Did it stop at 567 microns?

 

The contractor wanted to stop, inject some nitrogen and start the evacuation again.... so the "test" stopped at 567 microns.

 

The Mitsubishi install instructions are legitimate.

Sounds like the system is tight and dry but why not let the pump run longer, should easily be able to get a new system down too 200-300 microns or less and hold below 500 microns for an hour or more with evacuation best practices.

 

Thanks very much for your reply.

The black and white paper that comes with the wall unit has the "new" spec. which is...

1 - evacuate to 500 microns
2 - do a rise test
3 - the vacuum should hold 500 for one hour

The exact words are slightly different. I tried copy and paste but it didn't work. I can try something else later so you have the exact words.

The pump is a new Navac 8Di it reads 3 microns with the micron gauge mounted to the on the pump. We are not seeing any tendency for the microns to go lower. I talked to second level Navac support and the reason for this was not clear.

Of course, I'm worried I have a small leak.


Thanks again,
-Henry

 

We are out making another vacuum after another nitrogen injection. You have a good point re the instruments etc. I can’t post pictures yet. What do they like jpg or png ?

 

Either jpeg or png

 

There is no sign it will pull to 200-300. Pulls to 455 and stalls. Pump is new but might a bit dirty oil at this point but it has slight tint

 

Is his evac setup similar to this?

 

Yes, like that but we don't have the two hoses.... just the one hose and the MG44 gauge you have.
I think we are having issues with the core valve.

 

Could be the schrader core in the side port of the vcrt is not depressing correctly or the core tool could be leaking.

 

Is this a new install or repair?

 

New install - the isolation valves have not been opened.... yet.

 

The Mitsubishi tech thinks most likely something not great with the valve core T.
He can't say no chance of a leak.

 

If it’s pulling down to less than 500 and holding less than 600 after 35 minutes it’s not a leak in the system. There’s probably a leak somewhere in the valve core tool, hose or micron gauge connection.

 

That seems to make sense... two people are saying close to the same thing.
I guess the point is if there was a leak it would rise faster than 1-2um/minute.
Thank you guys for the support. The contractor is beating me in to submission.
We are going to try another decay test tomorrow after pulling all night on it.
If the leak is so small that the system works for years, then maybe it is okay.

 

If there was a leak it once the system is isolated for decay it would rise to several thousand microns within a couple of minutes and eventually rise to atmospheric pressure in less than 35 minutes.

If it was moisture it would rise to 1000-2000 microns and stall there.

He will know when his vacuum rig is tight because it will pull the system down below 500 microns in 5 minutes with 1 hose within 15 minutes it will likely be below 200 microns.

 

You fellows might see more in this graph than I can.
The core-tool T fitting is suspect for 2 reasons:
1 - the large jump when it is turned off from on
2 - the shape of the decay (time and amount) - not sure you guys would know much better then my guess

 

He needs to leave the ball valve cracked halfway for the first few minutes and then actuate the ball valve a few times during evacuation. The ball valve passageway holds a pocket of air if this isn’t done.

He should still be able to pull lower than 400 microns while the pump is running though.

The decay looks fine

 

I've seen that on YouTube. At this point I about have to give up with these guys. I'm going to assume the leak if I have one is small, maybe small enough to not matter much for a few years. They got me on the phone with the distributor and he said basically the zero leak down is not really realistic. One micron per minute is pretty good (probably no leak) or means the instrumentation is the culprit for the decay and unable to pull down past 460.

 

It may be an issue with his evacuation set up, as others stated.

You have to be able to pull below 500 microns in order for it not to rise above 500 microns. However, if it does not go above 1000 microns, it is probably not a leaker.

There is a method called a triple evacuation. This involves an evacuation down to a certain level, flow nitrogen through the system, repeat and then pull a final vacuum.

 

We have done the triple evac. This is the last "leg" of it. At this point we've injected nitrogen and pump down about 5 times or more.
The vacuum pump is running overnight tonight. Right now it is "stalled" at about 800 which is not very good. Usually it pulls down under 500 when all the gaskets/etc. are not leaking. They bought a new hose, micron gauge and pump. Things improved a lot but the core tool valve is old.

