Update:

HAPB-3 was launched 5/12/2008 at 7:45 am and recovered ~3 hours later. Maximum recorded altitude was 103,274 feet. I hope you enjoy the Images. The videos are below in the Launch and Recovery sections of this document. Added Google Earth KML Path. Right click ove the url and "Save Target As", then open with Google Earth.

Flight Computer

I decided to stay with the STAMP for another launch and change a few things from last time. I will reuse HAPB-2’s PCS and SCS setup with some minor changes. I will not be using servos on this flight but I am reusing one of the connectors for the new transistor switch I have installed. Thats right, I am moving away from servo controls and going to the world of transistor switching for camera activation.

The transistor switch is very simple. It consists of 2 – 2 pin connectors, a NPN transistor, and a 1K resistor on a daughter board that I have attached to the main board. The right hand connector goes to the camera and the left hand one is connected to one of the original servo connections. Pin 12 use to be the control line for one of the servos but is now the control line to the transistor switch. It turned out that I had a 1" x 1" spot available on the main PCB so I made sure the new switch PCB would fit in that spot.

I have also upgraded my film camera to the Coolpix L15 digital camera. This looks like it will be a great camera to use due to its 8Mpix pictures and optical VR image stabilization. It even has a PAL/NTSC output jack which may come in handy for later flights. Since I will be using this new digital camera with a transistor switch I have adjusted the HAPB-PCS code accordingly and while I was there I figured I would clean up some code in other areas as well. I pulled out the read memory routine and wrote a seperate program for that function called ReadEEPROM. I also removed the external temperature routine since the probe I was using can not go down to -60F. This allowed me to increase my data gathering to 272 records.

Other changes are in the “Takepicture” code where I have replaced the conditional altitude statements that where used for triggering the film camera with simpler code that triggers the camera by placing Pin 12 in a high and then low state. If you want to know how to modify the CoolPix L15 to use a transistor switch I have written up the procedures along with schematic and sample code. The process was very easy and I explain it in my Modifing the CoolPix L15 document.

HAPB-1 will supply the housing and instrumentation panel. There is not much difference in the instrumentation panel from the previous flights. Since I wanted to reuse my first panel I only needed to reprint a new label and cut it down in size.

Before	After

The system is still stackable but I have cut out the unnecessary compartment seperators and I am using the PCB's themselve as the seperators. The entire system is much more compact then in the past flights measuring in at only 6"x6.5"x6". Pretty close to a 6" cube and weighing only 1lb-7oz including batteries and radio.

Telemetry System

I am using the same telemetry system from HAPB-2 with no modifications because it has proven to be very dependable. I am however using the same power source to run the PCS and SCS. This will cut down on the broadcasting time considerably but testing has shown that there is still plenty of time for recovery if this flight goes as well as the first two flights.

System Testing

Testing has been completed with the new capsule and electronics. I ran 4 testing lasting 3 hours each to verify functionality of the computers and they are working the way I intended. The new camera is working great and I have easy access to it so I should get some pretty good pictures this time around. In fact, it should exceed 272 high resolution images (~3.8 Megs each). The video will not be as good quality as the pictures but it should be interesting to see.

I will be using the new Lithium batteries from Energizer® and have determined that the primary computer will fail in ~5 hours but the telemetry system will last about 8 hours. I have also completed the installation of the new software on the new tracking laptop. I am able to track both the balloon and my truck on the same map. I have been trying to do this since the first launch but was not successful until now. I am extremely happy with the results of the testing this time and anticipate a very productive launch and recovery.

Completed all testing and installed new software on tracking laptop. The new laptop with UI-View32 and Precision Map v8 is working great. Both the car and the balloon will now be on the same map. This will make tracking much easier and give my chase team a better perception of where we are with respect to the balloons position.

Launch Day - 5/12/2008

The original launch was suppose to take place Saturday May 10, 2008. Friday I had checked the weather and path predictions and decided it was a go. Around 11:30 that night I remembered seeing an article about President Bush's daughter getting married at Crawford Ranch so I jumped out of bed and searched the web. Sure enough - she was getting married that Saturday. I had forgotten to look for NOTAMs! For those of you that are not familiar with NOTAMs (Notice to Airman) they are notifications filed with the FAA that describe airspace conditions over designated areas. I searched and found 2 in effect until Sunday night. Lake Whitney was off limits. I spent the next 2 hours looking for an alternative launch site but the predictions where not good. Every place had a lake landing so I decided to postpone the flight until Monday morning.

The first 30 minutes of the flight followed the general path of the prediction.  HAPB was slightly more north but heading east.  But then something unexpected happened.  She started ascending in a spiral pattern as if caught in an upper atmospheric tornado.  This started to occur around 60,000 feet and continued until it burst at apogee which was slightly over 103,274 feet.  I say “slightly over” because I received a confirmed reading at 103,274 feet with the next reading 20 seconds later at 101,524 feet.  During those 20 seconds she was still going up but I do not know by how much.  With a little bit of work I could figure out the final altitude by comparing the raw data against the video timestamps but that would still be calculated.  I have decided that HAPB-4 will include a max altitude algorithm so I will not have to concern myself with those 20 seconds between broadcasts.

I have drawn the predicted path verses the actual path and you can see that the spiral pattern changed the ascent phase considerably. If you look at just the descent portion of the two paths you can see that the descent generally followed the path of prediction including final turnaround just before landing.

I have generated several charts that came from the raw data received during the flight. I have also included the raw data in excel format in case you want to play with it yourself. It is more granular (20 second intervals) then the charts below that are based on minute intervals. Finally, I included a KML file for those that like Google Earth.

Altitude	Capsule Temperature	Horizontal Velocity	Vertical Velocity

The video camera captured some cool footage, much better then I anticipated.  I had my suspicions about the quality of a $50 digital video camera but it did a pretty good job.  The Earth really is an amazing planet – it looks peaceful and so very quiet.  It’s hard to believe that our atmosphere is so thin and extremely fragile.  Enjoy the 5 minute video as you ascend and descend with HAPB-3 to the top of the world.

Recovery

At one point it was coming down fast over Richland Creek reservoir. It cleared the water (whew!) but then took an unfortunate turn at 2949 feet and headed back towards the water. SPLASHDOWN! Right at the edge. From that point it took 30 minutes for us to drive to it. The whole time we didn't know if it was actually in the water or not. GPS data could have been off or the reservoir could have been low. We bit our fingernails and hoped for the best during that time. When we arrived we found the entire area largely fenced with barbed wires and NO TRESPASSING - GOVERNMENT PROPERTY - VIOLATORS ARRESTED signs and we couldn't see over the giant berm so we still didn't know.

Damage Assessment

We all stayed dry including HAPB-3. A small amount of water got in the capsule only at recovery (we verified this on the video). Pictures, video, and data was recorded from the time I started the computers until I turned them off over 3 hours later. What an exciting day!

Back