Build thread: Plasma Dart

The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
It's ok, you obviously need to figure out the logistics of the plasma cooling interface in the fins before you can start construction, otherwise there will be delays and cost overruns, etc...
By now SpaceX would have built and discarded (or blown up) about 4 prototypes....
 
Last edited:
Uh oh, any damage? Any estimates on when power will be restored? In some parts of SE PA the flooding was worse than Sandy.
 
Some tree damage. Someone coming to assess on Tuesday. They’re estimating power restore late tonight, but we’ve already been through multiple rounds of those promises and still no power, so we’ll see.

This was not nearly as bad for us as Sandy, but I would say it’s the second most damaging storm I’ve experienced here.
 
Damn, that sucks.

I was living in the eastern panhandle of West Virginia when that massive derecho came through tenish years ago. We were without power for a week, and didn't have a generator. And we had our own well, so no power means no water. Whenever it rained I would take off my clothes and go outside with a bar of soap. What we did have was a good privacy cover of trees.
 
PAYLOAD SECTION, PART 2

Because the payload section is BT60 and the plasma core is BT55, I need to adapt down. A small insert goes in the end:
Payload insert.jpg
And then the end is block-sanded as flat and flush and smooth as I could make it, which is to say *extremely* so.
Payload finished end.jpg
Pic shows a light coating of thin CA as well (sanded smooth of course).

I had exactly 2" of BT55 nose block left over from the Skywriter, which seemed to me just about the exact correct length. I inserted a dowel and an eyelet in the nose block, and then pushed the whole thing in half way:
Payload nose block.jpg
I also gave the block a wipe of thin CA, just to give it a bit of hardness and durability. Dunno if it really helps or not.

The whole assembly looks good, and it rock-solid.
 
You'll all be excited to learn that this might be my current favorite fin design:
1597094262831.png
However, I'm not convinced it goes well with the nose design. Seems a bit disjointed to me. I will have to think on that.

The expression you are looking for is "Don't quit your day job."
 
NOSE CONE FINISHING

This is now the third rocket where I'm using the same BT60 4.3" conical nose cone. I had originally designed for a longer one but BMS didn't have anything like that so I just added another of these to my order.

I am on a quest to perfect my nose cone finishing technique. I am not there yet. I always apply thin CA to my balsa cones to harden the surface, but I find it goes on rather uneven and lumpy, and I have a very hard time sanding it smooth. In the past, when I've really tried to sand it all the way to a glass-like smooth finish, I've ended up sanding through to the bare balsa in some places, needed another coat, and then the process repeats. And so I started no trying to sand it so smooth, and then applying a coat of CWF to fill in the surface irregularites, and then another thin coat of thin CA to harden that up. That works.

This time I thought I'd try a slight alteration. First, a coat of thin CWF:
Nose finish-1.jpg
My thinking was that this very smooth and even surface would take the CA more evenly, and enable me to sand it smooth. And so on goes the CA:
Nose finish-2.jpg
Not smooth at all. And so, another coat of CWF and then CA, and the final product...
Nose finish-3.jpg
...is *still* not glassy smooth. If I run my fingers over it I can feel that is it not smooth. However, it feels smooth enough that after filler/primer and paint it'll be fine.

However, I must conclude that the CWF before the first coat of CA didn't really accomplish anything. Won't be trying that again.

Here is where I put in another plug for BSI SuperGold CA. Virtually odorless, I'm able to do all this work inside without stinking up the basement or blinding myself with the fumes. Worth the extra cost for my purposes. I do apply it much more sparingly, not letting any more drip off the tip of the nose than necessary.
 
NOSE WEIGHT

Based on the OR model, I was unsure if I was going to need any nose weight. It seemed like putting a little in now could be beneficial and likely not hurt anything. The cone was already drilled for a dowel, so it seemed ready-made. After rooting around in my tin-o-miscellaneous-screws-and-nuts, I found a perfect pair which weighed in at about 0.2 oz:
Nose weight-1.jpg
I inserted it into the hole, nut-first (to put most weight towards the front, and found that the dowel hole possibly goes all the way to China. I stopped putting in the screw when it seemed "far enough", but not ready to emerge out the front of the nose. It's pretty far down in there, and I'm sure I could have pushed it further.
Nose weight-2.jpg
I slopped in some epoxy and it is well affixed. I didn't try to completely fill the hole with epoxy, no point.
Nose weight-3.jpg
Should give me just a bit of margin in case the tail ends up unexpectedly heavy (that seems to be a trend for me these days).
 
TWIST LOCK

This is now the fourth twist-lock I'm putting on a rocket. As always, it provides a secure and easy nose cone attachment for the payload section, allowing quick access to an altimeter, no dealing with friction fits. I'm glad I printed up a bunch of these when I first design them. Still working through my stock.

This being my fourth one, I have the installation process pretty well perfected. Here it is, installed:
Twist lock-1.jpg
Perfect. Or, well, not quite. When testing fit, I noticed:
Twist lock mistake.jpg
When locked, the nose cone separated from the body tube slightly, indicating that the piece in the body tube was slightly too close to the edge. This has never happened before, and the way I did the installation should have prevented it from happening. But there it is. I thought about leaving it like that, but since I knew how to fix it I realized I wouldn't be happy until I did.

(I'm also seeing some pitting at the edge of the nose cone that could use fixing. Ultra-closeups are not flattering).

The fix involved slightly reshaping the piece on the nose cone:
Twist lock fix.jpg
By thinning at the base with a needle file, and thickening the other piece with a few layers of foil duct tape, I was able to shift the opening slightly closer to the nose cone, and restore the proper fit when locked.

