Prototyping a spinning cup oil burner

I was worried a nitrile rubber O ring won't handle the rotational speed or have significant friction, teflon ring seals exist but stretch and need tools to re compress them when fitted. An ideal O ring material would handle high speeds, be low friction and be "off the shelf" at a bearing/seal shop. Teflon would be ideal if I can fit it without stretching.

To quote John Lee Hooker: "It's in him and it's got to come out!". I've been thinking about this idea for a few years now and it's time to actually build and test stuff to find out the unknown issues and hopefully solve them with an easy to build unit, with off the shelf parts.

If I make it hard to build or need exotic special order parts then there's no real advantage to the idea.

 

Again, I'm not convinced you need a friction fit o-ring and/seal of any sort. Close tolerances are just as good in low pressure apps. Hydraulic mechanisms use tight tolerances rather than seals in countless applications and we're talking some serious pressure in those systems.

If I was forced to, I'd use an oil-lite material to fab a near friction fit union for the oil feed. It would wear to a perfect fit and the oil impregnation would prevent binding.

 

I definitely do not think you need any exotic oring. Remember, it is not the RPMs that will cause trouble, but rather, the surface feet per minute that the o-ring travels. In this case, if the o-ring diameter is at most 1/4 inch OD (should be plenty big enough for the oil port) at 20000 rpm it will be traveling only about 1200 ft per minute or 15 mph. It will be bathed constantly in oil. Try just a plain vanilla o-ring. I am confident it will hold up.

Now consider as a comparison the sealed needle bearing you intend to use. The rather ordinary rubber seal it wil use will be traveling two or three times as fast and will be retaining the grease meaning scant lubricant will be coating it. We all expect it to hold up for thousands of hours. I think the small plain-Jane o-ring will do fine.

I must say, the idea of using an oilite seal seems good too. Should be very durable and not that hard to machine to close enough tolerance. My personal preference would be a simple o-ring trial, and doing something more exotic only if the simplest solution fails.

 

Feeling that my above reply lacked objective engineering basis, I checked on o-ring design parameters and found this :

https://wp.optics.arizona.edu/optom...s/sites/53/2016/10/Parker-O-ring-handbook.pdf

This reference is a rich source of information relating to design application and use.

Under summary limitations of use they state:
A. Rotary speeds exceeding 1500 feet per minute contact speed.
B. Anenvironmentcompletelyincompatiblewithanyelas- tomeric material.
C. Insufficient structure to support anything but a flat gasket.

They expand on this as you read through the document. Other o-ring manufactures provide similar documents.

 

I have that Parker handbook, it's THE best online hydraulic and sealing design resource book, and free too. I'll clarify what those needle roller bearings are:, it's a formed steel outer race containing the needles and cage with no seals of any kind and no inner race, it relies on the shaft to function as the inner race, so running it on a shaft of what's probably 303 stainless with a non hardened surface is a bit of a punt. They are good for 17K RPM with grease and 24K RPM with oil lube, so a tiny oil hole in the fuel tube is an option.



Like Al2O3 suggested, a larger cup is likely to reduce the required RPMs of the shaft. I think the RPM would reduce linearly with cup increase to give a constant rim surface speed.

At this point I'll test the unit in my last drawing on the off chance it works in spite of it's weaknesses such as excessively long unsupported overhang and the relatively poor fatigue life of stainless steel. If I have three of these $5 needle rollers: two either side of the drive belt and one near the cup end, it will hopefully have less harmonic vibrations than just two bearings.

Assuming it works as intended, there'll be no seals of any kind and only two-three $5 needle roller bearings. A cup with twice the diameter makes 15K RPM operation feasible. If it goes into a death wobble and deflocculates/defenestrates I can salvage the outer tube and cup and move to the mechanically more complex/strong unit with two larger ball bearings on the outside of the 18mm tube.

 

This has been a fascinating thread! Watching with great interest. I will be curious to see if harmonics become the major issue with such long shafts at these rpms. How long of a shaft do you plan?
I have been thinking about another approach. What if you used a small high rpm motor and mounted the cup essentially on the motor shaft. That would place the motor very close to the furnace. However- the motor would be surrounded by a tube that carried the combustion air thereby cooling it. You could use only the motor bearings for support. You could introduce the oil tube through the wall of the air tube and drip it into the back end of the cup as shown previously by JV:


The cup could be long enough to extend through the furnace wall. This approach eliminates the extras bearings and shaft harmonics. What's the rpm of a small trim router? It would have a 1/4" collet. I'm thinking maybe a cup with a very short 1/4" shaft right into the router? Place the router inside an appropriate tube with a reducer that goes down to the appropriate size for the cup.
Robert

 

I can't take credit for that rather nice sketch. My sad drawings are either done on MSpaint or a napkin...lol

Although this time around I'm consciously aware of the large scale (my opinion) of Mark's burner, I think something along the size of the dremel isn't necessary. If he decides to follow the route of the above design. You can buy rather small / high rpm motors off of ebay rather cheaply. Not something I would recommend normally, but if using the above design approach, something that small should be able to handle the weight of the cup. Heck..., even put one bearing after the motor to help out with mounting efforts.

 

I think this is sound idea. I don't see why the drive shaft has to be so long and so far out from the furnace walls. My siphon nozzle burner is about 10" long before the blower connection and about 6" of this is inside the furnace walls. Nothing outside of the furnace gets very hot the burner tube is cooled by the combustion air.

 

The belt driven shaft unit would have the motor outside the forced air tube and inside a sealed cover to prevent forced air leakage where the drive belt exits the air tube. There may be air cooling issues. If I lose 10cm (4") of length going through the furnace wall insulation, then another 20cm (8") from the furnace would give 30cm (12") total length as a ballpark figure to separate the motor (135 degree C winding insulation) and rubber belt from the radiant heat of the furnace.

The direct drive unit, if I did build one, would have a means of sliding it in the tube away from the furnace on shutdown to avoid radiant heat. I'm trying to buy as much of the hardware "off the shelf" as I feel the direct drive cup is going to be a real pain to machine, needing a lathe and a mill with an indexing head to do manually. The belt drive cup can be made with just a lathe and the tubing will be given a light polish with 2000 grit where the bearings will run and the machined cup heat shrunk onto the shaft. If the radial holes in the direct drive cup clog, the tube around the motor will fill with oil, all while being in close proximity to the furnace.

I'd be interested in seeing what a machinist would quote to make the direct drive cup in 304 stainless with a 40 degree included angle taper and a 1.5" opening. I expect it's going to be expensive, in the hundreds of dollars for a one-off.

There does seem to be significant power consumed flinging the oil off the cup: I can hear the motor bog down as the fuel begins to fling off the cup and those cheap motor tools are 50-100 Watts or so.

 

Mark-post the specs on the cup. How about a prototype in 12L14? BTW Home Depot has a trim router for $29.
R

 

I have some cold rolled, 4140, O1 and I think A1... Any of that useful...?

 

I've got plenty of 12L14 in 27mm diameter bar and maybe some larger stuff as well as 304 stainless bar which machines ok. The problem is this: this is the milling machine I have:



You can see the dividing head on the pallet in front of the machine at the furthest point from the camera, it weighs at least 300 lbs and I don't have a crane to put it on the table (#FirstWorldProblems) so I'm avoiding a direct drive cup for two reasons: it's going to be a pain for me to make and it's also going to be a pain for any home foundry user to get one made unless they have a well equipped machine shop. My intent is to have a unit that a home user could build with minimal exotic parts (cup unit).

 

Killer universal machine... I've been looking for a "large" hobby or small industrial version of that.

 

The bigger stuff often goes a lot cheaper at auction, it has no Bridgeport style quill but otherwise it's a lot better. That's not my one in the photo: mine came from a technical school and was mint condition. Then six months later I bought an identical one with the dividing head and sold it but kept the dividing head: it's specifically made for the mill with a gearbox for cutting helical gears.

 

Yeah Mark... The problem with your machine is it would look much better in my garage! I'll send a truck... (and a crane)

 

Better take a crate of WD40 so the distributor doesn't get wet as you cross the Pacific ocean .

 

Forgot about that.... Always something keeps me from getting my mill. If it's not my wife, it's location, price or already sold. Might have to consider a cheap import.

 

After struggle with the staff at the local bearing shop and four trips to two stores, I had three HK1210 needle roller bearings with 16mm O.D. and 12mm I.D.. Initially the first bearing shop manager sold me B1210 1" by 5/8" bearings despite me asking for 1210 bearings with a 12mm bore . A second trip later and I got home to find he's sold me 12mm I.D by 18MM O.D. HK1212 bearings. So I made a third trip in before closing time on a Friday afternoon and got a full refund after he denied taking the original receipt from me on the second trip. I made it to a second shop right on closing time and bought the right bearings after opening the box and measuring them to be sure this time.

So, in short, a 49cm or 19 & 1/4" long tube with three bearings spun up but had the death wobbles at the slightest provocation. A 30cm or 12" long tube with two bearings in oil lube, was able to be spun up to 27k RPM with no hassles and then run at 15K RPM with no problems or signs of trouble at all. I made sure to leave the milling machine out of the frame on the right . Of course this test was without a cup on the shaft but it was an encouraging step in the right direction.

 

So it appears that if you hold the fuel line strongly enough then it should be good to go... nice

I don't think the will throw you off that badly. As long as it's well balanced at least.

 

Looks good. Simple solution so far.