Vacuum good, but nitrogen charge leaks.

SeppW on Thu September 09, 2010 9:19 PM

Update: Left vacuum on overnight (8 hrs). Gauge read 15 in/hg next morning.

This is a brand new compressor, if that shaft seal is leaking, that blows big time!

This project is about done. Wound up replacing compressor, drier, compressor/condenser hose, evap tube, TXV (rear), and flushed rear circuit.

Added 11 oz of PAG 150 DEC and got everything buttoned up (NyLog blue on all seals, anti-sieze on threads, fittings good and tight, not stripped out or cross-threaded anyway.).

Charged system with 200 psi of nitrogen. High side gauge started dropping pressure. Checked for leaks using Big Blu, sprayed every connection, fitting, condenser, compressor, tubes, & hoses. Nothing; no bubbles, foaming or hissing.

Pulled a vacuum down to 29.9 in/hg (almost on the 30). Stopped at about 15 minutes into vacuum and shut off valves at manifold and turned off pump. Vacuum held for 15 minutes, restarted vacuum for 1 hour. Checked again, and vacuum held for 30 minutes.

Had to go out of town, so I charged system with 50 psi of nitrogen. Came back in 3 days, connected gauges and 0 psi on both gauges.

Do I have a leak because I'll be damned if I can find it?

Any recommendations on how to proceed?

Edited: Fri September 10, 2010 at 7:33 AM by SeppW

Dougflas on Fri September 10, 2010 12:31 AM

evacuate system to 29.9, put an oz or two of refrigerant in the system. Follow that uo with 50 to 75lbs. N2. Use a sniffer. The leak should show up. I would chech the compressor front seal carefully. Place a shower cap over the front of the compressor. After a few hours or overnight, put the sniffer probe in there.

1stbscout on Fri September 10, 2010 12:43 AM

yes you have a leak..

If I had to guess I would check the shaft seal on the compressor.

iceman2555 on Fri September 10, 2010 9:11 AM

Were your gauges attached to the system during this 'down' period of time?
Could it be the shaft seal of the new compressor...sure....anything is possible....and if the shaft seal has no lube on it...yes...it could possibly be the leaker...does this mean the seal is bad...why not charge the system and operate it for several hours and retest the system. It is normal for shaft seals to seep refrigerant for the first 8-10 hrs of operation. If the shaft seal was not prelubed during installation, there is a good chance that the seal does not have an adequate coating of lube to insure total sealing.
When adding lube to the compressor it is advisable to invert the compressors after lube has been added to allow lube to flow downward and saturate around the shaft seal area.
Sometime ago, one of the major OE compressor manufactures, conducted a test concerning shaft seal leaks. Various models of compressors were removed from the assembly line and tested for leakage. Any that failed this test were repaired and retested. These units were then placed on storage racks. These units were tested after 30 days.....and then retested at 30 day intervals. After 60 days, several of the models were indicating seepage at the shaft seals, after 90 days some of these seepage units had become 'leakers'. The units that indicated leakage, were removed for further testings. Some were disassembled to inspect the shaft seal area and some were lubed and placed on test stands. These units were charged and allowed to operate for a period of time. Removed and retested. All units indicated that the shaft seals were once more truly functional.
Best bet is to locate the leak in your system. Unfortunately it may be that the system must be made operational to fully test the system.
Or it simply could be the case that your system simply leaks nitrogen...this may not be a true indication of a system leak.
The test for vacuum leakage is not measured in hours...it is measured in minutes. No more than 2 in/hg in 5 minutes is the standard test.
Keep in mind that when the system is left in vacuum for hours, that standard atmosphere pressure greatly exceeds the internal system pressure (vacuum) of the system. This could simply be a migration of higher pressure (atmosphere) to a negative pressure (vacuum). Your system still maintained a vacuum....it never returned to atmosphere pressures.
Connect a can of refrigerant to your system....introduce this into the system......operate the engine to introduce more temperature/pressures into the system and then test for leakage. 50-60psi of pressure should be sufficient for test. There is no need for tremendous amounts of pressure.

-------------------------
The strongest reason for the people to retain the right to keep and bear arms is, as a last resort, to protect themselves against tyranny in government.
Thomas Jefferson

NickD on Fri September 10, 2010 11:52 AM

If these are R-134a ports, should be good for about 3 seconds with those quick coupler gauges attached before leakdown occurs.

Using a hose removal-less process between vacuuming and charging, as you are closing the vacuum pump valve, should be opening the refrigerant valve to prevent air from entering the system.

emsvitil on Fri September 10, 2010 4:59 PM

Do the pressure test with the gauge set alone and see if it's the gauge set that's leaking.......

-------------------------
Ed
SoCal

SeppW on Thu September 30, 2010 8:47 PM

Following iceman's advice, I said fuggit and charged the system (with refrigerant machine). $200 for 4 lbs, ouch! Had to get it done as I was going on a trip and wanted air. The tech was impressed with work. Anyway, after it charged, I was burnishing the clutch and the guy ran over and asked what was going on, I just shrugged my shoulders. He came around a felt the vents, yes it was blowing cold, very cold. I also checked for several days with a sniffer for leaks and nothing.

Anyway, I left on the trip and ran the A/C practically full time. 1200 miles in three days. Worked like a charm, good and cool.

I did notice the compressor never cycles; runs constantly. However, it doesn't chatter like the ol' original Harrison R4 did, but it does make some noise at engine idle, like a slight knocking. Noise disappears at higher RPMs or is drowned out by engine noise. I thought the idler pulley was ready to come off when I first noticed it.

After I returned home, I was finally able to put the gauges on after I had a helluva time finding a seal for one of the hoses. Blew the seal out earlier when I was purging the hose of air with nitrogen.

Ambient temp was 97F, humidity 35%; after setting-up and running test I got the following readings: 47 psi on low side, 275 psi on hi side, vent temp was 55F. Those results were within the humidity-based pressure/temp chart (of a properly charged system) that came with gauge set.

Do these readings make sense? Should compressor be cycling? Is the knocking a concern?

TIA

SeppW