Krushnic Effect - Again

[jqavins]

The Krushnic effect came up again in another thread, and it brought my old questions to mind again. And yes, I've searched the forum and Googled the question several times, including just now.

How does the thrust degradation change with recess depth? To start in particular, how much degradation is there at one caliber? Whenever I think about recessing an engine, I think that I'd better stay down to 1/2 caliber to be sure I'm well clear of Krushnic. But it may well be that one caliber is where the degradation first becomes even noticeable, and by sticking to half that I'm applying "margin on margin" unnecessarily.

The broader matter could be addressed through a number different (closely related) questions, such as:

• How much thrust have you lost with a one caliber recess? None at all? 10%? 50%?
• What are the 10% loss and 90% loss depths?
• How rapid is the onset?
• How does the nozzle exit diameter (relative to the tube diameter) effect those answers, if at all?
• Some other characterizations.
• Does the propellant matter? I just read, from one source only, that composite engines don't suffer from the Krushnic effect. I find that really hard to believe, but I would be guilty of scientific sin to dismiss it out of hand.

Ideally, I'd like to see a graph of thrust relative to nominal for several recess depths, and a family of such curves for different nozzle exit to tube diameter ratios (and propellants).

If anybody knows where I might find such data - any of it - please post a source here.

Thank you.

 

[SolarYellow]

I'm not sure how calculable it would be. And even if someone calculated it, I'd want to see their hypothesis tested. Someone needs to build a test stand and burn some motors.

 

[jqavins]

Oh, yeah, that's what I'm talking about. Test data first, then curve fit. Theory from first priciples would be cool, and not what I expect.

 

[Funkworks]

Ideally, I'd like to see a graph of thrust relative to nominal for several recess depths, and a family of such curves for different nozzle exit to tube diameter ratios (and propellants).

I think it would also depend on the recess diameter (not just depth), and possibly the thrust curve itself, in that the thrust at a given time could depend on the thrust at a previous time (previous flow, local temperature). Temperature is also where the kind of propellant would come in. I'm sure a complete CFD analysis (first principles) can handle this, but I don't not know if there's any good approximation with easier equations that can come out of it. That's all I have.

 

[aerostadt]

There is lots of stuff out there as a search on Google, NAR website, and TRF shows. There is some data in this report:

https://nar.org/member-only-reports... Motor Thrust (J. Dunbar, NARAM 54, 2012).pdf

 

[jqavins]

I saw that one, and gave up on it too soon. "Oh, that's about using the effect, not characterizing it." Dummy. Characterizing the thrust retarder with the intent to use it is no different from characterizing it with the intent to avoid it. Graph 1 is a step toward what I want.

EDIT: I computed the %degradation from the measured impulse, expressed the recess depth in calibers instead of millimeters, and shot trial trend lines from the first non-zero degradation to the end. The different R² values might indicate that a quadradic is the best bet, but with so little data it would not be safe to place any bets. Actually, if the experiments are extended until the degradation approaches 100%, the curve should flatten out, my gut feel thinks.

 

[Daddyisabar]

Blast the data. What I need is something to quell the nightmares. There I am, walking up to the RSO table with my awesome modified TLP sports scale rocket. Bullet proofed using light glass, substitute Baltic birch plywood and a recessed motor mount holding in my big, super powerful motor. SO COOL! I didn't use any uncool, stinking, performance robbing nose weight. I had moved the motor mount up to balance compensate the rocket and my computer stability sim agreed. AGAIN, SO COOL.

Upon inspection the RSO makes a thumb/pinky phone and gestures "Mr. Krushnic calling!" Rejected at the RSO table again I'm triggered and hastily retreat to my panic room, complete with therapy puppies, comfort food, mood lighting, music and aromatherapy. My mindsimed beer can would have been perfect protection for the inner tube behind the motor!

Frustrated, I cut off the fin can and long motor mount section, making the small scale fins and motor stick out up front. I had drilled holes in the glassed motor mount rings to reduce weight, so air can now pass through. Launched off the ground I experience a weird sucking effect, mysteriously holding down my powerful bird. In my dream the RSO again appears and finger phone gestures "Mr. Burnoilli calling." OH the horror, I shout out "what's this have to do with a brand of pasta?" But then a miracle happens, the powerful motor overcomes the mysterious force and the chopped rocket flies straight up with a gentle tumble recovery. Was it "GDS that saved the day" I cry. The RSO says "No, it was just the beginner's luck of a silly, tractor motored oddroc. Not a Triumph of rocket science by any means. Nothing cool to see here."

So crushed I awake from the nightmare, vowing eternal loyalty to 3-4FNC with unrecessed, reasonably powered motors.

 

[jqavins]

I awake from the nightmare...

only to realize that every reasonable RSO in real life understands the one caliber rule of thumb, and to realize that Herr Bernoulli is so easily kept at bay by a mere clothes pin.

