Date   
Re: Transmitter update

Jared Smith
 

I now have the payload programmed. I have much testing to do and need to try to optimize the memory a bit (I'm at 70% dynamic memory), but it seems to be working correctly.

Here's an overview of what happens after adequate power is on-board:
  1. Program initiates and makes connection to clock generator and GPS
  2. GPS is programmed (high power mode, high altitude mode, correct NMEA sentences defined, etc.)
  3. Wait for GPS fix (around 3 minutes cold start, a few seconds warm start).
  4. Upon GPS fix, set Arduino clock to GPS time. Turn on LED (indicates GPS fix).
  5. Get lat/lon/altitude. Encode telemetry and temperature. All are stored to variables.
  6. GPS is set to low power mode.
  7. If minute is :00, :10, :20, etc., initiate standard transmission. LED flashes during transmissions.
  8. If :04, :14, :24, etc., initiate telemetry transmission. Then loop back to #1.
I wired up the solar array today and will check voltage tomorrow (a nice sunny day!).

I then need to figure out how to get all of this together in a way that it will hang from a tether with the solar panels pointing up, while also being VERY lightweight. I'd be happy to hear ideas.

One thing I'm not sure of is orientation of the GPS antenna. It's a very short antenna. I'm not sure if it should optimally point up or horizontally. Anyone know?

Gary, do you have your components? Let me know if you'd like to coordinate to get stuff from me. And I can send you Arduino sketches any time.

Thanks,

Jared



Re: GAS team balloon stuff

Tyler Griffiths
 

That was my thought too. Spinning. 

Tyler 


On Sat, Jun 17, 2017 at 10:51 PM Kevin Reeve <kevin.reeve@...> wrote:
Good thoughts Jared.

We do know the GPS is the one that has flown as part of the Byonics MT1000 on all flights we have done.
We should check to see if it has come loose in the MT1000.

Another thought is the payload/tracking box is spinning rapidly and thus causing issues with receiving accurate GPS signals. The doppler effect per se.

Kevin



From: <BARC-HAB@groups.io> on behalf of Jared Smith <jared@...>
Reply-To: "BARC-HAB@groups.io" <BARC-HAB@groups.io>
Date: Saturday, June 17, 2017 at 10:30 PM
To: "BARC-HAB@groups.io" <BARC-HAB@groups.io>
Subject: Re: [BARC-HAB] GAS team balloon stuff

I thought that the uneven terrain and fast approach velocity (around 60 MPH) caused signal interference that caused the inaccuracies, but that doesn't align with the data.

Attached is a Google Earth view of the reported descent. You can see that it's rather linear from the near the top of the descent (by the way, it fell at 175 miles per hour after burst!) to where it should have hit the ground NE of Paradise. The problem is that the descent should have slowed as it got into thicker air and approached the ground.

Attached is a chart that compares altitude over time for the May 20 vs. June 17 flights. May 20 is normal, but today's flight increased its descent rate below around 15000 meters. Unless there's some other factor in play, this simply isn't possible. The red line indicates what I think was the actual altitude.

This suggests that the GPS was reporting inaccurate altitude for probably the last 15-20 minutes of flight - for every single position update until well after it was on the ground. I don't think that signal disruption could have caused that. I think this suggests a programming or internal issue, perhaps due to the cold temperatures.

I'd be happy to hear any other theories.

Jared
 

On Sat, Jun 17, 2017 at 8:38 PM, <ag1t@...> wrote:
Before 2000, the government degraded the signal (called Selective Availability).  Bill Clinton removed the limitation via Executive Order in 2000.

Most good GPS devices can be accurate within a meter now both vertically and horizontally.    

http://www.gps.gov/systems/gps/performance/accuracy/

Gary


--
Tyler, Sent from Gmail Mobile

Re: Transmitter update

Kevin Reeve
 

Josh,

If you need one of these byonics T3 minis for building your digi let me know.
Kevin


From: <BARC-HAB@groups.io> on behalf of Josh Jensen <kd7wrc@...>
Reply-To: "BARC-HAB@groups.io" <BARC-HAB@groups.io>
Date: Saturday, June 10, 2017 at 11:06 PM
To: "BARC-HAB@groups.io" <BARC-HAB@groups.io>
Subject: Re: [BARC-HAB] Transmitter update

Jared,

I was just thinking. If you want to give your payload an airborne test before you launch it, we could take it up suspended from my hexacopter. I made some good progress today, and could possibly take it on its first flight soon. 

Now I just need to work on some ham payloads. Probably my first will be a camera and transmitter, then a small digi to pick up APRS beacons from lost payloads, or relaying Bear 100 runner data.

Josh

On Jun 9, 2017 5:25 PM, "Jared Smith" <jared@...> wrote:
Tyler -

It's a single defined frequency per band - so no need to change based on regions like you have to with APRS or other VHF/UHF modes.

I do, however, have to add some logic so it won't transmit over certain countries where airborne transmissions are illegal. I'm not sure yet whether to do this geofencing based on lat/lon or grid square.

Mine will do 20 meters WSPR only. So it can be locked on the defined 20 meter frequency with a tuned dipole. The signal generator makes the slight changes in tones (4 of them total) to encode the almost 2-minute long WSPR transmission.

Some balloons are using 20 and 30 meter WSPR, but this makes the antenna configuration a bit more complex. And some are also using APRS with a separate transmitter. I may add APRS depending on what the weight comes to.

Jared


On Fri, Jun 9, 2017 at 3:41 PM, Tyler Griffiths <tyler.griffiths@...> wrote:
 I should say as it circles the earth. 


On Fri, Jun 9, 2017 at 3:40 PM Tyler Griffiths <tyler.griffiths@...> wrote:
So does it need to change  frequencies like a VHF or UHF  beacon. Or does HF work on a single frequency ?

Tyler
On Fri, Jun 9, 2017 at 9:17 AM Kevin Reeve <kevin.reeve@...> wrote:
And I think the whole Wisper thing is cool because it is on a worldwide recognized freq, and has a great chance of being heard even when across the ocean.

