Results 1 to 6 of 6

Thread: Coselle concept

  1. #1

    Coselle concept

    I have decided to use air pressure instead of steam on my little steam project. Yes I am loosing the expansion. But the simplicity and ease of running multiple compressors just makes the design more simplistic. But the concern of volume comes into play as well as increasing the PSI.

    Now at 1200psig I have more then enough Psig to get things moving and pass the production of a 4 stroke design. But the container I was planning on using is a Natural gas container letting me run into the 2500psig range.

    Then I have someone text me about the Coselle concept. While construction is much more complex. The design seems like it would provide a much safer container. Throwing this back into steam reference a Water tube boiler vs fire tube. The water tube spreading the load out is safer every time. It is safer and can handle much higher stress then the fire tube. Yet Natural Gas containers are still using a container design instead of a wound one. Probably for cost.

    Here is a very quick animation of a Coselle container.
    https://www.youtube.com/watch?v=tfcqdMtqVlY


    Right now I am looking at break line.
    http://www.autozone.com/brakes-and-t...ne/602041_0_0/
    3/16 line in 20 foot lengths rated for max load of 12500psig.

    At my current (still working) valve size with a 10 degree duration I need 4 (dry) gallons of volume to maintain 15000 rpm.


    Using the propane tank calculator
    http://www.engineersedge.com/calcula...calculator.htm

    I figured for a od of 3/16 and a 1/16 wall with 200 feet of length and .03 ends both elliptical I end up with 4.59 gallons of volume at a weight that will easily be under 20 lbs.

    I feel I can not only increase my pressure to around 2000psig safely. But I can increase the volume to to over 20 gallons using 6 compressors with a safety shutoff at 400psig feeding a manifold that then feeds into 6 directional 3000psig one way check valves filling the 6 200 feet of line going to a manifold with a ball valve(throttle) that then feed the 12 lines into the top and bottom of the working cylinders.







    I am far from an engineer and maybe there is something obvious to one of you I am overlooking.
    In my head this works out

    I am well within the working limits of the line. The volume is there, the pressure is there, the simplicity is there, even redundancy is built in in the event of compressor failure.



    I would like some feedback. The actual building is still a ways off. I have designed a more compact simplistic better sealing valve design and will probably design a few more before actually starting construction around 10/17. But, I would like feedback on if this Coselle idea is valid or if I would be better off using a more generic natural gas container.

    Thank you for any helpful information or constructive criticism.
    Last edited by dn24210r; 12-25-2016 at 08:00 PM.

  2. #2
    Lead Engineer
    Join Date
    Aug 2013
    Location
    Houston TX USA
    Posts
    421
    One possible issue with using the very small I.D. tubing might be maximum flow rate at which the air will be able to be delivered to the engine cylinders.

    At the same time, it will be necessary for your compressors to continuously deliver the volume of air at the rate required to run the engine; and, the inclusion of a storage vessel of any significant volume in the system would seem to be unnecessary or redundant.

  3. #3
    Flow rate I got down. .25 orifice(2 valves 1/8 each)x12 cylinders(6 double acting)

    At 6000rpm(max) at 1200psig I will be using .08oz of volume per cylinder. That's .96 ounce




    At 3000psig
    300psig through .25 orifice=14.626m3/h

    14.626m3/hx12(CYLINDERS)=175.512
    175.512/18(20 DEGREES TOTAL DURATION)=9.75067m3/h
    Convert to gallons=42.930908
    Convert to ounce=5495.156224oz3/h

    5495.156224m3/h/60(MINUTES)/60(SECONDS)=1.527oz per injection per second
    1.527ozx6000rpm=9158.5937067oz
    Convert to gallons=71.55gal per second



    Wait something is wrong I think.




    I think I am double dipping on the rpm. The orifice was figured open(all the time) .25 orficex12 cylinders then I took the 20 degrees of duration out of it(5 degrees before and 5 degrees after TDC/BDC) then I multiplied that by 6000rpm. But if it is open all the time I should not need to multiply it by the 6000rpm right?


    Someone please check this. If it is open all the time not factoring in the 6000multiplication then were talking 1.52 ounce/per second at 3000psig. or .7125 gallons per minute.

    I feel that 6 compressors that are putting out 7cfm/300psi will work making about 1800psig/min. The If I have 4.5 gallons that is a run time of 6.31 minutes and a recovery time of 2 minutes. It should work. Unless the 6000rpm thing comes into play.

    But I think I am factoring for full run time with the open orifice so the rpm should not be relevant. Right?

    Thanks for any help.

