How to make a Delaval nozzle

[old grayfellow]

Hi All,
I'd like to produce a "DeLaval" type nozzle which is essentially just a venturi sort of thing. It has a "convergent" section tapering from e.g. 3/8" diameter at the entrance down to e.g. 1/8" dia over a distance of like 2", and then a "divergent" section which again opens up from e.g. 1/16" to e.g. 3/16" dia over a distance of perhaps 3-5".
The nozzle can be made out of a simple piece of mild steel bar stock, and the constriction area does NOT need to have a rounded profile like a typical venturi would.

So my question is, how to best "machine" something like this?
For the convergent section I was thinking of just first using several twist drills of appropriate diameter and depth to remove the majority of material, and then finishing the taper using a stepless tapered drill bit chucked in the lathe to finish the profile.
The divergent section seems a bit more challenging though. What I had in mind was to use a metal file of appropriate profile chucked in the lathe to finish off the taper. Since this is such a small diameter and deep hole, I don't know of any way of turning it using traditional lathe cutters or a boring bar, and also don't know of any tapered drill bits with that kind of long and narrow taper. The file might take a while to remove that stock, but I don't know of any other type of cutter that would give the right type of taper at that diameter.


Any suggestions?

 

[just Dave]

Did you look at taper pin reamers? Might be close?

 

[DDoug]

Hi All,
I'd like to produce a "DeLaval" type nozzle which is essentially just a venturi sort of thing. It has a "convergent" section tapering from e.g. 3/8" diameter at the entrance down to e.g. 1/8" dia over a distance of like 2", and then a "divergent" section which again opens up from e.g. 1/16" to e.g. 3/16" dia over a distance of perhaps 3-5".
The nozzle can be made out of a simple piece of mild steel bar stock, and the constriction area does NOT need to have a rounded profile like a typical venturi would.

So my question is, how to best "machine" something like this?
For the convergent section I was thinking of just first using several twist drills of appropriate diameter and depth to remove the majority of material, and then finishing the taper using a stepless tapered drill bit chucked in the lathe to finish the profile.
The divergent section seems a bit more challenging though. What I had in mind was to use a metal file of appropriate profile chucked in the lathe to finish off the taper. Since this is such a small diameter and deep hole, I don't know of any way of turning it using traditional lathe cutters or a boring bar, and also don't know of any tapered drill bits with that kind of long and narrow taper. The file might take a while to remove that stock, but I don't know of any other type of cutter that would give the right type of taper at that diameter.


Any suggestions?

Post a drawing, with dimensions & tolerances please.

 

[just Dave]

Look them up. Do you have a Machinery’s Handbook? MSC sells them ,might have dimensions .

 

[Rickyb]

my drawer full of taper reamers don’t come close to any of the listed dimensions so I’d bore the larger taper with an 1/8” solid carbide boring bar, d bit profile. The divergent section with the same bar to 1/8” dia then make a tapered reamer to finish The 1/8-1/16”.

 

[DDoug]

Look at sandblast nozzles

 

[old grayfellow]

Thanks for the ideas.

I had seen reamers as an option but couldn't find any easily that had the right dimensions/taper for what I was after.

For a little background, what I'm trying to make is a "cold spray nozzle", for making an apparatus to accelerate metal powder to supersonic speeds for coating/making parts (interesting technology). You can't just go out and buy such things for under 5-6 figures, but they're at heart pretty simple, so therefore I must make one..
Anyhow, attached are a couple of diagrams from a paper that tests various measurements and ratios to optimize nozzle design, and also a preliminary model I made in sketchup (already out of date however). In the paper they test a number of different dimensional variations, so there's some flexibility in dimensional ranges (within limits).

I thought of sandblasting nozzles also, which would be an easy off the shelf solution, however the nozzle geometries, shapes, and lengths are not right to give the supersonic speeds needed. Seems you need a DeLaval type nozzle, with an optimal exit/throat area ratio in the range of 4 to 6.25.
For throat a tad over 1/16" dia, that would want an exit somewhere around 3/16 (if my quick math is correct). And the divergent section has to be fairly long, at least a couple inches (longer is better).

I'm at the design stage so the exact dimensions aren't fixed yet, but the above is a rough starting point.
I did also find a file made by Belsaw that looks to allow profiling the divergent section in one go.

Also I did initially think of a 1/8" throat for using larger tooling, but seems like the air consumption of the device at 200-400psi would be pretty huge with a 1/8" throat..

 

[Garwood]

Could you WEDM it in two pieces and stick them back together?

