Oil Leak in NH3 Discharge Line Flange

[NH3LVR]

This morning I arrived at the plant I am currently servicing. After a quick look around the engine room, I went up to the roof to try to fix a mist eliminater problem we have been having.
After some success at solving the problem, I saw the Facility Manager coming through the roof hatch.
He informed me that there was a pool of oil underneath a 12 Cylinder, two stage Vilter. Cursing silently as I figured the crankshaft seal had failed and I would not get home before Memorial Day weekend was over, I went to look at at the problem.
It appeared that all the oil was coming from the flange above one of the discharge valves. Not a problem, just get the manlift, valve it off, stick a pump on it, suck it down a bit, remove the valve, make new gaskets, and put it back together.
Start it up and it works fine- no leaks.
But it leaves me with a question. Of course there is a good bit of oil in the gas leaving the high side. But how can I have an oil leak from a discharge line, without a trace of an NH3 leak?
My customer would like to know and I am at a loss.

[Sledge]

My guess would be that you cannot. For a leak of suffcient size that it required immediate attention, I would be expecting to see substantial foaming, with ammonia mixed with the oil, definitely some serious stink.

I am wondering about oil leaking from an oil return line.

How do you top up the oil in the system. The plant I worked in that had ammonia had oil return problems all over, and solved it by draining oil off the receivers and topping up at the compressors, as part of the daily rounds.

Any chance someone was topping up and spilled, in a location that looked like the source was the discharge valve?

[NH3LVR]

My guess would be that you cannot. For a leak of suffcient size that it required immediate attention, I would be expecting to see substantial foaming, with ammonia mixed with the oil, definitely some serious stink.

I am wondering about oil leaking from an oil return line.

How do you top up the oil in the system. The plant I worked in that had ammonia had oil return problems all over, and solved it by draining oil off the receivers and topping up at the compressors, as part of the daily rounds.

Any chance someone was topping up and spilled, in a location that looked like the source was the discharge valve?

No, it came from the discharge valve upper flange, ran down the piping. dripped on the crankcase, and then down to the floor. I spoke to the boss and he remembers that the flange did the same thing when the machine was installed. They tightened it up then.
I am never surprised to see a drop of oil around a vertical valve stem, where oil will pool around the packing. But not in a vertical valve.
The oil returns are 15 feet in the air and on the other side of the machine.

[US Iceman]

Here are several possibilities.

1. We use exhaust fans to keep the engine room air temperature at a reasonable level.
2. The exhaust fans are usually mounted on the roof, or at high points on an external wall.
3. The air intake dampers are either side wall mounted or roof mounted at an opposite end of the engine room from the exhaust fans.
4. Therefore, there are large amounts of air circulation through the engine room, especially at heights (normally not too much where the compressor motors are).
5. Ammonia vapor is light than air (if it has not absorbed too much water).
6. Since the ammonia gas is lighter than air and you have a lot of air circulation through the top of the engine room I would not be surprised if no one noticed a leak.

That's about the only thing I can think of.

[Josip]

Hi, NH3LVR

It is almost the same with compressor shaft seal leak.

You can notice oil leak but without ammonia smell and US Iceman explanation is reasonable with additional air stream coming from motor fan...

Best regards, Josip

[NH3LVR]

Good points brought up by the esteemed Gentlemen.
I have seen many seal leak without noticeable smell, but they are flooded witrh oil.
I shut off the engine room fan, and the screw next to the Vilter. Still no luck with the Sulfur stick.
I had a fleeting thought that it might have to do with the low head pressure and low discharge temp. My thought was the the oil mist might be coalescing in the line. But that seems a bit of a stretch. And it does not explain why the flange did not leak when the machine was shut down.
I thought the gasket might be wicking oil through, but it was hard and inflexible.
For you US folks, have a nice Memorial Day weekend. I will get home for part of it.

[US Iceman]

I have seen shaft seals leaking like Josip mentioned with no noticeable ammonia smell also. I did not remember those until he mentioned it.

Let's try a hypothetical situation for similarity to your flange leak NH3LVR.

The shaft seals are bathed in oil for sealing, lubrication, and cooling. Obviously, they leak with no ammonia smell (sometimes).

For the oil to leak out of the seal it has to be pressurized higher than ambient pressure. That is similar to your questionable flange.

In a discharge line we know it's common for oil to be present in the refrigerant leaving the compressor. If the gasket was leaking just enough to allow oil to bypass the gasket sealing surfaces, then this is similar to the shaft seal scenario.

The leak may be small enough to allow oil to escape under pressure, but the only thing coming out is oil. The oil might be providing a liquid sealing surface to prevent the ammonia from escaping.

That might be going out on a limb somewhat, but it's the best explainable scenario I can come up.

[NH3LVR]

In a discharge line we know it's common for oil to be present in the refrigerant leaving the compressor. If the gasket was leaking just enough to allow oil to bypass the gasket sealing surfaces, then this is similar to the shaft seal scenario.

The leak may be small enough to allow oil to escape under pressure, but the only thing coming out is oil. The oil might be providing a liquid sealing surface to prevent the ammonia from escaping.

That might be going out on a limb somewhat, but it's the best explainable scenario I can come up.

Since I have no better idea, I will go for it. Thanks Iceman

[Sledge]

A small leak in a crank seal still gives a little smell. A tiny leak that is only noticed from the stain on the absorpent towels we kept under the cranks, still left a trace odour that you could notice when you were doing charting, even in a well ventilated engine room.

A leak that would create a pool of oil under a compressor that required immediate attention, I would have thought would have left a lot of ammonia odour.

[ART KUHN]

In most compressors oil is pumped along the seal so when it runs there is only oil to leak and no Nh3.When it stops and pressure rises it wil run out of oil and starts to leak Nh3.Most inst. i work at there are more then 1 compressors so pressure will not rise but still when the pressure is above 0 bar eff. it will start to leak Nh3

Regards Art

A small leak in a crank seal still gives a little smell. A tiny leak that is only noticed from the stain on the absorpent towels we kept under the cranks, still left a trace odour that you could notice when you were doing charting, even in a well ventilated engine room.

A leak that would create a pool of oil under a compressor that required immediate attention, I would have thought would have left a lot of ammonia odour.