Unusual Hyd Pump Wear

Cavitation? (Just a WAG.)

Same here (cavitation)

If it is cavitation, clean or change any filters before - make sure no obstruction on inlet - elevate reserve tank if possible - port & polish intake ports & elbows - FWIW -

Just what is a WAG?

Wild @ss Guess

Well some times 40 years of experience can produce a quick and accurate answer. The indication was that much was checked and this was a reoccurring issue. Many miss the cavitation unless the audible sound is present. That is why on many of my posts I suggest the use of ultra sound. If they had listened to this pump with an ultra sound device and were properly trained the tell tail popping sound of bubbles would probably be heard. Others mentioned cleaning the intake filters which are a prime cause of this cavitation along with leaks that allow air into the system.

Ultra sonic checking is an excellent idea. I will see if we can arrange an extended field visit to perform that. I will add (updated information) we are seeing failures (some, not a high volume but some) of these during pre-delivery inspection, and others as long as 4 to 6 months out, up to a year in some cases. Note the wear may NOT be constant to the length of time; we have see long term cases with moderate wear and short term cases with significant wear of the plate. I mention the pre-delivery inspection because the manufacturer will be installing new systems including filtration, which I hope one could assume are clean, but then, one never knows.

It may be possible to detect cavitation by adding a Fourier analyzer app to your phone - hold it near a known good circuit, then near the suspect circuit - you may / should see a sig. difference in the frequencies and amplitudes.

I use one of these apps to find leaks in my compressed air lines & equipment - works very well.

Interesting.

Share links?

If you type 'Fourier Analyzer' into a smart phones app search, you should get a number of suitable apps, both free and paid.

Then if you open the app near a 'known good' ('quietish') hydraulic circuit, it might look something like this.

But if the circuit has cavitation, etc, it might look something like this.

The F. Analyzer plots a sound map - x direction is the frequency, y is the amplitude, or the power of the noise at that frequency.

Have not used mine for hydraulics , let me know if it works. Good luck.

I am interested in recording sounds made by the various bubblings that take place in a steel electrolysis vessel where the vessel itself if the cathode.

Is that a price in bitcoin in the upper right? hmmm...

You might need to use a mic you can attached to the vessel - or even one you can put in the vessel - but not sure how long it would last!

http://int.search.myway.com/search/GGmain.jhtml?p2=%5EBSB%5Exdm071%5ES18313%5Eth&ptb=8813C8F0-B3A5-4E4E-A118-9A7564C61FFE&n=782b4356&ind=&tpr=tbsb&trs=wtt&cn=th&ln=en&si=COqZnvyR1c8CFUUXaAodo78MGg&brwsid=451f7ec8-2cf0-40da-bbaf-c3e6973f59a6&searchfor=hydrophone&st=bar

Apps are there for smart phones to record, even free with the OS.

That price is in Baht - ~ 35 / 1 - so 179 baht is just under 6 US.

We can hear the sounds outside the vessel, even outside the foam insulation calorimetry enclosure - actually impressive pinging sounds at times. I suspect Arduino would love this, but only if the sounds could be superimposed to be a non-negative voltage (or convert to digital with a shield board), then log it, but my God, listening to 100 hours of that for a tell-tale "pop-ping" sound of a cavitation bubble, or localized hot spot bubble would be mind numbing.

I appreciate the link, my friend.

I am considering some type of cavitation or perhaps aeration issue. I am thinking some type of hydrostatic imbalance is involved, perhaps with the unique conditions of the special oil, temperatures and operating conditions.

Have you looked at particle analysis on any of the oil where this wear is showing up?

The thing that I find vexing is the isolation to the low pressure side of the pump, not the discharge side.

Maybe our buddy is right about cavitation. It is possible, especially if the fluid is contaminated with high vapor pressure product, or is wet, etc. Do the pumps become louder on long-term operation (say an hour or two after starting up)?