 

[DDoug]

Boron Carbide Venturi Nozzles | Sandblast Supplies

This venturi design will increase your rate of production up to 50% over straight bore nozzles. Shop Miles Supply for all your sandblast needs!

www.milessupply.com

 

[Scruffy887]

Do you really need this from steel? If so lots of luck to you.
I make both venturi nozzles and the wasp waist part from 360 brass and press them together. SHABLAM! Tiny wisp of 70 psi air generates 27" of mercury.
Yes, easily make my own taper reamers too.

 

[Limy Sami]

Make your own Dbit reamers - H&T 01 drill rod will do nicely, just run nice n slow.

It's worked for me (more times than I can remember)

 

[Erich]

400 psi air, metal powder accelerated to supersonic speeds, how long do you think a mild steel nozzle will last? Especially since performance is a strong function of the shape of the nozzle.

 

[Cole2534]

400 psi air, metal powder accelerated to supersonic speeds, how long do you think a mild steel nozzle will last? Especially since performance is a strong function of the shape of the nozzle.

Hopefully through the trial phase. =)

 

[old grayfellow]

DIY Dbit reamer sounds like a good way to go, especially if a piece of drill rod can be turned down to the right taper for a single cutter to make the whole profile in one go. Except not sure if I'll be good enough to cut the Dbit part accurately enough when we get down to the 1/16" end of the reamer (I have no access to CNC equipment, just WWII-era mill and lathe). Guess there's one way to find out..

About the vacuum/venturi point, this is something I still need to read up on for this particular application. With sandblasting I gather that the media reservoir is usually under atmospheric pressure and just gets pulled in by the vacuum from the venturi effect. But in the Fig 1 from the paper they mention a "Powder carrier gas inlet" as separate from the "Driving gas inlet". So I take that to mean that the powder has to be carried by pressurized gas as well. I was thinking to make the powder reservoir pressurized at the same level as the driving gas. If it wasn't pressurized, it seems like the powder might be driven in reverse despite venturi effect. 27" of mercury equates to around 13.3psi, but the driving gas delivery pressure is 200-400psi.

I will check into the sandblasting nozzles, thanks for that. Saw that they have some down as small as 5/64" which should be a good throat diameter. Will have to ask the vendor about the divergent profile and lengths available. Would be nice to buy something off the shelf that works.

About the durability point I'm not sure, but have a feeling it might be a different situation than when blasting abrasives. The principle of the cold spray device is that it actually sticks the metal powder onto the target object, so I almost wonder if we might not have the opposite problem of metal accumulating on interior of the nozzle rather than eroding it, especially when shooting powders of soft metals like copper or brass (but I don't know).

 

[old grayfellow]

Scruffy; I must make one clarification about the pressure vs. vacuum issue and possible backflow of the powder I mentioned:
Of course as you said you are getting 27" Hg of negative pressure using 70psi positive pressure with your venturi. But in the diagrams i posted from the paper (and their dimensions that I didn't show) they have their powder tube outlet well upstream of the throat which leads me to believe it would be in an area of positive pressure, rather than right in or very near the throat like a normal vacuum-generating venturi apparatus would do. So I'm guessing they're not able to utilize the venturi effect to suck the metal powder in and hence they are using a (pressurized) powder carrier gas.

 

[just Dave]

Maybe you need a core to cast /mold ceramic around.?

 

[Limy Sami]

DIY Dbit reamer sounds like a good way to go, especially if a piece of drill rod can be turned down to the right taper for a single cutter to make the whole profile in one go. Except not sure if I'll be good enough to cut the Dbit part accurately enough when we get down to the 1/16" end of the reamer (I have no access to CNC equipment, just WWII-era mill and lathe). Guess there's one way to find out..

About the vacuum/venturi point, this is something I still need to read up on for this particular application. With sandblasting I gather that the media reservoir is usually under atmospheric pressure and just gets pulled in by the vacuum from the venturi effect. But in the Fig 1 from the paper they mention a "Powder carrier gas inlet" as separate from the "Driving gas inlet". So I take that to mean that the powder has to be carried by pressurized gas as well. I was thinking to make the powder reservoir pressurized at the same level as the driving gas. If it wasn't pressurized, it seems like the powder might be driven in reverse despite venturi effect. 27" of mercury equates to around 13.3psi, but the driving gas delivery pressure is 200-400psi.

I will check into the sandblasting nozzles, thanks for that. Saw that they have some down as small as 5/64" which should be a good throat diameter. Will have to ask the vendor about the divergent profile and lengths available. Would be nice to buy something off the shelf that works.

About the durability point I'm not sure, but have a feeling it might be a different situation than when blasting abrasives. The principle of the cold spray device is that it actually sticks the metal powder onto the target object, so I almost wonder if we might not have the opposite problem of metal accumulating on interior of the nozzle rather than eroding it, especially when shooting powders of soft metals like copper or brass (but I don't know).

File the thin part of your Dbit