A New Friend and a New Lady

Moderator: Mike Everman

hagent
Posts: 413
Joined: Sun Feb 06, 2005 9:01 pm
Antipspambot question: 0
Location: Simi Valley CA

Re: re: A New Friend and a New Lady

Post by hagent » Tue Nov 15, 2005 2:49 am

Eric wrote:Did they say if anyone was involved in updating it, or if no one is doing it anymore maybe they could release the code.

Hell between all us we should be able to write our own damn program to specifications.

Somewhere I have visual studio... damn where did it go....

Eric
Hi Eric,

Here are the responses I got from Jose M Corberan.

Thank you for your comments

Regarding the modfication you proposed, you are perfectly right but we do not follow the development of the code any longer.

Best greetings from Spain and good luck with your devlopments

José Miguel

And

There has been no improvement to UFLOW since its creation and it is not possible to deliver the source code. Sorry!

Best regards

JM
Hagen Tannberg

ed knesl
Posts: 163
Joined: Wed Apr 13, 2005 5:27 am
Antipspambot question: 0
Location: Phoenix, Arizona

re: A New Friend and a New Lady

Post by ed knesl » Tue Nov 15, 2005 3:22 am

Larry, or anybody else, knowlegable in UFLOW !

I have few questions.

Where do all these input data come from ?
Like wall temp, initial temp and initial pressure ? Friction is set to Zero !?
They all differ for each individual pipe section, why ?
Are all these estimated values ?
How could the output have any scientific value, if input is just a guess.
Or, am I missing something ?

Thanks for clarification.

Ed
...Nobody is right, nobody is wrong...

Mike Everman
Posts: 5007
Joined: Fri Oct 31, 2003 7:25 am
Antipspambot question: 0
Location: santa barbara, CA
Contact:

re: A New Friend and a New Lady

Post by Mike Everman » Tue Nov 15, 2005 4:43 am

Hi Ed!
Well yes they are guesses originally, but you can converge on something like the right conditions for each segment by observing the first pop, and the velocties and temps that follow, to be input as the new starting conditions, and so on.
You're right, though, without knowing the initial pop pressure peak, it's gi-go. Though we should be seeing 160-180 kpa on small motors(?)
Mike Often wrong, never unsure.
__________________________

Mike Everman
Posts: 5007
Joined: Fri Oct 31, 2003 7:25 am
Antipspambot question: 0
Location: santa barbara, CA
Contact:

re: A New Friend and a New Lady

Post by Mike Everman » Tue Nov 15, 2005 5:04 am

Hagen,
For what it's worth, I'm almost done with the NUDiS pre and post processor in Excel. It's going to be better than UFLOW for input and automation.

A few things I've been working toward:

You can see temporal plots at any of 256 stations along the length.

You'll be able to enter geometry and watch the duct form as a background to the plots.

You can choose any time slice as the starting conditions for the next run. I can automate that so it converges on the true cycle conditions, a pseudo cyclic input of pops(!) This part is the most exciting to me.

Eventually, I want to get it to iteratively alter dimensions you choose, and goal seek for highest massflow, lowest partial pressure in the CC, or whatever metric floats your boat at the time...

I imagine making the heat release a non-linear input, based on the flamespeed of your fuel, and ignition temp.

Fun fun fun, but I have paying work to do right now. ;-( It should be a week before I can post a beta version. Thanks again to Sam for the neat 1d solver!
Mike Often wrong, never unsure.
__________________________

serverlan
Posts: 62
Joined: Tue Oct 05, 2004 11:31 am
Antipspambot question: 0
Location: Australia

re: A New Friend and a New Lady

Post by serverlan » Tue Nov 15, 2005 11:58 am

Hi Ed,

I can offer just one piece of info.

The colour of the flame indicates its temperature.
I can't remember the exact figure now, but lets say its about 2100K
If the intake temperature is 300K, then the combustion raises the temperature by a factore of 7:1.
If it burns fast enough to achieve this temperature before the pressure disperses, we could expect a similar increase in pressure above atmospheric.
Thus the initial burn pressure could be up to 700000 whatever-the-metric-units-are.