Turns out that the foil tape wrap is a very good method for tightening up the fit of the lock; slick and malleable and creates a nice smooth grab. Doesn't need to be tight, just enough so that it'll stay put if shaken.

Now to go clean up the base of the nose cone. :rolleyes:

Here's the completed payload section. In my OR model for this rocket I separated the payload section as a separate stage, so I could override the mass and CG of the whole assembly. Came out pretty close to the model. I'll probably see bigger variations in the rear.
Payload finished.jpeg
 
(I'm also seeing some pitting at the edge of the nose cone that could use fixing. Ultra-closeups are not flattering).
There's a saying among QA inspectors. (At least among some of them.) "Everything looks bad at 10×." (Meaning one should only use a 5× eye loupe for inspections.)
 
Quite true. Normally I don't inspect this closely; only did it for the gap and then noticed the other stuff. I hand-rubbed some CWF into the chips, added yet another wipe of thin CA, and it looks better.
Nose edge fix.jpg
I suspect that the twisting of the nose cone against the CA-hardened BT edge is taking its toll. I applied some more CA around the seam with the shoulder to harden it against further damage.
 
Well OpenRocket has thrown me for quite a loop. I think it's a bug, but I am investigating.

I have two versions of the rocket, one as single-stage and one as two-stage, with the payload section as the second stage. The outline of the rocket is the same in both cases.

The CP of the two-stage version is 2" further back than the one-stage version:
1597361557699.png
That makes a yuge difference in stability. I was working with the two-stage version, and thought I had 1.74 cal stability. The single stage version shows 0.65 cal stability. I would tend to suspect that the single stage version is more accurate, just on general principle. Shouldn't they be the same?

If the single-stage version is accurate, then I'm going to need a good bit of additional nose weight, which is very annoying. No other solution I can see. The profile of the rocket is pretty conventional, but I guess the strakes on the plasma core are doing me in. I wonder if they're being over-weighted, though, given that they are shielded behind the transitions up front (at least somewhat).

Need to work the numbers and see how much new nose weight I'm looking at. Argh.
 
Well OpenRocket has thrown me for quite a loop. I think it's a bug, but I am investigating.

Need to work the numbers and see how much new nose weight I'm looking at. Argh.

If you thread a hole into the nose cone and epoxy in a 1/4-20 threaded stud you can then add / remove weight as needed.

On my X-wing I actually threaded the composite insert that came with the nose cone I got from BMS.
 

Attachments

  • 002.JPG
    002.JPG
    257.8 KB · Views: 25
  • 007.JPG
    007.JPG
    169.5 KB · Views: 25
  • 009.JPG
    009.JPG
    220 KB · Views: 24
  • 010.JPG
    010.JPG
    159.9 KB · Views: 25
The twist-lock is already affixed to the base of my nose cone. I'll have to work around that, unfortunately no threaded stud will be possible.
 
OK, things are not as bad as they seem.

I made some corrections to the single-stage model, which was a bit out of date, and also updated the strake materials with a better guess than what was there (heavier, and pulling the CG forward). I end up with about .9 calibers. A JL Alt3 in the payload compartment at .35 oz will pull me to 1.4 cal (almost exactly 10% of rocket length) on a D12 or E30, and 1.25 cal on an E15. I can live with that.

So, if these numbers hold up (still more verification work to be done), then I'm OK with the altimeter and if I run without the altimeter I'll put an equivalent weight in there. If my tail ends up heavier than expected (likely, but unknown degree) then I may need a bit of additional weight. Not terrible. Still very annoying though.

Interestingly, with the extra weight, the D apogee went down, while the E apogees went up (small amounts, but noticeable.)
 
OK, things are not as bad as they seem.

I made some corrections to the single-stage model, which was a bit out of date, and also updated the strake materials with a better guess than what was there (heavier, and pulling the CG forward).

Did those updates also move the CP rearwards? I can see lots of things that would move the CG around, but if those two models had the same dimensions, it seems like the CP should have been in the same place.
 
Did those updates also move the CP rearwards? I can see lots of things that would move the CG around, but if those two models had the same dimensions, it seems like the CP should have been in the same place.
Ignorance on my part is bliss, but doesn’t keep me from throwing out ideas!

In simulations, does MASS effect CP? In “real life” adding mass without any OTHER CHANGES reduces velocity off the rod/rail, and at lower velocities fins are less efffective hence might change CP.

So when you go to the two stage sim, is it adding a motor or making some other change to the rocket as a whole?

I would assume in Open Rocket when the system separately does stability for the stack (motors in sustainer and booster or boosters) and for sustainer separately.

Have you tried any SIMPLE two stage designs, and gotten different calibers stability when simmed as single stage (I guess just put in a dummy weight for the forward motor when simulating single stage) vs as two stage? Maybe “something” in the program makes it treat multistages differently than single.
 
In my OR model for this rocket I separated the payload section as a separate stage, so I could override the mass and CG of the whole assembly.
The double stage model does not include a motor in the upper stage. The upper stage is a dummy that's only there so Neil can make the overall rocket's dynamic characteristics more accurate. So the liftoff weight is no different in the two cases.

On the other hand, a test of the software with with another, simpler two stage stack compared to a single stage with the same overall design sounds like a good idea.
 
In my opinion the CP of the full stack should be the same as the single-stage version, period. It's the same rocket. It has occurred to me, though, that descent will be screwed up, because OR thinks two separate pieces are coming down. So I'm probably going to abandon this two-stage approach to doing mass overrides.

For now, I trust the single-stage model more, so I'm going to convert back to it. I figured out how to accurately back-annotate my measured payload section into the single-stage model. Whatever nose weight I need, I will add.

I will pursue this problem in OR with the devs.
 
Back
Top