PET third stage COTS tank

.


Before:


During


After




Betterbottle makes 5 and 6 gal PET water bottles. I could only find a source for the 5 Gal, so that is what I tested.

Empty 674 gm
MR with water of 29 with peroxide 39!
Burst at 125 PSI.
Long period of plastic deformation at around 100 psi
It burst in the thin center with the ends remaining substantially thicker.
This means one could probably prevent the banana deformation and increase the pressure with a center overwrap.

Way out of the box presure system...

Another out there concept... this won't work with the muiulti segment tank of the last wild idea, but will work for a single tank.

In a H2O2 rocket 80% of the propellant volume is H2O2, so you need to pay more attention to the H2O2 side than the fuel side.

Leave a small space at the top of the H2O2 tank. In this space put a solid silver screen/catalyst hung from a thermally compatible string. Unwind the string into the tank letting the silver decompose the H2O2 pressurizing the tank, raise and lower the catalyst to control pressure.

Fuel is pressurized with a bladder from the gas in the main oxidizer tank.

Main propellant valve is a simple burst disk.

Way out of the box...

A 3rd stage made of 3L soda bottles surrounding a 3U Cube sat.
3L soda bottles are good to 100PSI with a Mass ratio of > 40 when used with H2O2.

The "goodness" of normal rocket tanks are usually rated by the PV/mass. The best aerospace composite tanks, like the Scorpius pressurization tanks used by Armadillo in the 180sec super mod can have mass ratios of better than 9 at 2200 PSI working pressure. If you could scale this to a 200 PSI tank then you get a water filled tank mass ratio of >80. The fundamental problem is for the small sizes we are talking about you can never get that low. The tank wall would be paper thin. At small sizes the issues are minimum gage not minimum strength. (Hard to beat a soda bottle 0.25mm thick. )

As soon as you go up in size bigger than minimalist nano sat launcher then properly built aerospace quality tanks are a must.

I also suspect that when fighting minimum gage issues going up in pressure for a tiny launcher can have benefits in motors size, expansion ratio, flow path sizes etc....
A big puzzle with too many knobs.

UPDATE:
Here is a link to the specs for high quality aerospace tanks...

A salute to the Military etc...

For my U.S. followers remember the Americans that voluntarily offer to pay the ultimate price in defense of this nation. Reguardless of how you view the current conflicts, you can not help but be humbled by the voluntary sacrifce these men and women give in our name.

Give them something, I contribute every year to http://www.lbeh.org/

List of significant parts....

In trying to determine the scope of a Nano sat launcher project I generated the following list or physical hardware. There is an equivalent scope of work for Ground support, Software, Logistics, Regulatory issues etc...

1. Nose cone/aero shell
1. Aero shell separation system
2. Nano sat mount
1. Nano Sat ejection system.

3. Avionics:
1. GPS/ MEMS IMU Integrated system
2. Avionics control
3. Telemetry system
4. Battery System

4. Third Stage
1. Tanks (assuming tanks are the structure)
2. Valves and Actuators
3. Pressurization System
4. Motor
5. Motor Thrust transfer structure
6. TVC Actuators.
7. Separation mechanism
8. Fill and Drain Umbilical system.
9. Roll Control TVC

5. Second Stage
1. Third Stage Support Structure
2. Battery for second stage actuators.
3. Tanks (assuming tanks are the structure)
4. Valves and Actuators
5. Pressurization System
6. TVC Actuators.
7. Motor
8. Motor Thrust transfer structure
9. Separation mechanism
10. Fill and Drain Umbilical system.

6. First Stage
1. 2nd Stage Support Structure
2. Battery for first stage actuators.
3. Tanks (assuming tanks are the structure)
4. Valves and Actuators
5. Pressurization System
6. TVC Actuators.
7. Motor
8. Motor Thrust transfer structure
9. Fill and Drain Umbilical system.
10. Launch Hold down mechanism

Spacex....

Congratulations to spacex.
The scope and scale of what Elon and his troops have accomplished is just breathtaking.
I am in stunned awe.

I raise a toast to a most unreasonable man!

Testing results... sort of...

We went out to FAR setup the teststand, and the little stainless motor.
The temperature was really cold (45 F) and the cat pack never warmed up. By the time it would have been warm, it was dead, due to contaminated peroxide.