 

Interesting about outgassing - makes sense but probably not my issue now, been using the hose for two days. I think the new vacuum hose is a Yellow Jacket. I was just out there, it does have a smell to it.
All the experts (you guys, Mitsubishi Tech and DSG tech in Sacramento CA) are saying the most likely issue is the instrumentation. I think the only thing not new at this point is the core valve.
Right now 1AM - the pump is out there running and sitting at 829, getting worse if anything. Not sure what good it is doing?
So it is definitely not pulling down quick like it has before.
The company is not happy with me and asking is it okay to charge the system. I don't like this and I said I wanted them to fix their instrumentation.
What you guys are saying makes sense.
As far as I remember on data center air conditioners or CRAC units they pulled a vacuum and held with no rise overnight. Why is that not normal for this mini-split, the Mitsubishi documents seem to go along with that. I don't want to start a fight, just asking at 1AM.
I really appreciate the comments and info. from you all.

 

In case someone wants to view the Mitsubishi spec. for vacuum and hold. Here is part of the page.

 

Was it leak tested at 600 psi nitrogen for 24 hours as Mitsubishi specifies?

 

Yes, we did that. I have a graph. At night/morning went down to 552 and then back up to 594. I'm working on posting the graph.
We then let it go another overnight.

 

Test the vacuum rig.
First attach the vacuum gauge directly to the pump. It should pull down below 100.
Then attach the hose and gauge and and evacuate those.
Finally attach hose, VCRT, and gauge.
That process will identify the weak spot.

 

Very good makes perfect sense. I think we have sort of done that but then hooked on to the core tool with micron gauge and ball valve and it was what it was because there were no other core tools. We will do that again this morning if I can talk him into doing that. I've asked them to bring a better or new core T fitting.

 

duplicate post

 

I will try to insert the 600 pound pressure test below here.

 

Question to you all.
Nitrogen pressure goes up and down with temperature.
I've been using calculators online - there are a number of them.
The change is about 1 psi to 1 degree F. Is this way off or correct?
I've been using this one.
https://www.oceanhvac.com/nitro.php

 

The equation is as follows:

P2= (p1*(t2/t1))

You need to convert *f to *k

48* = 282*k
10* = 260*k

200 * (260/282) = 184psi

The calculator you’re using is fine, that site was made by a member here at hvac-talk.

 

Another observation. The installer company technician explains that the "shop" doesn't really supply equipment to them. They have to buy their own. The shop might supply vacuum pump oil and that's about it. My gut feeling is they are not paying some of the guys that are coming here very much. They use hoses/etc. for a long time and don't check them. E.g. they first showed up with hoses that had gaskets visibly worn, distorted or torn up. I pointed that out. The next day hoses arrived with new gaskets. My feeling is the people on this forum are very attentive to the condition of the instrumentation they use. In your business, poor quality equipment not maintained really doesn't work out. Probably no new info. in this post

 

My posts might be getting sort of boring at this point. However, it might be interesting for you all to know what I did before they initially arrived for the install.
1- got all furnace ducts removed $X
2 - got blow-in insolation removed $X
3 - installed crawl ways over the ceiling joists
4 - installed LED light string to go where they needed to go
5 - installed a LED spot light they could move around as needed
6 - installed a gable fan
7 - my wife made homemade pumpkin pie and handed that out one afternoon when they had 4 guys here

 

The graph doesn't give me a good indication if there is an issue because there are no temps. Since you have so many data points is it safe to say the gauge was connected the entire time? If I was the tech on site I would have a good idea what the temps did through that time period allowing me to make an experienced conclusion whether the pressure test passed. The pressures went down overnight and up as the day progressed. That is evidence the swing is temperature related.

Have them sweep the system with nitrogen, run the pump a bit to get warm and change the oil, then evacuate system. It should not be difficult to pull below 400 microns. Isolate the pump and watch a full decay. If it gets to 1200 microns there is a leak. If it stops before that, change the pump oil and keep evacuating.

 

Yes - at the time I was not fully aware re. the temperature relation and the importance of that. The two days were close in temperature, most of the line set is in the attic. The gauge was connected the entire time - I think it was we didn't want to break the seal. I tapped on the gauge for each reading to get the needle to settle. This system does not allow a sweep. The pressure side has only the cap with the shutoff valve behind it. I agree it should be easy to pull down to 400 microns. When we did the pump test it pulled down to 3 microns. That's on a new Navac 8Di pump.

 

For reference here is the full section on vacuum and hold in the manual. Note gauges in the picture with -14.7psig and further down says micron gauge must be used. The Mitsubishi documentation is a bit of a mess.