|
|
Post by OldGit on Jul 22, 2022 21:34:58 GMT 1
Well as you'd expect, it's been a heavy month for aircon issues, most were corroded condensers (mainly due to road salt), damaged condensers due to impact, faulty pressure sensors and one with corroded rear aircon pipes. Another tech had one that he couldn't find the leak on, even with my 'super sniffer' (Elitech ILD 200), nitrogen and UV oil.... he changed the condenser anyway 'because it's always that'.... Predictably it was back the next day (that's some leak!) and in my list of jobs... I suppose I gave it away with the thread title but it was the evaporator leaking, no signs of overcharging (i.e. not burst) the car is four years old, oddly passed the vac leak test which for R134RYF is very strict, anyway, it appears I was one of only two tech's out of eleven that had seen leaking evaporators before, this was on a Merc. A177, the sniffer went mad under the car near the condensate drain and again when shoved where the pollen filter should be...
It's very unusual, there's no corrosion or signs of impact / damage and the UV oil leak appears to indicate it's a joint onto the lower endcap, the others I've seen (a grand total of three excluding this one) has obvious causes.
Anyone else experienced unusual evaporator failures?
|
|
|
|
Post by valhalla on Jul 22, 2022 22:47:20 GMT 1
That's an extremely unusual sort of failure for an evaporator. The most common, as you allude, is corrosion, and lots of manufacturers have had trouble with their material-specs for the aluminium alloy specification for these (condensate rots the evaporator from outside inwards). It isn't unusual (no Tom Jones jokes, Noberator......  ) for the diagnosis to be missed by a vacuum-test, as this is a poor way of assessing leak-rates on matrices - you have to use pressure and/or smoke to find these, as vacuum will self-seal the problem.
Your problem on this one car car would suggest that it had a manufacturing fault in the brazing of the end-tank to matrix. Very rare, I would say!
|
|
|
|
Post by OldGit on Jul 22, 2022 23:16:58 GMT 1
Fully agree about the vacuum test - it's pointless, but part of the process enshrined in the legislation... The tech involved was 'qualified' in terms of having a F-Gas certification and a three hour 'training' on how to use 'the machine'.... other than that, well I guess that's what they pay me for.
It is a very unusual failure, especially after four years with no visible corrosion, i'll be cleaning it off and pressure testing it in a spare moment to find out where and why it's leaking - if nothing more than to satisfy my own curiosity.
|
|
|
|
Post by remmington on Jul 23, 2022 0:30:08 GMT 1
Fully agree about the vacuum test - it's pointless, but part of the process enshrined in the legislation... The tech involved was 'qualified' in terms of having a F-Gas certification and a three hour 'training' on how to use 'the machine'.... other than that, well I guess that's what they pay me for. It is a very unusual failure, especially after four years with no visible corrosion, i'll be cleaning it off and pressure testing it in a spare moment to find out where and why it's leaking - if nothing more than to satisfy my own curiosity.
Never changed one - diagnosed one failure - hissing noise when pressurised with nitrogen (but customer would not pay for it). Changed a few seals into them in my time - (bulkhead area) but never a matrix. I got robinair sniffer (hatefull thing). ------------- I now will not warrant any aircon leak testing unless you allow me to leave nitrogen in the system against a set of gauges for deadended for 24hrs.
Another question? How much nitrogen pressure so you use for testing? I use 3bar - someone told me this is not enough? |
|
|
|
Post by OldGit on Jul 23, 2022 7:14:35 GMT 1
The components are designed to operate at up to 35Bar so probably tested at manufacture up to 50Bar, I test at 20Bar. 3Bar is nowhere near enough to be honest, especially for a leak that takes months to deplete the charge, with those, I tend to just use 'Extreme Ultra' leak sealer as no-one will pay for days of testing on something like that.