I disassembled the cat pack and dropped the screens in the peroxide dregs from the Jug we loaded from and there was almost no activity.

I then got a jug from a different batch and tried that, no activity.

So I cleaned the screens in 68% nitric acid. I then tried them in both batches of peroxide. Batch #1 killed the screen almost instantly.Batch #2 was reactive, so I let a screen soak in batch #2 after 2 or three minutes the screen was no longer reactive. The only conclusion is that both batches of peroxide were contaminated, batch #2 less so.

As a Final test I took a fresh clean screen and the screen that sat in batch 2 and put both in my beaker of peroxide, there was a big difference in reactivity, something is definitely contaminated. My Peroxide supplier thinks it was phosphate and has redone his proces to remove phosphate.

The end result is that engine tests will have to wait till I can get new peroxide or get my spargeing plant up and running and make my own propellant. The current tall pole in that process is diesel generator problems/waiting for parts.

In the mean time I'm going to work on tank development and some guidance and control electronics.

Some thoughts on an orbital launcher.
Assumptions:

• All stages contribute the same DV (Bad assumption)
• 9500 M/Sec Delta V may not be enough for a really small vehicle.
  
• MR is achieved MR per stage including payload.
• We assume the MR is the same for each stage (again a bad assumption)
  

This table shows you the minimum ISP performance you need given an achieved MR and a number of stages. For Reference the Blue Ball with no payload and no legs had a MR of between 5 and 6. If my notes are correct the Falcon 1 first stage with no Payload and no 2nd stage has a MR of about 16.

MR is down the left side, number of stages is across the top.

	Minimum ISP
Stages	4	3	2	1
MR
2	349.3	465.7	698.6	1397.1
3	220.4	293.8	440.7	881.5
4	174.6	232.9	349.3	698.6
5	150.4	200.6	300.9	601.7
6	135.1	180.2	270.2	540.5
7	124.4	165.9	248.8	497.7
8	116.4	155.2	232.9	465.7
9	110.2	146.9	220.4	440.7
10	105.1	140.2	210.3	420.6
11	101.0	134.6	201.9	403.9
12	97.4	129.9	194.9	389.7
13	94.4	125.9	188.8	377.6
14	91.7	122.3	183.5	366.9
15	89.4	119.2	178.8	357.6
16	87.3	116.4	174.6	349.3
17	85.5	113.9	170.9	341.8
18	83.8	111.7	167.5	335.0
19	82.2	109.6	164.4	328.9
20	80.8	107.8	161.6	323.3
40	65.6	87.5	131.3	262.5

The second chart shows the vehicle gross lift off weight given a 1Kg payload and assuming that the empty stages weigh the same as the payload they are lifting.

		GLOW multiplier
	4	3	2	1
2	256	64	16	4
3	1,296	216	36	6
4	4,096	512	64	8
5	10,000	1,000	100	10
6	20,736	1,728	144	12
7	38,416	2,744	196	14
8	65,536	4,096	256	16
9	104,976	5,832	324	18
10	160,000	8,000	400	20
11	234,256	10,648	484	22
12	331,776	13,824	576	24
13	456,976	17,576	676	26
14	614,656	21,952	784	28
15	810,000	27,000	900	30
16	1,048,576	32,768	1,024	32
17	1,336,336	39,304	1,156	34
18	1,679,616	46,656	1,296	36
19	2,085,136	54,872	1,444	38
20	2,560,000	64,000	1,600	40
40	40,960,000	512,000	6,400	80

The two bold numbers are about where I think my notional vehicle will land on this chart. This was a 1Kg nanosat with 1Kg of support structure would have a gross liftoff mass of ~2200 lbs, and the engine performance is a relatively easy ISP of 232 (Easy in vacuume, a bit harder on the first stage) Total Vehicle empty wt would be between 500 and 600 lbs.

you can do some really gross things... like a monoprop 4 stage launcher... MR 6 Glow for 1Kg payload is 40000 lbs.

My weekend ......Frustrating

We went out to FAR and set up the test stand.
Everything checked out we put in the fuel (gasoline) and
were preparing to load peroxide. When we opened the cap
there was a black substance stuck to the cap, when we open more fittings
there was an unknown black substance. The tank had been a peroxide tank.
It sat in the garage sealed for 2 years. On Friday we hydroed the tank and say no sign of contamination. The rough drive out to FAR dislodged some kind of black/brown powder from the inside of the tank. The powered looks like Iron rust on close inspection and is very reactive with peroxide. So the test was scrapped. It makes me wonder if something is leaching out of the aluminum after being stored for a long period.

