Subcooling and leak repair

[johnbinsc]

Hi guys,

I think one of my Trane XL1400 2 ton split system heat pump units (installed 1999) is low on freon (R22), thus has a leak, probably slow.  This unit has two TXVs, one on the evaporator, one on the condensor.

The preferred method of charging is by measuring the subcooling in cooling mode. 

Question 1:  Are there any tricks to getting a good temperature measurement on the liquid line?  Can I get away with a standard instant read dial thermometer?  I was thinking about taping the thermometer shaft to the line and covering with some pipe insulation, but was wondering if there was some kind of good thermally conductive putty that I could use to enhance the thermal contact between the two.  Or is all that overkill?

Of course, now that it is November, finding a day above 65 deg F to do the subcooling measurement in cooling mode is not so likely.  I could also measure the subcooling in heating mode, but I don't have a chart for that. 

Question 2: Is it practical to set the charge on the system by using the subcooling in heating mode?

Question 2a: If so, do I measure the temperature at the liquid line exiting the coil in the air handler, or can I get away with measuring it as it enters the outside unit?

Question 2b: If so, are there charts for this?  If not, what value of subcooling is the target in heating mode? (cooling mode subcooling ranges from 6 to 11 deg F).

I will try to find the leak using a sniffer and/or UV (the system already has dye in it from when the first evaporator went bad in 2003).  If the leak is again in the evaporator I would rather try to use a leak sealer rather than replace the evaporator as the system is no longer under warranty.

Question 3:  Does anyone have experience with SureSeal and/or EasySeal?  Do they work?  Are there any negative consequences?  Is there any reason, besides the expense, not to add can of leak sealer to the system?

I know about needing to make sure there is no moisture in the system with EasySeal, but the system is still under pressure so I expect it is dry. 

And for the record, I just bought the kit to get my EPA 608 certification so I can do this in an environmentally sound and legal fashion.

Thanks in advance!

[AmTec Services]

There should be a heat cycle charging chart for that Trane. You will need a good thermo to charge correctly. An indoor outdoor thermo with a remote sensing bulb or thermister is ok in a pinch, but if your gona do it often you might as well get the right tools.

I dont like sealers, but thats JMHO.

..................AmTec

[johnbinsc]

Thanks for the response. Would I get the heat cycle charging chart directly from Trane, or is there another source? The service sheet has some heat cycle performance charts but I don't think that's what you are talking about. I assume the heat cycle charging chart is somewhat like the cooling cycle charging chart with bracketed subcooling target lines plotted versus a liquid temperature and a pressure.

I just bought a used Fluke 52 Series II digital thermometer on ebay to do the temperature measurements.

There is a long recent discussion of Cliplight Super Seal HVACR on hvac-talk.com with no clear conclusion. I guess I should find the leak first, see if it is repairable without replacing the evaporator, and if not, see how big it really is before making any decisions.

[ACtechGUY]

As far as taking a liquid line temp goes , just tape your temp sensor to the liquid line near the condenser and insulate it for the most accurate reading.

As for Charging in heat mode..... There is no way to get a totally correct charge in heating mode. All you can get is " it will heat" charge. Heat pumps can only be accuratly charged in the cooling mode. If you have the service facts for that unit, then you have the charging instructions for that unit in the heating mode too.

About sealers... I have a test going on a 3 ton watersource heatpump for a server room with a tiny leak in the middle of Evaporator coil. I am hopeful that it will stop the leak, but I don't know yet.

[AmTec Services]

AC tech guy;

What type of metering device is in that unit?

[ACtechGUY]

It Is a ClimateMaster with a TXV. Why dooos u bee a askin'???........:headscratch:

[AmTec Services]

Interested in how leak stop acts with txv's and filter screens in front of them. Let me know how it works out. I've not used it in a r 22 system. Only is automotive, with disapointing results. But, I learned long ago I dont know everything and am willing to add to my arsenal anytime!

....................AmTec;)

[ACtechGUY]

I can tell you , the stuff I used was a very thin liquid that when exposed to just air did not change in at all , it just stayed a liquid. All I can see is there is a reaction when this stuff combines with refrigerant and is then exposed to air.

[johnbinsc]

Hi guys, took a while but I am back on trying to fix this unit.

I am posting as a reply to this thread since the details on the system are already part of the thread.

I now have my EPA 608 universal certification, and have some R22 to recharge this unit which is low at the very least, and thus probably leaking, but a slow leak.

I started adding charge to it today in heating mode, based upon the expected head and suction pressures for the given outdoor temp from the service facts. Yes it seems to be pretty low. After about an hour, the compressor cut out and would not restart. A check of the resistances shows that the internal overload protector (IOL) is tripped.

So my question: Just how long should it typically take on a 2 ton Trane XL1400 for the IOL to reset?

I got tired of waiting so I took off the top and removed the thermal insulation jacket from the compressor, which was still quite warm several hours later.

Thanks in advance.

[ACtechGUY]

A overheated compressor can take several hours to reset. We usually run water over them for about 10 to 20 minutes to get them to reset.

It is obvious that you have overcharged the system, causing the compressor to overheat. Perhaps you have another problem that caused the system pressures not to match up with the charging chart. Make sure that the evaporator coil and condenser coils are clean.

A dirty evaporator coil in heat mode will result in a higher than expected head pressure, and a dirty condenser coil in heat mode will cause a lower than expected suction pressure.

I have found it very rare that system pressures match the chart in heating mode for some reason.

[cvx]

Hi John

Regarding your charging methods, I'm not sure I'm following you, but you say the compressor tripped an hour later - did you have the bottle hooked up and charging the whole time? Was this bottle standing upright (charging gas) or turned upside down (charging liquid)? An hour seems pretty long, especially if it was upside down and you were charging liquid. Then you definitely put too much in. Out here in California, I did HVAC for a refinery for about 6 years. Mostly heat pumps and computer room (liebert) units, a couple of chillers. When we added freon, we'd almost always charge liquid and often use hot water (poured over the bottle) to increase the charge rate. You charge to the outside coil and wait a bit before starting the unit. (You are charging liquid straight into the system, so you don't want to charge to the compressor cause compressors don't like liquid; liquid can't be compressed and damages the compressor valving) We'd also use a charging scale to accurately see how much freon we were putting in.

I'm guessing you haven't heard of the hot water trick, and thinking no body mentioned a charging scale... don't know where you're going to go from here, but I'd recommend a scale in all future ops. You can buy one ($$$) or just get creative with something cheaper, like a decent digital fish scale. Tie some line to make a place to hook and weigh the bottle around the handle or the holes in the base ring of the bottle, depending on how you're going to charge from the bottle. Either way, once hooked up, with lines etc.. and ready too go, BUT BEFORE YOU BEGIN CHARGING take a reference weight. Charge a few minutes, stop and weigh. Doesn't matter if all the crap is hooked up when you weigh, it is every time, the only thing changing is the freon going out the bottle and into the unit- that's the weight change. If you're doing liquid, stop often to weigh!! (like every 60 seconds, til you get a feel for how much is going in) You won't believe how fast it goes in, especially with hot water! Better a few extra stops and a good charge than not enough and an overcharge.

In this way you'll have a reasonably good idea of the amount you added. Somewhere in your data / name plate there should be some sort of capacity mentioned. I have been out of the AC stuff for awhile, but the guys here can probably chime in with more info on system capacity. One thing to remember (as i recall) is that manufacturer's freon system capacities are for equipment only, ie.. they do not take into account the length of your line set (the copper lines carrying freon to and from the building). I also seem to recall some companies as also having charges listed for standard 25 or 30 foot line sets,but i could be wrong.

I'm not sure of your compressor type, but if you've charged enough to trip the OL you may have damaged the internal valving in the compressor, something to think about. If so, you could bleed off, start operating again only to find a different set of circumstances because the compressor is now damaged. Hope not. Just thought I'd mention it.

Good Luck, hope it goes well!

cvx

[johnbinsc]

Hi ACTechGuy and CVX, thanks so much for responding:

I've added less than 2 pounds so far, and the nominal charge is on the order of 15 pounds. I am charging with vapor to be conservative. (When you DIY you can take your time.) So I doubt I have grossly overcharged the unit...yet. I think the unit shut down because it is undercharged even with the the small amount of additional freon I got in there. The symptoms that got me started on this were that it stopped heating. By the time I checked it in the past the compressor had already reset apparently.

BTW, the IOL did finally reset, so I haven't burned out the compressor...yet.

The evidence for undercharge is (all in heating mode):

outside temp 47F, head pressure 130 psig, suction pressure 10 psig

while the service facts sheet says that for indoor air temp 70F (close enough) and outside temp 47F,

the head pressure should be 200 psig, and the suction pressure 50 psig.

I am using the glass digital bathroom scale I bought at walmart a few years back as my refrigerant scale. Weighs to 0.1 pound precision, which is probably good enough for my purposes. The 30 pound cylinder started out new at 35.8 pounds. Down a couple of pounds from there so far.

As for speeding up the charging process, this afternoon I got out my old electric heating pad (the kind for sore backs etc) and strapped it to the tank with an old bicycle innner tube. This should allow me to gently heat the tank to speed the vapor charging process.

One additional thing I will check out tomorrow is whether the run capacitor, start capacitor, and start relay are working properly. It is my understanding that if the run capacitor was bad then that could cause the compressor windings to overheat.

My testing plan is to simply measure the current to all the different components. If the relay is working properly, there should not be any current to the start capacitor once the unit has started, right? And if the unit starts without any difficulty, the start capacitor and relay are probably working properly at startup also. The biq question is how do I properly test the run cap? I would expect that simply measuring the current on the different compressor leads should tell me something, but I don't know how big the run cap current should be compared to the main winding. Are there any rules of thumb for this?

Thanks again for your help,

John

[cvx]

Hi John!

Saw your post.

Couple of quick thoughts on your post

– remember, the IOL is a thermal device. It could reset independent of a winding issue. The only real way will be to try and start the compressor and run it for a short bit. As an electrician, I’ll tell you it’s never a good idea to short cycle (on/off) a motor for testing. Give it 30 seconds or more before shutting it of.

-- with only 2 additional pounds charged to a system that was underperforming, I don’t think you overcharged it.

-- with regards to testing the unit as a whole now that the OL is reset, first off, I’d just try and start it. A 30 second run will not be detrimental to anything. What it will do is verify the the basic condition of the motor. If the motor is toast, ie, non-operable, might as well know it asap instead of spending a bunch of time testing stuff.

-- as for testing caps, you can test the cap itself using a multi meter. Most Flukes, even the low end ones, have cap test features. I googled “ testing for bad run capacitor” and here are a few sites with info off that search. The top one looks real good, the others are kinda various forms of the same info – I just threw a few up. Same info is basically applicable to start caps, you can google it too, I didn’t.

http://www.hvac-for-beginners.com/capacitor-testing.html

http://www.arnoldservice.com/run_capacitors.htm

http://www.ehow.com/how_7378788_test-run-capacitors-heat-pump.html

http://toad.net/~jsmeenen/capacitor.html

The best way to test the start relay is to look for voltage across the relay leads:

SHUT: When the relay contacts are shut, there is no difference in potential (electrician mumbo – jumbo) and your measured voltage will be zero between the leads - EVEN THOUGH ELECTRICITY IS FLOWING THRU THE CONTACT. You can verify that voltage is present by taking either of your test leads to ground on the unit. You should see voltage to ground when you do this to either of the leads .

OPEN: When the relay contacts are open, you should be able to measure a voltage across the leads of the relay. (and yes, more jumbo, there is now a difference in potential). Secondary method of to verify this is that you’ll be able to see voltage to ground on one relay lead, but not the other. This is because the contacts are open and voltage is only on one side of the relay – the contact hasn’t closed to pass it thru to the other side.

However, what if you approached it from the standpoint it was decently charged and the low performance symptoms were caused by something other than charge? In that case you’d have added 2 pounds extra to a charged system, not enough, I think, to dump the compressor. Could your problem be a bad tvx on the heat side? It’s late an I gotta run, but explore that path too. The need for clean coils mentioned earlier by one of the other fellows is also an excellent point. Coils are major players and need to be in good health (clean, able to pass heat (ie .. h ave fin material )

K gotta go

No spell check, hope it’s ok!

cvx

[ACtechGUY]

First... you did check that the coils were clean right?? Good.....

Here is a simple test to verify low charge.... run the system in cooling mode ( provided inside temps are at least 70 degrees) .

Find something that you can compleatly cover the condenser coil( with like a plastic painters tarp.)

Put your high side gauge hose on the liquid line and and suction gauge hose on the suction service port , NOT the TRUE suction port.

Let the unit run for at least 5 minutes to make sure pressures have stabilized and gotten as high as they can be .

WHat are your pressures?? Does the head pressure still look low even with the condenser coil airflow totally blocked?? Does the suction pressure still look low stupid low??

If they do , then you are low on freon.....

NOTES::

1.With a blocked condenser coil at 50 degrees OUTSIDE YOU MIGHT GET AS HIGH AS 200# LIQUID PRESSURE.

2. If you have anywhere near 200# on liquid pressure then you should see at least 60 to 65 # on the suction side.

Next step.... Acceptance......

So I am low on freon..... Maybe we should find the leak ......

The bad news is, most modern fancoils tend to develop unrepairable leaks around age 10. It just sucks.....

I can go into how to pinpoint a leak without a 300 dollar leak detector ....

[johnbinsc]

Hi cvx and ACtechGUY,

Lots to respond to in your posts...but first an update.

I added more freon for a total of 4 pounds or so added and the unit has been running for a day and a half since without setting off the IOL. So it is happier with more freon. The system performance in heating mode as measured on the manifold gauges is not yet up to the specs given in the service facts sheet, but I haven't added yet more freon yet because: 1) it is damn cold outside right now, and 2) it has been pretty slow going adding freon. Regarding 2), I will probably have to resort to using hot water to raise the pressure in my r22 supply cylinder as you suggested.

This unit has dye in it from when it previously had a leak in the evaporator 5 years ago, so while I was checking the cleanliness of the evaporator coil (and the spaceguard filter) I shined my cheapo UV led flashlight on the coil and there is a definite yellow patch right in the middle of the top half of the coil (horizontally mounted air handler). A check of the downstream side showed no concentrated spot, so it appears the leak in the tube through the fins. There are some other less prominent spots of yellow here and there, but not sure if they are additional leaks or are just spatter from the main leak. There is also some dye visible in the condensate drain.

I looked up the replacement cost of the COL6784 evaporator coil: $877 at Patriot Supply

I'm not sure if the leak I described above qualifies a repairable or not. But it probably is not easily repairable (i.e. by soldering).

So that brings me back to my original question about the Nu-Calgon EasySeal sealant. Is this unit an appropriate candidate for a try at fixing the leaky evaporator with EasySeal?

I guess one question I need to answer is what the actual leak rate is roughly, and if it is small enough to be sealed with EasySeal.

As for testing the run capacitor, I don't have much faith in the validity of the very low voltage capacitance testing done by that function on some multimeters. And neither of my meters offer that feature. But using my clamp ac ammeter, I can see that the start capacitor is flowing no current when the system is running (meaning that the start relay is doing part of its job), and that the run capacitor has about the same current flowing as the run and common winding of the compressor motor, so it looks like the run capacitor is working OK also. And the compressor starts right up, so the start capacitor and start relay are doing the other part of their job as well. So I think the associated electrical parts are all OK.

I have two identical units side by side, so next I will put the gauges on the second one (that has never given me a bit of trouble) and see how that compares to the performance charts.

Thanks for your continued help.

