ProjectHAB | Experiments in High Altitude Ballooning

Adding a default stencil layer in Eagle

Posted on December 2, 2013 by Steve Smith

After all the work setting up a stencil and adding a layer to produce the Stencil Gerber in the last post, I wondered if there was a way to automate the process a little more. What was needed was a default Stencil layer to appear in Eagle when a new board was made. This is achieved by editing the eagle.scr file within the Eagle directory structure. Make sure you make a backup copy of the file before you edit it. I copied mine and renamed the copy BAK_eagle.scr_BAK so that it would be a simple step to rename back if I made a mistake and Eagle no longer started.

Edit the file with a plain text editor. On Windows, notepad would be a good choice because it doesn’t add formatting. Avoid Wordpad, Word and other editors that add invisible formatting to the text.

Open the file located in your \Eagle\scr\ directory called eagle.scr This is the script that is run when a new schematic, board or library part is defined. Find the BRD: section which should be at the top of the file. After the current settings and before the SCH: heading, add ‘LAYER 255 Stencil’. This will define a new layer numbered 255 called ‘Stencil’ every time you create a new board. Save the file and start Eagle normally.

Start Eagle and open any board you have made before. If you haven’t previously made a board, click on FILE-NEW-BOARD If you now open the layers dialog by clicking on the layers icon in the left hand toolbar or clicking VIEW-Display/hide layers, you will see there is a layer at the bottom of the list called Stencil. Any board that didn’t have the new layer before the eagle.scr edit was made will need to be saved before the next step.

To add it to a CAM file or create one fresh, click on the CAM icon or FILE-CAM Processor. You can add it to your normal CAM file too, click on FILE-OPEN-JOB. Click on ADD to add a section. In either case, name the section Stencil. Set the device for GERBER_RS274X. Set the filename to %N_stencil.gbr This will use your projects filename and append ‘_stencil.gbr’ to it. Make sure pos. Coord and Optmize are checked. Then go into the layer list and make sure that only 31 tCream and 255 Stencil are highlighted. Then you can save the job for later use.

Once you have set up the above, you can then call on the section in the job after you have designed a board and it will output a gerber file combining the solder paste layers along with your new stencil layer. Of course, if the job already creates standard gerbers, the stencil gerber will now be added. I would recommend that you preview the file before you send it to your fabricator and perhaps edit to taste as I documented in my last post ‘Designing a Solder Stencil with Eagle‘.

Please note: Editing configuration files can stop applications from working. ProjectHAB takes no responsibility for damages or corruption due to following these instructions. You do so at your own risk. If you are not confident editing files, please ask someone who is.

Posted in Electronics | Tagged board, Circuit, Design, PCB, Soldering, Stencil | Leave a comment

Designing a Solder Stencil with Eagle

Posted on December 1, 2013 by Steve Smith

With the completion of VAYU-NTX I wanted to explore the use of a solder stencil rather than use my component by component soldering approach. I haven’t got an oven yet but I believe I can still use the hot air station at a pinch. I had a chat to Mitch of Hackvana who will generously create the necessary gerber file for stencil production and in this instance I have gone ahead with that option – Thanks Mitch! However, I do like to know how to do these things for myself and I like to be able to customise them to my taste easily. This led me to investigate how to produce the necessary gerber file. I thought that it should be possible with Eagle‘s facilities and I was right. Although Mitch has produced a solder stencil for VAYU-NTX, I chose to use this PCB design to see if it could be done. I took screenshots to show how I managed it.

The first thing I did was to align the PCBs holes on a 10x10mm grid. This is to make the board compatible with the marvelous PCB Fixture block made by Arachnid Labs based on the Stencil8 system by Zach Hoeken. The idea is that you peg your board down and then lay the solder stencil on top. Because of the registration of the holes, the solder stencil is perfectly aligned with the PCB. You can then squeegee solder paste across to apply a thin layer of solder on each pad. I will be documenting my first attempts when I receive the next batch of PCBs and the stencil from Hackvana.

So, to make sure the PCB holes align properly, you need to set Eagle’s grid to 10mm:

VAYU-NTX v1.04 with 10x10mm grid in Eagle.