We need to add interior visual inspection to the check list for all stored tanks, even if properly sealed and stored.

We worked on the big generator, burning the stale diesel and replacing with fresh. (Still not running)

On the plus side we did get our primary storage container cleaned up.
(As a result the garage at home is in distress once again)

We tested some of the super chemical compatible K190 resin for peroxide compatibility. It responded almost identically to the easier to get Vipel F010. A slight etching of the surface.

We'll see if we can fabricate a new test stand tank this week, and try again next weekend.

My Weekend....

I just finished hydroing the test stand and setting the pressure relief valve.
I have to hook up two pressure transducers to the electronics and the Stainless motor is ready to test. We will probably go out to FAR sometime Saturday and test the motor, work on our generator, and clean out our container. The clean out the container part probably means we will be out there overnight Saturday night and return some time Sunday.

Ways to save the world... and more Rockets

Saving the World
Any long time blog reader reader has already heard my rant about the only way to really spread an American standard of living to the entire population of the world requires that we find new resources, not just recycle our soda cans. I believe that off planet resources will be a significant part of the solution for things like rare earth metals, platinum group metals, expanding humans into the solar system etc.... The other half of this is a cost effective green energy source. I'm still very skeptical about the ability of space based solar power to solve the energy issues here on earth. I think its much more likely that small modular fisssion reactors and hopefully fusion will solve the energy problems.

A few weeks ago I posted about LPP. since that time I have visited their facility. I went through the lab and got to watch several "fusion" shots. They have multiple different ways to detect neutrons and are clearly generating neutrons via D-D fusion. They have obviously spent a fair bit in building their lab, but none of it was wasted in any way. They have good instrumentation and a well setup shielded room, but the rest of the lab space itself could be generously described as spartan. Its obvious that the people working on the project are passionate about what they are doing and believe they have a pretty good shot at making it all work. As a result I've decided to personally invest some $ in their business. If you are an accredited investor give them a call.

Back to Rockets....
I've started in earnest to work on a plan for winning the Nano Sat contest. There was a big meeting to kickoff the contest last week and in my mind the only significant questions left about the contest will be

• How do they resolve the no government funding rule,
• Who will be the allied organization.

The overall goal of centennial challenges is to encourage innovation from individuals and companies that are not part of the traditional aerospace environment. The goal of the specific Nanosat prize is to try and kick start a low cost nanosat launcher. At some level the two goals are in conflict. There are many small SBIR sized companies that have received government funding to work in this area, and so excluding government contractors from the contest will limit the pool of people that might succeed, at the same time including people receiving significant government funding really violates the spirit of the centennial challenges.
Personally I would like to see the same no government $ rules that were part of every previous centennial challenge.


On the technical and regulatory side of the house many many questions remain.

This weekend I hope to finally fire my stainless printed motor.
Building more real hardware really requires that I get more details of my notional launcher nailed down.

A personal non rocket tale and warning.

This has NOTHING to do with rockets but I think its important.
For the last two plus years I've felt tired and somewhat depressed.
I just thought I was getting older and though it was normal to feel drained all the time.
Last year while I was driving toward the LLC I was not getting a lot of sleep, largely by choice.

After the LLC I was looking forward to getting my energy back but no matter how much sleep I got, I woke up feeling trashed. I could barely work past 9pm without becoming dangerous and/or unproductive.

A month or so ago My wife went up to Long Beach for 4 days.
On Sunday I slept by myself and Monday I felt terrible. On Monday I drove up to Long Beach and spent the night with her, Tuesday I felt fine. Tuesday night I was home alone and Wednesday I felt terrible. Wednesday night she came home and Thursday I was ok.
I pointed this out to her and she said "when you roll over on your back you snore and I elbow you to make you roll over."

A light goes on My 80 yr old father has Sleep Apnea so I call my Dr and explain the A-B-A-B experiment I had just done. He says he will schedule me for a home sleep study..... 10 days later its tied up in the insurance authorization never never land so I buy myself a recording pulse oximeter and record a nights sleep. I find that my O2 sat goes to less than 80% more than 20 times in a 6 hour recording.