 

[shockie]

The Krushnic effect came up again in another thread, and it brought my old questions to mind again. And yes, I've searched the forum and Googled the question several times, including just now.

How does the thrust degradation change with recess depth? To start in particular, how much degradation is there at one caliber? Whenever I think about recessing an engine, I think that I'd better stay down to 1/2 caliber to be sure I'm well clear of Krushnic. But it may well be that one caliber is where the degradation first becomes even noticeable, and by sticking to half that I'm applying "margin on margin" unnecessarily.

The broader matter could be addressed through a number different (closely related) questions, such as:

• How much thrust have you lost with a one caliber recess? None at all? 10%? 50%?
• What are the 10% loss and 90% loss depths?
• How rapid is the onset?
• How does the nozzle exit diameter (relative to the tube diameter) effect those answers, if at all?
• Some other characterizations.
• Does the propellant matter? I just read, from one source only, that composite engines don't suffer from the Krushnic effect. I find that really hard to believe, but I would be guilty of scientific sin to dismiss it out of hand.

Ideally, I'd like to see a graph of thrust relative to nominal for several recess depths, and a family of such curves for different nozzle exit to tube diameter ratios (and propellants).

If anybody knows where I might find such data - any of it - please post a source here.

Thank you.

If you poke some small diameter holes around the periphery of of the body tube, then the recess length doesn't really matter other than you will be torching the inside of the body tube.

Gordon Mandell has a good article on the Krushnik effect in one of the early Model Rocket magazined at:
https://www.ninfinger.org/rockets/ModelRocketry/ModelRocketry.htmlIt's in a column titled Wayward Wind, Nov 1969.

The late, great Jonathan Dunbar did a NARAM-54 R&D report titled, "A Study into reducing the thrust of Black Powder motors", which is really good.

It's in the R&D section of the NAR member area.

By increasing the recess length to 1.5 times the body tube diameter(18mm) he was able to reduce the total impulse from a C6 to a B6.

The A8, B6 and C6 all have the same nozzle geometry so I would make an un-educated guess that the reduction would be the same, about 50%.
So as you go from 13 to 18 to 24 to 29mm BP motors in minimum diameter tubes, you would see different reductions in total impulse.

I would assume the thrust reduction would depend on the engine, and the nozzle parameters, and the diameter of the body tube along with propellant type. CFD would help here.

Lindsay Audin did one of the 1st studies in the NAR Technical Review #1 at:
nar.org/pdf/TCR1.pdf

 

[jqavins]

Lindsay Audin did one of the 1st studies in the NAR Technical Review #1 at:
nar.org/pdf/TCR1.pdf

I re-read it today, before posting. It has nothing on the progression of degradation with recess depth. It, in fact, names that as a topic for further research.

 

[bjphoenix]

Someone needs to build a test stand and burn some motors.

I did that almost 55 years ago.

I found an old photo. This was for a school science fair project, I made a test stand with a small electric motor turning a big coffee can, and a motor mount that slid on a steel rod pushing against a long coil spring. I haven't found any photos of the test stand but it looks a bit like the one in a photo in an old RDC catalog. These tests were done with Estes 18mm motors, probably A or B since we didn't have C motors back then.

 

[Daddyisabar]

I promise no never create a secondary combustion chamber behind the motor's nozzle.

 

[Daddyisabar]

only to realize that every reasonable RSO in real life understands the one caliber rule of thumb, and to realize that Herr Bernoulli is so easily kept at bay by a mere clothes pin.

But would it look good for my really really good looking rocket to be propped up with a clothes pin on the pad? A fancy stand off or no go. Image is everything. Vanity at the pad requires the entire package to be good looking, even close up. None of this "oh it looks good ten feet away" stuff.

Push the rule of thumb envelope to over a caliber. Larger diameter tube with airflow, fast burn smaller motor with a smaller nozzle and flame. The dark arts are calling!

 

[jqavins]

I did that almost 55 years ago.

I found an old photo. This was for a school science fair project, I made a test stand with a small electric motor turning a big coffee can, and a motor mount that slid on a steel rod pushing against a long coil spring. I haven't found any photos of the test stand but it looks a bit like the one in a photo in an old RDC catalog. These tests were done with Estes 18mm motors, probably A or B since we didn't have C motors back then.

View attachment 569086

Exactly the sort of thing I wanted! Thank you!

 

[Gunstar]

Here is another wrinkle, what if you put a cone inside the recessed area, small opening at the top for the motor, open as big as the body tube at the bottom, would that change anything?

 

[rharshberger]

Here is another wrinkle, what if you put a cone inside the recessed area, small opening at the top for the motor, open as big as the body tube at the bottom, would that change anything?

Maybe, my Mindsim says it will, simply by reduction of area, might even count similar to a over-expanded nozzle. No data to back up my mindsim though.

 

[Daddyisabar]

A mindsim without data to back it up is....TOTALLY AWESOME!