This can be done by either clicking on the dotted icon in the top left corner or going to VIEW – GRID. Set the size to 10, the unit to mm and set the Alt value to Finest by clicking on the Finest button. It doesn’t matter what units are set for Alt. I had previously aligned three of the holes on the VAYU-NTX PCB to be on a 10x10mm grid so it was an easy job to select everything with the group tool and move the group whilst holding Alt so that the holes aligned with the grid. After getting it as close as possible, I clicked on each hole individually whilst holding the CTRL key. This aligns them exactly to the grid. If your PCB hasn’t already got 2.5mm holes aligned to the 10x10mm grid, you will need to modify your design to comply. The pegs that are supplied with the pegboard are 2.5mm so set this size in Eagle and place your holes where the lines cross. It’s best to keep the holes on the outer edge to minimise the solder that gets on the pegs. Of course, if you can afford the extra PCB size, you could lay an extra border around it to accommodate the holes.

The next step is to create a new layer for your stencil outline. Click on the layer icon on in the tool bar on the left or click VIEW-Display/hide layers. You can then click NEW in the dialog and create your new Stencil layer.

Creating a Stencil Layer in Eagle

I chose the name ‘Stencil’, the colour of Magenta to show up against the black background and the first fillstyle. Click on OK and again, OK and then on the wire icon. You can also click DRAW-WIRE. Before you draw, set the miter to 5mm in the toolbar and the width to 0. You can then draw a rounded rectangle or in my case, a square to represent the outline of the stencil. In our conversation, Mitch said he normally gives 30mm border around a PCB. For this example, I used roughly 20mm border for each side.

Drawing the Stencil Border on the new layer.

If you find it difficult to draw with the miter, set it to 0 and draw your shape with square sides. You can then set the miter to 5mm and click on each corner to make it rounded after the fact.

We now need to add the registration holes so that the solder stencil will lay immediately above the PCB. First, set the grid to .5mm. Then select the Circle tool and draw a 2.5mm circle with 0 width over the drill holes that are aligned with the grid.

Drawing the new holes

Here I have zoomed the image up and you can just see the magenta outline of the new shape. To make it exactly 2.5mm, start the circle in the center of the drill hole and drag downward to it’s lower edge. Align it with the bolder inside circle of the drill graphic. Once you have successfully created one shape, duplicate it as many times as needed and lay the duplicates over the holes you require.

Duplicating the hole shape

In the above image, you can see the duplicated shape. Use your middle mouse button to pan to the next hole. You then need to duplicate some more hole shapes and place them on the border of the stencil. This is easily done in the same way.

All new hole shapes placed

Note that I have placed the holes on the nearest grid position to the edge of the stencil outline.

Now what’s required is to output a gerber file for the stencil. To do this, click on the CAM icon in the toolbar or click FILE-CAM Processor. Because I use Hackvana for my PCB fabrication, I had their CAM file already loaded. This makes no difference and you can start from a blank file.

Setting the CAM file

I have named a new section to ‘Stencil’ since I intended to save the CAM Job after use. To create a new section, click on ADD and scroll to the right most section if not immediately visible. Set the device for GERBER_RS274X. Set the filename to %N_stencil.gbr This will use your projects filename and append ‘_stencil.gbr’ to it. Make sure pos. Coord and Optmize are checked. Then go into the layer list and make sure that only tCream and Stencil are highlighted. Then click on ‘Process Section’ and the gerber file will be created. You may see a yes/no dialog pop up regarding a drill file. If so, click on yes. I have not as yet found out what this means but it doesn’t seem to have a detrimental effect on the gerber files.

Next, you will need a program called Gerbv. Load the newly created gerberfile by clicking on FILE-Open Layer(s).

New file viewed in Gerbv

Using this program has two benefits. It allows you to check if your new stencil gerber file looks correct and affords very basic editing. In the above image, you can see me removing two pads by selecting them with a marquee (click’n’drag) and then pressing delete. You can individually click and delete shapes too. I edited these pads out because they are for the power supply and won’t necessarily be soldered up when the components are laid on the PCB. Once your edits are complete, you can FILE-EXPORT-RS-274(gerber) to save your completed stencil file.

Supply this file with your regular gerbers and you can order your Stencil along with your PCBs.

Posted in Electronics | Tagged Assembly, board, Circuit, Design, Layout, PCB, SMD, Soldering, Stencil | Leave a comment

Released!