Armed with my data I say dam the insurance and full speed ahead. So I go in and see a sleep doctor that schedules a formal sleep study. The result is that I was having 33+ apneas an hour,
and not just while I was on my back.

When this happens your body senses that something is wrong and sets off all the panic alarms and wakes you up. As a result anytime I got into REM sleep it was short lived and unproductive.
You also get to activate your fight or flight alarm hormones every night. It also explains why I would wake up a 3 am feeling exhausted yet unable to go back to sleep. My body was in panic alarm mode. I was inadvertently running the REM sleep deprivation experiment.

I got a CPAP machine on Wednesday this week. Its only been two days, but I feel 1000% better. I wake up feeling rested, I worked late in the shop last night and got a lot done.

I'm writing this so anyone that snores and is feeling tired will seriously consider getting a sleep study. I did not look forward to sleeping with the Mask and hose etc.. but the change in how

I feel has been so dramatic that I would do it in an instant even if it meant wearing the mask 24/7. So if any of you are feeling tired and run down, not yourself depressed think about a sleep study.

I realize that something once written is on the Internet forever and that this revelation
might cause me some grief in the future while looking for jobs, contracts etc.... but if just one person reads this and it prods them to get checked out it will be well worth it.

As a side note I hope this will mean a lot more progress on the rocket projects!

Weekend update...

The stainless motor and test stand are almost ready to test.
Waiting for some more silver screen to arrive so I can make small cat packs rather than wastefully cut up 6" packs.

I worked on putting some more storage and shelving into the garage.
(My son finally gave up on the fantasy that one might someday park a car in there)

Sunday I went out to FAR and worked on getting generator set up for Propellant processing facility. I've never had a gas motor beat me, alas the universe is telling be I'm not a diesel mechanic. Will try again next time on site.

I will be at the SSI.org conference on the 29th 30,31st so if your there say hi.

As a science geek I enjoyed the 4 part videos from LPP power they are trying to do fusion on a shoestring and are being very open about the process. A long shot but if something like this works it changes the world..... Part 1

Composite Compatability tests.

I soaked 8 composite samples in 85% peroxide for about 5 hours at 70F.
Top Row:

• Epoxy with no fiber
• Glass fiber in epoxy
• Kevlar in Epoxy
• Carbon in Epoxy

Bottom Row: Vipel 010 Chemically resistant vinyl ester

• Vipel Resin no fiber
• Glass fiber with vipel
• Kevlar with Vipel
• Carbon Fiber with Vipel.

Experimental setup.
I set out 6 polyethylene cups with 300ml of 85% H2O2.
I tied polyester cord to each sample and placed them in the H2O2 solution.
They were placed so 75% of the sample was covered.
The Pure resin and glass fiber samples with the same resin shared a cup.
I put a clear polyethylene bag over each cup.
I periodically observed the samples.
I left them in the sun for ~5 hours.


The bags of 4 of the sample cups were clear, the bags over the two carbon fiber samples has fogged condensation on the inside of the bag. This seems to indicate some H2O2 decomposition.
It was not enough to measure any significant volume change.
None of the samples had shown any significant h2o2 decomposition. They were filled to 300ml +/-10ml and at the end they were all 300ml +/- 10ml.

I then removed the bags off the top of each cup with a long set of tongs.

After the 5 hours I was hesitant to handle the sample or move the containers as I was possible I'd created a hazardous solution. So I could not weight the samples.

After 5 hours the epoxy samples all had a layer of something floating on the top of the H2O2.
The Vipel samples had no visible contamination of the h2O2.

Then one by one for each sample (except the pure resins)

• I grabbed the loose cord tied to each sample with the tongs, carried them over the the wire fence and hung them on the fence.
• I then shot the sample with a 243 winchester with 85gr bullets from about 25 ft away. The impact velocity was 900 to 1000 m/sec.

None of the samples showed any sign of detonation or any increase in energy release from the rifle shot. I did not shoot the pure resin as I expected the pure resin would shatter and not provide any additional information.

After determining that none of the samples detonated I then individually dumped each h2O2 container on the dirt. (Mojave dirt is a great decomposition catalyst.) No notable difference in energy or release was noted.

I then removed each sample from the fence and soaked them in water to remove the peroxide.
The samples are shown in the picture above. The Vipel samples showed some surface etching, ie they had an oxide layer on the surface, but showed no significant structural degradation.
The epoxy samples wer reduced to bare fiber. There was almost no structure left.

In the picture both bare resins look discolored, but the Epoxy is bubbled and distored, where the layer on the vipel is just a surface layer its very thin. It almost looks like the oxide layer aluminum gets after a lot of peroxide exposure.

The company that makes the vipel F010 has an even more resistant resin, K190 that is harder to get. The K190 is rated for 50% peroxide at 100F for years of service.
Their chemical compatibility guide

When I'd finished pouring out the peroxide in the 6 cups I had a 100 ml or so left in my transfer cup. So I put a bunch of raw kevlar with no resin in that cup. It had ZERO reaction of any kind, no bubbles, no discoloration, no steam on the bag etc...

First test part in the bag

My first tank part attempt is in the vacuum bag and should be curing. Its been 90 minutes since I mixed the resin and its still fluid and not getting warm yet....I hope I converted drops of hardner to grams correctly for the resin mix.

It is a cold evening so.....it could just be slow

Having resin that does not go off makes a nightmare.
In any case I doubt the first attempt will be pretty....

Update the resin in the mixing cup got hard, I peeled the vacuum bag blotter felt and peelply and the underlying composite is still tacky. (Its really cold and damp in San Diego today so I'll let it sit another 24hrs.)
.

Tanks the other hard part

The Rocket Equation has a term that is the rocket motor efficiency and a term that is the mass ratio. In a pressure fed system the biggest component of that mass ratio is the tank.
If I could make a tank identical to a 2L soda bottle only 5 to 10 times bigger that would be almost ideal. I could probably find a mechanism, to join a bunch 2L soda bottles and I may pursue that, but I'm also going to try my hand at composite tanks. Toward that end I made mold forms for a 6" diameter cylindrical test tank mold. I heated it up and waxed it with mold release and setup the vacuum bag etc.. got ready for my first composite test part.

I have something non-rocket to do tonight, but tomorrow I hope to try making my first composite part since the solar plane project.

The first tanks will have glued Mylar liners that are not H2O2 safe, but the flight tanks will have mylar/PET liners with no glue. The jury is still out if the tanks will be seamed via vacuum heat, sonic welding, or RF welding.In addition to the liner I also want the tank to be chemically resistant to the H2O2. I want any liner failures to not be immediately catastrophic.

I'm getting a bunch of samples and setting up to do compatibility testing next weekend out at FAR. I'm testing glass, kevlar and carbon fibers with normal epoxy, vinyl ester and special chemically resistant vinyl ester resins. My gut says that plain E glass with the chem compatible resin will be fine, I'm almost certain that carbon with epoxy won't be. I hope the Kevlar and chem resistant resin also work as the kevlar is a bunch stronger than the glass.

Its a multi part test, start with 30% and work your way up to heated 85 or 90%. Anything past 50% really has to be done remotely. I'm planning to soak samples in PET soda bottle bottoms. sitting in hot water on a hot plate. The sample will have a string and pulley so I can pull the sample out of the solution remotely and shoot it with the 243. I could set up some kind of hammer and anvil test, but I worry that if it goes bang, the hammer becomes a projectile.

I figure I'll get the preliminary screening done this weekend and the samples that pass will get the hotplate test in two weeks.

A few ideas and loose ends

This week I learned that the NASA Nanosat prize is funded, (I thought it was a proposal not a funded prize.) So this clearly puts a bulls eye on a project target. This process starts with a notional vehicle and works backwards from the last stage. With a 1Kg payload I think that I can build a three stage launcher with a gross liftoff mass of less than 1000lbs.
To do something on this scale every gram matters. I'm starting to designbthe best 3rd stage I can and works backwards... (If you do a good enough job you only need two stages)

One needs:
Efficent light high expansion motor. Working on small motors via the DMLS project.

