Featherweight Tilt checker/timer

the rocketry show
rocket sok
Onebadhawk
The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
cwbullet said:
Adrian,

Do you have an update on these?
Click to expand...

I got sidetracked by looking into other brands of microcontrollers when it looked for a moment like I might have an availability problem with the ones I currently use. That issue went away, so I'm finalizing the second round of prototype hardware designs now. There are several products I'm prototyping from the same panel, to save cost.
 
Adrian A said:
I got sidetracked by looking into other brands of microcontrollers when it looked for a moment like I might have an availability problem with the ones I currently use. That issue went away, so I'm finalizing the second round of prototype hardware designs now. There are several products I'm prototyping from the same panel, to save cost.
Click to expand...

AVR for the win! An Atmel 328P is a great all-around 16mhz @ 5v (8mhz @ 3.3v) chip.
 
"There are several products I'm prototyping from the same panel, to save cost."

Any hope there is a five or six channel Raven in there somewhere???
I could really use another channel or two for stage-seperation and/or air-starts...
PLEASE!
 
Sorry, no plans for that. Since you can do dual deploy and 2 airstarts or DD and a separation charge and an airstart, I think there are few flyers who would want to pay more for more channels.
 
FredA said:
"There are several products I'm prototyping from the same panel, to save cost."

Any hope there is a five or six channel Raven in there somewhere???
I could really use another channel or two for stage-seperation and/or air-starts...
PLEASE!
Click to expand...

Fred- AltusMetrum is almost done with an 8-channel expansion board for the Telemetrum.

https://www.altusmetrum.org/TelePyro/

David
 
"Since you can do dual deploy and 2 airstarts or DD and a separation charge and an airstart"

Sorta...
But not really....
Need two for the drogue.
Need two for the main chute.
Need one or two more for airstarts/sep-charges.

I won't waste money buying two channel units.
What's an extra output cost -- can't imagine a FET, a few passives and a pin on the header costs much -- has to be under a dollar.

Five is the new two.
Not everybody flies tiny rockets....
Glad somebody is listening....hello Telemetrum.
 
FredA said:
I won't waste money buying two channel units.
What's an extra output cost -- can't imagine a FET, a few passives and a pin on the header costs much -- has to be under a dollar.
.
Click to expand...

The cost is a lot more than that, because the microcontroller's pins are full on that end. Each channel gets an analog input for continuity checking along with an output driver. Not to mention that the board would have to be wider, driving down the boards that can be produced per panel, which drives up the cost, along with the extra terminal that adds cost. The FIP's deployment setting screen would get more crowded making it harder to read, and I would have to rework the deployment settings command interface, and the recorded data structure. Based on the data I've gotten back from customers, only a small minority, maybe 10%-25%, use more than two. So providing 4 is already into the territory of diminishing returns.

I know that all the above is not likely to make you any happier, but I'm taking the time to explain my rationale anyway so that users who wouldn't want the extra size, cost and complexity know I'm still looking out for them.

An expansion board (like what someone mentioned for the Telemetrum) could be done without adding much to the cost for people who don't need the extra channels, but it wouldn't be cheap. It's tough to make any separate board and sell it for under $30, particularly if the numbers are small. I think the other design work I have going on now will provide more value to my customers, so I'm focusing on that. For example, the subject of this thread would be a good way to add a safe airstart to complement the Raven's other channels.
 
Does your 3-channel unit sell for less....no, it got discontinued...
If your argument holds water, why isn't that offered for 20% less?
Nobody wants you to kill your 4-channel unit....just expand your product line.

I get the real reason is pins.
Can't change CPU's easily to get more.
Other arguments seem to be just rationalizing the lack of pins.

Was hoping your CPU would go EOL and you would end up with a more capable unit...with another pin or two.

Stealing a channel for airstarts make the airframe less safe since I now give up a redundant charge on the main and/or drogue.

Adding more CPU's (Altimeters) to the mix is plain wrong -- don't want two independant, un-cordinated decisions about burnout or other critical flight events. Want ONE computer in charge of the flight.

Sorry you fail to see the utility.
 
FredA said:
Does your 3-channel unit sell for less....no, it got discontinued...
If your argument holds water, why isn't that offered for 20% less?
Nobody wants you to kill your 4-channel unit....just expand your product line.

I get the real reason is pins.
Can't change CPU's easily to get more.
Other arguments seem to be just rationalizing the lack of pins.

Was hoping your CPU would go EOL and you would end up with a more capable unit...with another pin or two.

Stealing a channel for airstarts make the airframe less safe since I now give up a redundant charge on the main and/or drogue.

Adding more CPU's (Altimeters) to the mix is plain wrong -- don't want two independant, un-cordinated decisions about burnout or other critical flight events. Want ONE computer in charge of the flight.

Sorry you fail to see the utility.
Click to expand...

Hang on a sec. If you're looking for redundant deployment charges in order to increase the safety margin... you want two separate altimeters anyway. Using two channels on the same unit for redundancy only helps if there's a fault on one of the channels or a bad igniter. If the unit loses power or suffers some sort of general failure, you still completely screwed.
 
The other option I guess would be a rework of your board and use multiplexers / demultiplexers to get your 8 or 16 channels.

I'm not familiar with your uC so it might be a bit of a task (not to mention there will definitely be a form factor increase).

The other thing I could think of would be an expansion board, but obviously you'd have to let users reflash the Raven2 to deal with it I am guessing.
 
FredA said:
Adding more CPU's (Altimeters) to the mix is plain wrong -- don't want two independant, un-cordinated decisions about burnout or other critical flight events. Want ONE computer in charge of the flight.
Click to expand...

Besides the TeleMetrum, both the RDAS and UFC-3 can be extended with additional pyro channels. But they all come at a price.

In my book, it is not a disadvantage if the backup charges are controlled by a different altimeter - with one exception: If the backup charges are oversized, it is a little harder to ensure that the regular sized primary charge fires first. Especially when dissimilar altimeters are used, it will either require a rather long delay or lots of experience with this combination.

Reinhard
 
If the backup charges are oversized, it is a little harder to ensure that the regular sized primary charge fires first. Especially when dissimilar altimeters are used, it will either require a rather long delay or lots of experience with this combination.

Bingo...
With two altimeters you have NO CLUE, NO CONTROL over what happens timing-wise between the two.
I can't make oversized backup charges...and make sure the little fires before the larger.
I can't make sure they don't fire at the same time.
I can't even make sure they will play nice together.

Maybe I'm foolish to TRUST MY ELECTRONICS....but then again I'm an EE.
I figure if you don't abuse them...they should be the most reliable part of the rocket.
So -- ONE and ONLY ONE altimeter for me...
Just need one that can do the work required.... 5 charges.

No interest in the RDAS....
Don't really need 8 channels of expansion....don't want expansion connectors that can fail.
Only want one more channel.
 
If you're an EE then why don't you just design your own altimeter, solder it up, and make it do whatever you want? Cheaper, better control, etc. ad nauseum.

Just my two cents.
 
FredA said:
Maybe I'm foolish to TRUST MY ELECTRONICS....but then again I'm an EE.
I figure if you don't abuse them...they should be the most reliable part of the rocket.
Click to expand...

I too am a EE. I work for a company that makes computers that fly airplanes. EVERYTHING that flies the plane is redundant. Dual, triple, and even quad redundant

Options would be multiple e-matches in the same charge - if one altimeter/match fails, another can still try to fire the charge.

Or the Raven has options for a channel to have a delay from a set point, such as fire 1 second after apogee. So you could have one altimeter set to fire at apogee and the Raven fire a second later for the larger charge. For the main, use different altitudes

But "stuff" happens, and I would want some redundancy. I believe the issues on timing can be worked around

But again, just my $0.02 worth. And normally it is only a penny for my thoughts!
 
If you're an EE then why don't you just design your own altimeter, solder it up, and make it do whatever you want? Cheaper, better control, etc. ad nauseum.

We know EE's are the best kind of engineers....and can do everything well...
But it's just a matter of time.
MUCH more efficient if I can convice a manufacturer to make a small change then roll the whole enchilada.

Priorities...just like Adrian has "other" product priorities.
 
jms said:
If you're an EE then why don't you just design your own altimeter, solder it up, and make it do whatever you want? Cheaper, better control, etc. ad nauseum.
Click to expand...

Perhaps a simple timer device could be connected to a Raven output. It would fire a charge then wait a second to fire a backup charge. That would allow one output/event to be used for two of the events.

-- Roger
 
Another device adds complexity and also prevents the Raven from detecting continuity on the charges.

That said, our big projects use a Raven with its outputs tied to a firing board, which triggers MOSFETs to dump a 7.2V RC pack into the explosive bolts
 
Maybe to steer this thread back to where it's supposed to be:

Adrian, how is the tiltometer coming along? Any pictures or anything of the protoboard you'd like to share?
 
FredA said:
If the backup charges are oversized, it is a little harder to ensure that the regular sized primary charge fires first. Especially when dissimilar altimeters are used, it will either require a rather long delay or lots of experience with this combination.

Bingo...
With two altimeters you have NO CLUE, NO CONTROL over what happens timing-wise between the two.
I can't make oversized backup charges...and make sure the little fires before the larger.
I can't make sure they don't fire at the same time.
I can't even make sure they will play nice together.

Maybe I'm foolish to TRUST MY ELECTRONICS....but then again I'm an EE.
I figure if you don't abuse them...they should be the most reliable part of the rocket.
So -- ONE and ONLY ONE altimeter for me...
Just need one that can do the work required.... 5 charges.
Click to expand...

first and formost im just asking questions to try and understand your thought process.

so do you not agree with the TRA L3 cert requirement of independent redundancy? it is my understanding you have to use seperate altimeters to protect against a failure that could effect all deployment events. im asking becouse i have started to put dual units in my larger L2 vehicles.

do you not have a degree of control with your user inputs. backups are set to apogee plus x seconds or feet agl.

use my L3 cert flights for example: ARTS2 set at apogee / 1000ft
raven2 set at apogee+2 sec / 700ft.

this works great. chutes deploy backups go off after.
(just not enough airflow to inflate 15ft round)
 
Last edited:
jms said:
Maybe to steer this thread back to where it's supposed to be:

Adrian, how is the tiltometer coming along? Any pictures or anything of the protoboard you'd like to share?
Click to expand...

i agree hope your getting these tested soon i have a couple stage flights comming up and my Ravens need a companion for the 25 - 30k agl ride at airfest :dark:
 
"not agree with the TRA L3 cert requirement of independent redundancy?"

You got that right...
There are many L3 cert rules that are, IMHO, for L3 certs only...
But that's a subject for another thread...

Enough for now...Adrian's mind is set...I'll move on...
I look forward to his Tilt Checker...
 
FredA said:

Enough for now...Adrian's mind is set...I'll move on...
I look forward to his Tilt Checker...
Click to expand...


jms said:
Maybe to steer this thread back to where it's supposed to be:

Adrian, how is the tiltometer coming along? Any pictures or anything of the protoboard you'd like to share?
Click to expand...

It's coming slowly. On Saturday I tried and failed to talk to the new accelerometer chip on the tilt checker prototype for reasons that are still unknown, but I suspect are assembly/soldering errors. 0.5mm pitch leadless parts are a challenge for me for hand-prototyping. I have a nice new logic analyzer that verified that the microcontroller is outputting the right signals to the part, and I triple-checked the part datasheet, so I'm stumped as to what the real problem was. I have updated the designs to improve my chances of soldering success, and I'll take another crack at it with the new board, along with the gyro chip that I wasn't able to test in this version because of a different bonehead board layout mistake.

Also on this prototype panel are The Sparrow, which is a new, simple and inexpensive altimeter that's designed to set a new standard for ease of installation and use, a new product for gyro-stabilizing high-speed sailplanes doing dynamic soaring, and a new product I'm excited about that was conceived in a why-didn't-I-think-of-this-earlier moment on Saturday. I really want to get that one done in time to test at Mile High Mayhem. I also want to add a board design for the Hummingbird to this panel of prototypes, but I haven't started that design yet so I'm not sure I want to delay the rest of the prototypes for it. In other news, a new production run of the Power Perches should be arriving from surface mount assembly house today, just in time since I ran out of stock yesterday. The Raven3 production run should be coming back from surface mount assembly this week as well, but that will take a bit longer to get them programmed and tested.
 
Yeah 0.5 pitch is a nightmare, I don't have the steady hands necessary for it. PM me the accelerometer if you like, I just worked on a project using an accelerometer that I had to fight with, we may have had the same fight on our hands.
 
Finally I'm about to submit my next panel of prototypes:

newboards.jpg


The bunny has no less than 3 magnetic switches, BTW.
 
I would buy the bunny. Not joking either. Just watch where you put the USB port :y:
 
You choose the bunny when you need to multiply something, and use the giraffe for a longer delay.
 
...fine, I'll ask it.

What are they (the bunny and giraffe)?!!?
 
I have a question and I want to follow this thread.

Do your interface programs only run on PC?

How big are they?

I have a Mac Pro so I have partition running Windows for other stuff and I would like to fit it on the partition if it is not to big.

Scott
 

Similar threads

Adrian A
Featherweight Blue Raven Development Thread: Recorded time-series data
2
Replies
33
Views
3K
Earache
E
I
Tilt Angle Comparison from Flight Data
Replies
0
Views
608
ihbarddx
I
Adrian A
New Featherweight GPS phone app available for Android and iOS
2 3
Replies
73
Views
6K
jgarbus
J
Adrian A
Blue Raven development thread: Flight simulation
Replies
1
Views
1K
Adrian A
Adrian A
Adrian A
Featherweight Blue Raven Development Thread: Deployment logic
6 7 8
Replies
228
Views
25K
Cooperr
Cooperr

Latest posts

Back
Top