During the time I was waiting for the parts, I was still thinking of other options that could salvage the original Stem/Seat assembly.
I thought of a 3 piece plastic seal that could be made on the rapid prototype printer. Unfortunately the printer was out of commission for some time (as we were waiting for a shipment of building material) so I started experimenting with melting plastic substances.
The coffee can cover that I previously used has the right level of hardness. It is not too firm and not too soft. I also discovered that a heat gun would melt it. The problem would be to get it in the brass cup.
Maybe another solution would be a glue gun? The material seemed firm enough and the pointy end of the nozzle would be able to inject the material in the restricted space. I went ahead and was able to form a fairly smooth surface by spinning the Stem/Seat assembly in my press drill right after injecting the glue and before it was completely solidified.
I went through the reassembly procedure, pressurization and…
… another failure.
The plastic from the glue gun is not hard enough and deformed under pressure.
Finally I received my new parts from my friend. After making a small jig and cutting the flats on the brass seat using a belt sander and some files this is the result.
With the necessary components ready, it was time to assemble the stem, the plastic seal and the brass cup.
At this point, I was a bit hesitant to cross-drill a hole through the brass piece and the stem because it seemed to me that the firing pin had nowhere to go. The larger diameter was preventing the pin going forward and when the hammer would hit the pin to open the valve, everything else will be stacked against CO2 and spring pressure. Following that train of thought I decided that only some Loctite was required to seal and keep in place the stem.
Loctite being a product that I don’t normally use, I had assumed that if I left it to cure over dinner time (a couple of hours) it would be long enough.
I discovered the hard way that curing is more in the order of 24 hours. Furthermore, it is impossible to have two bearing surfaces at the same time due to tolerance stacking and this was confirmed immediately after reassembly and pressurization.
Guess what? It was leaking again.
This time it was leaking less with the hammer on the firing pin than when it was cocked. This made me realize that effectively the larger diameter on the stem is not touching the shoulder in the valve block when the plastic seal is sitting on the valve seat. It is there only to prevent metal to metal contact when the seal is worn off.
The CO2 pressure pushed on the stem end (where the Loctite not fully cured was displaced) and leaked around it. With the hammer resting on the firing pin the flared base of the stem was pushing on the plastic seal, slowing the leakage but with the hammer cocked, the stem was moving forward until it stopped on the shoulder inside the body recess, and the gas flow was increased making more noise.
I was so intent in replicating the parts as they were originally made, that I overlooked the fact that with a sliding fit I only needed a blind hole in the brass piece instead of hole drilled all the way through.
The brass was drilled through only to give access to the end of the stem to push it out.
In my case a blind hole in the brass part would have solved my stem sealing problem.
Anyway, since it was too late because the part was made, I had to find a solution with the components as machined.
This time I wanted to get it right so I cleaned everything and reapplied the Loctite that would be left to cure for more than 24 hours. The plan was to add the crosspin as on the original assembly and it would also be Loctited. That would prevent any movement. To make sure that it would have no possibility of leakage, I finally plugged the end of the stem in the brass (spring side) with a dab of Goop.
The picture is a bit blurry. I should have checked it before reassembly but I forgot. So this is the best I have to show.
During the curing of the different adhesives, I continued to examine the valve body for other potential problems such as particles from my previous attempts or imperfection on the valve seat.
Being fairly deep it is difficult to have a good look at it but it seems that I was seeing some marks in the photo.
From a ½” plastic rod I decided to make a tool to polish the seat, then I discovered that the diameter deeper in the hole is smaller than my plastic cylinder. With an X-acto knife I started to shave it down and used a file to smooth out my cuts. To make sure that the end was square, I placed the rod in my press drill and pushed the tip of the cylinder against a sheet of sandpaper placed on a block. Once it was done, with a punch, I cut some disks of 400 and 600 grit sandpaper that were glued on the tip.
A few rotations by hand with the 400 then the 600 grit sandpaper seemed to do the trick and polished the seat.
Now the moment of truth was approaching and I was feeling anxious after all those previous failures. Nevertheless everything was reassembled. After inserting a CO2 cartridge and cocking the pistol, the screw was tightened.
Then, nothing! Not a noise!
It seemed that it worked. I pulled the trigger and was rewarded by a loud bang and a flawless recocking of the pistol.
Having used a generous amount of silicone grease and also some Crosman Pellgun oil, there was still a possibility that it would leak once all that lubrication was cleared out of the system. So I loaded it with pellets and proceeded to shoot the gun until it ran out of gas. I obtained about the same number of shots (32) as my other Crosman 451, so I took this as a good sign.
I repressurized it and shot another set but when it stopped recocking I left the residual CO2 in it and stored it. After that, I shot the pistol once every day to verify that it was holding and it was.
After a week still pressurized, I removed the cartridge and reinstalled a fresh one for another shooting session. At the end I left the CO2 in it and stored the pistol for another week. After that time I checked the airgun, it was still holding the pressure.
Mission accomplished, I am now confident that my Crosman 451 is fully functional.
In the meanwhile, the material for the rapid prototype printer was received and my pieces for the 3 part seal were made. Unfortunately the option has not been tested yet because the pistol is working and I am not too keen to take it apart now.
In theory it should work. The plastic from the high resolution printer is fairly dense and can be sanded to obtain a smooth finish. The parts fit snugly with each other and silicone grease should make the minute spaces in between practically leak free.
The following picture shows the pieces together to make a complete seal, the pieces scattered to show the individual components and another seal installed in the brass cup. The idea was to keep the original Stem/Seat assembly as taken out from the pistol and adapt a new seal in the cup. The two small pieces go around the stem under the flared shoulder and the ring can be sled over the large diameter of the stem to end up in the space left to fill the cup. Essentially the raised ring of the seat in the brass valve body should exactly fit on the surface of the one piece plastic ring of the seal.
I know that there are replacement kits sold for some Crosman models that come without a brass part; the plastic part is shaped as to fulfill the same combined functions as the brass cup and the seal and is directly attached to the stem.
I think that would be a good approach here too. When working on those different options I didn’t know yet that I would be successful (with the machined parts) before the 3D prints were ready. Because of the delay in receiving the material for the rapid prototype printer, this option has not been tested either.
To be prepared I made 3 versions. Both parts that have two tone of gray (in the pictures below) were to be used with the shaft made by my friend and the one that looked transparent would be used with a new threaded stem.
In all cases the end was plugged so the air couldn’t leak around the stem.
For the ones that would be assembled with my friend’s stem I made the dome version so it could be tried as machined. I was not sure if the dome would interfere with the CO2 tube inside the cavity so I made another one that is flat like the brass part but it would require shortening the stem a bit.
Underneath is a picture of the 3D printed result. The plastic part at the center has the dome; the other two have a flat end. The part on the left is oriented to show where the spring sits and the one on the right shows the sealing face that will contact the valve seat in the body. This one has also a smaller hole to be tapped to match the thread of a machined stem that is not done yet.
The preferred approach with this option would be to have a threaded stem and a plastic part that has the corresponding threaded hole. That way, it should be easy to take apart to replace the plastic component.
So if I was starting the project again these would be the only drawings I would need:
This project has also been a journey of discovery. It seems that no homemade solution is available but the one closest to it would be the 3 pieces of rapid prototype printed seal.
They are so small that it shouldn’t be too expensive to make and it has the advantage of using the existing Stem/Seat assembly without having to take it apart.
On the other hand, and before doing anything else, there might be in fact one homemade solution. It is the result of the whole journey and a combination of what I learned through it.
This is something I would try first if my other pistol starts leaking. Instead of removing the compacted worn seal, I would leave it in the cup as it would provide a firm base. To fill the gap between the old seal and the tapered flange of the stem I would use the plastic material of the coffee can cover. It would probably need 2 layers of plastic washers and, as experienced before, they will be stretched and deformed after passing over the larger diameter in the stem. I would then use a heat gun to practically liquefy the layers and make them flow to fill the space under the tapered flange. After this application of heat, the bearing surface of the top washer should become smooth again and hopefully the lowest layer would also adhere to the old seal underneath.
If it works, the problem is solved with a minimum of fuss and at practically no cost. There is nothing to lose to start with that option.