questions about blast compression

[Greg O'Bryant]

I was up last night tossing and turning in bed for about three hours thinking about pulse jets, of all things. Any ways I came up with some good ideas and I was wondering if anybody could help me out with a few things. My first question is, does anybody know at what cylinder pressure a car engine starts to have problems with pre-ignition? and second at what velocity would a gas have to be moving to creat that preasure if it where directed into a closed ended chamber? Can you see where I am going with this? I think that maybe the reason why a pulse jet never reaches a good precompression is that it ignites before it has a chance. Now this leads me to my last question. Has anybody ever tried or herd of using a timed fuel injector on a pulse jet. I mean injecting fuel right at the time that the engine is at it's peak pressure. Just so much that I don't know about?

[pezman]

I have done some playing around with using blast compression into a dead end. The results are impressive, but getting something that does more than single-shots (like 1 pulse every 3-4 seconds) takes some work.

You have hit on one of the two main compression techniques that are likely to be used in a pulse-jet -- momentum change. The other technique is sonic choking -- a restriction in a tube will allow gas to pass at only the speed of sound, causing pressure in the larger tube behind to build up. Sonic choking is the mechanism that Bruno Ogorelec relies on in his proposed BCVP design. In a valveless, the line of demarcation between the two techniques is normally blurred -- since valveless engines have an unchanging gometry, there is always a constant path to the outside of the engine.

As far as injection is concerned, it should be doable, but it takes away a certain amount of elegance. It takes some time for all the gas in a charge to burn anyway, so I'm sure that you can design a decent high compression engine w/o the need for injection (i.e. a high comprssion design does not necessarily need to delay the introduction of fuel). Injection adds a certain amount of complication and seems more like an optimization than a requisite part of a good design.

As far as pre-ignition is concerned, the holy-grail of pulse-jets is pre-ignition's more extreme cousin -- deonation. Make a pulse-jet with tubes the right size, the right fuel and high enough compression and you will have a pulse-detonation engine. Then you can retire on your patent royalties.

The issue of elegance aside, I have given some thought to scavenging fuel injection parts for a pulse-jet. Injectors, fuel rails, pressurization pumps etc. should be readily available from scrap yards. For me, it's just a matter of having the time and energy to pursue rather than day-dream it. I think there are a few posts about FI Js and if you make one, you will be admired by many, reviled by some but you definitely won't be ignored ;).

[Ogge]

Greg,
A prototype test rig is nearly finished for testing my blast compression intake design engine. We still have to add the ignition system to get the engine to start the oscilation with blast compression.

When lit from the exhaust, it just operates as a blow torch. There is not a pressure wave powerful enough to cause blast compression in the 6 BCI tubes.

The 6 parallel tubes are blast compression reflection chambers from triggering the follow on cycle. The reflectors are variable with the treaded rods.

The design also has a variable CC volume and different bolt on exhaust extension for testing different comfigurations for best preformance. It also contains resonance chambers to enhance the pumping in mixture from the intakes. We hope to get it finished and operational soon.

[Bruno Ogorelec]

does anybody know at what cylinder pressure a car engine starts to have problems with pre-ignition?

Greg, as Ben has pointed out, the pre-ignition (or rather self-ignition) depends very much on the fuel. In fact, the ‘octane number’ of gasoline fuel is determined by testing its propensity for self-ignition. The easier the self-ignition, the lower the octane number.

Of course, ignition does also depend on pressure -- and a number of other factors. But, you can forget about pressure as the key mechanism in the pulsejet because it is not. Perhaps you can somehow introduce pressure as a key factor here – as I tried with the BCVP concept -- but it is a very complex undertaking, full of pitfalls.

You can look at ignition as a matter of statistics.

“Ignitionâ€

[Mike Everman]

Adam lives! Thought we'd scared you off for good. I'd like to apologize for that. Good craftsmanship, man, and welcome back.

Greg, think of the pulsejet cycle as just like jumping on a trampolene. It's remarkably similar. Your jumps must come at just the right time to feed the cycle, and can't work if the jump is instantanious, too long, or too near the extremes of the cycle. Think of the extreme downward position when bouncing as the high pressure peak. The system has it's own resonance and for it to work, you need to start your jump so it's in phase with what wants to happen naturally. In the case of trampolene jumping, that means applying the push before you buttom out.
In the pulsejet, the pop starts somewhere near the point that the engines' pressure reaches ambient, as it's climbing up naturally in order to feed the cycle.
Faaascinating, no?

[Greg O'Bryant]

Hay people; thanks for all of the input. I'm sorry for any confusion that I may have caused by my questions. As I read through my post and everybodies replies I can see that I wasn't completely clear about a few things. So if anbody is interested I will try to shed some light on where I'm going. First I should say that the concept engine that I'm thinking of shoulden't be considered a pulse jet according to the traditional deffinition, but rather a blast compression engine that would have very high precompression, or even a PDE engine. That is why I was interested about pre-ignition. I'm thinking up a blast wave compression engine that would be ignited "detonated" by compression. I know that in cars the octane rating is an inportant factor, but to what point? I'm not a deisel mechanic but looking at some simple engine schematics I notice that some deisel engines maybe all , I don't know, compress their air before they inject the fuel. This way they achieve compression ratios higher than you would ever be able to get in a normal car engine becuase of pre-ignition. Look at a turbo engine that has a 30 to 1 compression ratio, the fuel is injected after compression. This is not the case in a pulse jet. This is why I'm asking, as backward as this may sound, and it is just my opinion, I think it may be a limiting factor in a pulse jet's performance. Bruno I particularly liked the input on the ignition source. Very interesting insight into the ignition of a pulse jet versus conventional engines, and if I ever have the time and money to actually try building my engine it may prove to be very valuable. Thanks

[mk]

[...] I know that in cars the octane rating is an inportant factor, but to what point? [...]

There was a conversation about -- does someone remember thr topic?

My way of thinking, briefly:
The octane number indicates the knocking resistance of the fuel or rather final fuel/air mixture when being (extremly) pressurized, as it happens in two- or four-stroke engines at the squeezing plates/areas/zones. There the distance between the piston and the cylinder head is below 1.0mm, rather around 0.5mm. These areas are doing trick by adding turbulence to the fuel/air mixture and forcing the mixture to concentrate around the spark plug electrode.
Overall "ball-shaped" molecules (as n-heptane here, which also is octane in the atomes sum) do feature a higher knocking/detonating resistance than "in-line" molecules (as iso-octane here) and do have higher self-ignition temperatures.
Knocking is not wanted, because it causes damages of the whole crank assembly, especially the bearings and the piston.
Despite it also leads to insufficient power values, because of the pre-ignition. Then the engine likes to "block off" at the RPM where knocking begins to appear (wildy).

Sorry for hijacking.

[pezman]

If history is any indicator, this thread will degenerate and maybe even get hostile ;). Everyone has their own built in assumptions about pulse-engines, the most common point of contention is that a lot of folks assume that a pulse-jet engine must rely on the Kadenacy effect, acoustics etc. to "pump" the engine. If you say pulse-jet, then a lot of folks will assume that the engine must rely on resonance etc. and not long after, there will be ferocious arguments where folks are talking past each other. Other designs are possible, and I guess it can be debated by hair-splitting semanticists whether these other designs are true pulse-jets.

You are right that diesel engines are high compression. My (limited) understanding is that compression ratios are typically in the 16/1 to 30/1 range, depending on design. If you are contemplating those kinds of pressures, you may need something other than blast compression to attain them. There is a proposed design out there that has PJ's driving a wheel in a circle. The chambers are pumped up by a piston that is moved by a cam in the center of the wheel and this compresses the fuel and then opens the port of the engine when it's time for a pulse. In a design like this, high pressure fuel injection would be handy -- pressurize the air, introduce the fuel at just the right time and open the tail when things are ready to rock.

Another concept out there is a free-piston engine. These are actually pretty cool. The most modern designs have a magnet in the middle of the passes through a coil, thus forming a generator. The generator provides feedback on engine position and velocity and a varying load is placed on the generator to control the engine. With a high-compression free-piston engine, there is afine line between getting 30:1 compression and driving the piston through the cylinder head. It is only recently that readily micro-processors with enough real-time processing power for controlling such a process have been available.

Note that both of these designs use a physical separation between the compression force and the fluid the drives the compression force. In blast compresion, there is no such separation -- the motive force comes from the face of a combustion wave (or an acoustic wave that precedes it). Typically, the fuel-air charge is ignited against a boundary (e.g. combustion chamber wall) or some sort of restriction (e.g. sonic choke), so one side of the charge becomes an increasingly powerful compression of the remaining fuel-air mixture -- which is nice because as pressure increases, more force is required.

So in blast compression, it's the burning of part of the fuel that drives the compression cycle. If you are witholding fuel until late in the cycle, I think that you will have a hard time getting enough force to compress the air. Starting rich and getting leaner as the fuel is consumed and pressure goes up seems the way to go. There might be some benefit to starting a little lean and then topping off the charge when things get going, but that seems like an optimization.

Build a little blast compressor just for fun. A device that shoots a blast into a dead-end chamber is easy to build. Leave some space around the throat where you point the blast tube into the chamber in order to let the pressure escape. If you build it right, you'll be amazed at the power that even a very small amount of propane will deliver.

