What I have been doing (Novice endeavours with rockets)

[Baldor]

A lot of time without posting, but I have been reading you every day.

 

After some rushed attempts last summer solstice, I decided to make some serious but leisurely progress.

 

 

This were some of the rockets I made last solstice.

 

First I needed was some real tooling, so:

 

 

And of course, I had to learn to machine. All my experience hands on was a few hours learning to face a lot of years ago. All my experience was with CAM software, complex geometries but not challenging to machine.

 

So, this is my first attempt to a 15mm core burner tooling:

 

 

The spindle is stainless steel, all the other parts are aluminium. I must redo the spindle, I'm having troubles machining the draft correctly. Nozzle diameter is 7mm.

 

So, now that I had the tooling, I needed some way to test the motors and get meaningful data and not only rely on eyeballing the launches.

 

 

Almost all material used was scrap I had laying around. I only purchased the 40KG load cell, and the sparkfun openscale . I need to calibrate it more preciseley, but for now it´s enough to make comparisons. Error not seems to be more than 10g.

 

So, time to make some motor and test it.

 

Test A:

 

15mm core burner.

60/30/10 BP, milled, pine charcoal.

Total fuel (including delay): 31g

Total weight: 64g

Sample rate: 80SPS

 

 

 

20180304_TestA.txt

 

 

 

 

 

 

[Baldor]

Now, a very similar motor with top fusing:

 

Test B:

15mm core burner.

60/30/10 BP, milled, pine charcoal.

Total fuel (including delay): 35g

Total weight: 67g

Sample rate: 80SPS

 

 

 

20180304_TestB.txt

 

Seems something have gone wrong. Max force arround 3500g, more than double the burning time, and some stuttering around max.

 

I have a previous test, bottom fussed as the first, and the results are consistent with the test in the previous post. I must try some more top fused motors, and see if this was a dud.

 

¿Some ideas explaining this last test?

 

¿Some sugestions in general?

 

To Do:

 

-Load motors to 1/2ID above spindle only. No delay.

 

-Make spindle without coned nozzle.

 

-Try hotter BP.

 

 

[pyrokid]

Very nice work! A lathe opens a world of possibilities!

[stix]

So first lets get this straight - when we are referring to "top fusing" we mean inserting the fuse all the way into the core toward the top of the motor (bulkhead) then igniting, and bottom fusing means igniting near the nozzle end?

 

I guess looking at the video it does look like that, with the tube on Test B directing the ignition into the core. The strange thing is, and I see your confusion Baldor, is that you would think the results would be the opposite. In that normally ignition further up the core would result in more initial pressure and therefore faster burn time. Ignition at the nozzle end means it may take a while for the flame to propagate up the core and therefore slower burning. Very puzzling.

 

The other thing to look at would be if the fuel density was the same, or perhaps there was nozzle erosion - do you make your nozzles consistent - did you measure the nozzle id after the tests? Both will affect pressure and therefore burn time.

 

For what it's worth, I checked the data and both motors had a "similar" total impulse of around 18-20 Newton Seconds. ie. Average thrust x burn time is the same as 1/2 average thrust x twice the burn time. Nevertheless, it's good to be consistent.

 

Also, although the delay segment was included in the fuel weight, my analysis shows that Test A produced a specific impulse of around 65 and Test B around 52, but we can already see that in posted graphs and data. An efficient black powder motor and fuel should produce around 80. But, like I said, if we subtract the delay segment, the specific impulse figures would be better.

 

Specific Impulse is the Total Impulse divided by the Fuel Weight. (Using the same units - in this case Newtons).

 

Sorry I can't offer any other useful info. I've not made that many BP motors, Sugar Rockets are my thing. Hopefully someone else can shed some light on the issue and give some advice for moving forward.

 

btw. Nice Lathe - wish I had one

 

Cheers.

Edited March 5, 2018 by stix

[Baldor]

Yes, top fusing near the bulkhead, bottom near the nozzle.

 

Yesterday I made another test, this time with no delay. I'm still formatting the results, but total impulse is consistent, max force was around 6500g, and the start and end of the curve were almost vertical, like a top fused motor should look.

 

All the test have been done with the some batch of BP. Consolidated to 6000PSI. I don't see erosion in the first motor, but see some erosion or cruft in the other two. Must measure to decide what is. In my todo list is making some more spindles to compare geometries.