Kevin



From: <BARC-HAB@groups.io> on behalf of Main <kevin.reeve@...>
Reply-To: "BARC-HAB@groups.io" <BARC-HAB@groups.io>
Date: Friday, June 9, 2017 at 9:09 AM
To: "BARC-HAB@groups.io" <BARC-HAB@groups.io>
Subject: Re: [BARC-HAB] Transmitter update

I think your selling yourself short on the programming skills. If you have that much working as of last night, you have made great progress and accomplished much, especially converting lat and long to grid square.

Kevin



From: <BARC-HAB@groups.io> on behalf of Jared Smith <jared@...>
Reply-To: "BARC-HAB@groups.io" <BARC-HAB@groups.io>
Date: Thursday, June 8, 2017 at 11:29 PM
To: "BARC-HAB@groups.io" <BARC-HAB@groups.io>
Subject: [BARC-HAB] Transmitter update

I have a basic prototype up and running with the WSPR transmitter. A photo is attached.

It puts out only 10 milliwatts (much less than that by the time it gets to my antenna) and was heard in several places across the US tonight on 20 meter band well after sunset. I track reception reports using WSPRnet.org. A screen shot of those hearing me is attached.

The furthest was W0PAS in Eastern Ohio. That comes out to 155,590 miles per watt!!! Not bad for a radio the size of a matchbook.

Thanks Gary for helping me test it out.

The GPS receivers came today, so that's my next project. I've learned my programming skills leave a lot to be desired! I programmed up the telemetry module yesterday. It converts NMEA lat/lon to grid square, and encodes the altitude and temperature into the telemetry format.

Jared

--
Tyler, Sent from Gmail Mobile

--
Tyler, Sent from Gmail Mobile


PCB designer? Weight issues.

Jared Smith
 

Does anyone have experience with PCB design?
My balloon payload finalized, I'm now at around 20 grams or so. This is 5 grams heavier than I want for optimal altitude. I can lose unnecessary resistors, etc., but this might gain me at most 1 gram.

My heaviest component by far is the Si5351A breakout board at 5.3 grams - https://www.etherkit.com/rf-modules/si5351a-breakout-board.html You'll notice that it's on a thick PCB, it has 3 outputs (I only need 1), and it has 5v to 3.3v conversion on-board (I'm strictly 3.3v). 

It should be 3.3v and one output only, with header and edge mount SMA pads. On a thin PCB, with 3.3v only and one output (header and edge-mount SMA pads), I think this could come in at 1 gram or less. I think this would be a very popular board that you could make some $$$ off of.

Is there anyone here or do you know anyone that would be interested in this. I'd fund the prototype boards, but need someone smarter than me to do the design. I have the full PCB design files and schematic for 3 different 5v/3.3v boards that would get you most of the way there.

Another possibility to lose weight is to most of my solar cells and add a solar upconverter/charger. It takes 9 panels to make 5 volts. They're very light weight (I think 4.5 grams for the full array - plus I need some infrastructure to keep them in place), but with a 



Re: PCB designer? Weight issues.

Jared Smith
 

D'oh. Hit Send a bit too early.

Anyway... I could go from 9 to 4 (maybe 3 or 2) solar cells with a SPV1040 step-up converter / charge controller - http://www.st.com/en/power-management/spv1040.html - (or similar) to charge the super capacitor to 5 volts. It would take a bit longer to charge it, but it should maintain the charge long enough for the regular transmissions. This would also need a custom PCB (at least I can't find an existing charger breakout board that is in the 0-2 gram weight range).

If I could pull off both of these adjustments, that should put my payload well below 15 grams which would be optimal for a higher altitude float.

Jared


On Mon, Jun 19, 2017 at 4:06 PM, Jared Smith <jared@...> wrote:
Does anyone have experience with PCB design?
My balloon payload finalized, I'm now at around 20 grams or so. This is 5 grams heavier than I want for optimal altitude. I can lose unnecessary resistors, etc., but this might gain me at most 1 gram.

My heaviest component by far is the Si5351A breakout board at 5.3 grams - https://www.etherkit.com/rf-modules/si5351a-breakout-board.html You'll notice that it's on a thick PCB, it has 3 outputs (I only need 1), and it has 5v to 3.3v conversion on-board (I'm strictly 3.3v). 

It should be 3.3v and one output only, with header and edge mount SMA pads. On a thin PCB, with 3.3v only and one output (header and edge-mount SMA pads), I think this could come in at 1 gram or less. I think this would be a very popular board that you could make some $$$ off of.

Is there anyone here or do you know anyone that would be interested in this. I'd fund the prototype boards, but need someone smarter than me to do the design. I have the full PCB design files and schematic for 3 different 5v/3.3v boards that would get you most of the way there.

Another possibility to lose weight is to most of my solar cells and add a solar upconverter/charger. It takes 9 panels to make 5 volts. They're very light weight (I think 4.5 grams for the full array - plus I need some infrastructure to keep them in place), but with a 




Re: PCB designer? Weight issues.

Josh Jensen
 

Is Brandon Tibbetts on this list? He's done some PCB design work before.

Josh

On Jun 19, 2017 16:14, "Jared Smith" <jared@...> wrote:
D'oh. Hit Send a bit too early.

Anyway... I could go from 9 to 4 (maybe 3 or 2) solar cells with a SPV1040 step-up converter / charge controller - http://www.st.com/en/power-management/spv1040.html - (or similar) to charge the super capacitor to 5 volts. It would take a bit longer to charge it, but it should maintain the charge long enough for the regular transmissions. This would also need a custom PCB (at least I can't find an existing charger breakout board that is in the 0-2 gram weight range).

If I could pull off both of these adjustments, that should put my payload well below 15 grams which would be optimal for a higher altitude float.

Jared


On Mon, Jun 19, 2017 at 4:06 PM, Jared Smith <jared@...> wrote:
Does anyone have experience with PCB design?
My balloon payload finalized, I'm now at around 20 grams or so. This is 5 grams heavier than I want for optimal altitude. I can lose unnecessary resistors, etc., but this might gain me at most 1 gram.

My heaviest component by far is the Si5351A breakout board at 5.3 grams - https://www.etherkit.com/rf-modules/si5351a-breakout-board.html You'll notice that it's on a thick PCB, it has 3 outputs (I only need 1), and it has 5v to 3.3v conversion on-board (I'm strictly 3.3v). 

It should be 3.3v and one output only, with header and edge mount SMA pads. On a thin PCB, with 3.3v only and one output (header and edge-mount SMA pads), I think this could come in at 1 gram or less. I think this would be a very popular board that you could make some $$$ off of.

Is there anyone here or do you know anyone that would be interested in this. I'd fund the prototype boards, but need someone smarter than me to do the design. I have the full PCB design files and schematic for 3 different 5v/3.3v boards that would get you most of the way there.

Another possibility to lose weight is to most of my solar cells and add a solar upconverter/charger. It takes 9 panels to make 5 volts. They're very light weight (I think 4.5 grams for the full array - plus I need some infrastructure to keep them in place), but with a 





Re: PCB designer? Weight issues.

Jared Smith
 

He's not, but I've copied Brandon on this message to see if he's interested, or if he'd like to join the list - https://groups.io/g/BARC-HAB/

I think I can figure out which components can go away on the Si5351A board and order a prototype, but this level of electronics is definitely not my specialty.


Thanks,

Jared


On Mon, Jun 19, 2017 at 4:30 PM, Josh Jensen <kd7wrc@...> wrote:
Is Brandon Tibbetts on this list? He's done some PCB design work before.

Josh

On Jun 19, 2017 16:14, "Jared Smith" <jared@...> wrote:
D'oh. Hit Send a bit too early.

Anyway... I could go from 9 to 4 (maybe 3 or 2) solar cells with a SPV1040 step-up converter / charge controller - http://www.st.com/en/power-management/spv1040.html - (or similar) to charge the super capacitor to 5 volts. It would take a bit longer to charge it, but it should maintain the charge long enough for the regular transmissions. This would also need a custom PCB (at least I can't find an existing charger breakout board that is in the 0-2 gram weight range).

If I could pull off both of these adjustments, that should put my payload well below 15 grams which would be optimal for a higher altitude float.

Jared


On Mon, Jun 19, 2017 at 4:06 PM, Jared Smith <jared@...> wrote:
Does anyone have experience with PCB design?
My balloon payload finalized, I'm now at around 20 grams or so. This is 5 grams heavier than I want for optimal altitude. I can lose unnecessary resistors, etc., but this might gain me at most 1 gram.

My heaviest component by far is the Si5351A breakout board at 5.3 grams - https://www.etherkit.com/rf-modules/si5351a-breakout-board.html You'll notice that it's on a thick PCB, it has 3 outputs (I only need 1), and it has 5v to 3.3v conversion on-board (I'm strictly 3.3v). 

It should be 3.3v and one output only, with header and edge mount SMA pads. On a thin PCB, with 3.3v only and one output (header and edge-mount SMA pads), I think this could come in at 1 gram or less. I think this would be a very popular board that you could make some $$$ off of.

Is there anyone here or do you know anyone that would be interested in this. I'd fund the prototype boards, but need someone smarter than me to do the design. I have the full PCB design files and schematic for 3 different 5v/3.3v boards that would get you most of the way there.

Another possibility to lose weight is to most of my solar cells and add a solar upconverter/charger. It takes 9 panels to make 5 volts. They're very light weight (I think 4.5 grams for the full array - plus I need some infrastructure to keep them in place), but with a 






Re: PCB designer? Weight issues.

Jared Smith
 

As a quick update, I've been working with Jason from Etherkit and he's designing a breakout board for the clock generator - https://github.com/NT7S/Si5351APicoBoB/blob/master/pcb/Si5351APicoBoB.pdf It should be MUCH smaller and lighter than the existing board - I'm thinking at most 2 grams (if you don't add the headers and wire it directly).

He's going to upload to OSH Park after he's designed the PCB and I'm going to order at least a few soon. Is anyone else interested in some? Gary?

Now I'm considering designing a breakout for a small solar charge controller so I can use fewer panels. The idea is to charge the supercapacitor with the minimum number of solar cells possible while also ensuring that there's enough juice to get through the 2 minute transmissions.

Jared


On Mon, Jun 19, 2017 at 6:01 PM, Jared Smith <jared@...> wrote:
He's not, but I've copied Brandon on this message to see if he's interested, or if he'd like to join the list - https://groups.io/g/BARC-HAB/

I think I can figure out which components can go away on the Si5351A board and order a prototype, but this level of electronics is definitely not my specialty.


Thanks,

Jared


On Mon, Jun 19, 2017 at 4:30 PM, Josh Jensen <kd7wrc@...> wrote:
Is Brandon Tibbetts on this list? He's done some PCB design work before.

Josh

On Jun 19, 2017 16:14, "Jared Smith" <jared@...> wrote:
D'oh. Hit Send a bit too early.

Anyway... I could go from 9 to 4 (maybe 3 or 2) solar cells with a SPV1040 step-up converter / charge controller - http://www.st.com/en/power-management/spv1040.html - (or similar) to charge the super capacitor to 5 volts. It would take a bit longer to charge it, but it should maintain the charge long enough for the regular transmissions. This would also need a custom PCB (at least I can't find an existing charger breakout board that is in the 0-2 gram weight range).

If I could pull off both of these adjustments, that should put my payload well below 15 grams which would be optimal for a higher altitude float.

Jared


On Mon, Jun 19, 2017 at 4:06 PM, Jared Smith <jared@...> wrote:
Does anyone have experience with PCB design?
My balloon payload finalized, I'm now at around 20 grams or so. This is 5 grams heavier than I want for optimal altitude. I can lose unnecessary resistors, etc., but this might gain me at most 1 gram.

My heaviest component by far is the Si5351A breakout board at 5.3 grams - https://www.etherkit.com/rf-modules/si5351a-breakout-board.html You'll notice that it's on a thick PCB, it has 3 outputs (I only need 1), and it has 5v to 3.3v conversion on-board (I'm strictly 3.3v). 

It should be 3.3v and one output only, with header and edge mount SMA pads. On a thin PCB, with 3.3v only and one output (header and edge-mount SMA pads), I think this could come in at 1 gram or less. I think this would be a very popular board that you could make some $$$ off of.

Is there anyone here or do you know anyone that would be interested in this. I'd fund the prototype boards, but need someone smarter than me to do the design. I have the full PCB design files and schematic for 3 different 5v/3.3v boards that would get you most of the way there.