Really light tanks.....I've purchased all the large beverage containers I can find and am weighing and pressure testing them. So far the best tank is a 2l soda bottle. 100+PSI and mass ratio of > 40. A single soda bottle is not really enough. Any scheme to group them together is heavier than the bottle. So I'm looking at maybe fabricating my own tanks. The idea H2O2 tank would be mylar lined (mylar is another version of PET like in soda bottles) with chemically compatible resins and kevlar. Normal Epoxy and normal Carbon fiber are not happy with H2O2. All the tanks at this scale are up against minimum gage issues so whatever pressure you get in the tank will be more than really needed. In concept these seem very similar to the H2O2 tanks shown by Richard Speck of Micro-space at the 2007 LLC. I've got most of the materials to try this on the way. The biggest unknown issue is how to build the liner. If I can make the process in patent #3661675work that would be ideal.

Really light valves and actuators... probably all based on brushless model airplane servos. one can get lighter brushed parts, but must trade against the requirement to put them in a pressure box and the smallest brushless ones seem to be a win.

Guidance.... more on that later...


Regulation... This one is hard because the regulatory framework that applies to orbital launches is tailored toward something like a big Atlas. My current concept is to make the 2nd and 3rd stages really light and fluffy. Almost all plastic, including the valves and plumbing. Only the thrust chamber will be non-plastic. The first stage will probably be metal. The concept is to launch off shore so that thrid party safety from the first stage is constrained by physics and the 2nd and 3rd stages are too light to survive reentry. The real trick will be proving that and path taken by the 2nd stage will either burn up or land within the 1st stage constrained by physics zone. IE if it shoots straight back down its not going so fast so it might burn up. If it flies perfectly by the end of its burn then its going to fast to survive reentry. The question is what happens between these two zones. My nominal vehicle meets all of the amateur limits except the no orbit allowed rule. IE less than 200Klb/sec and orbital altitude less than 150km. (Rules say you only need to go around once.) more rambling to follow as I figure things out.

Anyone in SoCal with access to a 14" diameter 20" long vacuum oven that I could use to try the process described in the patent above would be really helpful.

More Printed Motor news...

The DMLS machine that is most common is an EOS270. The company that makes these machines has recently started adding some new materials. The new Material I'm most interested is Aluminum for chambers, and maybe Inconel 718 for a possible turbo pump.

I'd gotten some chamber quotes a year or so ago and it was still too expensive. When I had GPI quote my design again this year it was half of what last years quote was and I ordered the part on the spot. (In Stainless) GPI was also very helpful in making geometry suggestions that woul d improve the quality of the resultant part.

After placing this order I thought I'd do some more price research and sent the same file to Morris Tech to quote. The response from Morris was very competitive and its clear that this industry is seeing significant price declines. (Neither vendor knows what the other quoted) I hope its a sustainable trend and not a bloody battle to bankruptcy.

I really don't like beating vendors against each other because any long term relationship needs to be win-win and you can't do that by abusing your vendors. While it might be tempting to beat vendors prices against each other, its not some thing I do.

I also asked both Morris and GPI about the availability of Aluminum.
GPI indicated that they would soon be running Aluminum and Morris indicated that they had already run some Aluminum parts and that in the next few weeks they would be running some more aluminum test parts.

Since I was doing interesting stuff as an individual inventor not a big corporation Morris offered me the chance to add a single part to the Aluminum test run for approximately their direct cost. Its an offer I can't refuse. So I will soon have small DMLS chambers in two materials!

The Aluminum Motor is actually a bit bigger than the stainless motor as it needs to be cooled the whole length with a stainless cat pack support thermally isolated from the cooled aluminum chamber as a sleeve inside.

So if you need DMLS parts there are at least two really awesome DMLS vendors in the U.S.

http://www.morristech.com/
and
http://gpiprototype.com/

Why the motor has horns.

If you look at this post from July. You will see how the finished motor will look. Rather then spend lots of $$ making the simple upper section I only had the lower part printed with DMLS. The upper section where the cat pack will go is fabricated in the normal way from a sanitary fitting. It will and then be welded to the motor bottom.

The horns sticking out the side are the fuel feed. The fuel goes in the curved part and a 8-32 Set screw with a orifice drilled in it goes down the straight section. Then the end of the straight section gets plugged. You can buy predrilled orfices in 8-32 set screws from Mcmaster Car.

Here is a picture of all the bits before welding:

The two small parts are the machined elbows to connect the vertical feed tubes with the Chamber top.

Stainless Chamber bottom on the way.

I received this picture today: