Insane Hydraulics
Bold and audacious blog about fluid power
This article is a follow-up to my previous post on how to DIY single-acting spools for mono-block DCVs, where I discuss the main principle behind the single-acting spool of a typical lever-controlled mono-block directional control valve. A single-acting spool does not close the pressure by-pass gallery when it is shifted in the direction that connects the work port to the tank, thus allowing gravity-powered loads to be lowered without raising system pressure (which would be the case if you used a double-acting spool with one of the work ports plugged), while also keeping other spools active (i.e., capable of soliciting the pump's services as needed).
While this principle is perfectly valid, and I firmly stand by my thesis that you can turn a classic double-acting spool of a mono-block DCV into a single-acting one by widening the by-pass groove in one of the directions (you can simulate this using the controls under this interactive drawing), it is super important to note that a practical application of this principle can be tricky—much trickier than it may seem when you look at my idealized drawing. So, this is what I will be talking about today, and, as usual, I will be presenting a real-life story that taught me these nuances "the hard way".
Check out these mono-block DCVs:
The one on the left is an original Walvoil SD-5, and the other one is a Chinese clone. Around here (the Alentejo region) such two-spool models are universally used to operate compact front-end loaders on tractors, and we try to keep both of these models in stock - a good one and a cheap one - to be able to satisfy all kinds of clients. I must say that I've never had a single issue with the Chinese model - not once - their quality is actually pretty decent, especially for the price.
Most of the smaller front-loaders tend to use double-acting cylinders for both the arms and the bucket, and therefore require no special spools. However, that time I had a client who wanted to replace the DCV of the front loader on one of his older tractors, and that particular loader was built with single-acting cylinders for lifting the arms - so he asked me if I could find him a two-spool solution, and he also made a specific request - he needed the single-acting function to work with port "B" (the one that's next to the lever box) - for the lever movement "to make sense" (lever in = arms up, lever out = arms down, because this is a move-line spool).
Now, I already knew that to change a spool on an SD-5, I needed to use an original Walvoil valve because the spools and the cavities in Chinese models were about 0.005 to 0.003 mm tighter for some reason, which made fitting Walvoil spools in the clones virtually impossible, and so my logical solution was to fetch an original two-spool SD-5 and an original single-acting spool. What could be simpler, right? Just a matter of choosing the right spool.
This is the point where I will allow myself to drop a critical remark about Walvoil's technical literature. It is not bad at all, but it often feels incomplete because a lot of the stuff that they manufacture and sell is not in their docs, and not all of their docs are available on their site. Let me show you: if you google for a Walvoil SD-5 catalog, you will get a link to the official website where you can download this document (3rd edition from 2012). You will not find a single-acting spool for the B port in it (see the available spool options on page 10), only spool No. 3, which is for port A.
Theoretically speaking, if you dig further through this file, you will eventually find a single-acting spool for the B port, but only for the version with a right inlet. This version has the pressure by-pass gallery located on the spring side of the spool, meaning it can't be used on "normal" SD-5s with a left inlet. I call them "normal" because we've always stocked only left inlet models.
But then, if you google some more, you'll find a short 12-page SD-5 file called "compact catalog", stating that it "...shows the product in the most usual configurations..." And there, you will find a similar page clearly giving you a part number for the single-acting spool for the "B" port - spool No. 4, part number 3CU1235120. Why can't this file be easily found on the official SD-5 page? Why is there no mention of spool No. 4 in the other catalog? I guess we'll never know.
Anyway, I had the part number in hand, so I checked our stock, and... bummer! We only had spools No. 3 here. There were a couple of No. 4s in Lisbon, but I could only get them the next day, and I wanted to solve the man's problem fast. So, I didn't have the right spools, but I knew how to DIY single-acting spools, and I had a lathe in the shop, so I had a brilliant idea: take one of them clones and convert it. "It'll take ten minutes," I told the man.
Now, the first tricky part when you are doing such a conversion is knowing how much you can hack off of a spool. Once again, the idealized drawing with wide gaps that I used to illustrate the principle is not very helpful here, because it can give you an impression that you can safely grind off a full spool's travel worth of spool, which in most cases would be too much. So, I first measured the width of the groove on a No. 3 spool I had in stock, and then I widened the groove to the same width on the "clone spool," only in the other direction:
My DIY angle grinder attachment for the lathe made short work of it. I quickly reassembled the valve, gave it a quick test with compressed air and... saw that it didn't work! That is, not entirely. It did work, but it was also "overshooting." In other words, it would work as intended up to about 75% of the spool's travel, but when the spool would shift all the way, it would vent the work port to tank when it was supposed to be pressurized. It would also shut off the by-pass gallery (which would cause the pump pressure to rise to the relief valve's setting) when the work port was supposed to be tanked. My initial thought was, "I should place a shim between the spring cups to limit the spool's travel and be done with it." But then I thought, "It probably wouldn't feel right to have two levers on the same DCV with different travels. It should be working with the standard spool travel, shouldn't it? After all, the short catalog that included the part number for the No. 4 spool did state it was containing '...most usual configurations...'" There must be something about the clone I am not seeing right. So, I went back to the counter and told the man that he would have to come back tomorrow, when the original No. 4 spool would arrive, and take the pricier Walvoil SD-5 after all. I decided to wait for the original spool and compare it to my DIY spool to see how they differ.
The new and shiny spool arrived the next morning, and I was surprised to see that the two spools were virtually identical. By pure luck the cut on my DIY spool exactly matched the original:
Note that the Walvoil spool has a narrow tapered band for throttling, while the cloned spool doesn't. But if you consider the total shut-off width - it's exactly the same on both spools.
Then, I assembled the No. 4 spool in the original Walvoil body, and it actually worked, but not perfectly. The pressure half of the travel was no longer by-passing the work port to tank. I even tested it on our test bench and confirmed that I could get to high pressure without internal leakage. The tank half of the travel was keeping the by-pass gallery open, but at the very end of travel it would still throttle it a little: the pressure in the P line would increase to some 40-50 bar (at 16 l/min). Not ideal, but functional. So, I set the relief and delivered the valve to the client.
Now, why did the clone was over-shooting so badly? First - I decided to check the spool travels - and indeed, I saw that while Walvoil centering kits were providing about 5.2 mm of travel, the clones did a whopping 6.1 mm, and that's a big difference!
Then, I tried original Walvoil positioning kits on the clone and saw that it was behaving a lot better now, but it was still overshooting slightly at the very end of travel. Obviously, there was a tiny difference in the configuration of internal channels between the cloned and the original Walvoil bodies. Once again, the most obvious solution would be to limit the spool travel, but neither of the catalogs mentioned anything about that.
So, I searched some more and discovered that some of the sites that sell No. 4 spools for SD-5s also include a note that this spool requires a special positioning kit with a shorter stroke of 4 mm (part number 5V08105022). Finally, all of the pieces of the puzzle fell into place. Why is there no mention of that in the official SD-5 PDF? Only God knows! But in my notebook, Walvoil just got another frowning face emoji next to its name.
I also know now what I'll do if the client who got the valve from me complains that it's bypassing: I'll simply reduce the spool travel by a millimeter or so.
So, here are the two main takeaways: