# Thread: Rising hot air question

1. ## Rising hot air question

See the attached photo:
Christmas carousel.jpg
This windmill-powered ornamental carousel is driven by hot air rising from the candles. When I install new 10-inch candles it initially refuses to rotate. After the candles burn down a few inches it starts rotating, and gets faster and faster as the candles continue to burn down. My educated guess is that the hot air is continuing to accelerate as it rises, and that as the candles burn down there is more elevation above the flames in which the hot air can accelerate before reaching the blades. Does this look like a good hypothesis?

2. Perhaps the greater rise time for the hot column has more time to heat the surrounding air producing a greater mass flow rate.

3. I wonder if the temperature of the spinning device itself has anything to do with it.

Sent from my iPhone using Tapatalk

4. Originally Posted by Hornblower
My educated guess is that the hot air is continuing to accelerate as it rises, and that as the candles burn down there is more elevation above the flames in which the hot air can accelerate before reaching the blades. Does this look like a good hypothesis?
I was thinking something similar - but about more air being pulled into the rising column and adding to the mass of air pushing the blades.

Also, does it need lubricating?

5. I think the increased “chimney” is the right explanation. Increase in speed due to increased column below the fan. The walls of the chimney are not solid but the inertia of the surrounding cold air. In school physics a microphone flame exhibits the same effect.

6. Originally Posted by Strange
I was thinking something similar - but about more air being pulled into the rising column and adding to the mass of air pushing the blades.

Also, does it need lubricating?
My thought as well. Might be interesting to tweak the blade angles and see what happens.

7. "Entrainment" of air.

8. Originally Posted by Trebuchet
My thought as well. Might be interesting to tweak the blade angles and see what happens.
If you're going to tweak the blades, you might as well form them into airfoils, so no hard angles to disrupt air flow.

9. Order of Kilopi
Join Date
Oct 2005
Posts
26,925
I agree with the "higher air speed" explanation, and the "entrainment of higher mass flux" explanation is really the same because the density of the air is pretty much constant, so the mass flux just depends on the constant density times that all-important speed.

10. Order of Kilopi
Join Date
Feb 2005
Posts
11,637
I wonder if an invisible vortex might be at work, hairy ball and all that.

Spacing the candles just so may be an influence?

11. Order of Kilopi
Join Date
Oct 2005
Posts
26,925
Where's the angular momentum coming from?

12. Order of Kilopi
Join Date
Feb 2005
Posts
11,637
Location of a fan/heater off frame perhaps. A lot of things have to come together

13. Turbulence with that greater distance might help get the wheel turning, overcoming the static coefficient of friction. I still think the increase in cross section with time is the key. Air entrainment, as mentioned, plus heat conduction along the column and a little extra turbulent flow might explain it. A narrow straw-like tube may be one way to test it.

Iphone

14. There is a party trick. You hold a tube over a candle. The real “chimney” accelerates the flow even more than the natural effect. It’s also the cross section expanding as George says. You only get rotation of the flow above the fan, caused by the fan, as in a windmill.

15. I think we are collectively getting a good overall look at this. After thinking about it overnight, I can see that the hot air is being forced along as it rises by the buoyancy force, so it will accelerate or get more air around it moving, or both. Tonight, with the candles burned down a couple of inches, it was starting to turn intermittently.

This ornamental device has minimal friction with no lubrication. The rotor is supported by a pointed rod into a conical receptacle at the bottom, and is prevented from toppling by a loose-fitting sleeve bearing at the top. If I remove the fan and give the rotor some slow spin, it takes a very long time to spin down. With the fan in place and no convection it stops quickly as a result of the air drag.

I have experimented with varying the pitch of the blades. With the blades at high pitch, meaning a low angle of attack in the rising air, it rotates slowly. Reducing the pitch increases the angle of attack and increases the rotation speed up to a point. If I try to go too far it will stall, just as an airplane does if the nose is too high.

16. Originally Posted by Hornblower
I think we are collectively getting a good overall look at this. After thinking about it overnight, I can see that the hot air is being forced along as it rises by the buoyancy force, so it will accelerate or get more air around it moving, or both. Tonight, with the candles burned down a couple of inches, it was starting to turn intermittently.

This ornamental device has minimal friction with no lubrication. The rotor is supported by a pointed rod into a conical receptacle at the bottom, and is prevented from toppling by a loose-fitting sleeve bearing at the top. If I remove the fan and give the rotor some slow spin, it takes a very long time to spin down. With the fan in place and no convection it stops quickly as a result of the air drag.

I have experimented with varying the pitch of the blades. With the blades at high pitch, meaning a low angle of attack in the rising air, it rotates slowly. Reducing the pitch increases the angle of attack and increases the rotation speed up to a point. If I try to go too far it will stall, just as an airplane does if the nose is too high.
and like a sail boat the fan tip speed can be higher than the airspeed. It's a force balance

17. Order of Kilopi
Join Date
Oct 2005
Posts
26,925
Originally Posted by publiusr
Location of a fan/heater off frame perhaps. A lot of things have to come together
I meant for the candles-- I don't think they could produce a strong vortex because there isn't angular momentum.