As an aside, I think that super-high compression is probably not necessary. The cycle that is used by a PJ is inherently "constant-volume-ish" and my understanding is that it is inherently more efficient than a Brayton cycle (e.g. turbine or ramjet), so even modest compression ought to pay big dividends.

BTW, I like Bruno's take on combustion -- it's something that "blossoms" rather than "happens"

[hagent]

Forgive me if I'm wrong but detonation should not have a lot to do with compression and more to do with the perfect air/fuel mixture. A fuel air bomb just releases fuel into the air and then a small charge explodes at the rear of the fuel air cloud to start the big bang. It looks quite impressive, you can see the pressure wave propagate through the air and destroy the target like a nuke.

Hagen

[pezman]

Detonation depends on a lot of factors, such as fuel type, ratios of fuel vs air, the containment geometry, the degree of turbulence in the fuel-air mixture etc.

Compressing the mixture will generally promote detonation. In fact, one of the most common detonation mechanisms is overdrive -- using a small detonation to initiate a larger one. Also, lowering the initial temperature of the fuel-air mixture will help promote detonation.

Here's a link to a nice overview document on detonation if you are interested.
http://www.gexcon.com/index.php?src=han ... Bchap6.htm

[RG Rhodes]

Has anyone tried running an existing working pj on a hydrogen/air
fuel mix?

This mix really likes to detonate- very energetic.

I would think that the cyclic rate would remain about the same,
but would expect some of the charge to detonate rather than burn
normally, thus creating a stronger pulse, and thus more thrust. No?

RG

[Mark]

There might be a need for some exotic alloys if you are burning hydrogen and pure O2. I tried introducing pure O2 into my running side port Logan and in just a second I ate a hole through the opposite side of my side port, much to my dismay. Guess I needed better mixing.
Mark

[RG Rhodes]

Hello Mark,

I'm going to try water cooling the FWE/Reynst I'm working on
in hopes of getting consistent running. Cooler chamber should
also result in a denser charge. With the cooling system in place,
I might try hydrogen/air. Just wondering if anybody has tried it
on a full-size engine.

The
gasoline engine and the gasoline to fuel it were developed together,
as the jet turbine and it's jet fuel, and likewise the diesel engine
the diesel fuel to go with it. In purely theoritical terms, what would
an ideal pulse-jet fuel be?

RG

[pezman]

You're right that hydrogen is one of the more detonable gases. Acetylene is even more detonable. However, getting more common fuels like alkanes or jp to detonate eficiently is the holy grail of Pulse Detonation Engine research.

Yor post basically described "deflagration to detonation transition" (DDT), where a wave in a tube goes from a conventional deflagration (like in a pulse-jet) to a detonation. Detonation is essentially a soliton phenomenon, where the speed of the flame front exceeds the speed of sound in the gas. This means that any fuel in the path of the wave gets burnt before it can move -- pretty much the ultimate constant-volume combustion process.

There is a group at Wright Patterson that has published enough information for a good hacker to build a hydrogen PDE out of a modified four cylinder engine. It's a pretty ingenious design, but it's a DDT design, so it's perfornace is way below the theoretical max for a detonation engine.

Here are "plans" for the Wright Paterson engine:
http://www.pr.afrl.af.mil/divisions/prt ... 2-0474.pdf

Here's a some light reading on the WP PDE and a cool photo:
http://www.space.com/imageoftheday/imag ... 30804.html

Here's an article that suggests that they now have it working on conventional fuel:
http://www.pw.utc.com/shock-system/flightsoffancy.html

[RG Rhodes]

Thanks Pezman-

I've seen a little bit about these and this is a good intro to the
problem. Pretty ineresting, pretty hairy!

Thinking that one of the problems which make the thing so difficult
is that the designers have been tasked to build a detonation
engine but fueled with turbine engine fuel. This is an economic
necessity but introduces the complexity we see.

I keep thinking in the back of my mind that if we could tailor a
fuel specifically for our purposes we might be able to come up
with some pretty good results with the engins we have, and no
moving parts!

I was away from the site for a year and what a year it has been!
Out of the primordial soup of this site patterns, ratios, and formulae
are beginning to emerge, like crystals precipitating out of a
super-saturated solution. I am just extatic about this- the bloody
hoodoo gadget is yielding and beginning to give up it's secrets. Much
progress. I'm thinking that given the state of momentum obtaining
now that it is perhaps not too early to begin thinking about big thrust.
Blast compression configurations? Special fuels? I think that now
that the ball is well and truely rolling its time for performance.

whatcha think?

RG