 

[lakeroadster]

Here is another wrinkle, what if you put a cone inside the recessed area, small opening at the top for the motor, open as big as the body tube at the bottom, would that change anything?

Likely it will help right up to the point that the cone catches on fire.

 

[rharshberger]

A mindsim without data to back it up is....TOTALLY AWESOME!

Its actually quite liberating.....free the mind!

 

[Cape Byron]

A mindsim without data to back it up is....TOTALLY AWESOME!

Nothing a good swing test won't solve.

 

[shockie]

https://www.rocketryforum.com/threa...ay-with-1-2-body-tube-diameter-recess.165704/

 

[SolarYellow]

Bringing this back up.

Thinking about near-MD construction using composite motors. All the real testing linked above seems to be based on BP motors where the nozzle is already inset in the case of the motor itself.

It makes intuitive sense that a composite motor with a protruding nozzle would have different results or characteristics with respect to the Krushnic effect, but I haven't yet seen any testing related to it. In particular, I'm thinking about a boat tail essentially glued to the back of the thrust ring. The result would be something like the CTI 29mm tapered/flush closure, but still with a thrust ring to push on the back of the fin can. I'm wondering how long it could be without losing impulse. And would including a few vents near the top of the boat tail help?

Also, it seems that static tests might deliver significantly different results than tests in which the air is flowing past the rocket body at possibly supersonic speeds, so how does one even test it, given the variation in output of production motors?

Apparently, this is something I've thought about previously:

I’d look at fabbing a 5-degree taper and end it at a 90-degree cone outward from the perimeter of the nozzle boss. Just as a guess that feels right. You could always build a test stand and burn some motors.

For clarity, this is what I was thinking. When I wrote the above, I was thinking of the green line, but I suppose an argument could be made for the orange line as well.

This is just to illustrate the principle. Obviously, the OD of the boat tail would start at the airframe OD, not the motor OD. (Okay, I was thinking "sub-MD/flying case" when I drew it.)

View attachment 608659

 

[jqavins]

Bringing this back up.

Thinking about near-MD construction using composite motors. All the real testing linked above seems to be based on BP motors where the nozzle is already inset in the case of the motor itself.

It makes intuitive sense that a composite motor with a protruding nozzle would have different results or characteristics with respect to the Krushnic effect...

Not to me. It seems to me, without data or aerodynamic models to back this up, that the rules of thumb should not be any different, when you define the recess distance as from the edge of the MMT to the nozzle. I don't see any way that the fuel grain location or the threaded edge of the casing can matter.

In particular, I'm thinking about a boat tail essentially glued to the back of the thrust ring. The result would be something like the CTI 29mm tapered/flush closure, but still with a thrust ring to push on the back of the fin can. I'm wondering how long it could be without losing impulse. And would including a few vents near the top of the boat tail help?

Regarding the vents, I think I've read somewhere that does, indeed, eliminate the Krushnic effect (if the vents are adequately sized). Which makes sense to me, since the Krushnic effect arises from low pressure in the MMT behind the nozzle, and vents such as you suggest would let outside air in, alleviating the low pressure.

Also, it seems that static tests might deliver significantly different results than tests in which the air is flowing past the rocket body at possibly supersonic speeds, so how does one even test it, given the variation in output of production motors?

With seriously reduced thrust while stationary, how do you propose to reach the high speeds that you speculate would alleviate the thrust loss? Are you thinking of a two stage with only the sustainer motor recessed? Or just one motor in a cluster? (I suspect that high speed would not help, but it doesn't matter if you can't accelerate to reach it.)

 

[jqavins]

Regarding the vents, I think I've read somewhere that does, indeed, eliminate the Krushnic effect (if the vents are adequately sized). Which makes sense to me, since the Krushnic effect arises from low pressure in the MMT behind the nozzle, and vents such as you suggest would let outside air in, alleviating the low pressure.

Poking around, yes, yes I did read that. In post #9, right here.

 

[cbrarick]

I once saw someone use the Krushnic effect on some recessed D motors, then did a "air start" on the pad of a central J black motor at 1 second. This was entirely planned to work that way, and it was cool. Lots of noise and smoke then as it died off, the central J motor hit and the rocket took off! Very cool.....

 

[sghioto]

Did this experiment back in June of 2015 about the same subject:

Picked up a 3 pack of C6-0 Estes engines and static tested today with the motors recessed 1 and 2 calibers. Results indicate about a 15% loss of impulse at 1 caliber and a 37% loss at 2 calibers.

 

[jqavins]

Lovely. It's interesting that the reduction in peak thrust at two calibers recess is substantially less than the reduction in average thrust and total impulse. Do you think there's anything to that, or is an anomaly in the measurements?

 

[sghioto]

Do you think there's anything to that, or is an anomaly in the measurements?

I think that may be due to that peak burst recorded at the end of the burn which might be an anomaly.