Filtration within hydraulic system subject to pressure pulses

[hydraulicsman]

We have observed some contamination anomalies within a few hydraulic systems that are subject to pressure pulses & varying fluid velocity.
What are peoples opinions on the best way to filter a system that is subject to a cyclic flow/pressure regime?
The fluid employed is an ISO32 mineral oil.
Some suggest creating filter banks where the filters are placed in series.
Any thoughts/ideas would be appreciated.
Thanks

 

[IceStationZebra]

The effectiveness of the filter media goes down in the presence of pressure and flow variations. The best way to deal with this is a separate dedicated filter loop. If that is not an option you will have to go to a filter with a finer rating, and this may require more filters to deal with the added pressure drop.

ISZ

 

[hydraulicsman]

Hi,
Thanks for the response.
Does anyone know how well a depth filter may work in this enviroment? Also Pall do a range of common radial types but with a special layering method for the fibre called Ultipleat. They claim it works better than the normal element style for cyclic regimes. Anyone tried these?
Thanks

 

[IceStationZebra]

I have no experience with the Unipleat, though I can see how it might help in your situation. In a typical filter the pleats would tend to "breathe" during pressure pulsation, which would tend to block some of the flow area making the problem worse.

Not sure what you mean by a depth filter, do you have an example?

ISZ

 

[Cbrunner]

Pressure pulsations are the enemy to filtration. The surge of fluid will cause filters to release trapped contaminants, not to mention the possibility of activating the bypass valve located in most filter housings.

There are a few different methods to deal with the problem. One is to correct the surge/pulsation problem (although this is usually not a cost effective option). Grossly oversize the filters to deal with the surge of fluid, and create less pressure drop across the filter media.

The most effective filtration would be a high beta ratio Kidney loop type filtration setup (as ISZ has already recommended). Here smaller filters can be coupled with a smaller pump and long run times to create nice clean oil.

 

[hydraulicsman]

Thank you for the suggestions.
Regarding the kidney loop idea, although excellent for cleanliness, incorporating a pump into this system to offline filter would be viewed as a parasitic loss.
ISZ - Here is a link to a manaufacturer of depth filters -

http://www.cjc.dk/products/fine-filters/

 

[Cbrunner]

Your depth filter is a kidney loop.

 

[hydraulicsman]

I realize it's a kidney loop. It was more just to show ISZ the idea of the filtration media stack. That's the part I am interested in.

 

[Cbrunner]

My apologies. If the parasitic loss of an off-line filtration system is such a concern, can you add an additional pump to your existing configuration - say off the opposing end of the motor?

I can appreciate the want for a high efficiency system, but from a cost perspective, what are the installation and maintenance costs of both systems (loop vs depth filter). I would be interested to see if you could get the oil as clean with a depth type filter as you could a kidney loop (at the same relative operating costs).

Depending on how your system would respond to back-pressure, you could run a gear type flow divider to power a kidney loop - that would even out the pulsations a bit.

 

[IceStationZebra]

I assume the filter is a low pressure unit plumbed so the oil goes through it just before the reservoir? And the usage is cyclical which causes the flow and pressure pulsation?

One possible solution is to put the filter in the high pressure portion of your circuit where it is much less cyclical. I have a 250hp hydraulic power unit set up with a 3um filter between the supercharge pump and the main pumps. Any excess oil not used by the pressure-comp main pumps is dumped over a 45psi check valve back to the reservoir. Constant flow and pressure, but a little more expensive that the typical return filter.

On a different 200hp system, which had flooded inlets instead of a supercharge pump, I had 3um filters between the pump outlet and the point of use. Not quite as ideal because even though the pressure was constant the flow still varied, but it didn't change that fast due to an accumulator installed near the point of use for that purpose.

As far as a depth filter vs. surface filter, I would think that they would both loose efficiency due to flow and pressure cycles. The question is which suffers more, and that is probably more supplier/design dependent than the general technology.

ISZ

 

[hydraulicsman]

Thanks for the opinions.
I feel it may help if I elaborate on the system slightly.
Basically the prime movers are cylinders, there is an external force applied to the cylinder to induce a pumping action... the output from the cylinders is controlled by a an algorithm which fires valves to control the machine stability and send pressure to storage accumulators. Therefore the system is relatively pulsy all the time. The accumulators have the effect of dampening the pressure pulses / valve firing induced shocks, but not completely.
To keep conversion efficiency high we prefer to err away from power consumers.
Anyway - Thank you and I'll most likely run some evaluation tests on different filtration set ups.

 

[Cbrunner]

Is the device that draws from the accumulators a relatively constant draw device, or does it surge in fluid demand as well?

 

[hydraulicsman]

The fluid draw from the accumulators is constant.

 

[Cbrunner]

Put an inline pressure filter there - solves most of the pulsation...