slow evacuation below 1000 microns in a system that exhibits very little rebound

archibald tuttle

New oil in a new pump and did a quick test with micron gauge that the pump pulled down below 100 microns almost instantaneously when attached to the gauge alone.

So after pressure testing this 2 ton single head panasonic system, started evacuation using the robinaire 15600 with gas ballast. So I ballast until we're in the 10's of thousands of microns. I wans't timing, but my recollection is that took a minute or two, and then I closed the gas ballast and with 10 minutes its below 1000 microns. But then it crawls form there for two hours down to 720 microns.

I didn't stop to ck. rebound which probably would have been the smart move but I kept thinking it would get where it was going and that it wouldn't just keep going down, even if slowly if it was overburdened with ice (it was cold. The system had been recovered not much left,and immediately capped while we took advantage of the need for repiping to move it to an aethetically and operationally superior location but still with line lengths of like 15 feet. it had stopped working about 9 months after installation but the tenant never mentioned the problem for a year after that until the cost of back up fuel oil became so noticeable) or leaking.

So two days later it has rebounded to only 780 microns.

That just doesn't seem like a system that is heavy with frozen moisture. Although I'm waiting for another reading today as it is even warmer and didn't drop below freezing last night, just to confirm. But under the current assumptions and to get the discussion started because I was hoping to charge this thing after work tonight (another moonlighting affair), I'm trying to figure out why I could be having so much trouble getting down toward the nameplate and recently observed micron rating of the pump.

I'm using a single 6' 3/8 appion megaflow to the pump with cores removed.

thanks, brian

Ironman

It can be very slow to get all of the refrigerant to boil off when it’s condensed in the compressor’s oil sump or in the accumulator. Heating the surface will greatly speed up the process.

If the vacuum is holding but not decreasing, that’s an indication of moisture still present in the system.

As I’m sure you know, the colder it is, the slower the evacuation process. Again, warming the compressor crankcase will greatly speed it up.

archibald tuttle

@ironman

i was warming the evaporator with a heat gun but I couldn't get to the compressor or accumulator without pulling the cover but that would have helped it seems.

the other thing I can't figure out is why there isn't more problem with extracting the oil when evacuating. I guess it is not particularly volatile so that is less of an issue and i certainly wasn't observing changes to the level of oil in the pump and everywhere i've looked that is not cited as a problem but it always makes me nervous because it is so difficult to gauge oil and oil transit in these closed systems (other than once the compressor blows . . . ).

pedmec

Your not extracting oil while evacuating. Your removing moisture from the system. the colder it is outside the thicker the oil in the compressor will be. this will make it harder for the refrigerant trapped in the oil to escape. That's why we recommend the crankcase heater to be on.

EBEBRATT-Ed

Look up the vapor pressure of water. When the temp gets low it is virtually impossible to boil off the moisture when it is cold out. You get down below 50 and it just won't move no matter how long you evacuate. Charge it and put a dryer in

archibald tuttle

@EBEBRATT-Ed

It's funny, i've seen for and against on adding a dryer to minisplits. What line and flow direction is indicated? I could get a female by male flare unit similar in form factor to the filter for an evac unit albeit capacity (?) if it can be attached at one of the valves at the 'condensor'/compressor unit. That's only adding one more flare in an easy to test and service location. Although I've been waiting for it to warm up and it's gone up 25 degrees. It's going to be 50 this afternoon; so I was going to try to pull the system down again and see if I can make it notably below 500 microns without significant rebound.

@pedmec fully aware the intention is not to extract oil. But, as @Ironman mentioned, there may still be refrigerant in the oil to go. don't know if 50 degrees still favors heating crankcase? hadn't though about the onboard crankcase heater and i gotta look at the manual as to whether that can be triggered independent of compressor operation with some button or combination of buttons on the remote, or i can pull the cover and use a heat gun at the base of the compressor if that is still indicated at 50 degrees.

it just struck me that crankcase heater or other external heatsource could approach a volatility or tendency for oil transport, esp. if the a micron vaccum is over the oil . . . but most of the oil loss issues i've seen are put down to oil circulating with refrigerant during operation and being trapped out in the system. But, if it is able to be pumped with refrigerant, are there conditions, e.g. too warm a crankcase, that could cause evacuation to take some of the oil that I should be cautious about?

this has actually always been a bugbear in my mind-albeit i mostly do what i'm told by folks who have done it far more often than I have and thus better understand the range of operation conditions and symptoms encountered. I appreciate, and try to follow, the quality advice I get here in this arena which is newer to me. never had this kind of trouble with R22, maybe just luck and/or-come to think of it-was always servicing air con so probably never pumping down even in shoulder season cold.

thanks,

brian

GGross

Since I have seen line driers mentioned here I would let you know that many mini splits do not allow in-line components to be added, including filters/driers. Specifically I know that all the LG units I sell are like this, to the best of my knowledge samsung, mitsu, and fuji also follow this rule. It would stand to reason that Panasonic is the same way and you should consult the manual for the unit to double check before installing a drier.

