Insane Hydraulics
Bold and audacious blog about fluid power
In this article I want to talk about some design details of the Hydraguide HGA series steering valves, which are super old-school:
We still (occasionally) get these from local farmers, usually under the pretext of something along the lines of "...we only installed new seals, and now the steering doesn't work, for some reason..."
Around here, these steering valves are universally known as "TRWs" because of the three-letter logo on the side of the housing:
I found a couple of catalogs online, and the oldest one mentions that Hydraguide is the Ross name given to hydrostatic steering systems. I actually looked it up: Ross Gear & Tool was acquired by TRW (Thompson-Ramo-Wooldridge) in 1964, which makes me wonder how old this design really is. Later, these valves became part of the Parker group when Parker Hannifin acquired TRW's Ross hydraulic motor and hydrostatic steering business in 1993. In 2018, Parker divested the whole Hydraguide product line to QCC, LLC - their long-time trusted partner. This is where these steering valves, apparently, still "live" as of now (November 2025).
The steering valve that I have here is, actually, somewhat special - it's a 27A0900002-1977 (if 1977 corresponds to the manufacturing year, it's older than me!). The special thing about it is the A09 size, which is never mentioned in the standard HGA lineup. I suppose it's one of the customary OEM shenanigans (Ford, in this case).
The poor thing had definitely been abused - the shaft ball was in the wrong hole, the spring washer was on the wrong side of the thrust bearing, the seals had been replaced but the steering was still leaking badly, and the spacer had been, apparently, hammered (God knows why). Well, no problem! All I need to do is disassemble and then reassemble it correctly, and as I do that, I'll be talking about points I find interesting, and I'll even show you a short video at the end.
The first thing I want to say is how ingenious this design is, especially considering that this steering valve was designed at a time when no CAD software was around. The main idea is, of course, the same - it's the good old hydraulic follower, only in this design, the oil is being distributed not by a rotary spool, but by a linear spool (which also happens to rotate, but the oil distribution is handled by its linear movement). The axial displacement of the spool is controlled by the spiraled groove on the input shaft, which interfaces with the spool via a single steel ball:
I didn't take any close-ups of the spool, but you can (kind of) see it on the left side of this picture. It is quite complex - and it needs to be because it is, essentially, a 6-way 3-position proportional valve that channels oil not only around it, but also through its splined center bore (images taken from this service manual):
But the coolest thing about this valve, in my opinion, is the orbital motor that provides the rotary follower action. Unlike modern steering valves, which switch the orbital timing when the steering direction changes, this rotary group has a fixed timing, provided by a thin commutator plate with a couple of cavities. This plate is rotated by the pin located at the end of the drive link. It is, essentially, a super-thin orbital motor, and the two lines that feed it are the hole through the center of the rotor gear and the channels around the bolts! Let's have a closer look - I think this part truly deserves it. This is the body and the spacer:
The two lines that go to the orbital motor are the spline bore through the center of the spool/shaft and the hole on the face of the body (the one between the threaded holes). Note how the flower-like groove on the spacer channels the flow from/to the face hole around the bolts This is how it looks when it is in place - the center bore is line A, and the seven holes around the bolts are line B.
The rotary group goes on top of the spacer, and it is then covered by the manifold that channels the oil to the seven orbital cavities. Finally, the commutator plate goes on top:
The two cavities of the commutator plate sliding over the seven slots of the channeling manifold provide the fixed timing. I find this solution super interesting. Here you can see the commutation cavity through the hole in the center of the manifold:
Now, let me show you (and possibly my future self) how to correctly connect the input shaft to the spool. First, put the ball into the correct cavity of the spool (if you're not sure, it's the smaller one; the other is for the ends of the clip spring). Next, put the spacer and the spring washer over the spool, and then "thread" the spiraled cavity of the input shaft into the spool:
Then use the torsion bar to make a gap between the spool and the thrust bearing, and then stand the spool vertically, like so:
By the way - here's another clever solution - the long and thin steel torsion bar, which functions as a centering spring. You can see that I marked the spool splines aligned with the input shaft splines. Now all you need to do is drop the drive ring aligning it with the marks, like so:
And finally you can install the spacer and the torsion bar:
That is it - very easy. There is, of course, the matter of setting the correct position of the spool with shims - but this is something that I really want to show in a video - especially how an incorrect amount of shims can affect the operation of this steering valve, but before I do that, I want to say a few words about the design solutions that are pretty bad.
Take a look at these sealed plugs:
They are very hard to coax out - you need to come up with some small hooked levers for that. The one on the left can only be pushed out by removing the relief valve and jamming a bent wire through the hole that leads to it. And you do need to pull them out to replace the seals. I don't understand why the designers didn't make a threaded hole in each of them - that seems like a logical choice to me.
Then you have the front cover, which has no centering in relation to the body, and the four holes allow for a lot of play around the bolts:
This means that every time you mount it, you need to make sure that it is not crooked. I don't think it would have been so hard to machine a pilot that would align it automatically.
And to conclude, my friends, I would like to show you the shim-based adjustment of the spool position. The shims between the outer race of the thrust bearing and the front cover determine the position of the spool in relation to the body. If the spool is offset incorrectly, one of the steering directions becomes "more dominant" than the other (to the point of having a tendency to steer on its own). It's actually all in the service manual, but since a picture is worth a thousand words, I decided to make a short video that shows what I am talking about: