My latest Thermojet

Moderator: Mike Everman

multispool
Posts: 125
Joined: Thu Jan 19, 2006 10:59 pm
Antipspambot question: 0
Location: UK

My latest Thermojet

Post by multispool » Thu Jul 12, 2007 12:31 am

Eric's free design
Making the fuel lines for it now. Was surprised to find that the UK brake tubing won't silver-solder. It's Cupro-nickel. Must be the Nickel thats causing it.

Built it with the hope of being able to start with no to little air supply.
Attachments
thermo_jet.jpg
thermo_jet.jpg (61.19 KiB) Viewed 6523 times

Dave_G
Posts: 128
Joined: Tue Jun 08, 2004 4:33 pm
Antipspambot question: 0
Location: USA

Re: My latest Thermojet

Post by Dave_G » Thu Jul 12, 2007 2:41 am

multispool wrote:Eric's free design...
Built it with the hope of being able to start with no to little air supply.
Very nice execution!

I have a 3-inch clone and a 5-inch Melenric Original that I bought from Bryan Seegers of M-Dot Aerospace, and they both start readily with a lit match, applied to a single inlet no less.

You should have no trouble starting this without air.

Regards,

Dave

multispool
Posts: 125
Joined: Thu Jan 19, 2006 10:59 pm
Antipspambot question: 0
Location: UK

Post by multispool » Thu Jul 12, 2007 10:11 am

Hi Dave,

That sounds encouraging!

The only diversion from plans is that I took Eric's advice to take the inlets flush to the cone. (hopefully for better breathing). My understanding is that this shouldn't alter inlet tuned path length's, we shall see.

The requirement for no air starting cylinders is to keep down the load of stuff I have to carry to Turbine demos. I build Pulsejets just for a break from turbines (and of course, I find them interesting).

You know those cheap tyre compressors that run from the car battery. It would be nice if I could use one of those if this engine does need a small amount of air to start.

Dave_G
Posts: 128
Joined: Tue Jun 08, 2004 4:33 pm
Antipspambot question: 0
Location: USA

Post by Dave_G » Thu Jul 12, 2007 3:48 pm

multispool wrote:...The only diversion from plans is that I took Eric's advice to take the inlets flush to the cone. (hopefully for better breathing). My understanding is that this shouldn't alter inlet tuned path length's, we shall see...
I'm sure it'll work just fine. The 3-inch (J3-200) engine I have is the same way because the guy who built it didn't know any better...he'd only seen one from the outside and just assumed the inlets were supposed to be flush. It runs just fine this way...it doesn't produce a lot of thrust, and it's just slightly harder to start than the J5-200, but it runs. Eric has a lot more experience with Thermojets than I do, I just have the two motors...

You know those cheap tyre compressors that run from the car battery. It would be nice if I could use one of those if this engine does need a small amount of air to start.
You could try it, but I've never used compressed air or spark to start these. Probably the simplest, handiest and most sure-fire way of starting these is to use a handheld propane torch. With the little engine I've found that if I don't let it warm up properly before advancing the throttle it will blow itself out.

I think you'll be pleased with what you've built. The most critical area seems to be properly sizing and positioning the injector nozzles. I once gave Mike a copy of the Thermojet manual which details this, which I think he posted here, somewhere? If you can't find it P-M me and I'll send you a copy.

Jim Berquist
Posts: 1396
Joined: Thu Oct 06, 2005 1:34 am
Antipspambot question: 0
Location: DEMING NM 88030

Thermojet

Post by Jim Berquist » Thu Jul 12, 2007 4:02 pm

Multi spool:


Eric built me a little thermojet. It's intakes are a little smaller then in the plan. I can get it to do everything but start without air. All I need is little shot to get it rolling!

Try one of these! It works for me! 8)

You can get them battery powered as well.

Jim
Attachments
inflator.jpg
inflator.jpg (12.86 KiB) Viewed 6423 times
WHAT TO FRAP, IT WORKED![url=callto://james.a.berquist]Image[/url]

larry cottrill
Posts: 4140
Joined: Sun Oct 05, 2003 1:17 am
Antipspambot question: 0
Location: Mingo, Iowa USA
Contact:

Front End Tuning

Post by larry cottrill » Thu Jul 12, 2007 4:56 pm

Dave_G wrote:
multispool wrote:...The only diversion from plans is that I took Eric's advice to take the inlets flush to the cone. (hopefully for better breathing). My understanding is that this shouldn't alter inlet tuned path length's, we shall see...
I'm sure it'll work just fine. The 3-inch (J3-200) engine I have is the same way because the guy who built it didn't know any better...he'd only seen one from the outside and just assumed the inlets were supposed to be flush. It runs just fine this way...it doesn't produce a lot of thrust, and it's just slightly harder to start than the J5-200, but it runs.
It works out this way because the acoustic path length from the front plate to the intake flare is MUCH more important to basic tuning than the intake length itself. Here's why:

Imagine starting with what you have now and gradually working the intake pipe farther in (the opposite direction to your actual mod). As you lengthen the pipe inward, the mass it will pick up and move out becomes smaller. Taken alone, this should raise the frequency. However, you've also increased the pipe impedance, which taken by itself would lower the frequency. So, there is a degree of compensation - probably not perfect, of course, but working in the right direction.

What you see in running a "flask" (e.g. a valveless engine front end seen in isolation) in UFLOW1D is that increasing the diameter of the neck (our intake pipe) raises the resonant frequency, increasing the length of the neck lowers the frequency, increasing the diameter of the belly of the flask lowers frequency, and increasing the overall length (belly plus neck) while keeping the neck the same lowers the frequency. What this boils down to is:

The fundamental resonant frequency of the flask is essentially determined by just two factors: 1. The mass in the belly of the flask that will be evacuated or entrained through the neck; 2. The impedance of the neck. That's it. anything which gives you more mass to move will lower the frequency. Anything that will provide less impedance through which the mass moves will increase the frequency.

This gives us a clue about another characteristic of the intake pipe. Suppose we have an engine that runs, but for some reason, we think the intake pipe is too small for the best possible performance. How do we proceed to make the pipe larger, without throwing the engine out of tune? The answer is to increase the diameter of the pipe WITHOUT decreasing its impedance. To do this, the length must be increased to compensate for the larger diameter! Only a program like UFLOW or NUDiS can tell you exactly how much, but roughly speaking, the length will need to be increased about the same percentage as the area was increased. This is inexact because of boundary layer effects (and undoubtedly, other factors).

For "conventional" valveless designs, as long as we keep the "flask" (the engine front end, including some of the chamber mass) at the same resonant frequency as the "mortar" (the engine main tube, excluding only the intake pipe) at reasonable assumed internal gas temps, we will have an engine that starts and runs.

The classic valveless pulsejet is like a father-son "three-legged race" where the boy is only six years old: To run, the two parts of the team have to do exactly the mirror image of one another with exactly the same timing, even though the masses and strengths are vastly different. Of course, this is a simplification, but true enough to keep in mind as a crude model.

L Cottrill

multispool
Posts: 125
Joined: Thu Jan 19, 2006 10:59 pm
Antipspambot question: 0
Location: UK

Post by multispool » Thu Jul 12, 2007 11:40 pm

Thanks Jim,
I'll get one of those if I run into starting troubles!

Hi Larry,
Thanks for the clarification. Yes I follow that apart from, I'm not sure how the inlet impedance is measured (what units?) I seem to recall that inlet pipe, aspect ratio was good at about 6.25:1 but surely, just as in electrical circuits, impedance varies with frequency. In an hypothetical design, do we arrive at a calculated operating frequency then choose an inlet diameter to obtain the best inlet pipe impedance (thats the way I think I would go about it as the inlet length from flare to the chamber front (as you say) is determined by the fundamental frequency, so would essentially be fixed (at any given temperature).
Alternatively, much easier, just calculate the critical lengths and if its impossible to arrive at near a 6.25:1 aspect ratio for the inlet pipe, then redesign, or use two or more inlets.

larry cottrill
Posts: 4140
Joined: Sun Oct 05, 2003 1:17 am
Antipspambot question: 0
Location: Mingo, Iowa USA
Contact:

Electrical vs Thermoacoustic

Post by larry cottrill » Fri Jul 13, 2007 1:16 pm

multispool wrote:Hi Larry,
Thanks for the clarification.
Is that what that was? ;-)
Yes I follow that apart from, I'm not sure how the inlet impedance is measured (what units?)

Units? Ha!

There is no "measuring" it, except by its effect on mass motion. But that doesn't matter very much. What's important is to recognize that it has a profound effect, and that it is a property that can be controlled by design. Its real value is that knowing how it works, at least roughly, can keep us from making really bad guesses about "what to do next".
I seem to recall that inlet pipe, aspect ratio was good at about 6.25:1 ...
Yes, but that is mostly predicated on having a high Q for the pipe (another factor of unknown units ;-). I am not saying at all that this is unimportant, only that it is secondary to basic system tuning. What you are doing here (assuming that I understand what others have said ;-) is getting an "open pipe" frequency for the intake duct that will resonate happily with some odd harmonic of the fundamental frequency. In some designs (like the Thermojet, I think) this can greatly enhance performance when you get it right. In a design where the harmonics are suppressed (e.g. the FWE chamber), it's practically unimportant.
... but surely, just as in electrical circuits, impedance varies with frequency.
Only the reactance part of impedance is frequency dependent. That's the part you're playing with when you achieve the "open pipe" resonance just mentioned. The largest part of impedance in this case is plain old resistance (how much flow will the pipe handle at a particular pressure difference), which basically doesn't vary with frequency. Bruno has pointed out that we tend to think that it is easy to get hot gas moving through a straight pipe while in reality it is not.

The electrical analogy is not perfect, but is close enough to create some amusement, in "thought experiment" terms. One thing that seems hard to convince newbies of is the huge difference between wave speed and gas flow velocity in the pipe. Yet an astonishing fact is that a wire will propagate electrical signals at the speed of light while electrons move through copper wire at about the speed of a good racehorse. It is only in a large vacuum under large voltage differences (for example, within a big color TV CRT half a metre long at 30,000VDC) that electrons develop fairly high velocities.

A straight pipe is something like an antenna transmission line, having both capacitance and resistance, and therefore, both types of reactance acting out of phase. This is what makes a simple open pipe resonant at some particular frequency. Of course, a complication that distresses the simple electrical analogy is that in our case there is varying temperature within the pipe, which alters both the mass layout and propagation speed. Ha!
In an hypothetical design, do we arrive at a calculated operating frequency then choose an inlet diameter to obtain the best inlet pipe impedance (thats the way I think I would go about it as the inlet length from flare to the chamber front (as you say) is determined by the fundamental frequency, so would essentially be fixed (at any given temperature).
Basically, yes, this is an acceptable method. The greatest difficulty I find in design (with tools like UFLOW1D, since I am not very math friendly) is getting the internal temps set up right. It is all guessing, based on what you've seen work before. Of course, it wouldn't need to be guessing if you were doing this in a big research lab with full-bore instrumentation, but that's not where we are.
Alternatively, much easier, just calculate the critical lengths and if its impossible to arrive at near a 6.25:1 aspect ratio for the inlet pipe, then redesign, or use two or more inlets.
Yes, in most cases that will work fine. You have to keep in mind that on these small engines, using two pipes doesn't let you divide the area by 2 - you have to be a few percent bigger than that if you really want equivalent impedance. You can actually reach a point of diminishing returns where, say, using three intakes doesn't let you make them much smaller than if you're using two. This is an example of a breakdown in our electrical analogy (where you can just keep adding resistances in parallel and get proportionally higher total current with each iteration).

L Cottrill

Eric
Posts: 1859
Joined: Sat Oct 04, 2003 1:17 am
Antipspambot question: 0
Location: United States
Contact:

Post by Eric » Fri Jul 13, 2007 10:20 pm

Nice engine multi. I cant wait to see it run :)

The self starting is mostly a matter of the injector entrainment.

I dont like protruding intakes with small engines, with the large engines it would greatly reduce the flow of air out the intakes, but with the small engines they arent quite big enough to get any gain out of it.

With the intakes smooth against the wall it will allow air to flow out easier, which in turn will help more air flow in. With small engines that really helps since they run at higher frequencies and will let them take in more fuel and air for more power.

Eric
Image

Talking like a pirate does not qualify as experience, this should be common sense, as pirates have little real life experience in anything other than smelling bad, and contracting venereal diseases

multispool
Posts: 125
Joined: Thu Jan 19, 2006 10:59 pm
Antipspambot question: 0
Location: UK

Post by multispool » Sat Jul 14, 2007 9:52 am

Hi Larry,

Ah yes! Many empirical relations based mainly on previous working designs seems to be productive, whereas pure theory often fails in this particular application.

Hi Eric,

I read between the lines that the fueling arrangement could be critical to easy starting. I wish I was able to test run these engines at home (would make for easy modification) but the neighbours are already fed up with the odour of jetA1 fumes eminating from my place!

I have attached some pics of the fueling arrangement. See the slot milled in the base rail. This allows positional injector adjustment from about 1/4" outside the inlet flares to about 2" inside. I have a feeling that I might need to go to smaller diameter fuel tube. What I have is 1/8th inside diameter but very thick wall, probably need to change to s/s thin wall stuff.

