Building my own compressor- Need filter advice

[pescador775]

4130 is a good choice. I would guess that heat treatment is not necessary. Don't recommend you chrome plate it, but if it is necessary--like for use at sea-- tell the plater to use a 24 hour cook off to prevent hydrogen embrittlement. There is nothing special about a filter or condenser used to treat NITROX. It needs to meet requirements for high pressure service. A length of 12-14 inches should be about right for a final filter. Place a nylon sock in the canister and pack the chemicals using felt pads for separators. It is helpful to back up the pads with plastic disks. Use a hole saw and drill to make ventilated disks from a bucket lid.

 

[pescador775]

I made some filters from 4130 about 30 years ago using the same dimensions and wall thickness, 2 inch, 1/4 inch. I still have one. The steel works well as a final filter but when used for water separation the interior wall tends to rust. If I remember correctly, the burst pressure for such a vessel is about 15,000 psi.

 

[LakeTravis55]

I am still working on the compressor and BargainMarge has been very supportive. Pesky, I removed the original 3hp military engine, modified the framework and installed the 5 hp Honda engine. You were correct about the binding that occurs when you bolt on the honda engine to the S-W flamethrower compressor. I ended up making a spacer between the engine and compressor flanges out of .044 copper shim material. That worked to eliminate the binding, but Marge later told me that all she does is sand down the output shaft of the honda with the engine running and it slides into the compressor eliminating the binding without having to use a shim at all. Anyway, I am trying to attach the coil of 1/4" OD stainless tubing to the existing 37 degree flare coming out of the compressor. I tried to flare the stainless and all it did was push the tubing out of the flaring holder. After some reading, I found that a regular flaring tool won't do stainless (and was 45 degrees anyway). The only 37 degree flaring tool I can find anywhere on the internet, including Ebay, that says it will handle stainless steel tubing is almost $500. That kind of eats up the budget pretty bad just to make two 37 degree flares in stainless! Suggestions?

 

[pescador775]

Don't worry, the Ridgid 37 degree flaring tool shows up on EBay for around 50 bucks. Trust me, this tool will do the job. Retail is about $120. I forgot about the sanding trick. It works mostly with Honda engines for some reason.

Yesterday, I was talking on the phone and during the conversation I heard a "pop" over the line. My friend was testing a new SW compressor. During the call, he shut it down under pressure. After a few minutes the second stage burst disk blew. Apparently, some HP air leaked back from the third stage causing the problem. I've never heard of this before but I suppose it could happen if the third stage intake valve was leaking. Probably a fluke.

 

[pescador775]

You probably should go ahead and sand the shaft. Use metal abrasive cloth sometimes called "emory".

I just realized that Ridgid makes a lot of different flaring tools. You need the Ridgid model 377 flaring tool which has an orbiting head and automatic let off.

 

[LakeTravis55]

I just wanted to comment on Pesky's suggestion of using a Ridgid 377 flaring tool to make the 37 degree flares. After monitoring Ebay and Craig's list for a while and not seeing one, I finally went to my local tool store in south Austin and bought the tool. What a GREAT tool! It makes beautiful flares and automatically disengages when it completes the flare. It's very smooth and almost effortless, even in stainless. It was worth every cent of the $100 that it cost. Just for the record, the 37 degree flare is also used for automotive racing AN fittings ( which I also play at). "Mopars or no car."
Thanks for all the help guys.

 

[iain/hsm]

If determined to find an alternative to spending the $400 on a production filter shell just some design pointers on our current design standards.
1. A minimum burst pressure of 25,000 psig including
2. A minimum working pressure to burst pressure ratio of 4:1.
3. A minimum certified pressure reversal (loading cyclic test) of 100,000 cycles.

One of the reasons of failure of the older imported compressors and there filters has been the failure of the 1.8:1 and the 3:1 ratio designs allowed in Germany under the TUV classification rules when run over a 15 year cycle life. In dive/fire shops these filters have been used as both filter/separators. The pressure cycle when combined with auto dumps, can be up to 4 pressure cycles per hour. Add to this the 1000 hour per year use. You can see where the 15 year life design now comes from.

Incidently IMHO the 3000psig WP in 4130 steel 0.250 wall on a 2” id is insufficient due to the steel corrosion allowance. 2” OD calculates out OK.

Our works standard hand flaring tools for hardened stainless steel and the most suitable tool for the job is the Imperial Eastman 37° ROL-AIR Flaring tool model 400-F. This beauty includes burnishing rollers to polish the flare. Three rollers in the flaring cone roll out the 37° flare above the die block. This maintains the tube original wall thickness as required under Mil Spec. Do lubricate the cone well before use.

New price is $923.70 from http://www.newmantools.com/imperial/ft5.html. The pleasure in using it will last a lifetime. (well at least to my retirement) :14:
But I suppose you could also save more by buying a production filter shell $400.

 

[captain]

I have used tubing flared with a 45 degree flaring tool on 37 degree fittings for many years with no leaks or other issues. Again this is not rocket science. All these mil specs are all well and good when building aircraft and space ships. If you need to buy a $900 tool to do a flare or two what's the point in trying to do it yourself. After all about the worst that can happen is a air leak, it won't come crashing down out of the sky.

 

[oxyhacker]

Iains numbers here are a good reminder that when building DIY filter/separators, when it doubt (and especially when you can't do the math), aim small - a skinnier chamber will be usually able to handle more pressure than a fat one, and a smaller chamber will have less explosive force should it let go. It also means you will be changing the media more often, but this can be a good thing - media should not sit in the stack unused for too long, and many home compressors get so little use that the media sits in it too long. Now if the compressor is being used commercially, then that's another story - but then you really ought to have a proper stack and not a kludge!

Oh, here's some interesting numbers for seamless SS pipe (which can often be found on ebay quite cheaply)

Incidently IMHO the 3000psig WP in 4130 steel 0.250 wall on a 2” id is insufficient due to the steel corrosion allowance. 2” OD calculates out OK.