19XR/XRV "coffee can" float...

[Randy S.]

Witnessed today, a well running at about 70 percent, properly charged, flows at specs, tubes clean, brand new 19XRV suddenly starve the evaporator. Liquid level in the condenser was at 9 o'clock while it was starving, 49 degree leaving chilled, 67 entering condenser water. One of two units in series on the evap side. The OEM guy had the coffee can apart last week, and found absolutely nothing wrong, even checking with the factory to be sure the right parts and pieces were there. No debri, rags, pistachio shells, or anything laying around inside. Good gas flow through the bubbler line. Everybody scratching their heads.
I wanted to try raising the tower fan setpoint and see what that would do to evap approach, thinking maybe the coffee can war running too cold for the amount of hot gas available. Warranty, you know, so I can't stick my fingers too far into it at this point.....

As the saying goes, "to be continued"...

Anybody ever seen anything like this?

[freonrick]

i have not seen that one yet. what was the pressures and speed when that happen? since everything was ok internally it makes you wonder if maybe the compressor quit pumping or surge, but that would not explain starved evap. did you check the liquid level in the evap? how is it running now? is this one down stream from the other one? what is the chill water setpoint and actual leaving water at the time?
I have some xrv that is running 80 degree tower water and the evap approach is running at 2 degrees at 85% load.

[Randy S.]

This one is the last in the series. It runs just fine as you ramp up and then when you get to around 90% it starves the evap just like you had shut off a valve. No surging going on.
Evap approach goes from just under 2 degrees to as much as 15 when the refr temp drops to 35 or less.

[carriermechanic]

Hey Randy,
I work for carrier in dallas. I have seen the float stick at different times on a xrv before. I called warranty after I figured out it was the float. They asked for my serial number. I gave it to them and they told me that that particular machine had a problem with the float. They said that there were several machines shipped out with a smaller coffee can. They shipped me the new can and it was actually about 2 to 3 inches taller. I installed the float and never had anymore problems. This machine would run fine and sometimes it would not before I changed out the can. I know that we have been seeing these floats a lot here lately. They have re-designed them so they are not as sloppy with the up and down motion.

[carriermechanic]

I do not think that raising your tower water is going to help in this situation. As you probably know the colder your cond. water is the better, you save more money with colder cond. water. If you can maintain about a 20-25 degree split between your cooler and condeser, then you will save more money. Say you have 45 degree leaving chilled water then you want about 65 to 70 degree cond. water comeing in. And freonrick can save a little money if he could get colder cond. water at 85%. 80 degree water at 85% is at its max. I understand it depends were you leave and how much tower you have. I am talking about the perfect scenarios. This is what we try to sell when we do a complete install of towers and xrv chillers.

Keep me posted on what they do to fix your chiller.

[freonrick]

A larger can would make sence, when his machine is unloading his freon level would rise in the evap adding more weight on top of the can. I would guess the max depth of the freon level in the evap would determine the size of the can so it can still float with the bubbler line feeding under neath it. I agree on the tower temp. When we get high humidity the a marginal tower thats the best we can get. The rest of the year is ok.

[referteacher]

I had a situation several years ago on a 23XL that uses the same style "can" expansion device where the "top hat braze joint to the cylinder slotted slide came completely apart. Of course then the slide doesn't come up at all and the machine pumps down and goes off on "low evap temp".

Also, as carriermechanic says, the way to get the most value out of a variable speed chiller is to provide as much "cooling tower relief" as you can get. As he says, 65 degr is NOT a typo. That is what the manufacturers specs are for part load conditions.

As a matter of fact, that holds true for fixed speed chillers also.

[chilbrig]

The early XR's had a problem with the Floats sticking, but for the most part that has been corrected. Although there still might be one once in a while. If the factory guy pulled the float out to look at it, he should have gone ahead and changed.

[carriermechanic]

I hate to say it but I agree with chilbrig.

[Randy S.]

The OEM guy was well aware of the "different" coffee cans that were shipped from the factory. He spent quite a bit of time verifying that we have the correct parts, and also made sure that there were no machining marks or any other rough spots to interfere with the action of the float. It seems to work for a little while and then just sink. That's the deal that has us all in a spin.

It ain't over yet, and I'll be sure to keep posting with any new info as things progress.

Thanks again, and y'all have a good weekend, and a fine Fourth...

[subcooledone]

I've talked to Carrier on this, they had floats sticking and changed the design. Still looked cheap to me. I first noticed suction would drop all of sudden. shut the unit down and start it up, everyting looked great. Change to new float and still had issues. It seem that bubbler tube needs to be longer,at some conditions the can gets logged (the gas around the can).
I have raised head to the keep float up, seems to work. We are taking about keeping a unit on, not kw/ton. soon as I can shut the unit down I will modify. I know techs at carrier who have changed floats and eveything is cool but thats on lager frames. The one I was working on is frame 3, whats yours?
note: two chillers,same units and ones ok.
charge and flows are normal. Sometimes I have a 1* approach and sometimes it's 10*. % load not a factor.
Warranty guy told me about the mod.

[Randy S.]

About 275 tons a piece. I'm away from work for a few days, so I can't tell you the exact size of the barrels. Itty Bitty is all I can come up with from home. Twenty something, I think.

When I left Monday, I suggested that we raise the tower setpoint. The machines are not tubed exactly the same, and the one that is complaining is intended to be the lower temp evap of the two. Still on warranty, so I only get to "suggest" right now.

[subcooledone]

Back in the day when I worked for the big "c" I thought it was strange how the service guys got feed back. We had two offices about 50 miles apart and no communcation. I was working on an oil problem with the factory (I was doing R&D for them) and came up with a solution. The guys up the road didn't find out until the s/b came out (6 months later). I added caps on the oil circuit. It slowed the transducer reaction time when the refrigerant would burp from the oil, without them you would get a very intermittent oil faults. Most s/b come from service techs.
Now they have issues with the CAN and techs are trying to figure it out, that sucks! that hush hush B.S.
They had some thurst failures that didn't add up, shoes looked good but a complete failure. (frame 3's).The old king has been around for a long time and I've never seen a failure were the shoes didn't whipe out.
It's like Trane and the vfd-bearing issue or the coolant pump seals failing. If it wasn't for sites like this and knowing guys at the factory shops we would be screwed.
so all you factory guys-thanks for posting the dirt, it helps.
Randy, let us know what they do.
Thanks.

[Randy S.]

Apparently, there were some changes made to the coffee can at one point without the approval of engineering, so a few machines left the factory with "different" cans. I believe it was in the dimensions of the skirt of the float device. Our dimensions are the same as what the factory says should be in the unit.

So that's in the mix along with everything else.

[Randy S.]

This is the lowest of two evaps in series, identical compressors. One pass evaporators.

It all came down to excessive evap flow, and low load charging. A tower reset based on average current percentage
helps, too, but the evap flow and charge were the culprits.
Ended up right at nameplate charge, once we got the flow right.

Discharge superheat ends up at 11 to 16 depending on load.
Approaches right at specs.

Third party commisioning, by the way, on the pumps. Never take anything for granted...

Time for a cold one...