Cracked hydraulic pump on an 863 - how to avoid a repeat?

[LessWhole]

I have a 1997 Bobcat 863 which cracked the hydraulic gear pump. I found that and have a new pump on order, arriving in a couple days. Now I'm trying to figure out what could have caused this issue in the first place.

The crack goes directly through one of the bolt holes holding the pump onto the hydrostatic assembly, so my initial thought was that some water got in that bolt hole, froze, and broke the pump.

But lately I'm wondering if the relief valve is bad. Then the previous pump could have seen far-too-high pressures, and cracked the housing. I don't want to repeat that story again, if this is the case.

Questions:

• Is it plausible that the relief valve failed *closed*? Does that happen?
• Can I remove the relief valve and have it bench tested somewhere else? Or is it possible the issue is somewhere in my control block?
• If I have to test it myself, what's the best way? I don't have any pressure or flow testers on hand. Can I just buy a 6k PSI pressure gauge and a hose, and attach it to one of the quick connect aux ports or tilt cylinder, operate & check whether the relief valve is working?

I'm a bit afraid of that last approach, since it seems like if the relief valve is set to high I'd effectively deadhead the machine into the gauge and maybe blow up another pump.

Grateful for any help ya'll can spare.

 

[John C.]

I can only speak in general terms as I haven't had the pleasure of working on skid steers over the years. If I'm wrong about something, someone will be along to correct me. How many hours do you have on the machine and do you think that the pump is original to the machine? What kind of material is the pump casing made of, cast iron, steel or aluminum?

Gear pumps are the least expensive type of hydraulic pump used in equipment. That doesn't mean that they are not expensive. They do have a finite life cycle no matter what you do. Most of the time in my experience in heavy equipment they will run in the neighborhood of 5,000 hours with specified maintenance. A lot of times in harsh usage I see them fail at around 2,500 hours.

Gear pumps in general have a finite life generally measured in pressure cycles. The insides of the pumps wear out because of debris in the oil but the casings take a beating because of pressure spikes. A pressure cycle can be thought of as going from almost no pressure up to relief pressure and back down again. The pump casing and the bolts holding those casings together will actually try to swell, stretch or expand with the increased pressure and then contract when that pressure is reduced. A spike is when there is an almost instantaneous increase and decrease in pressure. Relief valves will catch and release those spikes if they are slow enough but the instant ones have to be absorbed by the pump casings and the bolts. That's why those casings are so thick and the bolts are always of the highest quality. Maximum pressures that gear pumps as a whole can run are generally in the 3,000 PSI range and those are rare. Most of those pumps run in the 2,100 PSI range normally. There is an engineering term that is something like modulus of elasticity if I recall it right, probably not the exact term. It is a measurement of how many times something can be worked before it changes state, basically it work hardens to the point of the crystalline structure in the pump body changing or the bolts not returning to their initial length and the part fails. That's why I ask about the hours on the machine.

So to make the pump last longer, the idea is to reduce the number of pressure spikes. That can be done with better hydraulic relief valves, possibly reducing pressure settings on those valves or changing how the machine is used. The relief valve settings are easy enough. Your idea of a pressure gauge is the move I would start with after checking the cleanliness of the oil and possibly changing filters and oil before installing the new pump. Check your pressures against factory specifications and make sure you are within the ranges given for each function the pump handles. Consider what the machine is used for and how it is used. Skid steers loaders are a bull in a china shop in order to be useful and most people run them using inertia in order to load the bucket. Slamming into concrete walls is usually considered to be normal use to most people. Material gets stuck in the bucket a lot and many people slam the bucket tilt functions to loosen and drop that stuck material. Sloppy joints in the implements are another source of those pressure spikes.

To sum all this up, your pump may have just reached its design life. You will need to check and possibly adjust system pressure settings. You should consider how the machine is used and possibly modify that use through operator training or operational procedures.

Good Luck!

 

[LessWhole]

Thanks John! Here are some answers to some of your questions, in no particular order. I appreciate your help so much!

• The service manual specifies 3000 PSI for the relief valve adjustment
• Total hours are ~6100 right now. I believe the pump was original; the manufacturing date on it was 1997 - so it's either original, or very nearly so. The bolts certainly felt like they'd never moved.
• The old pump casing was aluminum. I've already taken it out and disassembled it, attached some pictures at the bottom of the post showing the damage.
• The machine used to be used by a construction company, then we bought it. It's a household machine, now, not used for business. It gets single digit hours per week, if that, and probably what most would consider light work. Right now it's at our church, where we'll use it a couple times this winter to push some snow. So it used to get heavy use, but the last few hundred hours have been light.
• Unfortunately, I wasn't running the machine when the pump blew. An inexperienced friend was driving it around the parking lot for familiarity. So I don't know exactly what happened, but they certainly weren't lifting anything really heavy or running into piles of material.
• One suspicion I have is they turned on the aux hydraulics, then accidentally pulled the trigger on the handle, which would have deadheaded the machine into the disconnected aux quick couplers. Maybe that would have provided the final spike needed to blow up the pump?