Another possibility to lose weight is to most of my solar cells and add a solar upconverter/charger. It takes 9 panels to make 5 volts. They're very light weight (I think 4.5 grams for the full array - plus I need some infrastructure to keep them in place), but with a 







Re: PCB designer? Weight issues.

Kevin Reeve
 

So how many components must you solder on this new board he will upload to oshpark?  

Sent from my iPhone

On Jun 20, 2017, at 4:35 PM, Jared Smith <jared@...> wrote:

As a quick update, I've been working with Jason from Etherkit and he's designing a breakout board for the clock generator - https://github.com/NT7S/Si5351APicoBoB/blob/master/pcb/Si5351APicoBoB.pdf It should be MUCH smaller and lighter than the existing board - I'm thinking at most 2 grams (if you don't add the headers and wire it directly).

He's going to upload to OSH Park after he's designed the PCB and I'm going to order at least a few soon. Is anyone else interested in some? Gary?

Now I'm considering designing a breakout for a small solar charge controller so I can use fewer panels. The idea is to charge the supercapacitor with the minimum number of solar cells possible while also ensuring that there's enough juice to get through the 2 minute transmissions.

Jared


On Mon, Jun 19, 2017 at 6:01 PM, Jared Smith <jared@...> wrote:
He's not, but I've copied Brandon on this message to see if he's interested, or if he'd like to join the list - https://groups.io/g/BARC-HAB/

I think I can figure out which components can go away on the Si5351A board and order a prototype, but this level of electronics is definitely not my specialty.


Thanks,

Jared


On Mon, Jun 19, 2017 at 4:30 PM, Josh Jensen <kd7wrc@...> wrote:
Is Brandon Tibbetts on this list? He's done some PCB design work before.

Josh

On Jun 19, 2017 16:14, "Jared Smith" <jared@...> wrote:
D'oh. Hit Send a bit too early.

Anyway... I could go from 9 to 4 (maybe 3 or 2) solar cells with a SPV1040 step-up converter / charge controller - http://www.st.com/en/power-management/spv1040.html - (or similar) to charge the super capacitor to 5 volts. It would take a bit longer to charge it, but it should maintain the charge long enough for the regular transmissions. This would also need a custom PCB (at least I can't find an existing charger breakout board that is in the 0-2 gram weight range).

If I could pull off both of these adjustments, that should put my payload well below 15 grams which would be optimal for a higher altitude float.

Jared


On Mon, Jun 19, 2017 at 4:06 PM, Jared Smith <jared@...> wrote:
Does anyone have experience with PCB design?
My balloon payload finalized, I'm now at around 20 grams or so. This is 5 grams heavier than I want for optimal altitude. I can lose unnecessary resistors, etc., but this might gain me at most 1 gram.

My heaviest component by far is the Si5351A breakout board at 5.3 grams - https://www.etherkit.com/rf-modules/si5351a-breakout-board.html You'll notice that it's on a thick PCB, it has 3 outputs (I only need 1), and it has 5v to 3.3v conversion on-board (I'm strictly 3.3v). 

It should be 3.3v and one output only, with header and edge mount SMA pads. On a thin PCB, with 3.3v only and one output (header and edge-mount SMA pads), I think this could come in at 1 gram or less. I think this would be a very popular board that you could make some $$$ off of.

Is there anyone here or do you know anyone that would be interested in this. I'd fund the prototype boards, but need someone smarter than me to do the design. I have the full PCB design files and schematic for 3 different 5v/3.3v boards that would get you most of the way there.

Another possibility to lose weight is to most of my solar cells and add a solar upconverter/charger. It takes 9 panels to make 5 volts. They're very light weight (I think 4.5 grams for the full array - plus I need some infrastructure to keep them in place), but with a 







Re: PCB designer? Weight issues.

Jared Smith
 

You'll have to solder:
- Headers (if you want them - I'll solder wires directly to save the weight)
- 2 capacitors (SMD)
- 2 resistors (SMD)
- The TCXO clock (small, but only 4 corner pads)
- The Si5351A. It's VERY small (only 3mm) and has 8 legs, so is likely to be a tricky solder. I've ordered a finer soldering iron tip!

This photo gives a good sense of scale for the components - https://cdn-learn.adafruit.com/assets/assets/000/018/751/medium640/adafruit_products_solder1.jpg?1407866334 The TCXO and Si5351A are the two components between SCL and CLK1.

You can see how with only these few components that the board could be very small and light. Jason should have the PCB design done soon.

I think I'm also going to swap out my supercapacitors with this one - http://www.digikey.com/scripts/DkSearch/dksus.dll?Detail&itemSeq=231174428&uq=636335947477185122 It has lower capacitance (1.5F vs 2.5F), but saves a full gram of weight.

I think with these changes I should be pretty close to my target of 16 grams.

Jared


On Tue, Jun 20, 2017 at 9:50 PM, Kevin Reeve <kevin.reeve@...> wrote:
So how many components must you solder on this new board he will upload to oshpark?  

Sent from my iPhone

On Jun 20, 2017, at 4:35 PM, Jared Smith <jared@...> wrote:

As a quick update, I've been working with Jason from Etherkit and he's designing a breakout board for the clock generator - https://github.com/NT7S/Si5351APicoBoB/blob/master/pcb/Si5351APicoBoB.pdf It should be MUCH smaller and lighter than the existing board - I'm thinking at most 2 grams (if you don't add the headers and wire it directly).

He's going to upload to OSH Park after he's designed the PCB and I'm going to order at least a few soon. Is anyone else interested in some? Gary?

Now I'm considering designing a breakout for a small solar charge controller so I can use fewer panels. The idea is to charge the supercapacitor with the minimum number of solar cells possible while also ensuring that there's enough juice to get through the 2 minute transmissions.

Jared


On Mon, Jun 19, 2017 at 6:01 PM, Jared Smith <jared@...> wrote:
He's not, but I've copied Brandon on this message to see if he's interested, or if he'd like to join the list - https://groups.io/g/BARC-HAB/

I think I can figure out which components can go away on the Si5351A board and order a prototype, but this level of electronics is definitely not my specialty.


Thanks,

Jared


On Mon, Jun 19, 2017 at 4:30 PM, Josh Jensen <kd7wrc@...> wrote:
Is Brandon Tibbetts on this list? He's done some PCB design work before.

Josh

On Jun 19, 2017 16:14, "Jared Smith" <jared@...> wrote:
D'oh. Hit Send a bit too early.

Anyway... I could go from 9 to 4 (maybe 3 or 2) solar cells with a SPV1040 step-up converter / charge controller - http://www.st.com/en/power-management/spv1040.html - (or similar) to charge the super capacitor to 5 volts. It would take a bit longer to charge it, but it should maintain the charge long enough for the regular transmissions. This would also need a custom PCB (at least I can't find an existing charger breakout board that is in the 0-2 gram weight range).

If I could pull off both of these adjustments, that should put my payload well below 15 grams which would be optimal for a higher altitude float.

Jared


On Mon, Jun 19, 2017 at 4:06 PM, Jared Smith <jared@...> wrote:
Does anyone have experience with PCB design?
My balloon payload finalized, I'm now at around 20 grams or so. This is 5 grams heavier than I want for optimal altitude. I can lose unnecessary resistors, etc., but this might gain me at most 1 gram.

My heaviest component by far is the Si5351A breakout board at 5.3 grams - https://www.etherkit.com/rf-modules/si5351a-breakout-board.html You'll notice that it's on a thick PCB, it has 3 outputs (I only need 1), and it has 5v to 3.3v conversion on-board (I'm strictly 3.3v). 

It should be 3.3v and one output only, with header and edge mount SMA pads. On a thin PCB, with 3.3v only and one output (header and edge-mount SMA pads), I think this could come in at 1 gram or less. I think this would be a very popular board that you could make some $$$ off of.

Is there anyone here or do you know anyone that would be interested in this. I'd fund the prototype boards, but need someone smarter than me to do the design. I have the full PCB design files and schematic for 3 different 5v/3.3v boards that would get you most of the way there.

Another possibility to lose weight is to most of my solar cells and add a solar upconverter/charger. It takes 9 panels to make 5 volts. They're very light weight (I think 4.5 grams for the full array - plus I need some infrastructure to keep them in place), but with a 








Re: PCB designer? Weight issues.

Brandon Tibbitts
 

Yes I'd be interested in helping, but have been loaded with quite alot in general. I am experienced some in layout, and should be able to put something together if needed. Unless there is a quicker/easier route that is already in progress, i'd be happy to try though.

Were you thinking of .030 thick pcb or .062? I've used express PCB to get things done in the past and typically just design the board to the size I need, and cut them out, getting multiples out of the fixed size sheets they offer.(PCB 3.8 x 2.5 x .062 thick) The software is all manual place and routing, but for simpler circuits not too bad. Is there a desired connector for J's, or just solder pads to hard wire into? 


There are a couple of newer layout design software packages I've been looking at trying out. Things are rapidly evolving in PCB EDA and multiple open source options. Maybe this would be a good time to try one..


Brandon - KD7IIW

On Wed, Jun 21, 2017 at 12:43 PM, Brandon Tibbitts <tibbs327@...> wrote:
Yes I'd be interested in helping, but have been loaded with quite alot in general. I am experienced some in layout, and should be able to put something together if needed. Unless there is a quicker/easier route that is already in progress, i'd be happy to try though.

Were you thinking of .030 thick pcb or .062? I've used express PCB to get things done in the past and typically just design the board to the size I need, and cut them out, getting multiples out of the fixed size sheets they offer.(PCB 3.8 x 2.5 x .062 thick) The software is all manual place and routing, but for simpler circuits not too bad. Is there a desired connector for J's, or just solder pads to hard wire into? 


There are a couple of newer layout design software packages I've been looking at trying out. Things are rapidly evolving in PCB EDA and multiple open source options. Maybe this would be a good time to try one..


Brandon - KD7IIW

On Tue, Jun 20, 2017 at 4:35 PM, Jared Smith <jared@...> wrote:
As a quick update, I've been working with Jason from Etherkit and he's designing a breakout board for the clock generator - https://github.com/NT7S/Si5351APicoBoB/blob/master/pcb/Si5351APicoBoB.pdf It should be MUCH smaller and lighter than the existing board - I'm thinking at most 2 grams (if you don't add the headers and wire it directly).

He's going to upload to OSH Park after he's designed the PCB and I'm going to order at least a few soon. Is anyone else interested in some? Gary?

Now I'm considering designing a breakout for a small solar charge controller so I can use fewer panels. The idea is to charge the supercapacitor with the minimum number of solar cells possible while also ensuring that there's enough juice to get through the 2 minute transmissions.

Jared


On Mon, Jun 19, 2017 at 6:01 PM, Jared Smith <jared@...> wrote:
He's not, but I've copied Brandon on this message to see if he's interested, or if he'd like to join the list - https://groups.io/g/BARC-HAB/

I think I can figure out which components can go away on the Si5351A board and order a prototype, but this level of electronics is definitely not my specialty.


Thanks,

Jared


On Mon, Jun 19, 2017 at 4:30 PM, Josh Jensen <kd7wrc@...> wrote:
Is Brandon Tibbetts on this list? He's done some PCB design work before.

Josh

On Jun 19, 2017 16:14, "Jared Smith" <jared@...> wrote:
D'oh. Hit Send a bit too early.

Anyway... I could go from 9 to 4 (maybe 3 or 2) solar cells with a SPV1040 step-up converter / charge controller - http://www.st.com/en/power-management/spv1040.html - (or similar) to charge the super capacitor to 5 volts. It would take a bit longer to charge it, but it should maintain the charge long enough for the regular transmissions. This would also need a custom PCB (at least I can't find an existing charger breakout board that is in the 0-2 gram weight range).

If I could pull off both of these adjustments, that should put my payload well below 15 grams which would be optimal for a higher altitude float.

Jared


On Mon, Jun 19, 2017 at 4:06 PM, Jared Smith <jared@...> wrote:
Does anyone have experience with PCB design?
My balloon payload finalized, I'm now at around 20 grams or so. This is 5 grams heavier than I want for optimal altitude. I can lose unnecessary resistors, etc., but this might gain me at most 1 gram.

My heaviest component by far is the Si5351A breakout board at 5.3 grams - https://www.etherkit.com/rf-modules/si5351a-breakout-board.html You'll notice that it's on a thick PCB, it has 3 outputs (I only need 1), and it has 5v to 3.3v conversion on-board (I'm strictly 3.3v). 