 

Specific impulse was 57... There is a lot of room for improvement. I will make two batches of BP, one 60/30/10 as the one I used, and one 75/15/10. I will test different combinations, and when I find one that doesn't cato, stick to it. It could be different proportion depending of spindle geometry.

 

Your input is very informative. You are pointing me were to look.

 

And the lathe.... These chinese lathes are not very expensive, but need continuous maintenance and can only make light cuts. Search for CJ18A for example. Problem could be shipping to Australia.

[Baldor]

I repeated the top fusing test, with better results.

 

 

 

 

15mm core burner.

60/30/10 BP, milled, pine charcoal.

Total fuel (including delay): 32g

Total weight: 60g

Sample rate: 80SPS

Total impulse: 18.2Ns

Specific impulse: 57.7s

2018030C_TestC.txt

The curve seems more normal for top fusing. Am I right?

The bump at 0.075s seems to be the moment the paper for the fuse is expelled. I found the paper tube intact after the test.

Also, after the end of thrust, it doesn't go to zero. It drops to 3.4N, and keeps droping slowly for a time. Seems I used a little to much BP over the end of the spindle, and this "delay" is showing as a little thrust.

One think I don't like of the test stand is the rubber support for the engine. In the video you can see some movement. In the next test it will be replace by a HDPE sleeve 1/3 the length of the motor. The sleeves are easy to make, and I can make an specific one for any tube size.

For now I will keep the spindle geometry the same, and try to dial the BP. Tomorrow I will make a 200g batch of 60/30/10, pre-drying the KNO3 and the charcoal. Last batch made a clump in the jar. I will make also a 200g batch of 75/15/10, and try different mixes of both to find how hot BP I can use without catos (Tests will be performed out of the stand, of course)

And a question: I can get vineyard charcoal instead of pine, but its 4x the price. Should I see a big diference?

 

[OldMarine]

Vineyard coal got a very bad rating from Danny Cregan's tests.What other woods do you have available? (Don't know your location)

[stix]

That last Test C is pretty good. I agree that there must have been too much bp over the end of the spindle, which caused the last third of the thrust curve to have not much thrust. Although, it still did have around 300grams but heading down.

 

So we know you are using a 15mmID nozzle throat. What is the ID of the tube casing, also the length of the spindle etc.?

[Baldor]

Nope... 7mm nozzle, 15mm ID casing. :-)

 

The nozzle is conical, about 40º (20º per side), from 7mm to 12mm.

[Baldor]

OldMarine:

 

I'm in Spain. What I can get is wood for the fireplace, but are usually "hard" woods. Holm oak, olive.. The softer ones are pine and vineyard. Maybe I can get my hands on some common beech (Fagus sylvatica) but I don't know if will be an improvement over pine. For now, I'm buying this http://www.granvelada.com/es/materiales-para-hacer-incienso/1858-carbon-vegetal-en-polvo.html

[stix]

Oh, Sorry. All this time I thought from your previous posts that you are working with a 15mm ID nozzle... Ok, thanks.... hmmmmm...

 

Then given a reasonable known "standard", that is, nozzle diameter is preferable to be 1/3 of the diameter of the internal casing ID, one could put forward that 7mm does not create enough pressure. Alternatively 5mm would give more pressure.

 

7mm = lower pressure and perhaps part of your current problem. Sure, you can increase the fuel burn rate via a hotter mix - but I think smaller nozzle, shorter core = more efficient.

 

Others may disagree.

 

btw. I wouldn't call yourself a "novice". You've shown careful planning in what you have done. Not many people bother to do that.

Edited March 7, 2018 by stix

[Mumbles]

Vine charcoal in Europe has always been sort of a different beast that vine charcoal in the US. It might be worth a shot.

[Baldor]

Stix, I had the assumption tha 1/2 ID was the standard for BP core burners. The rocket tool sketcher ( https://www.amateurpyro.com/forums/files/file/115-rocket-tool-sketcher-by-pudidk/ ) asks for 7.5mm nozzle. If I had troubles making a 7mm spindle, making a 5mm on will be a nightmare. Must give it a try before I finish this round of tests.

 

Also, spindle lengt is 7.5xID. Spindle length should not affect pressure inside the chamber. Both chamber volume and chamber surface increases linealy with the spindle length. Idealy, a longer spindle will mean more gas volume at the same pressure. If my intuition isn't wrong, this will mean a higher effective pressure in the nozzle.

 

About the planning... This comes with the profession. I have been a mould designer for more than 20 years, the last five the technical office chief. Fortunately, now I'm in a more relaxed job. :-)

 

Thanks Mumbles. I will try with vine and compare. I don't feel like making a TLUD now, maybe in the future.

