V_write_record:
' Write data with time to flash drive
SEROUT TX\CTS, vBaud, [noparse][[/noparse] $8,$20,0,0,0,24,CR, HEX2 hour, ":" , HEX2 minute, ":", HEX2 second, "," , degC.BIT15*13+32,DEC3 ABS (degC/10),".", DEC1 ABS degC,",",DEC3 (RH/10),".",DEC1 RH,CR, LF, CR]
Later on this evening I'll put the other support code for the Vinculum/thumb drive back into the program with the current set of variables, and use the things Phil and I were talking about back on 2/25 to get the initialization right. Around this point in the thread http://forums.parallax.com/showthread.php?p=711123
The support routines in outline are the following, and these will be called from the initialization or from the main loop:
' -----[noparse][[/noparse] Vinculum/Thumb Drive]-----
'
V_purge:
V_Get_Data:
V_TimeoutExit:
RETURN
V_synchronize:
' Send 'E' to start the synchronization process.
' Expect 'E' and a CR for the response, loop until true or error out
' Expect "e" and CR for the second respons
RETURN
V_Check_Drive:
' check if a thumb drive is plugged into the Vinculum, loop until true or error out
RETURN
V_short_commands:
' Vinculum short command mode
RETURN
V_Open_DataFile:
' Thumb drive data file will always append to the same name
RETURN
V_Close_DataFile:
RETURN
V_write_record:
' Write data with time to flash drive,
SEROUT TX\CTS, vBaud, [noparse][[/noparse] $8,$20,0,0,0,24,CR, HEX2 hour, ":" , HEX2 minute, ":", HEX2 second, "," , degC.BIT15*13+32,DEC3 ABS (degC/10),".", DEC1 ABS degC,",",DEC3 (RH/10),".",DEC1 RH,CR, LF, CR]
RETURN
··· Yes, the data will be analyzed using Excel. The flash drive stores the data in 'Notepad' format, undelimited (e.g. 23.4,37.2) From there we'll cut-and-paste it to Excel, then un-delimit
the data, analyze it, and graph it. The 23.4,37.2 format is fine. We''ll certainly go with whatever you guys decide.
The way the data is now going to be recorded on the flash drive means that you
can bypass opening the file in Notepad and cutting and pasting.
The reason is that each data field is now separated by a comma. Therefore,
you can now use the comma delimited mode for directly importing the file.
When you open Excel as a new spreadsheet, the easy way to bring the data in is:
1) File/Open
2) Choose the Datafile.txt on the flash drive
3) Double click on that text file
4) The 'Text Import Wizard window will open (See attached JPG image)
5) Click Next
6) Be sure that the box next to "Comma" is checked. If it is, you should see
the time stamp in column 1 of the Preview.
7) Click Next
8) For each column of data, choose General for now. I didn't see Time as a choice.
9) Click Finish and the spreadsheet should have the elapsed time in column 1
10) If column 1 in the spreadsheet doesn't display hours[noparse]:minutes:[/noparse]seconds,
select column 1 and click on Format and choose Cells. Then under category
select Time and then the right Type.
Column 2 in the spreadsheet will display the recorded temperatures and
column 3 will display the recorded Humidity.
Please be aware that unless you manually delete the older data and start with
a 'clean' flash drive, what gets imported into Excel will be the accumulation
of data that was stored each time the Stamp was powered off and on as well
as each time you pressed the RESET button.
The only way to distinguish between these is that each new group of data
will begin with a time stamp of 00:00:00. There is no way to tell what
day the data was taken. To do that you need a backup battery connected
to the DS1302 RTC. Time is not on your side to get that feature into this
version of hardware. Perhaps next year.
phil
Post Edited (phil kenny) : 3/8/2008 1:34:39 AM GMT
·· Thanks! That's a much simpler way to import the data directly from the flash drive into Excel, then delimit·the data·into columns without commas.
("Columns Without Commas"... sounds like the title of a classic movie, or a great novel!)
Would you care to define "final circuit board". I, for one, presumed the Board of Education was going to be the final platform. Not to say I wasn't a bit concerned about it being the final platform, but you may be in for a world of trouble trying to use a PCB, unless you have some set plans on how you can do it cleanly, neatly, and without any problems. I'd be happy to list the problems I'm aware of that might come about.
There is certainly less danger in using a pre-configured prototyping PCB, if that's what your plans were. In that case, you can forget the above.
Regards,
Bruce Bates
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Programming can't be all that difficult, it's nothing but 1's and 0's
Post Edited (Bruce Bates) : 3/8/2008 4:21:23 PM GMT
Before you begin soldering, please post a schematic of the parts going on the Radio
Shack board. Even a hand drawn one that was scanned would be a great help.
Bruce, I could be mistaken, but my impression of their board is that it will just be an
ordinary perf board with copper-clad holes and roughly the same size and shape of
the white solderless breadboard you see on the BOE.
No custom PCB intended. They plan to stuff the parts into the thru-hole board and hand
wire the interconnects.
Not sure about the mounting details. Maybe Mark can fill us in on those when
he posts the schematic.
How many Radio Shack boards do you plan to build?
More than just one, I hope.
If you make several, it will give you spares as well as enable more then one person
to be involved in the hardware/software debugging process.
phil
Post Edited (phil kenny) : 3/8/2008 4:40:15 PM GMT
I've not seen a schematic diagram of your circuit, but it seems to be pretty stable.
I am not sure if it is already present or not, but you should have a pulldown resistor (~=10kOhms) installed on the scl line of the Sensirion (p0).
How did you deal with the issue of protecting the thumb drive mechanically?
What kind of printed circuit board are you using? Is it a custom board printed for your circuit, or is it a perf board with a grid of copper plated holes?
Think very carefully in advance how to arrange the components, and it doesn't necessarily have to be an exact duplicate of how it is done currently on the white protoblock.
PS: For reference, here from the RTC_ROCKET_ROBOT thread is the recent top view Andrew provided. What I mean about rearranging is that the RTC chip and the Sensirion could change places, and the RTC could be rotated 180 degrees. That would result in fewer crossed wires.
I've attached another version of the program. This one re-incorporates the thumb drive.
I don't have a Vinculum thumb drive here, so you will have to tell me if it works or not, and I hope Phil can look it over too.
When it starts up, it will try to execute the synchronization routine up to 8 times. And it will try the disk check up to 8 times.
It will print out two numbers on the debug screen:
sync=?
disk=? not found
It will print "not found" only if it can't initialze the disk. I hope it will not print that!!!
In any case, I'd like you to record the numbers it prints after the sync= and disk=. It will be a number from 1 to 9. A small number is better, because it means the disk is ready after fewer tries.
If the robot is not attached to the computer, the LED will flash once at startup, and then it will flash out two number in Morse code. Please write those down, because those are the same two numbers for sync= and disk=
If the disk is initialized correctly, it will start off the new data file by writing the word, "Daedelus" into the file, and then it will go into recording data line by line as is listed in the program header from the discussion yesterday.
··· The final circuit board I mentioned is (tentatively) a 'blank' Radio Shack printed circuit board the same size as the broadboard on the Parallax
circuit board. It has the copper 'channels' (I don't know what they're really called) etched into it, along with holes where the wires go, and are soldered.
After you asked I began to wonder: does Parallax make a 'blank' circuit board for the same purpose, of the same size?·What are the options for
permanently·attaching the wires, sensors, and·Datalogger to a printed circuit board that's laid on in the same configuration as the Board·of Education?
To be straight-up honest with everyone, I had even considered putting a drop of super glue into each of the holes on the breadboard (to attach the wires,
sensors, etc.)... as a last (minute) resort. I know, I know: "SACRILEGE!" you're thinking. It was just a thought (brain cramp?!)··
Tracy, I'm going to download the latest version of the program as soon as I 'send' this reply to Bruce. Thanks so much; I'm really eager to see how it runs!
And thank you for carelessly sacrificing your time so we have it in time for practice tomorrow. You are too kind...··· -·· ---·· ---·· -.-··..·· -.·· -..
Christopher just beckoned: "Dinner's ready!" I'll report back after dinner (chicken, rice, and squash tonight!) and after I run the program.
Best regards from NH,
Mark and the Rocketeers
Bruce Bates said...
Mark -
Would you care to define "final circuit board". I, for one, presumed the Board of Education was going to be the final platform. Not to say I wasn't a bit concerned about it being the final platform, but you may be in for a world of trouble trying to use a PCB, unless you have some set plans on how you can do it cleanly, neatly, and without any problems. I'd be happy to list the problems I'm aware of that might come about.
There is certainly less danger in using a pre-configured prototyping PCB, if that's what your plans were. In that case, you can forget the above.
Although I only have the Datalogger hardware and none of the rest of the
robot, the latest code you posted looks like it never gets the "E" when
trying to initialize the Vinculum. The contents of the Debug window are
attached as a jpg.
In trying to pinpoint the problem, I've put Debug statements inside the
V_Get_Data and V_Synchronize subroutines to print out the entire 12
bytes of the buffer. On every pass the buffer contents remain zero.
To verify that I don't have a flaky Datalogger I reloaded the program
I used for storing data on the flash drive. It ran fine.
···· I ate·a hasty and tasty dinner, then hurried in to try out the new program. It runs exactly as we had hoped,·showing elapsed time, temperature,
humidity, sample number... even Daedelus' name. AND it still talks to us in Morse code! I couldn't be happier! Thank you. I feel that we are over a hump
bigger than you could ever imagine, especially with a report and video teleconference with NASA looming on the near horizon (and a houseful of eager
young programmers-roboticists-rocketeers in less than 15 hours.)
There was an initial glitch, similar to what we experienced a while back. At first the program would run just fine connected directly to the
computer. But when it was run independently (by switching it on to initialize it, without the serial cable attached), it 'hung up' again and·refused to·initialize.
I scratched my head; opened the program; then really scratched my head;·then puzzled over it logically, like a physicist-programmer as·Bruce·suggested.
I reasoned that it was 'hanging up' when it was initializing. Then I read Tracy's code, trying to figure out where to begin, as lost as ever.
Then I remembered that we had added a longer PAUSE before, to give the (Datalogger? Flash drive? Viniculum?) longer to 'boot up.' I added the change shown
below-- and miracle of all miracles-- it seemed to do the trick! I was dumbfounded! The very first 'fix' I tried seemed to work. I have absolutely no real idea
why except for the logic (?) I presented above.·Can you please confirm that this psuedo-fix is not so 'pseudo. The robot seems to run fine and initialize
independent of the computer just fine now. But expert, independent validation is certainly called (or rather, asked) for. Here's what I did:
Reset: · OUTS =· %0001001000000000·· ' indicator led is on. rts (p9) high to start, tx (p8) low to start. · 'fedcba9876543210 · DIRS =· %1111001111111101·· ' p1 is input for sensirion data, pa and pb input for rx and cts · DEBUG CR,"top",CR · PAUSE 10 · GOSUB RTC_init· ' reset the clock to 00:00:00
* PAUSE 1000······'---- MARK ADDED this line 08 March 08 ---- *·GOSUB RTC_init· '---- MARK ADDED this line 08 March 08 ----
I suspect (based solely on hunch and nothing more) that if we change the first PAUSE to 1000 and just delete the second PAUSE and
GOSUB commands it would run just fine. At this·point, though,·I'm not gonna mess with karma, kizmet, ying, or yang. It works, and so I'll
leave it for now (though I'm sure before the evening is over I'll copy the program and then delete the second PAUSE and GOSUB commands and
run it "just to see what happens." Stay tuned for details (or sparks!)
Thanks boatloads, Tracy. I'll keep you posted, and I'll look for everyone's comments or critiques.
·· Is the "pre-configured prototyping PCB" that you mention the same as the Radio Shack PCB that·I mention below? I'll try and post a picture so that
everyone knows what we're thinking of using. Does Parallax make an alternative that's laid out exactly like the Board of Education, and fairly easy
to solder? I don't think any of these kids have ever soldered before!
Thanks,··
Mark, the Rocketeers, Daedalus, and Bandit the Rocket Dog
Bruce Bates said...
Mark -
Would you care to define "final circuit board". I, for one, presumed the Board of Education was going to be the final platform. Not to say I wasn't a bit concerned about it being the final platform, but you may be in for a world of trouble trying to use a PCB, unless you have some set plans on how you can do it cleanly, neatly, and without any problems. I'd be happy to list the problems I'm aware of that might come about.
There is certainly less danger in using a pre-configured prototyping PCB, if that's what your plans were. In that case, you can forget the above.
The pre-configured prototyping PCB's of which I spoke are not sufficiently adaptable as a "replacement" for the BOE Board (in it's entirety) as I had envisioned. I was perhaps thinking one step beyond what you had planned. At least one element (RESET, battery power, etc) present on the BOE Development Board was missing on all the boards I reviewed. I guess that shows just how unique and targeted the Parallax boards are, and to their credit. Sometimes my visions go beyond the requirements, but the thought was there
Regardless of that, after 3-4 hours searching for what I had HOPED to find, no such luck. At this juncture I would recommend going ahead with the Radio Shack PCB or any other one which fits your needs for the replacement of the breadboard area, as you had planned.
If I find anything really "outstanding", I'll report back with the details, but don't hold your breath because of me. You'd look funny as Blue Max <er> Blue Mark
Regards,
Bruce Bates
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Programming can't be all that difficult, it's nothing but 1's and 0's
Just to point out, I have actually soldered before, I just haven't done it in a while. Attached is a picture of a·perfboard that looks similar to ours, it may not be the same size or shape. Also, I found an interesting link about how to solder a perfboard @ http://itp.nyu.edu/physcomp/Tutorials/SolderingAPerfBoard
See you at practice today,
Andrew
Mark in NH said...
Bruce,
·· Is the "pre-configured prototyping PCB" that you mention the same as the Radio Shack PCB that·I mention below? I'll try and post a picture so that
everyone knows what we're thinking of using. Does Parallax make an alternative that's laid out exactly like the Board of Education, and fairly easy
to solder? I don't think any of these kids have ever soldered before!
Thanks,··
Mark, the Rocketeers, Daedalus, and Bandit the Rocket Dog
Post Edited (Andrew Mahn) : 3/9/2008 12:10:03 PM GMT
·· This "Blue Mark"...! The picture that Andrew just uploaded is very close to what we thought about using (Radio Shack PCB). Keep in mind that it only
replaces the broadboard, not the entire Board of Education. The switches, etc. would remain on it. That brings up a key question: how do we wire the
sensors, wires, etc. to the Parallax board? As I (mis)understand it, the Board of Education is essentially a junction box that the wires are routed
through temporarily, until a project is complete and ready to be permanantly hardwired (soldered.) That's where we had proposed replacing the BOE
with a more permament PCB. Thoughts, anyone? Are we barking (or 'Mark'-ing) up the right tree? Does Parallax make their own PCB that is designed to
be permanently wired, after prototyping is done...? I'll browse through their newest catalog again at breakfast.
Bruce, did you read that the robot's program is virtually done?! That's great news, isn't it? What a big hurdle to be over! You guys have been absolutely
awesome with your steadfast encouragement, patience, and support! Thanks so much. We'll report back in later today at practice.
The Board of Education platform is intended as a programming facility, a pre-prototyping testbed, a means for performing experiments, and most importantly, it was designed as a classroom tool for demonstration of PBASIC Stamp concepts and for students to implement their ideas. It was never really intended as a permanent solution for projects which are soon headed for outer space
Parallax offers no PCB which replaces any of their programming boards, as far as I am aware. BTW, it's "breadboard", not "broadboard", but that may just be a typo, I wanted you to be aware of the actual name of the white area with the holes on the BOE Board.
As far as the permanent wiring is concerned, you're on your own with that one. This is one reason why I'd thought about a complete replacement for the BOE altogether, but that is not in the cards for now. What you may find under that white breadboard may be a surprise, or it may be just what you expect it to be. I've never tried to pry one off, so I can't assist you there.
Just as a matter of information, I've read every message that's been posted to these Rocket-Robot threads. The fact that you haven't heard much from me recently, only attests to the great work done by Tracy, Phil and others. "Too many cooks spoil the broth" and that goes DOUBLE for programming and troubleshooting. Had I sensed that either Tracy or Phil were having a problem· with either (I rather doubt that they would), I'd have been more than happy to help out.
I am extremely happy the robot program is near to being finished, but let's not have the launch party quite yet. There is probably more testing to do, as I'm sure you will see in the coming days ahead.
Regards,
Bruce Bates
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Programming can't be all that difficult, it's nothing but 1's and 0's
Post Edited (Bruce Bates) : 3/9/2008 1:59:39 PM GMT
I'm very happy the program is working. For a spaceflight, it would be nice to know really why something works, and not depend on miracles. But that is a detail that can be put on solid footing before the NASA engineers ask you if you are confident it will work 100%! At least it is working now, we hope.
My feeling is that transferring the circuit to a solder perf-board could create more problems than it solves. I won't go into why I think that unless you ask. By the way, the term "breadboard" comes from the early days of radio, I think, when an experimenter might actually used a wooden breadboard with nails sticking up to make connections.
At this stage of the game I do think that you can "harden" the white plugin breadboard. And yes, when everything is set, tack the components and the wires down with a little dab of glue. My personal favorite is GOOP adhesive, with the little dab taken on the end of a toothpick.
I'm basing my comments on the last photo that Andrew posted. http://forums.parallax.com/attachment.php?attachmentid=52620
1a) Exchange the position of the SHT and the RTC
1b) Turn the RTC 180 degrees.
That will make the signal connections to those chips straight and parallel.
Do the wiring with the wires lying flat against the breadboard, not looping high up.
2) Pull the data logger off the board and turn it 180 degrees (per PhiPi's suggestion), so that the thumb faces inward.
The connections will have to be made underneath the thumb drive while it is off the board, so thinking will be required!
NOTE: It is easy to change the pin assignments in the program, if that will simplify the wiring.
3) Think through the power supply connections, again to keep them neat, neat neat.
4) The enemy will be shock and vibration. When everything is set, think about what could shake loose and how to prevent that.
·· Thanks for your post. I was·glad to hear that you agree that soldering to a perf-board isn't our best option, at least at this point. I was strongly leaning
toward not soldering since I've never really done any perf-board soldering. But when·said that·it could create more problems than it solves, that clinched it.
As Bruce mentioned, 'broadboard' was a typo. I knew it was called a breadboard... but I never really knew why! What an education in the history of
electronics we're getting!
So you think we could use 'Goop' (or 5-minute epoxy?) to tack down the compenents and wires safely (enough)? The humidity sensor and RTC
seem pretty·snug·in the holes already; it's the wires I'm more concerned about. I'll get some Goop and experiment with it so I get an idea of its
viscosity, drying time, etc. We use a lot of 5-minute epoxy·in the rocketry programs, but I've never really used Goop a whole lot. I'll try it and then report
back.
So if we exchange the position of the SHT and the RTC,·rotate the RTC 180 degrees, and turn the Datalogger 180 degrees·to make the signal connections
straight and parallel, will we have to re-arrange the wires? Or do we simply change the pin assignments in the program (or both)? We'll be sure to·lay the
wires flat against the breadboard and not loop them, like you suggested. I'm not sure what you meant by "Think through the power supply connections"
(though I did get the point about neat, neat, neat...!) I like to see things orderly and precise, and easy to follow visually. Which specific connections are
the power supply connections?
Your point that we·cannot always·not depend on miracles to happen is·well taken.·The rocketeers and I·fully understand that·you and our friends on the
forum did most of the programming while we watched and learned. And learn we have!·I am working to put that detail on solid footing like you suggested, and
I am 100% confident the robot will work!
I cannot begin to tell you what a miracle I believe it was to find someone like you, and Phil, and Bruce to help us with the programming, and to help me mentor
these fine young students. I am not overstating the point one bit when I say, "Each of you has restored my faith in humanity." There really are kind and generous
people in the world, and you are them. Thank you, from the bottom of my heart, for all the time and effort you have generously given to (y)our project.
Here are some of the features of GOOP (aka Amazing GOOP) from their web site:
1. One part (no mixing)
2. Cures completely in 24-72 hours
3. Consistency:
··· 1) Thick but will run on vertical surfaces when wet
··· 2) Rubbery texture when dry
4. Dries clear
5. Max. Temperature 150 degrees F
- - - - -
I was hoping to find the actual chemical composition of GOOP, as I'm always concerned when using an "unknown" product·in what effect it might have on the surface(s) to which it will be applied. Unfortunately, none was to be had, nor was an MSDS* statement available.
Hint: When the chemical composition isn't readily apparent, one can often find an MSDS* data sheet, which MUST show the exact chemicals or substances used in the product.
*MSDS = Material Safety Data Sheet
Regards,
Bruce Bates
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Programming can't be all that difficult, it's nothing but 1's and 0's
Post Edited (Bruce Bates) : 3/10/2008 3:41:13 AM GMT
If you make those changes to the arrangement of parts, I believe the wires will come out in better, neater positions, even when you leave the pin assignments exactly as they are. At least for the SHT and the RTC. What I suggest is that you write it down on paper first. Make a diagram on grid paper similar to the picture below. (The diagram below shows a round part and two wires, but that of course is not on your board. ) The grid matches one for one with the holes on your breadboard. Make several copies, because it may take several tries and all the rocketeers may have a crack at it. Email them all a copy of the grid.
Then with a pencil draw in the SHT and RTD with where their pins are going to be plugged in., and the fun begins. Draw in the lines for the connections to the signals p0 to p15. And also to the power supplies Vss and Vdd. The game is to find the neatest arrangement. My suggestion about swapping the positions of the SHT and RTC (and rotating the RTC 180°) would I think come out better and keep the same pin assignments. But there are other ways to do it. The pin assignments are not written in stone, but keep in mind that in the end you will want all of your robots to have the same wiring (one single program version!). Now, I have not thought about how it will turn out with the thumb drive. You listened to a lot of suggestions about how to deal with the mechanical issue earlier in the thread. Once the rocketeers have settled on how to mount it so that it will be mechanically solid, they can deal with the neatness of the wiring. There is where it may prove useful to change a couple of the pin assignments.
The reason I like goop is that it has a rubbery consistency. (It is basically a rubber compound mixed in a nasty solvent -- always use with good ventilation!) It sticks to almost anything, it is viscous so it doesn't run down into the holes but can be pulled off clean if need be, and it is a good insulator. I like 5 minute epoxy too.
>So if we exchange the position of the SHT and the RTC, rotate the RTC 180 degrees, and turn the Datalogger 180 degrees to make the signal connections
>straight and parallel, will we have to re-arrange the wires? Or do we simply change the pin assignments in the program (or both)? We'll be sure to lay the
>wires flat against the breadboard and not loop them, like you suggested. I'm not sure what you meant by "Think through the power supply connections"
>(though I did get the point about neat, neat, neat...!) I like to see things orderly and precise, and easy to follow visually. Which specific connections are
>the power supply connections?
Forgive me if I'm stating the obvious, but you know that you're going to need to take special care to secure the batteries driving this thing, right? Under the acceleration you're going to get when you take off, a 9V battery can·easily weigh more than a half pound (and quite a bit more than that when your ejection charges go off) and standard battery snaps will definitely not hold it in place. You do not want a loose battery shaking around inside your payload bay in the middle of your careful work.
The battery will need to be held with zip ties (two of 'em, perpendicular to each other) onto a hard surface. Tie 'em down tightly. Alternatively you can get a plastic battery holder that will secure it along the plane of the payload sled (or whatever you're using), but you'll still need to use a zip tie to keep it from flying off of that.
The thumb drive isn't going to stay put in the socket either, or at least I wouldn't bet on it. That will need to be tied down as well somehow. Remember that even if you have a slow, smooth liftoff, when your ejection charges go off, you will get some pretty good acceleration spikes, and possibly another one when the payload section reaches the end of the shock cord. I've attached a little graph of the data from one of my flights about a year ago - that was on a K800, which gave a solid 10G spike at takeoff, but you can see that the two ejection charges were nothing to sneeze at either.
·· Thanks for reminding us of the obvious (which, because we were so immersed in programming, I put on the back burner.) Of course! G-forces! (Geeeeeee...!) We'll definitely have to secure the battery, and the flash drive, extremely well because we're launching on an Aerotech K1100T-15. Last year the rocket came off the launch rail like a rifle shot. In the photo you can even see 'Mach diamonds' in the flame coming out of the nozzle... before the rocket is even off the rail! In our TARC competition rockets, which use a much smaller engine (G80-7 is the largest) we secure the PerfectFlite altimeters with zip ties, as you suggested for our larger rocket.
I handdn't thought much about the shock forces from the ejection charges/ ejection events. There are five to consider, and any one could dislodge a launch- or data-critical component. The ejection events are:
1) engine ejection delay (15 seconds) = drogue 'chute AND
2) drogue altimeter ejection at apogee = (back-up for engine delay grain)
3) back-up drogue altimeter at apogee (back-up for the back-up)
4) main ejection charge set for 1,200' ASL
5) back-up main ejection charge set for 600' ASL -
6) Four parachutes total: drogue, robot, nose cone, and main· parachute.
I'll mull this over as I drive Christopher to Boy Scouts (I also stay busy as a Scoutmaster.) Thanks for your post. Hope to hear more from a fellow HPR rocketeer!
Just don't tell the Washington County 4H guys or especially the Madison West team that I reminded you. I fly with those guys.
With all of those backups, it sounds like there's going to be a lot of action in the air. Be sure that your team is careful after landing, in case there are any live charges left due to things not firing properly. We had one go off in someone's hands after a failed flight two weeks ago. No harm or injury, but make danged sure that all of the charges have fired before you allow a crowd near the thing after landing.
Post Edited (sylvie369) : 3/10/2008 10:16:14 PM GMT
Comments
I think you left out the two colon symbols.
Maybe like this:
Post Edited (phil kenny) : 3/8/2008 1:00:11 AM GMT
Later on this evening I'll put the other support code for the Vinculum/thumb drive back into the program with the current set of variables, and use the things Phil and I were talking about back on 2/25 to get the initialization right. Around this point in the thread http://forums.parallax.com/showthread.php?p=711123
The support routines in outline are the following, and these will be called from the initialization or from the main loop:
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
·· Hmmmmm. This is over·my head...
Standing· ·-...···-.--·
Mark··········· b········· y
··· Yes, the data will be analyzed using Excel. The flash drive stores the data in 'Notepad' format, undelimited (e.g. 23.4,37.2) From there we'll cut-and-paste it to Excel, then un-delimit
the data, analyze it, and graph it. The 23.4,37.2 format is fine. We''ll certainly go with whatever you guys decide.
Thanks,
Mark
The way the data is now going to be recorded on the flash drive means that you
can bypass opening the file in Notepad and cutting and pasting.
The reason is that each data field is now separated by a comma. Therefore,
you can now use the comma delimited mode for directly importing the file.
When you open Excel as a new spreadsheet, the easy way to bring the data in is:
1) File/Open
2) Choose the Datafile.txt on the flash drive
3) Double click on that text file
4) The 'Text Import Wizard window will open (See attached JPG image)
5) Click Next
6) Be sure that the box next to "Comma" is checked. If it is, you should see
the time stamp in column 1 of the Preview.
7) Click Next
8) For each column of data, choose General for now. I didn't see Time as a choice.
9) Click Finish and the spreadsheet should have the elapsed time in column 1
10) If column 1 in the spreadsheet doesn't display hours[noparse]:minutes:[/noparse]seconds,
select column 1 and click on Format and choose Cells. Then under category
select Time and then the right Type.
Column 2 in the spreadsheet will display the recorded temperatures and
column 3 will display the recorded Humidity.
Please be aware that unless you manually delete the older data and start with
a 'clean' flash drive, what gets imported into Excel will be the accumulation
of data that was stored each time the Stamp was powered off and on as well
as each time you pressed the RESET button.
The only way to distinguish between these is that each new group of data
will begin with a time stamp of 00:00:00. There is no way to tell what
day the data was taken. To do that you need a backup battery connected
to the DS1302 RTC. Time is not on your side to get that feature into this
version of hardware. Perhaps next year.
phil
Post Edited (phil kenny) : 3/8/2008 1:34:39 AM GMT
·· Thanks! That's a much simpler way to import the data directly from the flash drive into Excel, then delimit·the data·into columns without commas.
("Columns Without Commas"... sounds like the title of a classic movie, or a great novel!)
Mark
Can you send me the latest version of the program so I can use it before we meet for practice tomorrow?
Thanks,
Andrew
···The·latest version of the program, without·the subroutines that Mr. Allen and Mr. Kenney have been discussing (above)
are posted here on the forum (9 or 10 posts before this one.) They suggested that we wait until these new subroutines are·fully
intregrated into the program, and the Dataloggger and program are fully functional (sometime this weekend, I believe.) Please check the
e-mail I just sent you about the Flight Readiness Review report and pre-flight arming sequence. Christopher came home from school
yesterday and he will help you with it during practice tomorrow (1:00-4:30 PM.) He'll also be updating our team binder (as you should be)
by downloading the information from the forum. He'll also help you with soldering the crystal to your RTC chip. Did they come yet?
Keep up the good work!
...··· .····· .···· you!
Mr. Kibler
·· Do you think we should hardwire (solder) the sensors, Datalogger, and wires to the final printed circuit board yet?
It will be the same size as the Parallax Board of Education and it will replace it on the robot and nose cone module.
We have·five, 3-1/2 hour·Sunday practices left to get the project 'tight' (as in, 'completely done') We have some rocket
construction to finish up, pre-flight arming and launch sequences to write, the program to dial in, a Flight Readiness Review
report and video teleconference with NASA, and a few other odds and ends (travel arrangements, parent meeting, build a
box to ship the rocket in, etc. (we're carrying the robot on the place with us... we hope!) Then, it's launch time at NASA!
"Flame protection shield activated..." step 64 in the pre-launch procedure!)
Thanks!
Mark and the Rocketeers
Would you care to define "final circuit board". I, for one, presumed the Board of Education was going to be the final platform. Not to say I wasn't a bit concerned about it being the final platform, but you may be in for a world of trouble trying to use a PCB, unless you have some set plans on how you can do it cleanly, neatly, and without any problems. I'd be happy to list the problems I'm aware of that might come about.
There is certainly less danger in using a pre-configured prototyping PCB, if that's what your plans were. In that case, you can forget the above.
Regards,
Bruce Bates
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Programming can't be all that difficult, it's nothing but 1's and 0's
Post Edited (Bruce Bates) : 3/8/2008 4:21:23 PM GMT
Before you begin soldering, please post a schematic of the parts going on the Radio
Shack board. Even a hand drawn one that was scanned would be a great help.
Bruce, I could be mistaken, but my impression of their board is that it will just be an
ordinary perf board with copper-clad holes and roughly the same size and shape of
the white solderless breadboard you see on the BOE.
No custom PCB intended. They plan to stuff the parts into the thru-hole board and hand
wire the interconnects.
Not sure about the mounting details. Maybe Mark can fill us in on those when
he posts the schematic.
How many Radio Shack boards do you plan to build?
More than just one, I hope.
If you make several, it will give you spares as well as enable more then one person
to be involved in the hardware/software debugging process.
phil
Post Edited (phil kenny) : 3/8/2008 4:40:15 PM GMT
I am not sure if it is already present or not, but you should have a pulldown resistor (~=10kOhms) installed on the scl line of the Sensirion (p0).
How did you deal with the issue of protecting the thumb drive mechanically?
What kind of printed circuit board are you using? Is it a custom board printed for your circuit, or is it a perf board with a grid of copper plated holes?
Think very carefully in advance how to arrange the components, and it doesn't necessarily have to be an exact duplicate of how it is done currently on the white protoblock.
PS: For reference, here from the RTC_ROCKET_ROBOT thread is the recent top view Andrew provided. What I mean about rearranging is that the RTC chip and the Sensirion could change places, and the RTC could be rotated 180 degrees. That would result in fewer crossed wires.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
Post Edited (Tracy Allen) : 3/8/2008 5:20:51 PM GMT
.... ..
I've attached another version of the program. This one re-incorporates the thumb drive.
I don't have a Vinculum thumb drive here, so you will have to tell me if it works or not, and I hope Phil can look it over too.
When it starts up, it will try to execute the synchronization routine up to 8 times. And it will try the disk check up to 8 times.
It will print out two numbers on the debug screen:
sync=?
disk=? not found
It will print "not found" only if it can't initialze the disk. I hope it will not print that!!!
In any case, I'd like you to record the numbers it prints after the sync= and disk=. It will be a number from 1 to 9. A small number is better, because it means the disk is ready after fewer tries.
If the robot is not attached to the computer, the LED will flash once at startup, and then it will flash out two number in Morse code. Please write those down, because those are the same two numbers for sync= and disk=
If the disk is initialized correctly, it will start off the new data file by writing the word, "Daedelus" into the file, and then it will go into recording data line by line as is listed in the program header from the discussion yesterday.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
··· The final circuit board I mentioned is (tentatively) a 'blank' Radio Shack printed circuit board the same size as the broadboard on the Parallax
circuit board. It has the copper 'channels' (I don't know what they're really called) etched into it, along with holes where the wires go, and are soldered.
After you asked I began to wonder: does Parallax make a 'blank' circuit board for the same purpose, of the same size?·What are the options for
permanently·attaching the wires, sensors, and·Datalogger to a printed circuit board that's laid on in the same configuration as the Board·of Education?
To be straight-up honest with everyone, I had even considered putting a drop of super glue into each of the holes on the breadboard (to attach the wires,
sensors, etc.)... as a last (minute) resort. I know, I know: "SACRILEGE!" you're thinking. It was just a thought (brain cramp?!)··
Tracy, I'm going to download the latest version of the program as soon as I 'send' this reply to Bruce. Thanks so much; I'm really eager to see how it runs!
And thank you for carelessly sacrificing your time so we have it in time for practice tomorrow. You are too kind...·
Christopher just beckoned: "Dinner's ready!" I'll report back after dinner (chicken, rice, and squash tonight!) and after I run the program.
Best regards from NH,
Mark and the Rocketeers
I'm going to download the latest version of the program as soon as I send this reply to Bruce. Thanks so much; I'm really eager to see how it runs!
And thank you for carelessly sacrificing your time so we have it in time for practice tomorrow. You are too kind...·
Christopher just beckoned: "Dinner's ready!" I'll report back after dinner (chicken, rice, and squash tonight!) and after I run the program.
Best regards from NH,
Mark and the Rocketeers
Although I only have the Datalogger hardware and none of the rest of the
robot, the latest code you posted looks like it never gets the "E" when
trying to initialize the Vinculum. The contents of the Debug window are
attached as a jpg.
In trying to pinpoint the problem, I've put Debug statements inside the
V_Get_Data and V_Synchronize subroutines to print out the entire 12
bytes of the buffer. On every pass the buffer contents remain zero.
To verify that I don't have a flaky Datalogger I reloaded the program
I used for storing data on the flash drive. It ran fine.
Still working on trying to find the hangup.
Maybe Mark will have better luck.
phil
···· I ate·a hasty and tasty dinner, then hurried in to try out the new program. It runs exactly as we had hoped,·showing elapsed time, temperature,
humidity, sample number... even Daedelus' name. AND it still talks to us in Morse code! I couldn't be happier! Thank you. I feel that we are over a hump
bigger than you could ever imagine, especially with a report and video teleconference with NASA looming on the near horizon (and a houseful of eager
young programmers-roboticists-rocketeers in less than 15 hours.)
There was an initial glitch, similar to what we experienced a while back. At first the program would run just fine connected directly to the
computer. But when it was run independently (by switching it on to initialize it, without the serial cable attached), it 'hung up' again and·refused to·initialize.
I scratched my head; opened the program; then really scratched my head;·then puzzled over it logically, like a physicist-programmer as·Bruce·suggested.
I reasoned that it was 'hanging up' when it was initializing. Then I read Tracy's code, trying to figure out where to begin, as lost as ever.
Then I remembered that we had added a longer PAUSE before, to give the (Datalogger? Flash drive? Viniculum?) longer to 'boot up.' I added the change shown
below-- and miracle of all miracles-- it seemed to do the trick! I was dumbfounded! The very first 'fix' I tried seemed to work. I have absolutely no real idea
why except for the logic (?) I presented above.·Can you please confirm that this psuedo-fix is not so 'pseudo. The robot seems to run fine and initialize
independent of the computer just fine now. But expert, independent validation is certainly called (or rather, asked) for. Here's what I did:
Reset:
· OUTS =· %0001001000000000·· ' indicator led is on. rts (p9) high to start, tx (p8) low to start.
· 'fedcba9876543210
· DIRS =· %1111001111111101·· ' p1 is input for sensirion data, pa and pb input for rx and cts
· DEBUG CR,"top",CR
· PAUSE 10
· GOSUB RTC_init· ' reset the clock to 00:00:00
* PAUSE 1000······'---- MARK ADDED this line 08 March 08 ----
*·GOSUB RTC_init· '---- MARK ADDED this line 08 March 08 ----
I suspect (based solely on hunch and nothing more) that if we change the first PAUSE to 1000 and just delete the second PAUSE and
GOSUB commands it would run just fine. At this·point, though,·I'm not gonna mess with karma, kizmet, ying, or yang. It works, and so I'll
leave it for now (though I'm sure before the evening is over I'll copy the program and then delete the second PAUSE and GOSUB commands and
run it "just to see what happens." Stay tuned for details (or sparks!)
Thanks boatloads, Tracy. I'll keep you posted, and I'll look for everyone's comments or critiques.
Mark and the Rocketeers
·
·· Is the "pre-configured prototyping PCB" that you mention the same as the Radio Shack PCB that·I mention below? I'll try and post a picture so that
everyone knows what we're thinking of using. Does Parallax make an alternative that's laid out exactly like the Board of Education, and fairly easy
to solder? I don't think any of these kids have ever soldered before!
Thanks,··
Mark, the Rocketeers, Daedalus, and Bandit the Rocket Dog
·· I was having a similar problem but it "seems" to be remedied (I say that cautiously.) Take a look at·what I posted two posts and let me know what
you think and/or if it works. I'm not getting the strange data (-40 degrees, etc.) that I see on your debug screen (unless I accidently knock a wire
loose.) Let me know what happens when you lengthen, or add the 'reset' subroutine PAUSE to 1000 if you·try it.
Mark
The pre-configured prototyping PCB's of which I spoke are not sufficiently adaptable as a "replacement" for the BOE Board (in it's entirety) as I had envisioned. I was perhaps thinking one step beyond what you had planned. At least one element (RESET, battery power, etc) present on the BOE Development Board was missing on all the boards I reviewed. I guess that shows just how unique and targeted the Parallax boards are, and to their credit. Sometimes my visions go beyond the requirements, but the thought was there
Regardless of that, after 3-4 hours searching for what I had HOPED to find, no such luck. At this juncture I would recommend going ahead with the Radio Shack PCB or any other one which fits your needs for the replacement of the breadboard area, as you had planned.
If I find anything really "outstanding", I'll report back with the details, but don't hold your breath because of me. You'd look funny as Blue Max <er> Blue Mark
Regards,
Bruce Bates
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Programming can't be all that difficult, it's nothing but 1's and 0's
Just to point out, I have actually soldered before, I just haven't done it in a while. Attached is a picture of a·perfboard that looks similar to ours, it may not be the same size or shape. Also, I found an interesting link about how to solder a perfboard @ http://itp.nyu.edu/physcomp/Tutorials/SolderingAPerfBoard
See you at practice today,
Andrew
Post Edited (Andrew Mahn) : 3/9/2008 12:10:03 PM GMT
·· This "Blue Mark"...! The picture that Andrew just uploaded is very close to what we thought about using (Radio Shack PCB). Keep in mind that it only
replaces the broadboard, not the entire Board of Education. The switches, etc. would remain on it. That brings up a key question: how do we wire the
sensors, wires, etc. to the Parallax board? As I (mis)understand it, the Board of Education is essentially a junction box that the wires are routed
through temporarily, until a project is complete and ready to be permanantly hardwired (soldered.) That's where we had proposed replacing the BOE
with a more permament PCB. Thoughts, anyone? Are we barking (or 'Mark'-ing) up the right tree? Does Parallax make their own PCB that is designed to
be permanently wired, after prototyping is done...? I'll browse through their newest catalog again at breakfast.
Bruce, did you read that the robot's program is virtually done?! That's great news, isn't it? What a big hurdle to be over! You guys have been absolutely
awesome with your steadfast encouragement, patience, and support! Thanks so much. We'll report back in later today at practice.
Mark and the Rocketeers
The Board of Education platform is intended as a programming facility, a pre-prototyping testbed, a means for performing experiments, and most importantly, it was designed as a classroom tool for demonstration of PBASIC Stamp concepts and for students to implement their ideas. It was never really intended as a permanent solution for projects which are soon headed for outer space
Parallax offers no PCB which replaces any of their programming boards, as far as I am aware. BTW, it's "breadboard", not "broadboard", but that may just be a typo, I wanted you to be aware of the actual name of the white area with the holes on the BOE Board.
As far as the permanent wiring is concerned, you're on your own with that one. This is one reason why I'd thought about a complete replacement for the BOE altogether, but that is not in the cards for now. What you may find under that white breadboard may be a surprise, or it may be just what you expect it to be. I've never tried to pry one off, so I can't assist you there.
Just as a matter of information, I've read every message that's been posted to these Rocket-Robot threads. The fact that you haven't heard much from me recently, only attests to the great work done by Tracy, Phil and others. "Too many cooks spoil the broth" and that goes DOUBLE for programming and troubleshooting. Had I sensed that either Tracy or Phil were having a problem· with either (I rather doubt that they would), I'd have been more than happy to help out.
I am extremely happy the robot program is near to being finished, but let's not have the launch party quite yet. There is probably more testing to do, as I'm sure you will see in the coming days ahead.
Regards,
Bruce Bates
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Programming can't be all that difficult, it's nothing but 1's and 0's
Post Edited (Bruce Bates) : 3/9/2008 1:59:39 PM GMT
I'm very happy the program is working. For a spaceflight, it would be nice to know really why something works, and not depend on miracles. But that is a detail that can be put on solid footing before the NASA engineers ask you if you are confident it will work 100%! At least it is working now, we hope.
My feeling is that transferring the circuit to a solder perf-board could create more problems than it solves. I won't go into why I think that unless you ask. By the way, the term "breadboard" comes from the early days of radio, I think, when an experimenter might actually used a wooden breadboard with nails sticking up to make connections.
At this stage of the game I do think that you can "harden" the white plugin breadboard. And yes, when everything is set, tack the components and the wires down with a little dab of glue. My personal favorite is GOOP adhesive, with the little dab taken on the end of a toothpick.
I'm basing my comments on the last photo that Andrew posted. http://forums.parallax.com/attachment.php?attachmentid=52620
1a) Exchange the position of the SHT and the RTC
1b) Turn the RTC 180 degrees.
That will make the signal connections to those chips straight and parallel.
Do the wiring with the wires lying flat against the breadboard, not looping high up.
2) Pull the data logger off the board and turn it 180 degrees (per PhiPi's suggestion), so that the thumb faces inward.
The connections will have to be made underneath the thumb drive while it is off the board, so thinking will be required!
NOTE: It is easy to change the pin assignments in the program, if that will simplify the wiring.
3) Think through the power supply connections, again to keep them neat, neat neat.
4) The enemy will be shock and vibration. When everything is set, think about what could shake loose and how to prevent that.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
Post Edited (Tracy Allen) : 3/9/2008 4:58:13 PM GMT
·· Thanks for your post. I was·glad to hear that you agree that soldering to a perf-board isn't our best option, at least at this point. I was strongly leaning
toward not soldering since I've never really done any perf-board soldering. But when·said that·it could create more problems than it solves, that clinched it.
As Bruce mentioned, 'broadboard' was a typo. I knew it was called a breadboard... but I never really knew why! What an education in the history of
electronics we're getting!
So you think we could use 'Goop' (or 5-minute epoxy?) to tack down the compenents and wires safely (enough)? The humidity sensor and RTC
seem pretty·snug·in the holes already; it's the wires I'm more concerned about. I'll get some Goop and experiment with it so I get an idea of its
viscosity, drying time, etc. We use a lot of 5-minute epoxy·in the rocketry programs, but I've never really used Goop a whole lot. I'll try it and then report
back.
So if we exchange the position of the SHT and the RTC,·rotate the RTC 180 degrees, and turn the Datalogger 180 degrees·to make the signal connections
straight and parallel, will we have to re-arrange the wires? Or do we simply change the pin assignments in the program (or both)? We'll be sure to·lay the
wires flat against the breadboard and not loop them, like you suggested. I'm not sure what you meant by "Think through the power supply connections"
(though I did get the point about neat, neat, neat...!) I like to see things orderly and precise, and easy to follow visually. Which specific connections are
the power supply connections?
Your point that we·cannot always·not depend on miracles to happen is·well taken.·The rocketeers and I·fully understand that·you and our friends on the
forum did most of the programming while we watched and learned. And learn we have!·I am working to put that detail on solid footing like you suggested, and
I am 100% confident the robot will work!
I cannot begin to tell you what a miracle I believe it was to find someone like you, and Phil, and Bruce to help us with the programming, and to help me mentor
these fine young students. I am not overstating the point one bit when I say, "Each of you has restored my faith in humanity." There really are kind and generous
people in the world, and you are them. Thank you, from the bottom of my heart, for all the time and effort you have generously given to (y)our project.
Together we make·one heck of a team!
Regards,
Mark and the Rocketeers
Here are some of the features of GOOP (aka Amazing GOOP) from their web site:
1. One part (no mixing)
2. Cures completely in 24-72 hours
3. Consistency:
··· 1) Thick but will run on vertical surfaces when wet
··· 2) Rubbery texture when dry
4. Dries clear
5. Max. Temperature 150 degrees F
- - - - -
I was hoping to find the actual chemical composition of GOOP, as I'm always concerned when using an "unknown" product·in what effect it might have on the surface(s) to which it will be applied. Unfortunately, none was to be had, nor was an MSDS* statement available.
Hint: When the chemical composition isn't readily apparent, one can often find an MSDS* data sheet, which MUST show the exact chemicals or substances used in the product.
*MSDS = Material Safety Data Sheet
Regards,
Bruce Bates
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Programming can't be all that difficult, it's nothing but 1's and 0's
Post Edited (Bruce Bates) : 3/10/2008 3:41:13 AM GMT
If you make those changes to the arrangement of parts, I believe the wires will come out in better, neater positions, even when you leave the pin assignments exactly as they are. At least for the SHT and the RTC. What I suggest is that you write it down on paper first. Make a diagram on grid paper similar to the picture below. (The diagram below shows a round part and two wires, but that of course is not on your board. ) The grid matches one for one with the holes on your breadboard. Make several copies, because it may take several tries and all the rocketeers may have a crack at it. Email them all a copy of the grid.
Then with a pencil draw in the SHT and RTD with where their pins are going to be plugged in., and the fun begins. Draw in the lines for the connections to the signals p0 to p15. And also to the power supplies Vss and Vdd. The game is to find the neatest arrangement. My suggestion about swapping the positions of the SHT and RTC (and rotating the RTC 180°) would I think come out better and keep the same pin assignments. But there are other ways to do it. The pin assignments are not written in stone, but keep in mind that in the end you will want all of your robots to have the same wiring (one single program version!). Now, I have not thought about how it will turn out with the thumb drive. You listened to a lot of suggestions about how to deal with the mechanical issue earlier in the thread. Once the rocketeers have settled on how to mount it so that it will be mechanically solid, they can deal with the neatness of the wiring. There is where it may prove useful to change a couple of the pin assignments.
The reason I like goop is that it has a rubbery consistency. (It is basically a rubber compound mixed in a nasty solvent -- always use with good ventilation!) It sticks to almost anything, it is viscous so it doesn't run down into the holes but can be pulled off clean if need be, and it is a good insulator. I like 5 minute epoxy too.
>So if we exchange the position of the SHT and the RTC, rotate the RTC 180 degrees, and turn the Datalogger 180 degrees to make the signal connections
>straight and parallel, will we have to re-arrange the wires? Or do we simply change the pin assignments in the program (or both)? We'll be sure to lay the
>wires flat against the breadboard and not loop them, like you suggested. I'm not sure what you meant by "Think through the power supply connections"
>(though I did get the point about neat, neat, neat...!) I like to see things orderly and precise, and easy to follow visually. Which specific connections are
>the power supply connections?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com
The battery will need to be held with zip ties (two of 'em, perpendicular to each other) onto a hard surface. Tie 'em down tightly. Alternatively you can get a plastic battery holder that will secure it along the plane of the payload sled (or whatever you're using), but you'll still need to use a zip tie to keep it from flying off of that.
The thumb drive isn't going to stay put in the socket either, or at least I wouldn't bet on it. That will need to be tied down as well somehow. Remember that even if you have a slow, smooth liftoff, when your ejection charges go off, you will get some pretty good acceleration spikes, and possibly another one when the payload section reaches the end of the shock cord. I've attached a little graph of the data from one of my flights about a year ago - that was on a K800, which gave a solid 10G spike at takeoff, but you can see that the two ejection charges were nothing to sneeze at either.
(You can see the video I made for a presentation here:· http://youtube.com/user/sylvie369· )
Post Edited (sylvie369) : 3/10/2008 6:14:29 PM GMT
·· Thanks for reminding us of the obvious (which, because we were so immersed in programming, I put on the back burner.) Of course! G-forces! (Geeeeeee...!) We'll definitely have to secure the battery, and the flash drive, extremely well because we're launching on an Aerotech K1100T-15. Last year the rocket came off the launch rail like a rifle shot. In the photo you can even see 'Mach diamonds' in the flame coming out of the nozzle... before the rocket is even off the rail! In our TARC competition rockets, which use a much smaller engine (G80-7 is the largest) we secure the PerfectFlite altimeters with zip ties, as you suggested for our larger rocket.
I handdn't thought much about the shock forces from the ejection charges/ ejection events. There are five to consider, and any one could dislodge a launch- or data-critical component. The ejection events are:
1) engine ejection delay (15 seconds) = drogue 'chute AND
2) drogue altimeter ejection at apogee = (back-up for engine delay grain)
3) back-up drogue altimeter at apogee (back-up for the back-up)
4) main ejection charge set for 1,200' ASL
5) back-up main ejection charge set for 600' ASL -
6) Four parachutes total: drogue, robot, nose cone, and main· parachute.
I'll mull this over as I drive Christopher to Boy Scouts (I also stay busy as a Scoutmaster.) Thanks for your post. Hope to hear more from a fellow HPR rocketeer!
Regards,
Mark and the Rocketeers
With all of those backups, it sounds like there's going to be a lot of action in the air. Be sure that your team is careful after landing, in case there are any live charges left due to things not firing properly. We had one go off in someone's hands after a failed flight two weeks ago. No harm or injury, but make danged sure that all of the charges have fired before you allow a crowd near the thing after landing.
Post Edited (sylvie369) : 3/10/2008 10:16:14 PM GMT