18. This is getting curiouser and curiouser. Last night it was just starting to try to rotate intermittently before I extinguished the candles. Tonight it started rotating freely before I had lighted more than two or three candles. I knew the action improved as the candles burned down, but at the start the flames were at virtually the same elevation as when I put them out before. I don't remember such a sharp change the last time I put in new candles. Maybe there is intermittent drag from dust bunnies or something similar under the rotors.

I can feel a big difference when I pass a finger through the air column at different heights. At 12 inches above the flames I feel hot spots about an inch wide with cool air in between. At 16 inches, the current height of the fan above the flames, it is cooler but evenly distributed. Clearly there is plenty of mixing and a broad column acting on the whole fan. When the flames were only 12 inches from the fan the hot columns were at best on the tips and sometimes between the blades.

19. there is also the wick variable which is to do with the heat input, and at twelve inches I would expect the environmental air flows too, which might change from night to morning. The steady state of any free- in- air candle must take some time to stabilise. Some versions of that device have hanging rods which strike bells when fast enough. Thus showing it is a true heat engine doing work.

20. Originally Posted by Hornblower
When the flames were only 12 inches from the fan the hot columns were at best on the tips and sometimes between the blades.
Have you tried aligning the candles to be all centered on the blades? Perhaps vertical airflow behind some of the blades will oppose normal rotation.

iPhone

21. Originally Posted by George
Have you tried aligning the candles to be all centered on the blades? Perhaps vertical airflow behind some of the blades will oppose normal rotation.

iPhone
I tried rotating it by hand those first couple of times, and it just stopped is if there was no convection happening.

22. I expect the engine is sensitive to external air currents. The version I used to have was smaller with small candles but a light breeze sideways would stop it for many seconds until the steady convection flow got going again.

23. Originally Posted by profloater
I expect the engine is sensitive to external air currents. The version I used to have was smaller with small candles but a light breeze sideways would stop it for many seconds until the steady convection flow got going again.
There was no cross current during any of these burns. I verified it with smoke from extinguished candles or matches.

24. Established Member
Join Date
Feb 2014
Posts
241
Originally Posted by Hornblower
This is getting curiouser and curiouser. Last night it was just starting to try to rotate intermittently before I extinguished the candles. Tonight it started rotating freely before I had lighted more than two or three candles. I knew the action improved as the candles burned down, but at the start the flames were at virtually the same elevation as when I put them out before. I don't remember such a sharp change the last time I put in new candles. Maybe there is intermittent drag from dust bunnies or something similar under the rotors.

I can feel a big difference when I pass a finger through the air column at different heights. At 12 inches above the flames I feel hot spots about an inch wide with cool air in between. At 16 inches, the current height of the fan above the flames, it is cooler but evenly distributed. Clearly there is plenty of mixing and a broad column acting on the whole fan. When the flames were only 12 inches from the fan the hot columns were at best on the tips and sometimes between the blades.
Plumes are complex. Minor factors can greatly influence the laminar to turbulent transition. Case in point:

thermal-plume-of-burning-candle-gary-s-settles--jason-listak.jpg

And yes, over an arbitrarily large area the force increases after the the transition (at least until the turbulent flow converts some of the kinetic energy to heat).

25. Originally Posted by VQkr
Plumes are complex. Minor factors can greatly influence the laminar to turbulent transition. Case in point:

thermal-plume-of-burning-candle-gary-s-settles--jason-listak.jpg

And yes, over an arbitrarily large area the force increases after the the transition (at least until the turbulent flow converts some of the kinetic energy to heat).
Good catch, VQkr. That image is consistent with what I felt with my fingers at different heights above the flames.

Last night I did an experiment by holding two candles with the flames at the same starting height but an inch or so closer to the center. The fan started turning immediately. It now appears that at the regular radius the narrow laminar columns are either barely touching the tips of the blades or missing entirely. The broad upper portion that is turbulent but still rising does not miss. My next test, after the current candles are used up, will be to make some temporary holders that will bring 6 candles in to the tighter radius. I will not make that permanent because I need to keep the flames a safe distance from the wooden structure.

26. Order of Kilopi
Join Date
Feb 2005
Posts
11,637
Maybe insert some vape/smoke to see what's going on.

27. Originally Posted by publiusr
Maybe insert some vape/smoke to see what's going on.
The challenge is to get visible smoke into the laminar column and nowhere else, without disturbing the air currents. I have tried with no success.

28. Originally Posted by Hornblower
Good catch, VQkr. That image is consistent with what I felt with my fingers at different heights above the flames.

Last night I did an experiment by holding two candles with the flames at the same starting height but an inch or so closer to the center. The fan started turning immediately. It now appears that at the regular radius the narrow laminar columns are either barely touching the tips of the blades or missing entirely. The broad upper portion that is turbulent but still rising does not miss. My next test, after the current candles are used up, will be to make some temporary holders that will bring 6 candles in to the tighter radius. I will not make that permanent because I need to keep the flames a safe distance from the wooden structure.
I did not have to go that far. I installed a new set of candles with them leaning in slightly. I did that by dipping the bottom inch of each one in hot water to soften it and then press it into the holder at the desired angle and let it cool and harden in that position. At the full 10" height the flames are about 1/2 to 3/4 inch closer to the center, and the narrow laminar columns are turning the fan without a hitch. The flames are still well separated from the wooden structure.

#### Posting Permissions

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