A few questions

What pump model is it?

What pto percentage are they using?

What is the max engine rpm setting?

Over what time period does this wear happen?

Can you share a picture of the damage?

As for cavitation that will show up on the vanes more than anything. It first shows as pitting but progresses to something that looks somewhat like disolving. Do these pumps have that kind of damage?

Vickers 2025 double pump (failure in the 20 cartridge, the 25 is not impacted), standard performance, and the Vickers 25 single pump, high performance model.

We are waiting on PTO and rpm information. The time frame has been anywhere from an hour to a year on the extreme ends. It appears to be averaging 3 to 6 months. A picture of an extremely worn plate has been attached to another post.

The vanes illustrate scratching and mild discoloration, but not pitting or metal removal/transfer/dissolving.

Okay. I'm very familiar with those pumps though in my case they are all new. Digger derricks for the double pumps and aerial buckets for the singles.

If it's not cavitation or excessive moisture contamination related, running the fluid way too hot or having too thin of fluid for the working temperatures and pressures can cause rapid pump wear issues.

The target temperature range I have always shot for in a hydraulic system is 160 - 200F for at least 50% of the operating time in normal operation.

Much below 160F and the fluid doesn't get warm enough to drive out any moisture it may collect and much over 200F most lighter grade general purpose hydraulic fluids will start to lose too much viscosity and lubricating characteristics when subjected to high loading conditions.

Yes, I am concerned about the temperature extremes. We have been informed these are being run at both extremely cold and hot temps, and with that special oil, I have a major concern it is losing adequate viscosity at hotter temperatures. Conversely, we have inquired if the manufacturer is supplying auxiliary heating to warm the oil at very low temps, and I have not yet received an acceptable answer.

A tank heater is a good idea in the winter months. In my experience most major fleets spec this as standard equipment.

Update: I have seen some additional pumps with the indicated wear. In addition, I have determined that the wear pattern moves with relation to how the pump is assembled for operational rotation; that is, there are "forward" and "reverse" rotation pumps. The cam ring/vanes/flex plates are inverted to change the rotation. The wear stays with the intake or suction side with the component inversion. Also, the vendor is using a special oil by Kendall ("Glacial Blue") in this application (Canada/US Pacific NW). We are obtaining a supply of this oil to attempt to duplicate the operating conditions. A photo is attached illustrating the most extreme condition; we are seeing this pattern of wear with varying degrees of depth of the milling of the bronze alloy.

Glacial blue is a good low temp oil. We have one major fleet client that used it.

That wear looks to be from the vanes themselves. I have heard of something like this that had to do with a clogged suction filter or strainer, I don't recall exactly which.

I'll run this picture past some of service guys to see what they say.

Also if you haven't found it yet Parker has a great trouble shooting guide for vane pumps.

https://www.google.ca/url?sa=t&source=web&rct=j&url=https://www.parker.com/literature/Vane_Pump/PDF%2520Literature/SERVICE/CatalogueTroubleshooting-HY29-0022-UK.pdf&ved=0ahUKEwiiq9PyhN_RAhVP_WMKHRK9BqAQFggcMAE&usg=AFQjCNGrlV7rj6keOZOSA6UBLiDxtlq2vQ&sig2=a72kwEYGyNEwWgQtXvTclw

Update: we have confirmed that a 15W oil is in use with these pumps. Also, we have some more equipment configuration information: Engine RPM is 1100, with a 120% PTO ratio for a pump RPM of 1320. Oil temperature is running at 145 deg. F. A concern is the use of a 1 1/2" inlet hose that is 17 ft. long. The engineer I am working with has a concern about sufficient fluid flow. Also, a 50 gallon tank is being used.

So, the next step will be an attempt to duplicate the operating conditions. I have worked with the company to configure a test stand to run the 15W oil at elevated temperatures (145 deg F as noted), and across a range of temps. If the failure is not noted, I will work to reconfigure the test stand to (as closely as possible) replicate the inlet hose setup (17 ft long 1 1/2" hose).