If the gas were to expand as it is burning (i.e. a slower burn) then that temperature would not be reached (expansion produces cooling) and the temperature would be lower and the colour different. So, at the moment I think that is a reasonable pressure estimate. Most people tell me I'm dreaming, that the measured pressure peaks are much less than that.

There are other questions of how quickly the engine will allow the dispersal of the pressure. A long thin one should slow down the expansion of the gas, and so more likely to produce a high pressure.
I guess large round combusion chambers might produce the high pressure in a small region only, which is insufficient to raise the pressure of the entire combustion chamber to the maximum.
Much will depend on how well the engine breathes - how big each new charge is.

But the always seem to produce the same colour. I just can't get past that fact.

Don

serverlan
Posts: 62
Joined: Tue Oct 05, 2004 11:31 am
Antipspambot question: 0
Location: Australia

re: A New Friend and a New Lady

Post by serverlan » Tue Nov 15, 2005 12:01 pm

Mike, you are a God!

Now for the really hard bit.

Can you put a T or Y junction, so that one pipe can feed into two? or more?

Don

Jonny69
Posts: 282
Joined: Sun Oct 23, 2005 8:14 pm

re: A New Friend and a New Lady

Post by Jonny69 » Thu Nov 17, 2005 3:16 pm

[smartalec mode] Metric = Bar or Pascals [/smartalec mode]

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

Re: re: A New Friend and a New Lady

Post by larry cottrill » Fri Nov 18, 2005 8:28 pm

Mike Everman wrote:Hi Ed!
Well yes they are guesses originally, but you can converge on something like the right conditions for each segment by observing the first pop, and the velocties and temps that follow, to be input as the new starting conditions, and so on.
You're right, though, without knowing the initial pop pressure peak, it's gi-go. Though we should be seeing 160-180 kpa on small motors(?)
A friend of mine used to say (quite cynically): "Measure it with a micrometer, mark it with chalk, cut it with an axe." It seems to me that what you and Eric want me to do with UFLOW is just the opposite path: "Measure it with a piece of twine, mark it with a grease pencil, cut it with a vertical mill." You want the tool to be a kind of performance predictor, based on guesses about an engine we've never observed running! I just don't use UFLOW1D in that way, or feel that it's necessary to do so. Probably, the refined version of NUDIS you're trying to wring out will be a lot more likely for that kind of application.

I think perhaps it's just the difference in philosophy between you engineers and us blacksmiths. All I really want is a relatively quick way to get something that will run the first time you build it. Is that so awful? Remember back about a year ago, when I threw the Short Lady out there, and Steve built it and it ran on the first go? Remember the oohs and aahs? Remember how someone (I think it was you) who said, "I can't get over how short this engine is."? That was done by (a) intuitively designing the pipe and refining it with UFLOW - WITHOUT an intake pipe!!!; and (b) designing an intake geometry based solely on the Hinote Criteria, with end corrections, making a guess for a diameter that might work. All the UFLOW input was educated guesses. The trick is that our guesses have to pass some kind of reasonableness test in our own minds (now, let's see, we know the chamber will be red hot, the tailpipe probably not so hot ...). That's exactly how it was done.

I will be happy to see what you come up with as you work with the new program. Until then, I'll continue to use UFLOW1D as the handy blacksmithing tool that it is. (I hope the engineers one and all can find forgiveness in their hearts for my admittedly crude methods ;-)

L Cottrill

ed knesl
Posts: 163
Joined: Wed Apr 13, 2005 5:27 am
Antipspambot question: 0
Location: Phoenix, Arizona

re: A New Friend and a New Lady

Post by ed knesl » Sat Nov 19, 2005 3:41 am

Larry,

That must be misunderstanding. I have never used UFLOW and was
damn confused how could you guys get it going.
It was a simple inquiry and request for clarification.

Speaking for myself I greatly admire your work, design and testing
of new concepts. Your participation in this hobby is essential !
Hope some day sombody brings up new, user friendly software.
UFLOW does not impress much in that regard.

Thank you again, Larry !
...Nobody is right, nobody is wrong...

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

re: A New Friend and a New Lady

Post by Eric » Sat Nov 19, 2005 6:04 am

All I was trying to say is that you could probably use it to tweak designs a lot more with it. Different design variables may have different effects at different performance values. Something that is really tuned in at a low power level may not be optimized for higher power levels.