Posted on November 28, 2013 by Steve Smith

I’ve been busy testing the VAYU-NTX PCB and cobbling together code to enable it to send telemetry. I have managed to combine Anthony Stirk‘s RTTY demo code and Matthew Beckett‘s NORB 3 code along with a few snippets of Chris Stubbs‘s CHEAPO code to create VAYU-NTXs firmware. As is always the case with these things, there were several problems along the way. Not least the hardware modification I needed to carry out to get the v1.03 board to accept Anthony’s PWM based routine. However, after I’d sorted that and added some features from the original VAYU code, I just couldn’t get the Satellites value to register properly. It turned out to be a rogue character that managed to slip into the ‘C’ formatting line in the code. I’m not sure how it got there but I didn’t know enough ‘C’ to spot it. It took the keen eye of Matthew Beckett to exclaim ‘Ah!’ and help fix the problem.

The telemetry currently sends the following information:

$$callsign, sentence counter, time, latitude, longitude, altitude, distance from launch, GPS lock (1/0), battery voltage, satellites – Checksum

Don’t forget to change the callsign and launch co-ordinates in the code before use. You will find the callsign in the parse_NMEA function. The two ‘$’ signs are required for decoding so leave those in. The launch co-ordinates are in the variable declarations at the beginning of the code. If you don’t know the co-ords. for your launch, you can find them by looking on Google Maps. Locate it, right click on the map and click on ‘What’s Here’ in the pop-up menu and the Lat/Lon will appear in the search bar.

VAYU-NTX v1.04 Eagle Preview

When testing the original v1.03 PCB, I found it useful to check the +VCC line a few times and the TXD and RXD lines from the GPS unit. I have added dedicated test points for this purpose. The +VCC and GND pads are above the GPS unit and the TXD/RXD pads are just to the left of the GPS unit near it’s top strip of contacts. Also new from v1.03, LED2 is now orientated the same as LED1 and I have added extra sized contacts on the crystal pads to accommodate a smaller SMD type 5052 package. The option for bridging the switch position is more clearly marked and (not shown above) the battery connections are shown on the silk screen.

I have ordered a fresh batch of PCBs from Hackvana with the fixes mentioned above. This is with an additional soldering stencil to allow more convenient assembly. I’ll document the process when I build the first of the new batch.

Links to all the design files can be found on the VAYU-NTX v1.04 page.

Posted in Payloads, VAYU-NTX | Tagged Assembly, board, Circuit, Layout, PCB, Programming, SMD, Soldering, Stencil | Leave a comment

Evolution

Posted on November 21, 2013 by Steve Smith

Tuesday 19th saw delivery of the VAYU-NTX v1.03 prototype PCBs from Hackvana. I am really pleased with the consistency of quality and finish on their PCBs. I chose a blue colour this time and I really like it. It’s not a traditional blue but leans towards green and I like the effect. Well done Hackvana once again.

VAYU-NTX v1.03 PCB Top & Bottom sides

I am also very pleased with how the silk screen print shows up. It is very easy to read.

During that afternoon, I assembled the first board. I used the same techniques I describe in a previous post. It takes quite a while so patience is of the essence, but I would rather take the time to get it right than ruin a board.

VAYU-NTX v1.03 PCB fully Assembled

I did have a little difficulty with the 100µH coil. The base was a little too large for the footprint I had accounted for on this board and I had to break a small piece of ferrite off to get it to fit properly. It doesn’t seem to affect the operation but I will source an alternative component for the final version. You can see in the above image, the capacitor behind the coil is at an angle. I eventually had to de-solder the coil altogether and re-seat it to get it to connect properly. Apart from that minor setback, the board went together relatively easy.

Later that day, I looked into trying some test code and after uploading the standard Arduino bootloader, I got one of the on board LEDs to blink.

VAYU-NTX v1.03 PCB Blinking test

So having ascertained that the ATMEGA329P was operational, I looked into running some RTTY code. Anthony Stirk had alerted me to some posts he had written on how to get this to work, namely his ‘Getting started with the NTX2b and the Arduino’ Part 1 and part 2. I realised very quickly that I had made a bit of a blunder on the VAYU-NTX circuit. I had connected the NTX2b’s TXD pin to a digital pin on the ‘328 that wasn’t PWM (Pulse Width Modulation) capable. This could be made to work with a potential divider circuit but not with Anthony’s new code.