It should be 3.3v and one output only, with header and edge mount SMA pads. On a thin PCB, with 3.3v only and one output (header and edge-mount SMA pads), I think this could come in at 1 gram or less. I think this would be a very popular board that you could make some $$$ off of.

Is there anyone here or do you know anyone that would be interested in this. I'd fund the prototype boards, but need someone smarter than me to do the design. I have the full PCB design files and schematic for 3 different 5v/3.3v boards that would get you most of the way there.

Another possibility to lose weight is to most of my solar cells and add a solar upconverter/charger. It takes 9 panels to make 5 volts. They're very light weight (I think 4.5 grams for the full array - plus I need some infrastructure to keep them in place), but with a 









Re: PCB designer? Weight issues.

Jared Smith
 

Brandon -

Thank you for the offer. It sounds like this board is being designed, so that should be taken care of.

To answer your questions though, I'd want anything on the thinnest PCB board readily available for weight savings, though Express PCB looks pretty slick.

I'm going to try out this solar upconverter/charge controller - https://www.tindie.com/products/onehorse/bq25504-solar-cell-lipo-charger/ This should allow me to drop most of my solar cells to feed 1 volt or so to the upconverter which will charge the 5v supercapacitor to 4.2v. And it manages the system power and pulls the plug at 3.27 volts - so I won't drain the supercap overnight and won't get into a start/stop loop with the Arduino at sunrise/sunset. It also has a high/low pin to indicate low voltage, so I could detect this before starting a transmission.

I'd be happy to hear any thoughts on whether you think this is a good idea or if there's anything I'm overlooking. I mostly need to ensure that there's enough juice between the smaller solar cells and the supercap to get through each transmission.

I'm at 5.8 grams for my 9 solar cells, so going to 2 cells should save around 3.5 grams. The charge controller is only .7 grams (around .4 grams if I order the thinner board from OSH Park and assemble it myself), so that's a savings of around 3 grams.

If my first flights work, I might try designing and fabricating my own custom payload boards - one tiny board to handle everything would be VERY small and MUCH easier to put together on a payload - and more reliable too without all of the wiring. This is what most of the very successful (global circumnavigation is a good measure) floating balloons go with. Some are as low as 6 grams (the weight of one US quarter) for the entire payload!

Jared


On Wed, Jun 21, 2017 at 12:43 PM, Brandon Tibbitts <tibbs327@...> wrote:
Yes I'd be interested in helping, but have been loaded with quite alot in general. I am experienced some in layout, and should be able to put something together if needed. Unless there is a quicker/easier route that is already in progress, i'd be happy to try though.

Were you thinking of .030 thick pcb or .062? I've used express PCB to get things done in the past and typically just design the board to the size I need, and cut them out, getting multiples out of the fixed size sheets they offer.(PCB 3.8 x 2.5 x .062 thick) The software is all manual place and routing, but for simpler circuits not too bad. Is there a desired connector for J's, or just solder pads to hard wire into? 


There are a couple of newer layout design software packages I've been looking at trying out. Things are rapidly evolving in PCB EDA and multiple open source options. Maybe this would be a good time to try one..


Brandon - KD7IIW

On Tue, Jun 20, 2017 at 4:35 PM, Jared Smith <jared@...> wrote:
As a quick update, I've been working with Jason from Etherkit and he's designing a breakout board for the clock generator - https://github.com/NT7S/Si5351APicoBoB/blob/master/pcb/Si5351APicoBoB.pdf It should be MUCH smaller and lighter than the existing board - I'm thinking at most 2 grams (if you don't add the headers and wire it directly).

He's going to upload to OSH Park after he's designed the PCB and I'm going to order at least a few soon. Is anyone else interested in some? Gary?

Now I'm considering designing a breakout for a small solar charge controller so I can use fewer panels. The idea is to charge the supercapacitor with the minimum number of solar cells possible while also ensuring that there's enough juice to get through the 2 minute transmissions.

Jared


On Mon, Jun 19, 2017 at 6:01 PM, Jared Smith <jared@...> wrote:
He's not, but I've copied Brandon on this message to see if he's interested, or if he'd like to join the list - https://groups.io/g/BARC-HAB/

I think I can figure out which components can go away on the Si5351A board and order a prototype, but this level of electronics is definitely not my specialty.


Thanks,

Jared


On Mon, Jun 19, 2017 at 4:30 PM, Josh Jensen <kd7wrc@...> wrote:
Is Brandon Tibbetts on this list? He's done some PCB design work before.

Josh

On Jun 19, 2017 16:14, "Jared Smith" <jared@...> wrote:
D'oh. Hit Send a bit too early.

Anyway... I could go from 9 to 4 (maybe 3 or 2) solar cells with a SPV1040 step-up converter / charge controller - http://www.st.com/en/power-management/spv1040.html - (or similar) to charge the super capacitor to 5 volts. It would take a bit longer to charge it, but it should maintain the charge long enough for the regular transmissions. This would also need a custom PCB (at least I can't find an existing charger breakout board that is in the 0-2 gram weight range).

If I could pull off both of these adjustments, that should put my payload well below 15 grams which would be optimal for a higher altitude float.

Jared


On Mon, Jun 19, 2017 at 4:06 PM, Jared Smith <jared@...> wrote:
Does anyone have experience with PCB design?
My balloon payload finalized, I'm now at around 20 grams or so. This is 5 grams heavier than I want for optimal altitude. I can lose unnecessary resistors, etc., but this might gain me at most 1 gram.

My heaviest component by far is the Si5351A breakout board at 5.3 grams - https://www.etherkit.com/rf-modules/si5351a-breakout-board.html You'll notice that it's on a thick PCB, it has 3 outputs (I only need 1), and it has 5v to 3.3v conversion on-board (I'm strictly 3.3v). 

It should be 3.3v and one output only, with header and edge mount SMA pads. On a thin PCB, with 3.3v only and one output (header and edge-mount SMA pads), I think this could come in at 1 gram or less. I think this would be a very popular board that you could make some $$$ off of.

Is there anyone here or do you know anyone that would be interested in this. I'd fund the prototype boards, but need someone smarter than me to do the design. I have the full PCB design files and schematic for 3 different 5v/3.3v boards that would get you most of the way there.

Another possibility to lose weight is to most of my solar cells and add a solar upconverter/charger. It takes 9 panels to make 5 volts. They're very light weight (I think 4.5 grams for the full array - plus I need some infrastructure to keep them in place), but with a 









Re: PCB designer? Weight issues.

Brandon Tibbitts
 

Yep, that sounds like you're on the right track. I'll keep my eye open for thin PCB options, and once everything is proven out, then could help put it all to one thin/tiny PCB. As you mentioned that is key for longer term reliability.

Simple GPS antenna work better horizontal, but not entirely deaf while vertical either..

Do you have a system/component overview? Or did I miss it somewhere. Also were you archiving the code somewhere?

Brandon

Re: PCB designer? Weight issues.

Jared Smith
 

Thank you Brandon.


I'll put together a GitHub repository for the code and will share it soon. You won't really be able to test it without the hardware, but I'm happy to share it.

The charge controller didn't work as expected. The idea was good (the board was only .6 grams), but I didn't realize that the IC has an input current limiter of 100mA. With the voltage upconvert (1.5v converted to 4.2v), this decreases current and didn't leave enough to keep things running. By the time I added enough solar cells to maintain adequate current, it no longer provided a weight saving to use the charge controller. I'm looking at different options - or just might go with a slightly smaller solar cell and run 8 or 9 of them.

I've been testing everything and it seems to work great. I have it sitting out in the sun this morning to see what happens in marginal solar conditions as the sun comes up. The WSPR mode is incredibly efficient - it was heard in Florida on 10 milliwatts last night. The new, smaller clock generator should be here in a couple weeks, then I think I'll have most everything necessary for a launch.

Jared




On Tue, Jun 27, 2017 at 11:48 PM, Brandon Tibbitts <Tibbs327@...> wrote:

Yep, that sounds like you're on the right track. I'll keep my eye open for thin PCB options, and once everything is proven out, then could help put it all to one thin/tiny PCB. As you mentioned that is key for longer term reliability.

Simple GPS antenna work better horizontal, but not entirely deaf while vertical either..

Do you have a system/component overview? Or did I miss it somewhere. Also were you archiving the code somewhere?

Brandon


Re: PCB designer? Weight issues.

Jared Smith
 

After realizing the previous charge controller wouldn't work, I decided to design my own. The schematic and PCB renders are attached. This uses the SPV1040 charge controller - http://www.st.com/en/power-management/spv1040.html - and is generally based on their example board - http://www.st.com/resource/en/application_note/cd00292052.pdf

The board is very small - only 1 inch X .5 inches. The components are mostly tiny 0603 size - so should be a fun challenge to solder. The IC is 3mmX4.4mm.

The S+/- pads are for the small solar cells. I should be able to use 2 of them in series to make ~1.2 volts. The circuit then upconverts this to 5 volts which charges the supercapacitor (CAP+/-) which functions as a battery (except this one works at -40 degrees). It then routes back to V+/- which will go to the Arduino raw power input. Most of the capacitors and resistors are used to set the controller charge values.

There's also a voltage divider on the output that goes to the A connector. This will allow me to measure the output voltage from the Arduino so I don't transmit or turn on the GPS if the voltage is too low.

Feedback or criticism of the design is VERY welcome. I figure I have around 50/50 odds that I designed this in a way that will actually work. I'll order up some PCBs from OSH Park soon and give it a go.

Thanks,

Jared



On Wed, Jun 28, 2017 at 8:06 AM, Jared Smith <jared@...> wrote:
Thank you Brandon.


I'll put together a GitHub repository for the code and will share it soon. You won't really be able to test it without the hardware, but I'm happy to share it.

The charge controller didn't work as expected. The idea was good (the board was only .6 grams), but I didn't realize that the IC has an input current limiter of 100mA. With the voltage upconvert (1.5v converted to 4.2v), this decreases current and didn't leave enough to keep things running. By the time I added enough solar cells to maintain adequate current, it no longer provided a weight saving to use the charge controller. I'm looking at different options - or just might go with a slightly smaller solar cell and run 8 or 9 of them.

I've been testing everything and it seems to work great. I have it sitting out in the sun this morning to see what happens in marginal solar conditions as the sun comes up. The WSPR mode is incredibly efficient - it was heard in Florida on 10 milliwatts last night. The new, smaller clock generator should be here in a couple weeks, then I think I'll have most everything necessary for a launch.

Jared




On Tue, Jun 27, 2017 at 11:48 PM, Brandon Tibbitts <Tibbs327@...> wrote:

Yep, that sounds like you're on the right track. I'll keep my eye open for thin PCB options, and once everything is proven out, then could help put it all to one thin/tiny PCB. As you mentioned that is key for longer term reliability.

Simple GPS antenna work better horizontal, but not entirely deaf while vertical either..

Do you have a system/component overview? Or did I miss it somewhere. Also were you archiving the code somewhere?

Brandon



Re: PCB designer? Weight issues.

Josh Jensen
 

Jared,

0603 components aren't that bad to solder. It's when you get down to 0402 size that they get to be a pain. 

Josh

On Jul 6, 2017 09:46, "Jared Smith" <jared@...> wrote:
After realizing the previous charge controller wouldn't work, I decided to design my own. The schematic and PCB renders are attached. This uses the SPV1040 charge controller - http://www.st.com/en/power-management/spv1040.html - and is generally based on their example board - http://www.st.com/resource/en/application_note/cd00292052.pdf

The board is very small - only 1 inch X .5 inches. The components are mostly tiny 0603 size - so should be a fun challenge to solder. The IC is 3mmX4.4mm.

The S+/- pads are for the small solar cells. I should be able to use 2 of them in series to make ~1.2 volts. The circuit then upconverts this to 5 volts which charges the supercapacitor (CAP+/-) which functions as a battery (except this one works at -40 degrees). It then routes back to V+/- which will go to the Arduino raw power input. Most of the capacitors and resistors are used to set the controller charge values.

There's also a voltage divider on the output that goes to the A connector. This will allow me to measure the output voltage from the Arduino so I don't transmit or turn on the GPS if the voltage is too low.