Is this setup near the mark?
Attachments
adjustment.jpg
adjustment.jpg (72.55 KiB) Viewed 6182 times
fueling.jpg
fueling.jpg (63.07 KiB) Viewed 6183 times

Eric
Posts: 1859
Joined: Sat Oct 04, 2003 1:17 am
Antipspambot question: 0
Location: United States
Contact:

Post by Eric » Sat Jul 14, 2007 3:42 pm

Looks good. If you want to put the injectors farther into the intake its best to have something more like 1/8" OD tube so it doesnt block too much flow.

Looks like it should fire up with no problem :)
Image

Talking like a pirate does not qualify as experience, this should be common sense, as pirates have little real life experience in anything other than smelling bad, and contracting venereal diseases

Mike Everman
Posts: 4930
Joined: Fri Oct 31, 2003 7:25 am
Antipspambot question: 0
Location: santa barbara, CA
Contact:

Post by Mike Everman » Sat Jul 14, 2007 8:36 pm

yeah, great setup!
Mike
__________________________
Follow my technical science blog at: http://mikeeverman.com/
Get alerts for the above on twitter at: http://twitter.com/mikeeverman

Bruno Ogorelec
Posts: 3542
Joined: Sat Oct 04, 2003 7:31 am
Antipspambot question: 0
Location: Zagreb, Croatia

Re: Electrical vs Thermoacoustic

Post by Bruno Ogorelec » Sun Jul 15, 2007 6:17 am

larry cottrill wrote:There is no "measuring" it, except by its effect on mass motion.
Larry, of course you can measure it. Impedance is drag. Drag is a force. Forces can be measured and expressed in the usual units. It just isn't very simple to measure. But, you can calculate it.

I used to have a simple formula for tube drag. It takes into account tube length and diameter, gas temperature and an empirical correction factor for the smoothness of the tube. For smooth round tubes, the factor has been worked out pretty accurately. Alas, I lost both the formula and the tube constant years ago and have not been able to reconstruct them, or find them in the reference sources again.

Maybe you can find it somewhere. It's a useful little tool. Makes it easy to work out, say, a twin-tube equivalent for your single tuber etc.

larry cottrill
Posts: 4140
Joined: Sun Oct 05, 2003 1:17 am
Antipspambot question: 0
Location: Mingo, Iowa USA
Contact:

Re: Electrical vs Thermoacoustic

Post by larry cottrill » Mon Jul 16, 2007 12:47 pm

Bruno Ogorelec wrote:
larry cottrill wrote:There is no "measuring" it, except by its effect on mass motion.
Larry, of course you can measure it. Impedance is drag. Drag is a force. Forces can be measured and expressed in the usual units. It just isn't very simple to measure. But, you can calculate it.

I used to have a simple formula for tube drag. It takes into account tube length and diameter, gas temperature and an empirical correction factor for the smoothness of the tube. For smooth round tubes, the factor has been worked out pretty accurately. Alas, I lost both the formula and the tube constant years ago and have not been able to reconstruct them, or find them in the reference sources again.

Maybe you can find it somewhere. It's a useful little tool. Makes it easy to work out, say, a twin-tube equivalent for your single tuber etc.
Bruno -

Good to hear from you on this! I didn't mean it was incalculable, only that we have no means of empirical measurement. HOWEVER:

I disagree that impedance is only drag in this case. Drag would be the "resistance" component. There is also force needed to accelerate the gas mass in the pipe, though in the case of the intake this seems pretty small. That inertial component is the "reactance" part of our total impedance. Like any good reactance, it will be highly frequency dependent. Would you not agree with this assessment?

All the best,

L Cottrill

Bruno Ogorelec
Posts: 3542
Joined: Sat Oct 04, 2003 7:31 am
Antipspambot question: 0
Location: Zagreb, Croatia

Re: Electrical vs Thermoacoustic

Post by Bruno Ogorelec » Mon Jul 16, 2007 1:51 pm

larry cottrill wrote:Drag would be the "resistance" component. There is also force needed to accelerate the gas mass in the pipe, though in the case of the intake this seems pretty small. That inertial component is the "reactance" part of our total impedance. Like any good reactance, it will be highly frequency dependent. Would you not agree with this assessment?
Of course. But, I'd say that this part is straightforward. It's just mass whose inertia has to be overcome. It should not depend on the character of the tube. "And thus ceases to be of interest in this context", as my late father would say.

Thanks for the good wishes. I hope this finds you well.

Post Reply