The solution was reasonably simple. I cut the line from the ‘328 to the NTX2b by baring the PCB trace and cutting with a sharp blade. I then ran a flying lead directly from the NTX2b’s TXD pin on the bottom of the board that could plug into the expansion header in an appropriate place. It turned out that I had made pin 3 a PWM line so that sufficed. In future revisions of the board, this issue will not occur since I have re-routed it appropriately for Anthony’s code, but more importantly, a PWM capable pin on the ‘328.

VAYU-NTX with flying lead

Once this small modification was carried out, and Anthony’s code modified to point to the correct pin, it fired up – to a point. Anthony’s code is based around using 5v to power the Arduino. VAYU-NTX runs on a boost regulator supplying 3.3v. This meant that the RTTY signal had a proportionally smaller shift, around 235Hz. Ideally, it needs to be around 450Hz. After a conversation with Anthony and Ed in the #highaltitude IRC channel, I managed to change the values in the code to reflect the different voltage and VAYU-NTX can now send RTTY telemetry sentences.

VAYU-NTX v1.04 RTTY Test

I will be continuing to update the PCB design whilst I get the code together. Another small improvement I have made is to add SMD pads for a smaller SMD crystal and flip LED2 round to be the same orientation as LED1. Why I placed it like that, I really don’t know. I have also added a marked jumper on the switch circuit to show the option to bypass the switch if required.

Now to interrogate the GPS module!

Posted in Payloads, VAYU-NTX | Tagged Assembly, board, Circuit, Component, Design, Layout, payload, PCB, Programming, Radio, SMD, Soldering, VAYU-NTX | Leave a comment

Spam OFF!

Posted on November 8, 2013 by Steve Smith

I’ve switched comments off on the blog. It’s gradually got more spam as time has gone by. Whilst I look into a way of preventing this, I will accept e-mails to steve (at) projecthab.co.uk

Posted in Spam | Tagged spam, time, waste | Leave a comment

Vayu-1 Re-launch

Posted on November 5, 2013 by Steve Smith

On November 3rd 2013, I had the pleasure of re-launching VAYU-1 from a local location, Shenstone Park. I chose this location because it is in very easy reach of my home but mostly because it is relatively high against the surrounding terrain. In fact, it is part of a hill that gives me problems when tracking HABs to the north and north west. There are several hazards such as power pylons nearby but Shenstone Park is above them.

Shenstone Park Location. Green arrow indicates. launch spot.

I prepared the balloon and payload over the hour leading up to the designated 12:30pm launch time whilst I waited for my friend Graham and his son Sebastian to arrive. This gave me time to carefully fill the 36in foil balloon with Helium to the point of buoyancy. I had previously weighed the payload and found it to be 38.7g. I added enough helium so that the balloon just lifted this weight. However, conditions were very blustery outside and I was concerned that without enough lift, the balloon wouldn’t clear the local obstacles upon launch so before leaving, I did give it some more helium. Graham and Sebastian arrived at around 12:15 and after a quick chat, we set off to the launch point. Although within walking distance, Graham drove us to avoid unnecessary strain on the balloon and payload. Within a minute or so, we were there. We parked in Old Road and entered from the top of the park. We made our way to a likely spot (indicated by the green arrow on the image above).

Unlike the Youlbury launch, I had managed to fill a couple of standard latex ‘party’ balloons with helium to act as tests, and this time we used them. The first test did manage to show us the wind direction but didn’t have enough lift to clear the surrounding fences, however it allowed us to fine tune our launch position to take advantage of a dip in the surrounding greenery towards the buildings in the corner of the park. I did attempt to set up a uStream channel before the launch but my unfamiliarity with the system meant that no one got to see it. For some reason, it didn’t record the launch either. Perhaps I should have recorded it manually and uploaded it later.

Ready for launch. School buildings visible in the background.

I remembered the experience of the Youlbury launch and how Dave Akerman waited for a lull in the wind before giving me the instruction to launch. Once ready, it took several minutes of waiting for the right window. While we waited, some interested onlookers, walking their dog, passed by but didn’t ask what we were doing. Finally, the wind dropped sufficiently for launch and at last, VAYU-1 was airborne!

Its Away! Finally, VAYU-1 is airborne!

We watched as it climbed and hoped it wouldn’t get caught in any of the surrounding greenery. It didn’t and kept climbing. It was a great feeling to see VAYU-1 disappearing towards the clouds. I hadn’t brought any tracking equipment with me so I was trusting the payload to have a good GPS lock. It did have as it turned out.