Edit: for reference this is what LG says, I am curious what your Panasonic unit says about this

"Components such as oil traps, solenoid valves, filter-driers, sight glasses, tee fittings, and other after-market accessories are not permitted on the refrigerant piping system between the outdoor units and the indoor units. LG Single Zone systems are provided with redundant systems that make sure oil is properly returned to the compressor. Sight-glasses and solenoid valves will cause vapor to form in the
liquid stream. Over time, driers will deteriorate and introduce debris into the system. The designer and installer must verify the refrigerant piping system is free of traps, sagging pipes, sight glasses, filter driers, etc."

pedmec

Oil will not leave your compressor due to a vacuum as the vacuum is the same thru out the system. There really ain't a high to low point in the system to have a pressure differential blow it out. I have read that under 200 microns the oil has a potential to vaporize. That would be the only way a non functioning system could lose oil with no refrigerant in the system to have migration. And even then there are people in my company who dispute that the oil will vaporize. I have had this exact conversation with other techs in my company. But we evacuate to 500 microns and then do our decay test.

As far as the suction line filter drier. I have installed it in a mini split years ago and didn't have a problem but this was before i knew that they didn't want anything in the piping. It does void the warranty. Be careful that if its a heat pump that your using a bi-directional filter drier.

archibald tuttle

Final question a for my retrial. Break the exiating 860 micron for 3 days vaccum with nitrogen and re evac or just start from where im at with outdoir temps of 50? Should i run crankcase heater or heat gun on the crankcase at that temp?

pecmsg

Pressurized with N2 for leak checking. 

Evacuate and break with the refrigerant used to 5PSIG

Evacuate again and break with refrigerant used to 5PSIG
Evacuate and recharge

archibald tuttle

@pecmsg

so did the N2 part of that cycle before previous attempt at evac where we got to 720 micron. rebounded to 860 after 3 days. for info sake i restarted pump and almost instantly it jumped to 1150 microns and then up to 1300 and remains unchanged for 15 minutes.

could restarting the pump have helped finally pull refrigerant out of oil? i'm kind of surprised at the extent of sudden rebound and that i'm not going down at all after 15 minutes but still in fairly deep vacuum.

of course if there is refrigerant in the oil i'm kind of hesitant to break the vacuum with refrigerant but i guess if I go right back to evac i'm not running the risk at 5 psig that a lot more will dissolve in the sump?

and if i am going to break with refrigerant does it matter if I charge gas since i'm going to evac again, so getting a slightly off azeotrope ratio isn't too problematic, right?

mattmia2

I assume the problem comes with the mix still left in the tank.

warmer is always better. You could probably put a space heater next to the outdoor unit and direct it inside and warm things up.

archibald tuttle

mystery continues. i warmed up the compressor base and got very little difference in results. i pulled the micron meter and checked the pump alone again and it right into vacuum down to 100 and falling. i was getting some anamolous results with the micron meter showing vacuum when i new there was none, almost as if there were a check valve in the 1/4" hose it was on. but it responded as expected to direct connection to the vacuum pump.

very weird. broke the vacuum with R410A to 5 psi per @pedmec and ran again and still wouldn't go below 900, rebounded to 1250. almost as if that were the vapor pressure of leftover refrigerant in oil that i can't seem to get out but that is just a guess.

taking the micron meter on and off for tests was a pia because I didn't want any air, i.e. including humidity, sucked into the schrader while backing off but i didn't want refrigerant going to micrometer, so i broke to just very slightly positive pressure and pulled it off. . it's on a schrader tee off the C&D core removal tool, i don't know why they don't put a ball valve on the tee on those tools as well. Already got two stacked so i had a ball valve to cut off the vacuum pump and nitrogen/refrigerant with the second one teed off the schrader tee in the first with core removed but i would need yet another core removal tool if i wanted no core and or at least an isolation valve on the schrader tee from the second core removal tool where the micron meter is attached.

i've read some people are stack three core removers for this setup. which some of this could be obviated if the tees on the removal tool had a ball valve as well although that thought gets me to my principal pet pet peeve. Al this stacking core removers is silly. While I think I can get them all tight it is remarkably fidgety, you gotta keep track of all these cores you're pull out. and the attachement is usually right next to the compressor unit and you gotta work out the angles so you can swing the valves and attach to the tees. For crying out loud, why don't the manufactures just put a vacuum rated valve instead of a schrader and if they are worried about stupid customers letting refrigerant out, they could take off the handle and put some kind of obscure wrench driver for the valve.

of course valves are yet another possible source of leaks but so are schrader cores . . . I'm tempted to jsut pull the cores and put a female by male flare valve on there but use a copper washer rather than o-ring style so don't have to worry about degradation of the seal, albeit why wouldn't that be equal concern with zoom lock or other press refrigerant fittings. or have they figured out some non 'rubber' approach to press seal?

nuff said. back to the breech.

mattmia2

Refrigerant system valves have a cap that goes over the steam, they are sealed with a cap rather than stem packing. They are also much more expensive than a schrader. You could stack what you need in npt or flare fittings off of a single port on the core removal tool.

archibald tuttle

@mattmia2 right, those valves are what is used for the release of refrigerant in a charged compressor/condensor/outdoor unit. And they aren't easily accessible for mistaken use. If those are the long term reliable solution, if they are worth $10 more than a schrader, given the service needs of these units seems well worth it to me to add another set even if the unit is going to be $20 more. And I assume they are high vacuum compatible although they wouldn't be vacuum challenged in normal commissioning since they wouldn't be opened until after the evac and would have pressure behind them to break the vacuum essentially instantaneously on opening.

At least on the short term side, there are plenty of pressure and vacuum capable ball valves that could be employed, albeit i'm not sure what the relative service life/likely degradation of packing is. In the nfn department, I feel like the schraders that sometimes develop leaks have components subject to failure with age so they are capped as both foreign matter prevention but also to provide a secondary seal using brass cap with silicone/rubber washer seal which is itself a long term failure item (and distinct from system valves whose caps are metal to metal.

Perhaps, if undisturbed, these flare caps for schraders are less likely to fail despite rubber seals, and the idea is to keep a good supply of new caps/seals on hand so if you are making the service visit 5 years down the pike you just replace it as matter of course.

ratio

The cap on the Schrader valve is the primary seal, hence the complaints about plastic caps. The core is just in there to give you time to get your hoses on before all the charge leaks out.

archibald tuttle

@ratio couldn't agree more re plastic caps but that's what came on the input tee for my vacuum pump (robinaire) and I kinda wondered about it but it I guess those caps can resist nature's abhorrence of vacuum.

still wondering about the longervity of 'rubber' seals which seem ubiquitous in that application.

mattmia2

You can replace them with flare caps that make a metal to metal seal, for the system at least. For the manifold set, brass caps with rubber or teflon seals are probably more convenient.

pecmsg

ratio said:

The cap on the Schrader valve is the primary seal, hence the complaints about plastic caps. The core is just in there to give you time to get your hoses on before all the charge leaks out.

Actually the caps are Dust Caps. Just to keep crap out!

GW

ratio said:

The cap on the Schrader valve is the primary seal, hence the complaints about plastic caps. The core is just in there to give you time to get your hoses on before all the charge leaks out.

I think plastic is cheese too, but normally there's small o rings in there, pretty solid setup, in my opinion

GW

I'd be tickled to see your vac set up. It pains me to see some guys still using old-school gauges.

If you're pulling on the condenser, why not use two hoses??

archibald tuttle

@GW

I agree. promptly replaced use of manifold in this circumstance with just a single ritchie/yellow jacket 40345. Manifold was all i had around to meter small interim refrigerant charge mentioned by @pecmsg.

I've definitely seen guys set up with two hoses to the pump. But I have so little trouble pulling down to 1000 microns or so and so much trouble going beyond that that my instinct is that the hose capacity wasn't the choke point. but maybe i'm not thinking correctly about that given that it still all has to pull through a 5/16 to 1/4" core remover. While I understand that is less length of constriction my instinct cut against adding another evac hose as solving my dilemma to get finishing vacuum.

GW

Are you connecting directly to a pump?

archibald tuttle

@GW absolutely, direct connected through C&D core remover. there was no manifold installed between pump and unit, I didn't have a second core remover on the tee on the first where I added a manifold to refrigerant tank for breaking the vacuum. between. but only one 3/8 appion megaflow line. not a double as you have there. My instinct was that not having two lines (although i've seen several folks recommend that set up) was my choke point because I make good progress to around 1000 and then it just stops going down (a little more or a little less which seems to correspond to envrionmental temperature at the moment with result it being slightly higher if warmer but range after 3 attempts between 720 and 1300 microns). Maybe me instinct is I have more refrigerant trapped in oil that I would have anticipated and as I pull some out more evaporates from oil . . . but I would feel like that can only go on so long, but it seems to go on 'forever' on this job.

Of course, maybe my instincts are just plain wrong. wouldn't be the first time.

brian

ChrisJ

pecmsg said:

ratio said:

The cap on the Schrader valve is the primary seal, hence the complaints about plastic caps. The core is just in there to give you time to get your hoses on before all the charge leaks out.

Actually the caps are Dust Caps. Just to keep crap out!