Ultimately, I guess my questions are these:

1. I know for a fact that the relief valve was working a few weeks ago. Is it plausible for it to fail, and fail such that it doesn't relieve pressure?
2. If it has failed closed, would I be okay hooking up a simple pressure gauge to a hose (like disconnect a tilt cylinder hose, attach a pressure gauge to it)? Or could I send a big enough pressure spike to blow up my brand new pump in seconds?
3. ... or should I just put the pump on, run the machine, and only worry if it feels wrong?

I guess you can tell how green I am at this! Software is my day job, but I try to keep myself somewhat mechanically minded. Just want to make sure I'm worrying about the right things here.

 

[LessWhole]

I should add - this is the post (on another forum) that has me scared: http://www.skidsteerforum.com/DesktopDefault.aspx?tabid=54&frmView=ShowPost&PostID=86010

It's a horror story of replacing a pump and then immediately blowing up the new one, but there was never really a resolution posted there. That's what's making me nervous.

 

[John C.]

Aluminum doesn't work harden in the same fashion as cast iron and steel. The failure on that pump is the common type I've seen over the years. The case failed because it likely just reached its total life cycles. Anything over 5,000 hours on an aluminum pump body is great life.

Relief valves usually fail because the springs holding them closed lose their strength and the pressures are then low. It isn't likely that the valve got stuck closed. There should be a place where you can plumb a gauge into the system to monitor the pressure. Follow the hose from the pump to the main control valve and see if there is a quick coupler of some kind that you can plug into. You should probably have a service manual or an experienced person with you before taking on adjustments. You can feather a control usually to bring pressure up slowly when you attempt the check and adjustment. That way if the relief valve for some reason doesn't work you can see it and let off the pressure before something breaks.

 

[LessWhole]

Thank you so much John. Regular end-of-life behavior is exactly what I hoped to hear.

I have the full service manual. It specifies two ways to check the pressure. One is to attach a Bobcat tool to the two aux quick couplers, and then restrict the flow using a valve to slowly raise the pressure, then you can see where the relief valve opens. Of course, the Bobcat tool is a $2k kit...

The other way is basically the same procedure, but using the hoses going to the tilt cylinder. It sounds like I can reasonably take off only one hose, attach a ~5k PSI pressure gauge, and feather the tilt control pedal to check where the relief break is. I'd do it with a quick coupler on the aux ports, but I don't think I can really feather those very well.

Adjusting the relief pressure is done by just turning a screw on the end of the valve in the control block - relatively accessible (in compact skid steer terms...)

Thank you so much for your help!

 

[John C.]

I read that post and note some problems. First off is that the original poster didn't specify where he got the new pump. I've seen some after market stuff that failed because of substandard materials and machining. He also didn't supply any info on what he had done before starting the machine with the new pump. Did the pump blow up instantly when the engine started or after he operated a function. The other issue for us on a web site is that we can only offer general information because we are not on site. Most of the things we do are based on visual information of the whole machine and experience with similar machines in the past. Your observation of the auxiliary circuits possibly being pressurized is a good thought. The main relief in the control valve would be able to see that if it had happened and opened so from here, I don't see that as a cause.

I would prefer plumbing into the tilt cylinder with some kind of Tee connection. That function can be feathered much easier. It is also the most likely source of hydraulic spikes.

 

[LessWhole]

I'll discard that post as noise, then. Really appreciate you lending your experience to help me here.

So the procedure for the pressure/relief test is to use a tee to get a pressure gauge inline with one of the tilt cylinder lines, then move the bucket around such that I slowly add load & pressure to the line with the gauge tee'd in. That makes a lot of sense, and it's less scary than connecting the gauge directly to a line (and disconnecting the cylinder).

Again, thank you so much. I'd love to buy you coffee if you're ever in Boise.

 

[John C.]

The wife has a couple of relatives there we sometimes go to see. I'll keep you in mind if heading that way.

Good Luck!

 

[heymccall]

The aluminum bodied gear pumps in all 3 of my TL150 machines cracked their mounting flange in the 5 to 6k hour range. Replacements still going strong.

 

[jacobd]

As others have said, an aluminum gear pump with 6000 hours on it has lived a full life. And if by some incredible circumstance the relief valve on your machine did fail in such a way as to cause an overpressure condition it's likely a hose would burst before the pump exploded.

 

[LessWhole]

Thank you all for your help. I installed the new pump last night, and everything seems great. No leaks so far. That pump is an enormous pain to install, but we got it on! Drive/lift/tilt all okay, relief valve working as expected. Ya'll saved me a great deal of trouble trying to confirm the relief valve was working without just testing it.

Now I've got one more issue, seems electrical this time. If you're curious, https://www.heavyequipmentforums.co...electrical-issues-preventing-aux-usage.92194/.