Insane Hydraulics

Site theme image

Komatsu LPD45+45 Pump from PC88MR-6 Excavator - Additional Information

This article is an addition to my post on the Komatsu LPD 45+45 pump from PC88MR-6. I want to talk about two important details that I forgot to mention last time, do a little bit of show and tell, discuss an interesting matter raised by a reader, and also leave a small tip on how to determine the correct o-ring width for a boss seal with a 45º chamfer.

So, after I published the Komatsu article, I received an email with a very alarming piece of information about the pump - "...the mini & maxi stroke limiter screws are very critical and dangerous to touch..."

I must say that I love getting emails like that. The email came from a PC50MR-2 owner, who had himself rebuilt the pump of his excavator, and I give him props for that! The man stated that unscrewing the min displacement screw of his pump would cause the swashplate to collide with the rotary group components, and sharing this information would, quite obviously, be important because after reading my post about the Komatsu pump stalling the engine, where I mention how I confirm the swashplate stuck at max displacement by "feeling" the adjusting screw, one could get an idea that it would be safe to "fiddle" with the displacement limiters at will.

First of all, even though -

- I still wanted to check and double-check if this "stroke limiter peril" was real for the pump I had on hand, lest I put in jeopardy an unsuspecting reader of mine (disclaimer or not). And, by the way, if anyone reading this has gone through a similar experience or has an input on the PC50 pump (or any similar pump that can carry the limiter pitfall - please, please, shoot me an email - I would love to hear about it, and we all could learn something new.

So, let us start with the cutaway view of the pump:

The cutaway shows the swashplate at maximum displacement, and then it also contains a thin-line outline of the plate at minimum displacement. You can tell that the swashplate and the respective limiters are depicted in a way that suggests that it is slightly angled at min and has some travel to go at max. Furthermore - the pump is officially named "LPD 45+45", but the hydraulic diagram from the service manual states the min displacement of 5+5 cc/rev and the max displacement of 41+41 cc/rev, which aligns with what is depicted in the cutaway view. But before we go into more details on that, let me get out of the way the two important pump details I said I'd add today:

First - the design of the displacement limiter screws and the body:

The screws are sealed with o-rings, which means you can adjust them leak-free (which is a good thing) and most importantly - the screws and the body are designed so that they have limited inward travel! In other words - you can not screw them "too much in" even if you want to. This is very important to know because if you are only used to "normal" displacement limiters, which are basically long set screws that can go all the way in - you may interpret the fact that one of these unexpectedly bottomed out (i.e. hit "something hard") after but a couple of turns as something abnormal.

Second - I said in my last post that the swashplate of this pump had no mechanical bias and required system pressure to stay on stroke - but this is not entirely true. When the pump is assembled, the spring inside the barrel does provide a positive bias to the swashplate for the very same reason the pistons exposed to the high pressure do - the offset axis of the swashplate. Just look at how the rotary group "drops" when the swash plate angle changes from zero to max displacement:

The bias force is pretty weak, but it is present, and when you assemble the pump, you do see the swashplate gain an angle when the barrel spring is compressed - which means that this pump does have some self-priming capacity after all (which in no way removes the need to bleed the air from the bleeding screw on top of the control before you start it after an oil change or an overhaul).

Anyhow, with that out of the way, let us get back to our displacement limiters. These are the max and min angles the swashplate can physically achieve:

You can't see much because the pistons are in the way, so let us leave just three pistons and have another look, first with the swashplate at min angle:

As you can see, as the swashplate swivels back to what essentially is a zero angle, the piston shoes, the retainer plate, and the ball guide are "floating" well above the swashplate surface - and thus no collisions are possible! I am quite relieved, actually. If you look closely, you will see a mark on the "nose" of the barrel (the part that goes inside the ball guide) which gives an idea of how deep the barrel "sinks" into the ball guide when the pump is assembled and the barrel spring is compressed. I checked this with old parts, I checked this with new parts, and, of course - me being me - I already arranged with the client to go and test the pump once the mechanic has it mounted on the excavator.

Now, the max displacement limiter is another deal - you can tell from the cutaway view that the notch in the ball guide, indeed, comes very close to touching the inner edge of the hole in the swashplate - and you can see this when you try it with real-life pats as well. I painted the notch red to see if I could detect any scratches or contact points - but no matter how I tried - I saw no marks on the paint. It does come pretty close though, but the gap is still visible:

After I assembled the pump - I used threaded rods to push the swashplate to the ends of travel and then rotated the shaft to see if I could feel anything "funny" - and, once again, saw nothing abnormal. The client collected the overhauled pump yesterday - so I should be testing it in a couple of weeks from now, give or take. I promise I'll report on the results of real-life trials even if I break the damned thing fiddling with them screws!

There's one last thing I wanted to talk about you today - the boss o-rings and how easily they can fool you if you are not careful. They surely fooled me! Allow me to explain what I mean by that. Here are the new parts that I got for this overhaul:

The "hard parts" were all good, but the seal kit... well... I think I used 4 or 5 o-rings from it at best. Even the shaft seal was of the wrong size. It seems like someone tossed a bunch of random o-ring seals in a plastic bag and then called it a day. But the seal that really trickled me was the o-ring for the flange boss (the one that sits in the chamfer of the pump case sealing the front flange) - it looked and felt so damn nice that I installed it without checking its width, and it was only when I started to tighten the front flange screws that I felt "something strange" and saw a gap between the flange and the body, which is never a good thing (especially when you are applying aftermarket parts):

So, I re-opened the pump and measured the size of the chamfer for the boss seal (something I should have done before closing the pump), and it was 6 mm, and, of course, the o-ring included in the "pirated" seal kit was 3.53 and not the required 3 mm:

And if you are wondering why I'm so sure that it had to be a 3-mm o-ring - here's a nice tip for you - for a 45º chamfer, a circle with a diameter that is half of the chamfer's width (the diagonal I'm measuring in this picture) will have a cross-section area that is exactly Pi/4 the cross-section area of the chamfer - in other words - about 79% - which is a perfect cavity fill percentage for an o-ring seal! So, aim for an o-ring that is half of the chamfer's width, and you will be "golden" every time!

The cross-section area of a 45º chamfer cavity is chamfer width (diagonal) squared divided by four, so a 6 mm wide chamfer has a cross-section of exactly 9 mm². A 3-mm o-ring has a cross-section of about 7 mm², but a 3.53-mm o-ring has a cross-section of 9.8 mm², so even if it looked OK, it was too thick for that chamfer! Here's a drawing that may help you figure out the math (if you ever get curious):