[Maserface]

For what its worth, standard "skyrocket" dimensions give a nozzle diameter that is 1/2 x the ID.

[stix]

For what its worth, standard "skyrocket" dimensions give a nozzle diameter that is 1/2 x the ID.

 

My apologies, yes, that's probably correct. That's why I should shut up when it comes to bp rockets

And that's why I can't have nice things

Edited March 7, 2018 by stix

[Maserface]

Stix. I didn't say it was the most efficient or powerful core geometry, just that it's common. You are absolutely correct that it may be more effective to modify the core geometry than the propellant.

[Baldor]

Maserface, I plan to play with both propellant and geometry. I enjoy working with the late, and the small spindles are something I have to work out. But the BP I'm using is very tame, 60/30/10 with commercial pine charcoal, and it's easy to modify. My idea is to have a base of data with the current geometry and progressively hotter BP, then start comparing different geometries against this data.

 

Next test should be a couple of motors with 75/15/10 BP, outside of the stand, of course, and see if they CATO. It one of them fails, I will mix 50/50 the tame BP with the hotter one, This will give me a 67.5/22.5/10. Rinse and repeat until I find an adequate composition for this tubes and geometry. Of course, if the motors doesn´t fail with the hottest BP, this means I have a lot of work to do with my BP.

 

Stix, you don't need to shut up. Even mistakes are informative, and as Maserface said, the standard doesn't mean the best, so yours wasn't a mistake.

[Arthur]

European vine charcoal is regarded as very hot, comparable to balsa.

[stix]

One thing to consider when testing, especially with smaller amounts of fuel, is what I call "Fuel Loss Compensation".

 

This only applies to vertical test stands.

 

As the fuel burns, it is consumed and therefore should be taken into consideration, that is, the load cell reading will be incorrect as the fuel is consumed.

 

No big deal in the end, unless you want to do "accurate analysis" of your recordings - perhaps worth thinking about though.

Edited March 10, 2018 by stix

[Baldor]

I thought about it, Stix, but 30g against a peak thrust of 5Kg don't seems much. Assuming a constant fuel burning rate (It's not constant, burning surface increases if perfectly lit), should be easy to subtract it from the results.

[stix]

You are correct Baldor in that it doesn't make too much difference.

 

But I like to be precise as much as I can be.

 

Do you Tare/Zero your measuring device first of after?

Edited March 10, 2018 by stix

[Baldor]

I tare it before, but then I subtract the weight of the motor in the spreadsheet. I ordered a few weighs for calibration, should have them next week, but the comparisons I made with other scales, I should be within a 5% error margin. I also weight the motor before and after, to see if the weight is consistent with the measurements I did when making the motor. (I put in a plastic glass a little more than needed BP, and weight it before stating pressing and after.)

[stix]

I tare after the motor has been mounted, meaning we start at point zero. When the motor has finished burning, the thrust becomes negative. My software then applies an algorithm which integrates the loss back into the data and redraws the graph, therefore showing a "true" representation of the thrust curve. I reckon 99.5% accurate, ie. 0.5% error margin or better.

 

In your case, the simplest way is to subtract 1/2 the fuel weight (15grams) from the mean/average thrust. This will get you a more accurate total and specific impulse, but the graph itself will never be a true representation of the test. A graph starting at 60-70grams doesn't look good in my opinion.

 

I'm sure that there will be some people falling off their chair laughing about my so called "accurate" recordings and whether it matters to the big picture. I think it absolutely does matter if you are doing small tests, ie. changing fuel mixes, nozzle design, fuel geometry....etc.

Edited March 11, 2018 by stix

[Baldor]

The reason I tare before is to see the weight of the motor before and after the burn in the collected data. This way I can compare with the weight measured in a precision scale. Once the data is in the spreadsheet it's easy to "tare" it again. It should also make easier to interpolate the fuel weight once I decide to do it.

[JMan]

Im not nearly into the data recording side as you two (although Id like to be Im just unsure how to set up a thrustplate and software to record it) but wouldnt you get much more accurate and real results if you had your vertical test stand so the rocket pointed up? That way the weight of the rocket is already zeroed and the fuel consumed (resulting in less downward force) would naturally be taken out.

 

Like I said I dont know how to set a stand up to accurately record but Ive thought about it and tried and thats been my plan (basically a seesaw with force table on one end and rocket on the other)

Edited March 11, 2018 by JMan