Feedback or criticism of the design is VERY welcome. I figure I have around 50/50 odds that I designed this in a way that will actually work. I'll order up some PCBs from OSH Park soon and give it a go.

Thanks,

Jared



On Wed, Jun 28, 2017 at 8:06 AM, Jared Smith <jared@...> wrote:
Thank you Brandon.


I'll put together a GitHub repository for the code and will share it soon. You won't really be able to test it without the hardware, but I'm happy to share it.

The charge controller didn't work as expected. The idea was good (the board was only .6 grams), but I didn't realize that the IC has an input current limiter of 100mA. With the voltage upconvert (1.5v converted to 4.2v), this decreases current and didn't leave enough to keep things running. By the time I added enough solar cells to maintain adequate current, it no longer provided a weight saving to use the charge controller. I'm looking at different options - or just might go with a slightly smaller solar cell and run 8 or 9 of them.

I've been testing everything and it seems to work great. I have it sitting out in the sun this morning to see what happens in marginal solar conditions as the sun comes up. The WSPR mode is incredibly efficient - it was heard in Florida on 10 milliwatts last night. The new, smaller clock generator should be here in a couple weeks, then I think I'll have most everything necessary for a launch.

Jared




On Tue, Jun 27, 2017 at 11:48 PM, Brandon Tibbitts <Tibbs327@...> wrote:

Yep, that sounds like you're on the right track. I'll keep my eye open for thin PCB options, and once everything is proven out, then could help put it all to one thin/tiny PCB. As you mentioned that is key for longer term reliability.

Simple GPS antenna work better horizontal, but not entirely deaf while vertical either..

Do you have a system/component overview? Or did I miss it somewhere. Also were you archiving the code somewhere?

Brandon



Re: PCB designer? Weight issues.

Kevin Reeve
 

Jared,

I have a temp controlled soldering station that includes the heat gun for SMD work/rework, and various tip sizes.  I also have some awesome indium solder. You are welcome to use it.

 There are some great SMD solder videos on Utube.  Some use solder paste across the pads, hold the component with tweezers and heat it up with the heat gun and push in place.

Kevin


From: <BARC-HAB@groups.io> on behalf of Josh Jensen <kd7wrc@...>
Reply-To: "BARC-HAB@groups.io" <BARC-HAB@groups.io>
Date: Thursday, July 6, 2017 at 10:14 AM
To: "BARC-HAB@groups.io" <BARC-HAB@groups.io>
Subject: Re: [BARC-HAB] PCB designer? Weight issues.

Jared,

0603 components aren't that bad to solder. It's when you get down to 0402 size that they get to be a pain. 

Josh

On Jul 6, 2017 09:46, "Jared Smith" <jared@...> wrote:
After realizing the previous charge controller wouldn't work, I decided to design my own. The schematic and PCB renders are attached. This uses the SPV1040 charge controller - http://www.st.com/en/power-management/spv1040.html - and is generally based on their example board - http://www.st.com/resource/en/application_note/cd00292052.pdf

The board is very small - only 1 inch X .5 inches. The components are mostly tiny 0603 size - so should be a fun challenge to solder. The IC is 3mmX4.4mm.

The S+/- pads are for the small solar cells. I should be able to use 2 of them in series to make ~1.2 volts. The circuit then upconverts this to 5 volts which charges the supercapacitor (CAP+/-) which functions as a battery (except this one works at -40 degrees). It then routes back to V+/- which will go to the Arduino raw power input. Most of the capacitors and resistors are used to set the controller charge values.

There's also a voltage divider on the output that goes to the A connector. This will allow me to measure the output voltage from the Arduino so I don't transmit or turn on the GPS if the voltage is too low.

Feedback or criticism of the design is VERY welcome. I figure I have around 50/50 odds that I designed this in a way that will actually work. I'll order up some PCBs from OSH Park soon and give it a go.

Thanks,

Jared



On Wed, Jun 28, 2017 at 8:06 AM, Jared Smith <jared@...> wrote:
Thank you Brandon.


I'll put together a GitHub repository for the code and will share it soon. You won't really be able to test it without the hardware, but I'm happy to share it.

The charge controller didn't work as expected. The idea was good (the board was only .6 grams), but I didn't realize that the IC has an input current limiter of 100mA. With the voltage upconvert (1.5v converted to 4.2v), this decreases current and didn't leave enough to keep things running. By the time I added enough solar cells to maintain adequate current, it no longer provided a weight saving to use the charge controller. I'm looking at different options - or just might go with a slightly smaller solar cell and run 8 or 9 of them.

I've been testing everything and it seems to work great. I have it sitting out in the sun this morning to see what happens in marginal solar conditions as the sun comes up. The WSPR mode is incredibly efficient - it was heard in Florida on 10 milliwatts last night. The new, smaller clock generator should be here in a couple weeks, then I think I'll have most everything necessary for a launch.

Jared




On Tue, Jun 27, 2017 at 11:48 PM, Brandon Tibbitts <Tibbs327@...> wrote:

Yep, that sounds like you're on the right track. I'll keep my eye open for thin PCB options, and once everything is proven out, then could help put it all to one thin/tiny PCB. As you mentioned that is key for longer term reliability.

Simple GPS antenna work better horizontal, but not entirely deaf while vertical either..

Do you have a system/component overview? Or did I miss it somewhere. Also were you archiving the code somewhere?

Brandon



USU launch Saturday morning

Jared Smith
 

I don't have any other details, but if any of you are available to help support this, that would be very helpful. I'm not sure yet if I'll be around. Kevin or I will send more details as we get them.

Jared

Re: USU launch Saturday morning

Tyler Griffiths
 

I may be at the one day class. But I do not teach till after lunch so I can probably be there in the morning. 

Tyler 


On Thu, Jul 13, 2017 at 7:00 AM Jared Smith <jared@...> wrote:
I don't have any other details, but if any of you are available to help support this, that would be very helpful. I'm not sure yet if I'll be around. Kevin or I will send more details as we get them.

Jared

--
Tyler, Sent from Gmail Mobile

Re: USU launch Saturday morning

ag1t@...
 

We won't be able to make it.  We will be out with the scouts hiking from Tony Grove to Bunchgrass that morning.  We will listen for it though.

  Gary