Update: Running tests on examples of the product under review. One pump demonstrated the beginning of the milling issue while under test; several demonstrated significant scoring of the bimetallic plate as was seen with customer returns. One had no wear or scoring, only demonstrating some discoloration of the bronze alloy; a light brown color. Upon further investigation, we determined that the 145 deg. F may not be completely accurate as a temperature at the inlet connection; we determined through some modeling and experiments that the temperature is closer to 160 deg. F. Therefore, we aimed for 160 deg. F at 2700 psi. Interesting to note that at this temperature and pressure, a high intensity noise with vibration was generated by the test stand, with an immediate drop of 200 to 400 psi when the noise started. When the pressure was reduced to 2000-2100 psi, the noise/vibration stopped. Upon returning to 2700 psi, the noise and vibration re-engaged anywhere from 5 to 90 seconds later. Also note the pump which did NOT demonstrate wear or scoring never generated this phenomena. My next two tasks will be to re-run the tests with the manufacturer's recommended oil, and also instrument one of the customer's trucks to gather real-time and place data for further investigation.

What is the auto-oxidation onset temperature of this when some peroxides are present, or when some old oil not is completely removed during swap out?

What is the autoignition temperature in air?

You may be changing the characteristics of the oil in such a way that moisture is present in one pump that is not present in the cooler one. That water, or some other substance, could flash off in hot spots and result in cavitation noise?

Does the oil show signs of discoloration, such as darkening due to "char" formation?

Final update: the organization I was working with has closed this corrective action based upon the tests I helped conduct, information we received from the customer and some internal investigation. Two main things we discovered are the intravane assemblies during machining were being measured is such a way that intravanes with excessive length (above the upper specification limit) could pass through the value stream undetected. We determined this was a significant part of the excessive wear, especially at the higher temperatures and pressures we saw. The manufacturer has changed the measurement process to capture this fault mode. The other issue was what was determined to be excessive length in the inlet plumbing at the customer application. We simulated this and we saw excessive wear once again. The customer has made changes in their system, and has also made improvements to reduce operating temperatures as required. With the improvements introduced by the organization and the customer, it appears that much of the issue has been dealt with, as customer outcomes have improved greatly and returns for the issue to the organization have been reduced by a significant amount. I thank you for all the input.

"... intravane assemblies during machining were being measured is such a way that intravanes with excessive length ..."

This is odd, could you explain this a little bit more?

Sorry for the delay. Some of the information is proprietary and I had to ask permission to discuss this. I'll speak in general terms. The intravanes are a 2-piece sub-assembly; I will be referring to the large outer part which has a nose ground to interface with the cam form of the pump's ring, and is ground to length. This length dimension interfaces with the bimetallic internal plates on this design. In the initial grind, the operator runs a sample grind sequence, then measures the parts on a CMM. To avoid having an operator do math, the CMM results printout gives a adjustment factor that the operator then enters into the grinder. This adjustment is based upon (in part) a calculation of length measurements, one of the outputs to calculate the adjustment is average length. The issue was, one or more of the length measurements could actually be over the upper specification limits for length, but the calculation for average could generate a value within limits. It looks like parts were slipping through that were above the limit but classified as within limit. Depending upon some other inputs (temperature, type of oil, others I can't get into), these were driving excess wear under some circumstances. The process was adjusted to avoid this measurement issue and some guidelines were issued regarding approved oils and operating conditions, and some design recommendations were issued to the customer. Hopes this clarifies.

Yes, isn't statistics wonderful. Instead of using a two-sided distribution next time, how about trying a one-sided distribution?

I told 'em something like that when I finished. Off to the next project...motorcycles this time.

Cheers. I knew something needed to be lopped off.

You might send them some of the Jerry Clower jokes:

"knock him out, John",

or "Marbles"