John

[ACtechGUY]

I would agree that the electrical parts are all fine. The compressor overheating was strictly due to low freon level , and not failing capacitors or anything else.

About putting in the leak seal..... It really can't hurt to try. I Have been asking around the supply houses and they hear from their customers that the leak seal works sometimes even on really bad leaks. What you have is NOT a really bad leak.

If it were you would lose the whole charge in a couple of days or hours(THATS A BAD LEAK!!)

If your leak is in the middle of the tube sheet, Unfortunately Your leak is not repairable by conventional means. That leaves 2 options..... replace or try the magic gunk.

If you were to take the front panel off the air handler and then take off the drip shield under the door , you could see the front tubing bends. You will likely find that where some of the tube bends go thru the rusted sheet metal, there will be tiny leaks. probably more than one. .That is how many Trane coils end their lives. These leaks are not fixable due to their locations.

[johnbinsc]

Hi guys,

It's been a month and I wanted to follow up. As of my last post Dec 8, I had added 4 pounds of R22 and the system had been running again in heating mode for a couple of days. Unfortunately, on 12/11 the compressor IOL had tripped again, so that gives some idea of the size of the leak, i.e. 4 pounds in 5 days in heating mode. I figured it was time to give the leak sealer approach a shot.

It took three days for the damn IOL to reset. (I was too lazy to disassemble the unit and take the thermal jacket off the compressor. I am also wondering if the sump heater might be slowing down the reset process.)

On 12/15, I added yet another 4 pounds of R22, ran the system for 30 minutes, did a QwikCheck acid test which came up looking fine (i.e. no indication of any acid). Then I added a can of Nu-Calgon EasyDry, just to be on the safe side and be sure the system is dry before I add the sealer. Ran the system another 30 minutes. Then I added a can of the Nu-Calgon EasySeal. Ran system for 1 hour.

Since this sealer supposedly needs air and moisture to work, and it was a very dry winter day, I did some things to increase the humidity in the air handler, i.e. I sprayed a bunch of water mist into the air in front of one of the returns, took a hot shower, placed a tray of water in the return plenum right before the air filter, etc. Don't know if this made a difference or not.

On 12/17, system was still working fine, so I put the gauges back on it for some performance measurements. Based upon the heating mode performance charts and the outdoor and indoor temperatures, the head was low by 40 psi and the suction was low by 18 psi. The liquid line temperature indicated there was no subcooling of the refrigerant returning from the indoor coil (remember we are in heating mode).

Adding about 3.2 pounds more R22 brought the head pressure to 185 psig, 10 psi of its 195 psig target, and the suction pressure to 59 psig, 7 psi above its 52 psig target. Liquid temperature was 92F, so the subcooling was now a respectable 4F.

Seemed close enough for the moment, so I buttoned the system up and then went on vacation for two weeks!

As of today 1/12/2011, the system is still working fine. The defrost cycles seem to be working OK, i.e. it defrosts and then gets on with heating. I get warm air from the registers when the OD temp is above 40F or so. I get not so warm air when the OD temp is in the 20s. It has been cold and snowy here so I have not tried to put gauges on the unit, and probably will not until spring time if it keeps working ok. In spring I will put it back in cooling mode on a warm day and check the charge with the subcooling performance chart.

As for the Nu-Calgon EasySeal, it seems to be working well so far after 3.5 weeks. I will update this thread periodically with status reports.

My reasons for going with the Nu-Calgon EasySeal versus the competing product were:

1) It is my impression that the Nu-Calgon EasySeal is a liquid or gas until it reacts with water and air, whereas the competing product is a solid in the can that is dissolved by refrigerant. Probably this makes no difference, but I prefer the concept of a liquid sealant.

2) The Nu-Calgon EasySeal supposedly requires both water and air to seal, rather than just water. Hopefully this makes it less likely to cause problems gumming up the compressor etc., but again, this may not make any difference.

On the down side for the Nu-Calgon products: they are both (EasySeal and EasyDry) charged with hydrocarbon propellant, so my system now has 6 oz of hydrocarbon refrigerant mixed in with the R22.