|
|
|
|
Post by remmington on Jul 24, 2022 8:19:19 GMT 1
The components are designed to operate at up to 35Bar so probably tested at manufacture up to 50Bar, I test at 20Bar. 3Bar is nowhere near enough to be honest, especially for a leak that takes months to deplete the charge, with those, I tend to just use 'Extreme Ultra' leak sealer as no-one will pay for days of testing on something like that. |
|
|
|
Post by remmington on Jul 24, 2022 8:20:52 GMT 1
|
|
|
|
Post by remmington on Jul 24, 2022 8:21:33 GMT 1
"tip your head to the left to view photos" TeeHee...  |
|
|
|
Post by OldGit on Jul 24, 2022 9:07:36 GMT 1
Try using your high side gauge..... and that bottle should be on a trolley!
|
|
|
|
Post by Joepublic on Jul 24, 2022 14:19:28 GMT 1
Try using your high side gauge..... and that bottle should be on a trolley! I can see a trolley, wheels at the bottom, handle at the top? |
|
|
|
Post by OldGit on Jul 24, 2022 15:41:30 GMT 1
Wheels are top & bottom, handles on the side! That's not a trolley, its the beginnings of a centrifuge.
I possibly didn't enlarge the picture |
|
|
|
Post by Joepublic on Jul 24, 2022 15:57:23 GMT 1
Wheels are top & bottom, handles on the side! That's not a trolley, its the beginnings of a centrifuge.
I possibly didn't enlarge the picture Local guy did some welding for me, works welding trains (I'm near Crewe), he told me about witnessing an acetylene bottle falling over, it went through 3 walls (two different buildings) before getting jammed and coming to rest - no one was hurt, but the photos of the damage remain in the training dept. They build the chassis with a curve so once they fully built the weight makes them straight |
|
|
|
Post by remmington on Jul 24, 2022 19:02:51 GMT 1
Try using your high side gauge..... and that bottle should be on a trolley! Don't think I have ever seen a OFN test kit with a high side connector or gauge ever. Just googled them - they all connect to the low side (from what I can see). The master cool set we have does not have a highside - it did come with two connectors - R134A R1234YF |
|
|
|
Post by OldGit on Jul 24, 2022 19:23:53 GMT 1
I'm assuming you use a 'machine' for your degas / regas jobs, if not, how do you diagnose pressure issues?
To be honest, I just use the OFN on my manifold and use the high-side gauge, when the system is not running the pressure is pretty much the same on both sides of the restrictor or compressor if no additional restrictor fitted.
It's the same principle as leak testing tyres & rims, get the pressure up to show the leak, it'll drop quicker at a higher pressure as there's more force to push it out so you'll see it / sniff it sooner. Even 8Bar is better than three, a system running normally at ~40 deg. C. will see 15Bar on the high side, enough to push refrigerant out of a small leak quickly enough to deplete it after a few days. Test at as high a pressure (subject to design limits!) as you can.
|
|
|
|
Post by remmington on Jul 24, 2022 19:44:28 GMT 1
I'm assuming you use a 'machine' for your degas / regas jobs, if not, how do you diagnose pressure issues? To be honest, I just use the OFN on my manifold and use the high-side gauge, when the system is not running the pressure is pretty much the same on both sides of the restrictor or compressor if no additional restrictor fitted. It's the same principle as leak testing tyres & rims, get the pressure up to show the leak, it'll drop quicker at a higher pressure as there's more force to push it out so you'll see it / sniff it sooner. Even 8Bar is better than three, a system running normally at ~40 deg. C. will see 15Bar on the high side, enough to push refrigerant out of a small leak quickly enough to deplete it after a few days. Test at as high a pressure (subject to design limits!) as you can.
Yeah both machines we have here - have both gauges - I think the newer Texa one will take OFN thru a yellow port in the back if you want. The older R134A Robinair one does not (I think). |
|
) for the diagnosis to be missed by a vacuum-test, as this is a poor way of assessing leak-rates on matrices - you have to use pressure and/or smoke to find these, as vacuum will self-seal the problem.
