Designing an AV Bay for DD

Since there’s been a lot of talk about DD and AV bays lately, I thought I should document the AV bay I built for my L2 project.

I would estimate there are as many permutations of AV bays as there are types of chocolate chip cookies! This is of course just one way to do it, and I offer this solely as food for thought.

The AV bay is contained in this Little Dog DD, purchased as a kit from Rocketry Warehouse. It is a 54mm design with a 38mm motor mount.




When I started out designing this, I discovered that thinking is hard work. Allowing for everything in advance – bolt clearances, sheer pin placement, etc – was tough.

Before I could design anything, I needed to know what dimensions I was working with. So, measured my Raven and PowerPerch down to the mm.

Then, I needed hardware sizes.

My background in testing parachutes has shown me that recovery systems break predominantly in one of two places: at the weakest component, or, adjacent to an excessively strong component. Hence, part of recovery engineering is strength matching throughout the system.

Since I was using 3/8” Kevlar as the shock cord, at 3600 lbs breaking strength, that became the median I would attempt to strength match.

I selected forged, medium strength steel eyebolts from McMaster. These have ¼”/20 threads. After some quick calculations, I found this would fail between 3360 lbs (root area method) and 3975 lbs (tensile stress area method). This bracketed 3600 lbs nicely.

However, the threaded rod to hold the AV bay together became a conundrum. Calculating strengths for two ¼”/20 rods, I came up with:

• Medium steel: 7950-6720 lbs.
• 18-8 Stainless steel: 4452-3762 lbs.
• Aluminum: 2760-2258 lbs.

Somewhat like Goldilocks, I found the medium steel to be too strong, and aluminum too weak. So I settled on 18-8 Stainless Steel rods.


Designing
With some real dimensions, I started out with some notional sled placements. When I overlaid the profile of the Raven3 with PowerPerch, I found that conventional didn’t cut it… the Raven either exceeded the dimensions of the coupler, or, the Raven's bolts came in contact with the rods:





Soooo, I had to become creative. The result that fit nicely was certainly not a conventional sled orientation; but, it fit.





So, I cut out the sled, and started drilling holes!



(sorry there's no actual build photos. When I built this I didn't even think about a build thread)

The completed AV Bay


Here’s the Little Dog:



And an enlarged view of the coupler section:


From top to bottom:

• Reflective Trim MonoKote
• 6-32 button head socket screw
• AV Bay vent
• Alignment index
• Shear pin tapped & threaded hole
• Airframe vent

To help make assembly easy, I chose to use a drilled Index Mark:




As the shear pins are broken, the airframe slides off the coupler/AV Bay. This also shows the indexing used:




It’s worth addressing the 6-32 Socket Screws. I chose these because they were positive, durable, and easy to use. I had found plastic rivets less convenient.

The 6-32 screws mate into PEM nuts on the inside of the coupler:

Also, the three 6-32 screws are laid out equidistant around the airframe – except for one screw, which is offset by 1/16”. This ensures that the AV Bay can only be screwed together one way, so it’s correctly oriented; which ensures that all the holes will line up correctly; which eliminates the need for a switchband.



Speaking about the holes, here is one of the three vents. The vents were drilled after the upper airframe was screwed to the coupler, so both the inside (coupler) and outside (airframe) match perfectly.

Also, because everything can only go together one way, I marked to outside with the location of the Raven’s magnetic switch. This has made activating the altimeter easy as pie, not the struggle which I’ve seen when the switch location is a bit uncertain.



Continuing with the removal of the AV Bay, here it is after removal of the lower airframe/fincan:

The arrows point to (L) the tapped & threaded hole for one of the three 2-56 nylon shear pins. Having experienced drag separation before, I am done with that particular rodeo, and use shear pins now. If you look carefully, you’ll notice the shear pins are offset from the same triangular arrangement as the rest of the rocket. This is to clear the rods and sled inside.

To the (R) is one of two holes for the J-Tek ematch wires. This hole is as small as possible. In the interests of simplicity, I chose not to use a barrier strip as is conventionally seen. This simple change has sped up AV Bay assembly in the field immensely.

Also, the nuts are from McMaster, called Flange Hex Nuts. I keep losing washers, and this is a one-piece washer/nut combo.



Here’s the AV Bay by itself.




After removing the Flange Nuts, the sled starts to slide out.




With the sled fully out, you can see how it is permanently epoxied to the lower coupler bulkhead.

This is part of what makes for easy assembly. I really like how the PowerPerch is an all-in-one solution, it makes for a very clean install.



The bottom of the sled.

Although the aluminum tube that slips over the ¼”/20 rod certainly helps support the sled, and make the assembly stronger, its primary purpose is to be a guide for the rod as the AV bay is slid together. I sourced the aluminum tube from (surprise!) McMaster.


I shot this to kind of help visualize the AV bay. It is completely assembled without the coupler.

The sled was deliberately cut ~ 1/64” too long, then sanded down until it is firmly compressed between the two bulkheads when they are seated on the coupler. This makes the sled a structural element as well as securing it better.

Rigging for flight


I use the surgical tubing method authored by Tony (tfish). I’ll make these prior to driving out to the launch. This helps make assembly for a second flight an easy task.




Assembly of the charge to the Raven takes about 15 seconds to pull the wires through, insert the wires into the terminal block, and screw down the terminals.




I have really liked directly wiring the charges to the PowerPerch. There’s less to go wrong by wires coming loose and it certainly makes for a clean looking bay. Also, it has been amazingly reliable.

Of course, to use this method you pretty much are locked into an e-match that has long wires. Since that's what I use anyway, it isn't an issue like it would be if I were using Q2G2s.


Whew! Well, there it is. I hope this gives someone some ideas. A sincere thank you for reading your way through this!



All the best, James

Well, as "food for thought" goes, that post was just about as tasty as them chocolate chip cookies! Nicely done.

I like how, even though this is well and thoroughly thought out, in the end it's all kept simple and clean. I also like how you think of things in terms of the strength of the weakest link. It seems that so many designs have ultra bomber components in one place, and another part might be a quarter of the strength (so why have all that extra mass and strength in the first area?). An eyebolt that is 5 times stronger than the shock cord makes no sense. You seem to get that.

One question - it seems your AV vent holes are quite large, no? What was your reasoning or calcs in determining the size/number you went with?

Thanks for posting such a detailed account of how you go about it. Interesting and informative (and well done I might add).

s6

Very nice! My only question is, how do you seal the area around the e-match leads where they pass thru the bulk head?

Heres a picture of the Av bay in my Binder Excel I use terminal blocks epoxied in place on the bulkheads, the eyebolts are stainless steel welded as used in sail fittings. This was upgraded to dual altimeters prio to doing my L2

Very nice AV bay. Keep posting.

stealth6 said:

Well, as "food for thought" goes, that post was just about as tasty as them chocolate chip cookies! Nicely done.

I like how, even though this is well and thoroughly thought out, in the end it's all kept simple and clean. I also like how you think of things in terms of the strength of the weakest link. It seems that so many designs have ultra bomber components in one place, and another part might be a quarter of the strength (so why have all that extra mass and strength in the first area?). An eyebolt that is 5 times stronger than the shock cord makes no sense. You seem to get that.

One question - it seems your AV vent holes are quite large, no? What was your reasoning or calcs in determining the size/number you went with?

Thanks for posting such a detailed account of how you go about it. Interesting and informative (and well done I might add).
s6

Click to expand...

Well, ha-ha, I was worried someone might notice that.

This is a case of something calculated out to an eyelash, and changed for a totally unrelated prosaic reason.

Anyway, I first refered to the Altimeter Port Sizing charts on the vernk website. These charts, as far as I can tell, are built around calculations based on a 1 psi differential at 68 degrees F, through an orifice with a discharge coefficient of 0.5.

I also ran the calculations based on expected performance of this rocket - ~10,000' AGL at the Black Rock - and a Discharge Coefficient of 0.45. (You can integrate Bernoulli's equation with the Continuity equation, or, you can use this handy calculator which does just that) A bit of research into implied back pressures, and shooting for a barometric resolution of 10' (0.0033 psi at that altitude) resulted in the need to flow 0.433 CFM. This resulted in three altimeter vents @ 0.0526". Not having a 7/128" drill bit, I settled for 1/16" in this 6" x 2.1" AV bay.

What I found is that for typical rocket flights, that aren't in extremis, that the Vernk calculations are a good approximation. This rocket needed vents slightly larger than the chart would imply, but not by much.

Sooo...anyway...I drilled the holes at 1/16". One of my goals was to be able to see the blue power light on the PowerPerch, I thought had positioned it to do so. The hole was too small, though, and I had to drill it out to 1/4".

Meh. How's that for embarrasing.

rocketkid88 said:

Very nice! My only question is, how do you seal the area around the e-match leads where they pass thru the bulk head?

Click to expand...

I have some modeling clay kept handy just for that. However, the hole is quite snug on the J-Tek wires, and after getting lazy and not bothering with the clay, I have since not seen any residue intrusion into the bay. I will say it took about three shots at drilling the hole, starting obviously small and stepping up until the wire would pass, to get the right size. If I'd just drilled what had looked right it would have been too big.


All the best, James

Fdog said:

"This is a case of something calculated out to an eyelash, and changed for a totally unrelated prosaic reason."

Click to expand...

Great quote. And I know exactly what you mean; I've done things like this before myself - planned everything out just so, and then at the last second just "wung it" for whatever reason.

And about that not using clay or somesuch to seal the hole. I just had a flight the a couple days ago where I had a similiar situation as yours - a very tight hole/wire that I didn't feel the need to plug - and I ended up with a fair bit of residue inside the bay anyway. I'm actually sealing those holes this morning with epoxy.

s6

stealth6 said:

Great quote. And I know exactly what you mean; I've done things like this before myself - planned everything out just so, and then at the last second just "wung it" for whatever reason.

And about that not using clay or somesuch to seal the hole. I just had a flight the a couple days ago where I had a similiar situation as yours - a very tight hole/wire that I didn't feel the need to plug - and I ended up with a fair bit of residue inside the bay anyway. I'm actually sealing those holes this morning with epoxy.

s6

Click to expand...

I agree. Almost a random occurrence on just how the pressure wave releases from the charge (and what it takes with it)--seal the holes.

stealth6 said:

Great quote. And I know exactly what you mean; I've done things like this before myself - planned everything out just so, and then at the last second just "wung it" for whatever reason.
s6

Click to expand...

I think we have all done that at one time or another.

James, I understand the desire to connect ematches to the altimeter, but I would have to disagree that it's faster in the field to do so. I use terminal blocks and prep the av-bay the night before. I never have to open the av-bay at the field. Just connect the ematch and shock cord and I'm done, no opening and closing the bay.

I understand what you're saying Handeman, in that case it certainly is quicker!

On the advice of one of my mentors I use a freshly charged battery every flight, so I pretty much have to open the bay anyway, especially between flights. And most of our launches are multi-day events, we'll drive out to the playa, camping & launching for 3 days.

All the best, James

Very nice work. Thanks for the posting

Fdog said:

Click to expand...

My only question is with this picture, when you insert the Sled in the coupler/avbay. How do you plan on hooking up charges on the other Bulk Head?

Looks good. One suggestion that will make it easier to remove the sheared pins after flight. Only tap the airframe holes for the shear pins, and clearance drill the coupler. Then you just unscrew the half with the head still on. The other stub will just fall into the bay.
-Ken

BLKKROW said:

My only question is with this picture, when you insert the Sled in the coupler/avbay. How do you plan on hooking up charges on the other Bulk Head?

Click to expand...

The eMatch leads are left ~10" long. I just stick them through the holes in the upper bulkhead.

When the sled is slid onto the threaded rod, initially the Raven is about 3" outside of the coupler. The main charges are attached at that time.

As the sled is inserted into the coupler, the eMatch wires bend into an "S" shape.

So, the assembly sequence is:

1. Fresh battery into PowerPerch
2. Thread and attach apogee charges
3. Thread main charges through upper bulkhead
4. Insert tip of threaded rod into aluminum tube on sled
5. Attach main charges
6. Fully seat sled and tighten flange nuts

Sorry I don't have a photo of this, I'll shoot one and post it.


All the best, James

new2hpr said:

Looks good. One suggestion that will make it easier to remove the sheared pins after flight. Only tap the airframe holes for the shear pins, and clearance drill the coupler. Then you just unscrew the half with the head still on. The other stub will just fall into the bay.
-Ken

Click to expand...

Hot tip! I'll give this a try.


All the best, James