You dont really have to measure anything with twine and cut using a mill, its all theoretical either way you go about doing it. Ultimately if you are using Uflow to test engine designs, and then build those designs you are building an engine based on guesses and aproximations. Thats not necessarily a bad thing but if you are going to do all that anyway why not try to extract every bit of potential out of the program and see if it leads to a better engine?


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

Carl Litzkow
Posts: 19
Joined: Tue Nov 08, 2005 9:29 pm
Antipspambot question: 0
Location: Gainesville, FL

re: A New Friend and a New Lady

Post by Carl Litzkow » Thu Dec 08, 2005 2:58 pm

Greetings to all who have participated so far,
I hope to have lunch very soon wih an explosive expert (yes his ID card does say Exlosive Expert). He was responsible for the bridge demolition in the movie "True Lies" with Arnold Schwartz....... and Jamie Curtis. They blew up a lot of the old causeway linking the Florida Keys. He is in our woodworking club and last time we talked promised each other a quiet pub and lots of time would be required. He has a lot of info I hope to pass on including conflagration software, explosive gasses @ supersonic speeds etc.

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

re: A New Friend and a New Lady

Post by larry cottrill » Thu Dec 08, 2005 3:18 pm

Carl -

That sounds really cool, but just a word of caution: Make sure what you present is really applicable in some way to propulsion theory, and doesn't come across as encouraging amateur explosives experimentation. That goes down pretty badly here (even in the Off Topics Forum). I'm sure you'll use good sense; this is just a "word to the wise".

L Cottrill

Mark
Posts: 10931
Joined: Sat Oct 11, 2003 10:14 pm

re: A New Friend and a New Lady

Post by Mark » Thu Dec 08, 2005 3:50 pm

Maybe we could use some shaped charged philosophy for the combustion in pulsejets.
Mark
Presentation is Everything

Mike Everman
Posts: 5007
Joined: Fri Oct 31, 2003 7:25 am
Antipspambot question: 0
Location: santa barbara, CA
Contact:

re: A New Friend and a New Lady

Post by Mike Everman » Fri Dec 09, 2005 12:06 am

Edit: I was responding to Bill's question on the end of page 1. Duh. I'll leave this as it pertains to UFLOW code and a NUDiS update:

Too busy to see straight, unfortunately. "doing the impossible for no money in zero time" is our motto.

Dr. Corboran cannot support UFLOW as it has been superceded and the grad student that wrote the interface in some obscure programming language is long gone. Also, the core of the solver is used by them in their automotive consulting, so he will not release the source code. Nice guy, but not happenin'. I'll bet he's surprised at the enduring interest in the thing.

I'm still working on the initial condition processing for the NUDiS pre-processor. Pressure pre-sets work, and automation of the initial conditions to include a new pop as Eric is talking about will be a feature once I get V and T presets working right.

I'm mostly falling into bed instead of being the PJ night owl these days. Vacation is coming up; I want to have something user friendly for the new year. So far, it's working great. Sam's got a great little solver here.
Mike Often wrong, never unsure.
__________________________

hinote
Posts: 1241
Joined: Sat Oct 04, 2003 1:54 am
Antipspambot question: 0
Location: Central California

Re: re: A New Friend and a New Lady

Post by hinote » Fri Dec 09, 2005 12:30 am

Mike Everman wrote: Too busy to see straight, unfortunately. "doing the impossible for no money in zero time" is our motto.

I'm mostly falling into bed instead of being the PJ night owl these days. Vacation is coming up; I want to have something user friendly for the new year.
Don't let the bastards get you down!! Say no, and if you have something worthwhile they'll come crawling back!!

I'm having serious hopes for your efforts with the NUdis pre-and post-processor.

I've had a fight with UFlow for several days now, regarding the intake optimization thingy (after another engine "failure to resonate", experience). When it works, it works OK--but I've only been able to get it to work in a very narrow window of operating parameters.

Mike--look for a private post about my current problems.

I'm committed to valveless PJ progress; if it's (occasionally) intermittent, I guess I'll just have to work that much harder.

Bill H.
Acoustic Propulsion Concepts

".......some day soon we'll be flying airplanes powered by pulsejets."

Post Reply