Graham, Sebastian and I made our way back home to evaluate whether or not to chase over a cup of tea. It was soon apparent that I had given VAYU-1 a little too much helium to land in the UK and was predicted to end up in mainland Europe. The signal from the payload was very strong, stronger than I expected even though we were relatively close. We followed VAYU-1’s progress on SpaceNear as it crossed the Thames multiple times.

VAYU-1 Track on SpaceNear

By this time, there were a few receiving stations and my own station was getting a fine signal. I feed the audio from an FT-857D into my PC to track. The frequency was good, just a few Hz below the designated 434.300Mhz at 434.299.87MHz. I had set the SpaceNear filter to only display VAYU-1 to keep the display clean. As it made it’s way parallel with the Thames, VAYU-1 climbed steadily with a few dips from turbulence. I was very pleased with it’s progress. All the while, I was chatting to the HAB community in the IRC channel #highaltitude Someone joked that I’d beaten my previous distance record already (1.44 kM on the first launch).

Progress was good and Wouter PA3WEG in the Netherlands was now receiving good telemetry.

Wouter PA3WEG receives VAYU-1 telemetry.

The prediction that SpaceNear provided fluctuated quite a bit from Germany to Poland and back. I would have been happy to get my payload into another country. Sadly, at around 13:38, VAYU-1 started a fast decent. I was amazed how fast it was coming down, reaching over 10 Meters per second at one point. Had the balloon burst? The frequency had changed dramatically for a few seconds but re-settled back to 434.299MHz

VAYU-1 Decent into the channel.

We watched as the receiving stations slowly lost the signal and finally F5APQ received the last telemetry. Despite what seemed a premature end to the flight, I had enjoyed the whole process and was very happy with the height and distance attained this time around. Graham and Sebastian were also very positive about the experience.

Vayu-1 Altitude Graph captured from SpaceNear.us

While I was collecting screenshots and enjoying the moment, Felix DL1SGP sent me a message on IRC stating that a plane had come very close to VAYU-1’s path. He later produced a graphic produced with Google Earth to show just how close it came.

Flightpaths Compared between Turkish Airlines Flight THY1963 and VAYU-1

The markings in blue are VAYU-1 and the dark track is Turkish Airlines flight THY1963. As you can see, it was a close thing. Quite possibly, VAYU-1 was affected by the plane’s turbulence, enough perhaps to burst the foil balloon. A dramatic end to my first successful flight.

Altitude Comparison created with a screenshot from http://www.flightradar24.com/

I went to FlightRadar24 and looked up THY1963, re-ran the flight track and paused it at the point where VAYU-1’s telemetry was lost. I overlaid the Google Earth graphic and compared the altitudes. Although cut short, I’m still very pleased with how it all went. I’m only disappointed that Felix and others didn’t get to track VAYU-1 on the day. Whether the balloon burst naturally or if THY1963 had anything to do with it, it was a great experience, and one I wish to repeat.

VAYU-2 is pretty much ready, although I’d like to add some more data to the telemetry. I have also ordered the VAYU-NTX pcbs that utilise the more stable NTX2b transmitter module in place of the RFM22b.

VAYU-1 flightpath on Google Earth along with THY1963

With thanks to Felix DL1SGP for the Google Earth KML files.

Posted in Launches, VAYU | Tagged Balloon, Flight, Foil, HAB, Launch, Pico, Terrain, Tracking, VAYU | Leave a comment

Vayu-1 Re-launch Announcement

Posted on October 30, 2013 by Steve Smith

I’m going to re-launch VAYU-1 on Sunday 3rd November 2013 from Shenstone Park in Crayford.

Shenstone Park on Google Maps (Satellite view)

Hopefully, all being well with weather etc. we will launch at around 12:30pm. The site is 36m above sea level and so gives nice clearance from local high points like pylons. Transmit details are the same as the first launch:

Balloon Type: Pico 36in Foil
Callsign: VAYU-1
Frequency: 434.300MHz+/- Drift
Mode: USB RTTY 50 7n2 Shift 450Hz
Launch site: Lat = 51.453388 Lon = 0.170352
Launch time: 12:30ish

Track on SpaceNear – Details on how to receive via radio on the UKHAS website.

Hopefully, VAYU-1 will attain a better altitude and distance than the first launch. Predictions are trending towards the SE Coast of the UK. Once in the air, I will run a few more predictions to see if it’s viable to chase and recover.

Wish me luck!

Posted in VAYU | Tagged Balloon, Experimental, Foil, Location, Pico, Telemetry, VAYU | Leave a comment

SMD Soldering Technique

Posted on October 26, 2013 by Steve Smith

During the building of VAYU-1 and VAYU v1.01, I developed a technique for soldering the very small SMD parts including the TPS16016 Boost Regulator chip. This device is a mere 2.5mm across and the pins are minute. Try as I might, I can’t find a syringe that will deliver a thin enough line of solder paste and I’m not in a position to obtain a solder stencil. So, I had to try and find a way that I could easily solder SMD parts myself. I am no stranger to crafts and have some bits and pieces lying around. I decided to try and apply the solder manually, with a pointed implement. I found this tool that came from a scalpel set I had many years ago.

Pointed Implement

I use an old SIM card carrier as a palette and put a small blob of solder paste on it from the dispensing syringe.

Solder paste

In this instance, this is much too much for a single 8-pin SMD chip. The pads on the board are cleaned if required to make them ready to receive the solder paste.

Clean pads ready to accept TPS16016 chip

Using a jewelers glass, I pick up a minute amount of solder paste on the very point of the ‘pointed implement’ and touch it on the pads, I then roll the implement so that the solder is deposited on the pads. There is a knack to this so you might like to practice on a few old boards first. Once you have enough solder paste along the contacts, you can tidy the line up with the implement by carefully nudging the paste around. You will end up with something looking like this:

Solder paste applied to the pads

In this instance, I have gone a little over the top. The first one I did had less solder on it and worked fine. This one caused a small problem, more of that later. I was up against failing light so I rushed, something I should have known better than to do.

Once the solder is in place, you can carefully place the component in the correct orientation on the pads. I bought a set of anti-static tweezers on eBay for this purpose. After placement, you can nudge the component so that it is in exactly the right place.

Component placed

I should have realised there was too much solder paste by this point but like I said, I was in a rush. This will still demonstrate the technique though. The next step is to use a hot-air soldering station to apply hot air on the solder paste. I set mine to 300 Deg and it seems to work fine. You can see the solder paste go shiny once melted and the surface tension should draw the solder to the contacts.
As I have said, there was too much solder in this instance and that led to solder bridges across several contacts, here indicated by the red circles. There were two bridges on the other side of the chip too.

Solder Bridges

My answer to this is to use a standard fine-tipped soldering iron and a dress making pin. I place the pin in the gap between the legs of the chip, heat them up and draw the pin out. This normally brings the excess solder out. Again, I’ll point out that this shouldn’t be necessary except for my impatience in getting the images together.

The Finished Soldering

I have slightly scratched the PCB with the pin by drawing too hard between the legs of the component but it won’t affect operation. I will be a lot more careful in the future. If I had of used half the amount of solder paste, I would have obtained perfect joints, as I did on the VAYU v1.01 prototype board. I hope others can learn from my mistake.

Posted in Electronics | Tagged Assembly, board, Circuit, Design, PCB, SMD, Soldering | Leave a comment

VAYU-1 1st Flight

Posted on October 26, 2013 by Steve Smith

On Sunday 20th October, I set off to Youlbury with my friends Graham, Theresa and young Sebastian to launch VAYU-1 for the Scout JOTA day. In the preceeding week, I’d checked the weather forecast and it consistently said that it would be cloudy with rain spells. When the time came, I took a gamble that there would be enough clear sky to launch. We made good time and arrived shortly after 12:00 despite a lengthy stop at services for a late brekkie.

We were very warmly greeted by Steve Hunter (M0SXH) and Richard who quickly introduced us to everyone else. Whilst we waited for Dave Akerman (M6RPI) and my friend Charlie Robson to arrive, we had a cuppa and a chat. I was keeping an eye on the weather and it was certainly clear enough to launch at that point. I begun to prepare and hung VAYU-1 on a window catch for a quick test.

VAYU-1 payload hangs in the Window for a quick test.

I showed Steve and Richard how to receive the telemetry on their radio kit and indicated what appears on SpaceNear. Shortly after this, Charlie arrived followed by Dave a little while later. Now everyone had arrived, I begun to prepare for the flight in earnest. Dave and Charlie assisted me with inflating the foil balloon and Dave showed me how to use ballast to measure the lift. This involved suspending a weight of known quantity under the balloon and then measuring how much lighter the weight was when the balloon was allowed to lift it. I used an open ended container filled with rice. As we filled the balloon to get the right lift, we threw in some other weights I’d made from small amounts of solder.

Getting the lift right

During this time, I was explaining to the assembled folk what High Altitude Ballooning was all about and what I was trying to achieve with VAYU-1. I explained that the reason I chose a small foil ‘Pico’ balloon for my first launch is so that I could go through the motions and get some experience without it costing a fortune. As in most endeavors, there is a steep learning curve and I wanted to make sure I had a good idea of the procedures and possible pitfalls involved in launching a balloon. I really appreciated having Dave Akerman assisting, a veteran of many launches.

Dave had brought his mobile tracking kit as well as the now famous Babbage the bear. In an interlude, Dave showed us how he had managed to get a Raspberry Pi board, batteries and a camera inside Babbage for his epic jump.

Having set up the balloon and VAYU-1 payload, it was time to launch. Sadly, the weather had turned a little sour. However, we went ahead. Steve went and grabbed some more interested Scouts whilst I led Dave, Richard and everyone else out to a field opposite the Scout complex. After a quick chat with Dave, I decided to launch from the adjoining field where the Scouts could watch from the fence. Dave followed me all the time and streamed video to the BATC website.

Procession to the launch

Trying not to disturb the sheep, I positioned myself as far down the field as possible. This was to ensure the balloon didn’t hit either the power lines running across the field or the line of trees making up the border. I watched Dave for my que to launch since he was in a better position to feel the wind than I was. He gave the word and I released my first flight with a push. Worryingly, the balloon didn’t climb very fast and we were all shouting ‘Up! up!’ at it in a feeble attempt to get it higher in the air. Luckily, it did start to climb faster and we watched it as it gracefully floated over the tree line.

We walked back to the complex and I realised that I hadn’t done any tests with small balloons, which I had planned to do. Note to self: Don’t get so excited! Once back in the complex, I explained to the Scouts in the adjoining tent how the tracking would be done. They had SpaceNear projected on a screen for everyone to see.

VAYU-1 on SpaceNear Tracker page being projected onto a screen.

As we made our way back to the hut to decide which way was best to chase VAYU-1, we had no idea of the difficulty it was having staying aloft. The rain had started and was forcing VAYU-1 back down. We realised what was happening when we tried to receive the telemetry back at the hut. There was no signal to be found. It was deduced that VAYU-1 had landed about one and a half kilometers away. A group of us assembled to see if we could retrieve the payload. Richard kindly gave us directions for the best way to exit the complex. Graham led the way and we set off to find VAYU-1.

By this time, the heavens had opened and I was very unsurprised that VAYU-1 had come down. The weight of the water on the balloon must have been quite high. We arrived in the vicinity of the landing where Dave reported a good signal. Sebastian and I set off in the rain to see if we could locate the downed balloon. We searched around for quite a while before I realised I’d got a helpful text message from Chris Stubbs detailing where VAYU-1 was likely to be. This was a great help and pointed us in the right direction. On the way, Steve phoned to tell us he and Richard had joined us to help in the search. I explained where Graham and Dave were parked and they headed for a rendezvous.

Sebastian and I were drenched by this time but we didn’t want to give up. I was very pleased hen Sebastian exclaimed ‘Is that it?’. I looked and indeed, Sebastian had spotted the balloon up in a Hawthorn bush. It was too high up to reach without aid so we tried to recover the payload with a flagpole, loaned from the field adjoining us. This proved ineffective but as luck would have it, Steve had let us bring a telescopic pole and a set of loppers from the Scout complex. Sebastian and I went back to the cars to collect them to find Steve and Richard pulling up in their car.

After retrieving the equipment from the car, Steve and Richard elected to join us in recovering the payload. Dave had reported loosing the signal from his car so I was a little concerned that I’d damaged the payload by trying to retrieve it with the flag pole. We got back to the landing site within a few minutes and Sebastian and I together got the payload out of the Hawthorn successfully. After Steve had taken a few pictures on his phone for posterity, I identified that I’d broken the power switch off of the regulator board. This had obviously cut the power but the payload was otherwise undamaged.

Steve & Sebastian recovering VAYU-1 – Photo Steve Hunter

Recovery successful! – Photo Steve Hunter

After removing the bits of Hawthorn from the deflated balloon and VAYU-1, Sebastian carried it back to the car. Despite the short flight and the rain, we were all smiling.

Recovery Team: Richard, Steve, Myself and Sebastian – Photo Dave Akerman

After I had separated the balloon from the payload, we all said our goodbyes. Despite a very quick flight and an extremely wet recovery, it had been lots of fun. I had been able to demonstrate a launch and the preparation beforehand and had managed a successful recovery. VAYU-1 will fly again soon.

I had lots of assistance in this endeavor. Firstly from Chris Stubbs, whose circuit and PCB VAYU-1 is based on. I had loads of help from all the guys involved in UKHAS which I appreciate immensely. A big thanks to Dave Akerman and Julie for coming along to assist. I was made extremely welcome by Steve, Richard and all the Scouts and staff at Youlbury. But the biggest thanks must go to my friends Graham, Theresa and Sebastian for getting me there and especially Sebastian for spotting the balloon!

Despite the short flight, I learned a lot on my first launch. The knowledge I gained will be invaluable for future launches. I also made some great new friends too!

Posted in Payloads, Tracking, VAYU | Tagged Flight, Foil, HAB, JOTA, Pico, Scout, Telemetry, Tracking, VAYU | Leave a comment

An added boost!

Posted on October 21, 2013 by Steve Smith

Before I regale you with the full VAYU-1 launch story, I would like to share with you a small success. If you are a regular reader of this blog, you will know that I took the original CHEAPO Tracker design by Chris Stubbs and modified it with a few additions that I wanted available for my future launches. This included designing a boost regulator to raise a 1.5V supply up to the board’s required 3.3V. I have been itching to try the circuit out and today I built it up. Just the boost regulator, since I figured that if that didn’t work, I would have to either modify the board or have a new design fabricated.

Vayu v1.01 pcb showing the built Boost Regulator

Vayu-1.01 Boost Regulator Close-up

This circuit was a real challenge to build. When soldering the TPS61016, there were two solder bridges created. I managed to remove one with a fine tipped soldering iron but the other wouldn’t budge until I got a sewing pin between the devices pins, heated it up and drew the excess solder out. I realise that I probably put too much solder paste on the contacts before heating with a hot air gun. I soldered the (rather large) coil next. This I did by placing a small amount of solder paste on the board, placing the coil in the correct place and blowing the hot air under the coil. I saw the solder melt which indicated it had affixed properly.

After soldering the final components of the regulator circuit, it was time to test it. I croc-clipped a couple of AA batteries in series to the power connections on the side of the board and took a reading with my DVM. 3.27V Close enough. I then tried with a single AA and got the same result. Perhaps one of the components is slightly out of tolerance but 3.27V is close enough not to matter in this instance. When I have a chance I’ll investigate the possibility of bringing it up to 3.3V exactly but modifying the surrounding components values.

Vayu-1.01 Boost Regulator Meter Reading

The next step is to build the rest of the board when my order of ATMEGA328p-TQFP arrives. Having built the original CHEAPO board, I feel that I’ve had enough experience to make a good job of it. The real secret ingredient is patience. It takes a while to get each component placed correctly on the pads with solder paste. Also, if one ends up soldered incorrectly, it needs to be dealt with patiently and in an organised fashion.

I certainly couldn’t solder the TPS61016 with a conventional soldering iron, not even with the finest tip. I couldn’t see the contacts close enough to place the bit properly. I use a cheap hot air soldering station that I purchased on eBay along with solder paste.

Atten 858D+ Hot Air Station

I have built several boards, all successful, with this method. I put a small amount of solder paste on the contacts on the PCB and tidy them up if necessary. I then place the component on the pads and orientate it properly. Then, I pass the hot air across the components pins and wait until the solder paste melts. Although the components and PCB get quite hot, I haven’t lost any yet.

Watch out for a further post of the completed build.

[UPDATE]

Since I had the components ready and the hot air station on standby, I thought I’d solder the rest of the small components on to the board in preparation of receiving the MCU. I left off the GPS unit, headers and crystal to give me enough room to solder the MCU properly.

VAYU v1.01 SMD Components

[UPDATE 2]

VAYU v1.01 Prototype fabricated by Hackvana

This is the fully finished, populated PCB. Challenging with all the many solder pads. Complete set of Docs available on the VAYU v1.02 page!

Posted in Payloads, Tracking, VAYU | Tagged board, Circuit, Component, Custom, payload, PCB, SMD | Leave a comment

ProjectHAB Twitter