The caps are even listed as secondary seals. That's why unlike cheaper tire valve stem caps all of ours have O rings in them, or at least should. I throw the plastic caps out and use brass.

mattmia2

ChrisJ said:

pecmsg said:

ratio said:

The cap on the Schrader valve is the primary seal, hence the complaints about plastic caps. The core is just in there to give you time to get your hoses on before all the charge leaks out.

Actually the caps are Dust Caps. Just to keep crap out!

The caps are even listed as secondary seals. That's why unlike cheaper tire valve stem caps all of ours have O rings in them, or at least should. I throw the plastic caps out and use brass.

The better brass ones have a flare and make a metal to metal seal.

archibald tuttle

@mattmia2 @ChrisJ and of course the system valves on the compressor unit have metal to metal seal as well although its not any flare standard by angle or thread i'm aware of.

ChrisJ

archibald tuttle said:

@mattmia2 @ChrisJ and of course the system valves on the compressor unit have metal to metal seal as well although its not any flare standard by angle or thread i'm aware of.

They're just standard SAE 1/4" 45 deg flares aren't they?

When blanking off a system I never trust the valve on my pump or my manifold and especially not rubber hoses. They all seem to leak, no matter how new.

I use a Robinaire 15600 pump, same as you and YJ Brute II with liquid filled gauges and a YJ digital micron gauge. Anytime I've tried to include any hoses, or the pump in that mix it never holds I need to isolate at the unit it self.

archibald tuttle

@ChrisJ great minds think alike on equipment . . . The caps im talking about arent the schraders. These are over the hex driven valves that isolate the factory charge in the outdoor compressor unit. They r way bigger than 1/4" and don't seem to correspond to any threading/mating standard I'm aware of.

pecmsg

Some minis used 5/16” flair fittings for access!

ChrisJ

archibald tuttle said:

@ChrisJ great minds think alike on equipment . . . The caps im talking about arent the schraders. These are over the hex driven valves that isolate the factory charge in the outdoor compressor unit. They r way bigger than 1/4" and don't seem to correspond to any threading/mating standard I'm aware of.

Ah yes.
I don't think there's any flare. There's just a chamfer to make it pretty and get rid of a sharp edge

GW

Brian if you can’t get below 1000 you may have a leak. Not sure how long you sat watching. Normally the numbers continue to drop, how fast is a different topic. But a dead stop- seems like a leak 

archibald tuttle

@GW I thought that might be the case, a very small leak, although we valved off the pump and left it sitting for two days with no appreciable rebound. maybe 100 microns or 10% in two days! If that is a leak, it is damn small.

so I had to leave this job and on to the next of commissioning a new 1 ton minisplit so i was only evacuating lines and indoor unit. I got off to an anamolous start, when engaging the 'oil check' run just evacuating the micron gauge and line with the pump isolated from the system with the ball valve on the C&D core remover I was not seeing those 10s of microns as I had in earlier tests with this pump but in the 100s of microns. The oil was clear as a bell so now i'm starting to suspect the pump, but I opened the valve to the system and the sudden temporary rush of exhaust, since the lines (and indoor unit) were at atmospheric pressure having just been cut and flared and installed, gave a sudden expulsion of perhaps an ounce of oil.

I initially had no difficulty filling the oil reservoir to the fill line and, in my first several hour attempt to evacuate the stubborn entire system that is the subject of this post, I observed the line to be stable where it belonged. When I went back a couple days later for another attempt, the unit sitting level showed it was toward the low side of the site glass which struck me odd but a I added the remaining ounce from the 16 oz. bottle. I noticed later during the operation of that cycle that the oil must have been 'hiding' somewhere as the line showed above the fill almost to the top of the site glass.

Now the literature specifies that overfilling is a problem as much as underfilling although it says it will be evidenced by discharge of oil through the exhaust and since I didn't experienced any oil expulsion when the pump first stablized at an overfilled level I presued I was still within a proper operation envelope and I didn't finger that as a possible problem.

sure enough, with the oil level right where it belonged after the oil exhaust event, had no problem with the evacuation. Got just below 100 microns with rebound and stabilizing around 300 for a half hour. So I'm going to go back and see if I now experience better results on the my first failed effort with this pump.

I'm going to feel really stupid if this was the problem

GW

Brian no fear- I went from being a plumber, to taking on boilers for many years, then to HVAC. I’m not ashamed to ask the most stupid questions. The people that make mean remarks are just unhappy people 😀

I’m curious if by chance you might’ve had a leak in your hose/connections.

ChrisJ

IMO the only stupid question is after someone pipes a boiler completely wrong and then argues that it'll be ok.

That's stupid.

Asking before you make mistakes is never stupid.
Even asking and then fixing your mistakes isn't stupid, just more work.

GW

you got me there