  4. #4
    Lead Engineer
    Join Date
    Aug 2013
    Location
    Houston TX USA
    Posts
    421
    You talk about a .25 orifice but your Coselle storage uses .0625 I.D. tubing and this will control its discharge flow rate regardless of your valve size.

    With regard to opening time:
    First, the total time open and flow rate is dependent upon how much total pressure you want in the cylinder at the end of its stroke. It appears to me, and I may be missing something, that, as opposed to a combustion expansion engine, for your engine the maximum engine power would be achieved with your intake and discharge valves each being successively open for just under 180 degrees of a cylinder rotation.
    Second, if this is a single acting piston design then you will be required to deliver the cylinder volume on each rotation; but, if it is a double acting piston design then it will require 2x the cylinder volume per each rotation.

    With regard to running time:
    If your compressors are not able to deliver at the full flow rate at your design pressure then the engine will not be able to deliver a constant level of power/speed over any duration of running time.

    Just as a suggestion, if you were still designing for steam then it would require calculating in terms of mass delivery "lb/sec, etc" but since you are now working with a gas "air" you can reduce your units conversions by simply using volume delivery "cu in/sec, etc".

  5. #5
    Here is my layout that I have been kicking around in my head. Just a whiteboard sketch to get things across.



    One motor running 6 pumps with 6 one way 3000psi check valves pumping into a manifold that fills the coselle cylinder that then runs into another manifold/fuel rails that then run into the cylinder valve head at the top and bottom of the cylinder.

    Here is a early version of the valve based on a open source engineering design. found in this link;
    http://tinyurl.com/zpan7rp







    But I did not like the idea of having the bump valve wear area right on the cylinder.


    Here is my newest more compact valve design


    The valve is more compact with a greater sealing surface and the wear area is completely in the valve. This allows the valves to be replaced as they wear down vs replacing the piston. There are two valves but I am only showing one. Also the top and bottom of the piston are shown if the thing is flipped. A bit of cross section is referencing this. This is just a sketch. I am working things out still.



    The valve is bump style and is opening 5 degrees before and 5 degrees after TDC/BDC. The cutoff is at 5 degrees down the cylinder stroke or roughly 2.77% of the effective stroke length and the same cylinder vol cutoff. The inner cylinder diamater is 2.25" and the central exhaust is at 90 degrees with a size of .20".

    This is a inline 6 engine using a jeep 4.0 short block as a basis with cylinder banks grouped as 1/6, 2/5, and 3/4. I fire both banks(1/6 etc) at the same time in this single stroke design. Every 60 degrees(50) it fires and exhaust offset by 90(ish) degrees. No oil is used within the working cylinder. I am still on the fence of using steam or not. The working piston rides on bearings at the top and bottom of the cylinder which allows it to come within a few mil(still working on expansion) of the cylinder walls. The working piston design is a dual inverted dome at top and bottom in a effort to lessen blow by. Seals at the top of the cylinder will prevent any oil entering the cylinder while allowing me to use oiled bearings on the working cylinders main shaft.

    The non powered slave cylinder within the shortblock retains all the oiling in which the crank, bearings, etc will remained oiled via a oil pump. This non powered cylinder is the basis for the main rod keeping it moving upon a vertical path allowing less friction.




    You are right though. This started out with me only wanting 1200psig steam and that was well within the safe working limits of 3/8 line. You can see how the design could easily use a horizontal lamont boiler getting rid of everything before the coselle tube.
    Trying to increase the air pressure to 3000psig to run air(still loosing power) has me relying on that manifold as a area.
    I have seen a few 3/8 line that advertises it is rated for 3000psig breaking pressure. But the fittings are not. I think I may need to look at scuba fittings which are easily rated for 3000psig.

    To stay on topic
    Lets assume that I have safe working pressure line that is 3/8 od with fittings that are also well within working pressure. Is a coselle concept still valid or would I be better off using a tank from a safety aspect. Obviously I am loosing working volume. But the safe(er) operation is what I am after. If I can contain a bursting line vs trying to contain a bursting tank then it is worth asking if the concept is valid or not.

    Once again the volume requirements are small. I do not feel the rpm is a factor as I think I have already calculated for full open operation then took 20 degrees out per cylinder. But i think I may be wrong also and need to include enough volume for that rpm max.
    Attached Images Attached Images
    Last edited by Kelly_Bramble; 12-30-2016 at 07:05 PM.

  6. #6
    I think I have found a way to get the pressure I want while having the safety I also want.

    sCO

    2

    So this whole working off air pressure thing is out.

    Steam from water is out. And steam from supercritical CO2 is in. Still working on understanding the basics of this.

    Topic closed as I am going way off topic with my little understanding of supercritical c02 generation.


    Thank you for all your help

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •