing already has a huge amount of refrigerant in it, and to require more, may in milder temps, run a sky high head pressure. It currently runs over 325 head on mild temperature days, and hate to see rise much higher than that.

When all these holiday engagements slow down, I hope to get some numbers. . .

Gary
18-12-2004, 09:57 PM
Well, something else I would like to throw in the mix here of discussion, is the amount of refrigerant charge. If I slow down the fan speed, [in my mind] it will require more refrigerant to run in colder temperatures. This thing already has a huge amount of refrigerant in it, and to require more, may in milder temps, run a sky high head pressure. It currently runs over 325 head on mild temperature days, and hate to see rise much higher than that.

When all these holiday engagements slow down, I hope to get some numbers. . .

Turn the fan blade over. You will NOT need more refrigerant.

Then report the following temps so we can see everything the system is doing:

Condenser air in temp
Condenser air out temp
Saturated condensing temp (from P/T chart)
Liquid line temp near condenser
Liquid line temp near metering device

Evaporator air in temp
Evaporator air out temp
Saturated suction temp (from P/T chart)
Suction line temp near evaporator
Suction line temp near compressor

Gary
18-12-2004, 10:41 PM
BTW, if you are charging by sightglass you are doing it wrong. Yes, I know it's the way everyone does it. Over the past couple decades I have removed more refrigerant from systems than I have put into systems. Everyone overcharges.

Gary
19-12-2004, 03:09 AM
Slow the airflow down enough for the saturated mixture to reach the sensing bulb. The TXV will pull back from the wide open position, and you won't need nearly as much refrigerant.

shogun7
19-12-2004, 03:13 AM
Effing moron.

Proximate Genus: Frosting.
Specific Difference: Uneven between circuits.

How does your question tackle the specific difference when it doesn't even satisfactorily indicate a proximate genus?
Oh stick it in your ear you sick baste88 ;) What I find really funny Marc is youve been all over the place with guesses as with Gary has been at least consistant but hey keep guessing you might hit the right solution :D

Gary
19-12-2004, 12:31 PM
An entirely plausible theory, and the frost pattern through the center lends credence to the uneven distribution part, however...

It seems unlikely that the TXV is throttling the refrigerant flow. If a saturated mixture were reaching the sensing bulb, then the (medium temp) TXV could be adjusted through its full range. As I recall, Busman was unable to open up the valve to further drop its superheat.

With only vapor reaching the sensing bulb, the superheat cannot be adjusted below the TD, although in this case the superheat seems to be slightly below TD, presumably because it is being influenced by the lower ambient surrounding the suction line.

In any case, the airflow needs to be reduced in order to achieve some rough balance.

Also, we need to look at the high side numbers, adjust the refrigerant charge, and then see what the numbers say about overall operation, as well as distribution.

An interesting project, for sure.

Gary
19-12-2004, 08:57 PM
If the valve SH is sitting on the border of conflict with TD (fourth circuit (SH index circuit)) then there wouldn't actually be a whole lot of stroke left for it to open through.

Which would mean that it is running wide open.

busman
20-12-2004, 04:36 AM
Well, I give you all credit, you've got your thinking caps on. I just got home from another family party, and it's waayyy too cold to be messin' around out there right now. Temps in the 20's F, with a 40mph wind gusts, near zero wind chill with snow to boot. I just locked out my heat pumps and I am firing on hydronic boiler at the moment.

The distributor tube lengths, are all exactly cut, and carefully measured. Now, as stated before, I could not find a distributor with the size tubing that Sporlan's software called for. They called for 3/16 tubing. Sporlan does not offer a 7 circuit 3/16, only in 1/4 inch. Using 1/4 tubing does not offer the pressure drop needed to properly distribute refrigerant. I used the 1/4 one, and used exactly and evenly cut 1/4 tube stubs to mount into the distributor, and then sweated the 3/16's into the 1/4 tubing. Unconventional, yes. Insertions were all at exact measurements, so that all would be even too. All 3/16's were debured, and inspected to allow even flow between all circuits. I'm starting to wonder here if maybe there is something going wrong in the distributor because of this "unconventional" usage of the distributor.

Gary
20-12-2004, 08:31 AM
It is entirely possible that the increased delta-P from refrigerant overcharge is affecting the distribution also.

It seems to me that we are putting the cart before the horse. This system has multiple problems, and we need to identify and resolve them in the proper sequence. The frost pattern is likely to be the last problem on the list.

We need to start with a full list of temperature measurements.

We already know that the airflow needs to be reduced and the charge needs to be adjusted. Who knows what other problems a full list of temperature measurements will reveal?

busman
20-12-2004, 01:08 PM
I'm not sure if we are going to be able to come up with the "perfect" conditions for this unit to work at and attain a "perfect" charge. The ambient temperatures that this unit will run at, is anywhere from +55F to down to +15F. With an outdoor coil half the size of the state of Georgia, it's going to be hard to find an extact charge. I've tinkered around with the thought of managing the liquid end somehow, but unsure how. Since this is a heat pump with a reversing valve. I could come up with a receiver, about 4 check valves, and create some kind of liquid "rectifier" of some sort, but thought that it maybe somewhat unpractical.

When I first put this unit online, I did have it overcharged. It shut off on high pressure late one night running in heat mode (mild evening). So I reduced the charge. Then last year, I added refrigerant because I opened up TXV, and it has run happly since, with the exception of the frost.

Little unsure about your delta P idea. Are you talking across the entire valve to evaporator inlet, or just post TXV across the distribution assembly?

Re-hashing some of the other ideas here too, not sure I've gotten the jist of the TXV superheat/full stroke open idea either. It seems as though this problem is persistant across most of it's operational temperatures, only just being more noticable when frosting occurs.

As for head pressures, it is mostly dependant again upon ambient temperatures it's operating at. It will varies from at the lowest ambient of operation about 200 psi to as high as 325 at it's warmest operation. Of course it's really nice in the summertime, because the head just barely even gets to about 200 on the hotest day, and bottoms out sometimes around 150 on the coolest of days.

shogun7
21-12-2004, 01:16 AM
Busman in a heat pump the out side coil that is too big can act like it has an insufficient charge, liquid line restriction, or TXV problems which can generally cause what is called the ZEBRA EFFECT, which are bands of frost like you describe. Now, some physics and heat pump basics, the warmer the outside air, the more capacity it has for moisture. Heat pump coils generally run 20 colder than the outside air temperature to facilitate heat exchange. So if the outside air temp is 35, the coil is 15 or so - plenty cold to lower the air passing through to dewpoint temperature and then convert the condensation to frost. Moderate outdoor air temperatures accelerate the frosting phenomenon, other things being equal. Now of course the only way to prove the is by taking the proper readings as Gary suggested, but I suspect that unless you install a smaller condenser you will not do much to alleviate the present situation. It seems to be working in a fashon, but you can try to do some of the recommendations that have been brought up with out replacing it :)

shogun7
21-12-2004, 02:43 AM
Which of course brings us all the way round back to the original question "Why the uneven frost pattern?"

Or were you hoping that you could hide the fact that you have no idea of whats going on by merely giving the symptoms an exotic name? Intellectual camouflage.

Well of course your just jelous because you didn.t think up the term EH bumbkin? Boy are you thick! the reason has already been explaned ...the outside coil is too big for the application!!!
DID YOU HEAR THAT? peanut brain? now if you want the exact cause then it's because of uneven refrigerant mass flow. You
see I don't need all the double talk to make me sound like you that is state many various reasons and hope I hit the right coard. By the way what are you afraid of? :eek: That I might be right?
:D

shogun7
21-12-2004, 03:00 AM
In what way is the coil too big?

You see, it's like asking the officer at the scene of an accident (officer shogun):
"what caused the accident, officer?"
and officer shogun answers:
"well, accidents occur when things go wrong."
And so you persist:
"what went wrong here, officer"
and officer shogun answers:
"Well, can't you see, dummy, there's been an accident"
Perfectly logical...What :D I hope I'm not keeping you up too late :D Listen I have to go work out now so why don't you get some sleep youv got a big day tomorrow ...Right!

Gary
21-12-2004, 11:31 AM
I'm not sure if we are going to be able to come up with the "perfect" conditions for this unit to work at and attain a "perfect" charge. The ambient temperatures that this unit will run at, is anywhere from +55F to down to +15F. With an outdoor coil half the size of the state of Georgia, it's going to be hard to find an extact charge. I've tinkered around with the thought of managing the liquid end somehow, but unsure how. Since this is a heat pump with a reversing valve. I could come up with a receiver, about 4 check valves, and create some kind of liquid "rectifier" of some sort, but thought that it maybe somewhat unpractical.

OR... You could flip the fan blade over and get me a full set of temperature readings.




When I first put this unit online, I did have it overcharged. It shut off on high pressure late one night running in heat mode (mild evening). So I reduced the charge. Then last year, I added refrigerant because I opened up TXV, and it has run happly since, with the exception of the frost.

It's running, but it isn't happy and it certainly isn't efficient.




Little unsure about your delta P idea. Are you talking across the entire valve to evaporator inlet, or just post TXV across the distribution assembly?

Re-hashing some of the other ideas here too, not sure I've gotten the jist of the TXV superheat/full stroke open idea either. It seems as though this problem is persistant across most of it's operational temperatures, only just being more noticable when frosting occurs.

As for head pressures, it is mostly dependant again upon ambient temperatures it's operating at. It will varies from at the lowest ambient of operation about 200 psi to as high as 325 at it's warmest operation. Of course it's really nice in the summertime, because the head just barely even gets to about 200 on the hotest day, and bottoms out sometimes around 150 on the coolest of days.

Pressures and temperatures by themselves don't mean a thing. As you hinted by relating high side pressures to ambient temperatures, it is combinations of temperatures that tell us what we need to know. In this particular case, subtracting the ambient from the SCT gives us the condenser TD.

Given a full set of temperatures, we can combine them in various ways to tell us everything about the system.

My books would greatly improve your skills. :)

busman
22-12-2004, 02:40 AM
Yes, I know, get the numbers. Wife is out shoppin' at the moment, I could only steal away long enough to read replies. If I don't get my honey-do list done before her arrival, then I may have some explainin' to do about my time and where it was spent. . . don't want to have to do dat. :rolleyes:

Will keep you posted. More company coming over tommorrow night. I guess they don't call it the 12 days of Christmas for nutin'.

Temprite
22-12-2004, 01:24 PM
My books would greatly improve your skills. :)

Gary I am interested in purchasing your books,but when I go to the payment section of the site,no window comes up telling me that it is a secured site.Normally if I am on a secured site this window appears.

Rick.

Gary
22-12-2004, 06:59 PM
Gary I am interested in purchasing your books,but when I go to the payment section of the site,no window comes up telling me that it is a secured site.Normally if I am on a secured site this window appears.

Rick.

The site shows "secure" for me, indicated by a closed lock at the bottom. Is anyone else having this problem?

Temprite
22-12-2004, 09:10 PM
I have gone in there and seen the closed lock.But when I used to run windows 98 as an operating system it used to bring up a window indicating a secure site.I presume my current operating system (windows xp) is the same.Sorry to be a pain but you can't be careful enough when you are giving credit card details over the net.

Gary
22-12-2004, 10:22 PM
Given that I am still on Win98, I don't know if XP has the same features. And quite possibly the window you refer to is an optional function of your browser, which may not be currently activated.

I can only assure you that your transaction will be secure.

Gary
22-12-2004, 11:05 PM
If the lock is locked.

Exactly so. The closed lock is the universal symbol for a secure site.

shogun7
23-12-2004, 05:32 AM
You'd have to be either an NT or a freak to get that one.



Indeed, according to the data so far available, the problem can only be uneven refrigerant distribution. Until that is sorted you likely can't start making any ultra-efficient adjustments to the refrigerant charge. Unless of course you throw another NT into the pot :)
Indead you twit that's what I have been saying all along, NOW you decide to come in from the cold :rolleyes: All I can do is feel sorry for you dip sh*t! I will pray that you recover from being constipated of the cerebril :o

Temprite
23-12-2004, 12:18 PM
Given that I am still on Win98, I don't know if XP has the same features. And quite possibly the window you refer to is an optional function of your browser, which may not be currently activated.

I can only assure you that your transaction will be secure.

Gary
You were right.
I enabled the function and as soon as I went into the products section it informed me that it was a secure site.
Thanks.

Gary
27-12-2004, 05:22 AM
Gary
You were right.
I enabled the function and as soon as I went into the products section it informed me that it was a secure site.
Thanks.

I get so tired of being right. It's a curse. :(

busman
27-12-2004, 05:30 AM
Well, now as the holiday schedule has calmed down a bit, I went to tinker about the unit this afternoon. Went to gather all the info that I could, and my thermocouple thermometer was not being co-operative. Seems that as the temperature of the instrument drops, it's reading increases. So, unless there's something new going on here with frost froming at 45F, I'm going to have to hunt down a new instrument.

busman
27-12-2004, 01:06 PM
Well, it's so nice to be in company of so many "cursed" and "blessed" people in this forum. Hopefully after 12 pages and over 100+ posts we'll all be right about something here! :D

Gary
03-01-2005, 12:51 PM
Did you turn that fan blade over yet?

Superheatman
04-01-2005, 09:53 PM
Just browsing and caught this posting of yours...if you havnt got around to sorting the problem yet then it is because you have your thermocouple plugged in back to front...just reverse the pins by 180 degrees and the thermometer should work fine,,,,

busman
08-01-2005, 12:16 PM
Thermocouple is plugged in correctly. I have made that mistake before. It is an instrumentation problem. I keep forgetting to borrow the other instrument to do the test. I thought that after the holidays things would quiet down enough to go and piddle around with this thing again, but still seems to be a struggle.

I tried slowing down the fan blade to see what would happen, and it didn't seem to make much difference at all.

I talked to one of my fellow techs around here about the problem. Question after question, and answer after answer to no avail. Finally came to the conculsion that if all else fails, you could always crimp down on the lines that are over feeding. Not the best answer in the world, but starting to look very enticing after all this.

frank
09-01-2005, 07:24 PM
I got called out to a Toshiba 13kw RAS unit in the week and noticed a similar sort of zebra frosting pattern when the unit had been running for 20 minutes or so. The system running amps were a little low and it may be a little short of charge. The problem was not related to the defrost and I'll leave the charge question until better weather.

busman
10-01-2005, 04:02 AM
When you say "zebra" are you talking about an un-even frosting pattern between circuits, or are you talking about just even frosting between circuits leaving you with stripes of frost on the coil?

Now that you mention it, we've had a case one other time with this un-even refrigerant flow thing before. It was a Bohn evaporator in a walkin cooler. Cooler held about +37 (sorry, just a bit above zero for you Celius chaps) If I remember correctly, it had three circuits. Bottom circuit froze up very badly. Big ice ball on the bottom of the coil. Someone along the way put a big drain pan under the factory drain pan to catch the dripping off of the ice that grew out from the coil. Thought, "hey must be no nozzle in the distributor" or something along that line. Guess what, everything was "textbook." Correct orifice size, TXV, charge was good, distributor orientation, on and on and on. . . Still to this day it does the same thing.

frank
10-01-2005, 09:53 PM
When you say "zebra" are you talking about an un-even frosting pattern between circuits, or are you talking about just even frosting between circuits leaving you with stripes of frost on the coil?

Even frosting between circuits

Gary
15-01-2005, 05:18 PM
I tried slowing down the fan blade to see what would happen, and it didn't seem to make much difference at all.

Did you flip the fan blade over? How do you know if it made a difference or not? Did you measure temperatures before and after? Without temperature measurements, this is meaningless.

You are looking for a single problem, with a single solution; A single magic bullet cure. Sorry, it doesn't work that way.

You have multiple problems, and the only way to solve them is to take a full set of temperature measurements and solve problem #1, then take a full set of temperature measurements and solve problem #2, etc., etc., etc., following a logical sequence.

Now you are going to start pinching tubes and make the system worse not better. You can then add pinched tubes to the list of problems.

Multiple sequential problem solving is what separates the techs from the wannabe's, and what it takes to follow the wannabe's around fixing their screwups.

Peter_1
15-01-2005, 07:07 PM
http://www.climaticcontrol.com/info/jjzgate/Infotec/Info-Tecs1-10/Infotec_7(rev1).htm

http://www.outokumpu.com/pages/Page____18588.aspx

Gary
16-01-2005, 04:30 PM
If the TEV and distributor dP exists with at least more than approximately 4K subcool then the problem indeed lies with the distributor and so something mechanically associated with the distributor would have to be changed to adjust the ballance of distribution. Right now, poor distributor performance is the only logical explanation for the uneven distribution. Crimping distributor tubes is logically a potentially pragmatic solution though not necessarily practical.

We don't know what the subcooling is, nor the delta-P or anything else, because we don't have the proper temperature measurements. If the refrigerant is completely boiled off before it reaches the coil outlet tubes, then all of those tubes will be at the same temperature, making it near impossible to adjust flow by crimping tubes.

Clearly the first step is to reduce load by reducing airflow. And we need full sets of temperature measurements to see what the system is doing each step of the way.

busman
17-01-2005, 03:56 AM
Well, the cobbler's kids have no shoes. Preach to everyone about maintaining their own systems, and yet I go three months without cleaning my own air filters.

Some success has just been scored this week. I just got a Fluke16 meter for buying some guys some lunch. Not too bad, $14 dollars later walkin' out with a $140 meter. Time has been crunched here lately with some other more important projects. I will hopefully be able to sneak outside sometime here soon.

Looking back here, you keep saying turn over the fan blade. How in de heck is this going to make any difference? I don't follow your train of thought. You would only be changing air flow. Air flow may change, but how will that effect uneven refrigerant feeds to the circuits? The TXV may open and close, but already on previous occaisions, with different refrigerant charges, and superheat settings on the TXV, and same results.

I will experiment and give a full suite of numbers when time allows . . . more sooner than later I hope. :cool:

Gary
17-01-2005, 10:35 PM
Well, the cobbler's kids have no shoes. Preach to everyone about maintaining their own systems, and yet I go three months without cleaning my own air filters.

Some success has just been scored this week. I just got a Fluke16 meter for buying some guys some lunch. Not too bad, $14 dollars later walkin' out with a $140 meter. Time has been crunched here lately with some other more important projects. I will hopefully be able to sneak outside sometime here soon.

Looking back here, you keep saying turn over the fan blade. How in de heck is this going to make any difference? I don't follow your train of thought. You would only be changing air flow. Air flow may change, but how will that effect uneven refrigerant feeds to the circuits? The TXV may open and close, but already on previous occaisions, with different refrigerant charges, and superheat settings on the TXV, and same results.

I will experiment and give a full suite of numbers when time allows . . . more sooner than later I hope. :cool:

You should never use the word "only" when referring to airflow. Proper airflow through both coils is critical on all systems.

In this particular case, you have a 3 ton TXV trying to control flow through a 10 ton coil. You have derated the coil by going to a smaller fan motor running at a lower speed, but it isn't enough.

Until the airflow is reduced to derate the coil enough to allow the TXV to modulate, the TXV cannot be adjusted properly.

When the TXV can be adjusted down to about 7F superheat, then you will find that you have WAY too much refrigerant in the system.

When the refrigerant charge is adjusted (to the proper subcooling) then you will find that the indoor coil is more efficient, as well as the compressor, the operating costs will be reduced and the whole system will have a longer lifespan.

And oh yeah... you also have a distribution problem.

busman
19-01-2005, 03:03 PM
Really quick here guys, I was able to take a quick reading the other morning. I didn't write down all the numbers, but a quick summary for ya: Suction line temp at just before the accumulator was 2F colder than ambient (23F) I took a reading off of one of the non-frosting lines, and I had similar reading. Frosting line was reading 19F at the outlet of the coil. I think we can call that a distribution problem. :)