Possible selection problem with a rotary screw compressor

[ashtree]

A couple of years ago i specified a rotary screw compressor package to provide service air to a small water treatment plant that i was building. The air requirements are relatively small but peaky , during periods where there are numerous valve operations. To give you an idea of air requirements the compressor has done about 360 hours in two years. The unit is about as small as a rotary screw goes and is classed as a 4kw unit.

The unit was selected primarily based on low noise output as there are residences quite close by. I had also used similar units in similar applications elsewhere without a problem, although these had been specified by people other than myself.

This unit recently destroyed itself with the compressor unit being full of rust and water but also no oil.

So we started talking with the manufacturer/suppliers and i got this massive tale of woe how the unit was the wrong selection and they need to run for extended periods and they need to get hot to vaporise the water etc. Maybe thats fair enough and a couple of suppliers told me the same story.

However i also spoke with an expert who has considerable service and overhaul on these units and he says " All that is correct, but these units use special oil that maintains lubrication with considerable amounts of water mixed in , so provided the oil is changed out regularly etc it should be okay."
As far as i can tell (obviously none of the operation staff want to know anything about any of this) the unit that has failed has had no oil changes (despite a 6 month/2000 hour requirement) and i am dubious about whether it has even had an oil level check whilst in service. This is obviously not good. However is this lack of maintenance the actual cause of the problem or is the real problem this short run time as the manufacturer suggests?

At least one of the other units that have worked elsewhere successfully gets regular oil changes and so far has had no problems despite operating on the same cycle times. Based on this and the low noise requirement i am tempted to go back with a rotary screw and "insist" on better maintenance practices.

What annoys me is that none of the manufacturers or suppliers say anything about the need for extended running when you buy these units nor is it published in any of their manuals or application notes.

Does anyone have an opinion or experienced similar problems?

Regards
Ashtree
"Any water can be made potable if you filter it through enough money"

 

[zdas04]

360 hours on the clock in 2 years? No oil changes or even verifiable checks of oil level? You were on borrowed time at 6 months, be grateful that you got 2 years. Screw compressor oil is hydrophillic. It wants to suck water vapor out of the production stream. As it absorbs water it becomes more viscous so it is harder to pump, loses lubricity so you have to pump more, and increases surface tension so larger drops will fail to coalesce (and you lose oil as aerosol). If you don't get the oil hot enough to cook this water off the oil then everything will rust while the compressor eats itself.

A common fix is to increase the storage capacity so that the compressor runs less often, but longer each time it starts. You want the temperature out of the screw to be above 205F for at least an hour each time it runs.

[bold]David Simpson, PE[/bold]
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[ashtree]

Thanks David.

That's what the manufacturers are saying, but there just is not that much air demand. In saying this i have others doing the same service without any problem
So is more regular oil changes an option or should i forget about rotary screw and go with a reciprocating and spend the cost difference on noise reduction.


Regards
Ashtree
"Any water can be made potable if you filter it through enough money"

 

[Compositepro]

A reciprocating compressor seems far more suitable, based on your description. If your peak flows are not too great or long lasting, an air tank may allow the use of a much smaller compressor.

 

[zdas04]

This is an application where a dry screw makes more sense than an oil-flooded screw. I want an oil-flooded screw to run 24/7 so that I can manage the oil over time. I prefer the dry-screw (over both flooded screws and recips) in this 2% run-time application because of the humidity in the air. I would avoid the oil-flooded screw because of oil management issues. I would avoid recips because of lubrication issues (if it sits for 98% of the time, humidity from the air is going to rust up the cylinders and cause bad things to happen).

[bold]David Simpson, PE[/bold]
MuleShoe Engineering

In questions of science, the authority of a thousand is not worth the humble reasoning of a single individual. Galileo Galilei, Italian Physicist

 

[georgeverghese]

Agreed, looks like you'd want to minimise reliance on lube oil for cooling and lubrication in this application - an oil free screw compressor running start stop would most likely be a better choice.

On the other hand, if you have a good sized air receiver, then you may get a decent run time for each start stop cycle on this oil flooded machine to heat up the lube oil and dry it out. Obviously, these valve actuators need to be able to operate at air pressures down to PAL when the compressor kicks in again.

Strange why this machine has no auto trip on loss of or low oil level ?

 

[ashtree]

I will probably go with a reciprocating type and go well down in size and capacity. The rotary screw type was a lot bigger than required but i could not easily get one that was any smaller. The plant is currently being sustained by a single phase unit purchased at the local hardware store and just plugged into the system.

Noise always has been and still is the issue however and the reciprocating compressor will need some serious noise reduction.

The existing rotary screw has twin 85 litre tanks but the usage is small except for a peak when changing the valves. The size of this peak varies a little but as the hours run shows the run time is quite small. So i doubt whether a large enough receiver could reasonably be used to get the run time required.

I am yet to investigate the question about why it did not trip on low oil. This was in fact my first question when i was told about it.
I am travelling to another plant in the next few days that has a similar set up and similar cycle time so i will attempt to see what is different at that site. The only difference i am aware of is that i know it has had the scheduled oil changes.

And thanks for the comments and feedback.
Regards
Ashtree
"Any water can be made potable if you filter it through enough money"

 

[ashtree]

Just as a follow up i was at one of the other plants today.
It has a 5.5 kw unit different brand but i would bet that much of the actual equipment is the same just painted a different color.
This unit is about 3 years old and has about 1100 hours total run time. Of this it had 282 hrs of pressure time and the rest was unloaded run on time.
This unit has had three oil changes in three years and i saw the service records and we checked out the oil currently in the machine. It still looks okay and volume is good.
Interestingly this machine has had two mechanical failures on the oil filters, with vertical splits developing on the crease lines where the filter wrench grips.

Regards
Ashtree
"Any water can be made potable if you filter it through enough money"