New Flight Computer - would you be interested?

[Stomper]

Sometimes unsolicited advise is not welcome, but this driver switch is a little under sized. 1.5Amps can easily be achieved with many batteries and matches/starters and if driver(s) get over currented and shorts (which is what it will do); The next time power is turned on, the result may be undesirable.

That is not enough.

I updated the drivers to 6 A N-Channel FETs ...

The Hardware is nearly ready for production. I think I will go to production on Monday.

Anyone who want to join the team for review and/or coding the modules?
I would offer the hardware at cost price for all the coders and helpers for the first version.

PM me if you are interested.

 

[BLH]

View attachment 373403
Search for VTS in TRF or on the Australian Rocketry Forum and you will find my Vertical Trajectory System. Here is the build thread: https://forum.ausrocketry.com/viewtopic.php?f=56&t=5324

The electronics do a frame-by-frame conversion of pulse width to voltage and I feed that into unused deployment channels on the telemetry for transmission and logging, all synchronous with the rest of the flight data.

If this worked.?.? Apogee ejection speed could be near zero, Correct? That would be cool and easy on rocket parts and chutes and such!

 

[OverTheTop]

If this worked.?.? Apogee ejection speed could be near zero, Correct?

Correct. Straight up would give almost zero velocity at apogee each time. Probably need to start thinking seriously about the possibility of backslides at apogee..

It works, sort of. It stabilises to a vector quite nicely. Unfortunately the vector is not straight up! It is about 30deg off vertical, and due to the auto-tune on the internal model of the inertial system. That's what I get for trying to use COTS parts without tweaking firmware. The next incarnation is likely to be based on a CleanFlight platform and will need the firmware "adjusted" to ensure the correct nominal path.

My biggest take-away from the flights so far is how little system gain is actually needed. I will be printing some more fins with smaller area to get the gain down further. Ideally what I fly it on should have strakes to get the airflow over the rear fins a bit more predictable too.

 

[BLH]

My biggest take-away from the flights so far is how little system gain is actually needed. I will be printing some more fins with smaller area to get the gain down further. Ideally what I fly it on should have strakes to get the airflow over the rear fins a bit more predictable too.

This is what I was kinda mentioning earlier... I think as speed goes up, system(rocket) gain will also go up and require control gain to go down. I know when I used to race RC planes, I would take off and land plane with full servo travel and then once the plane got going, my RC unit had a switch to reduce servo travel(gain) by 1/2. Otherwise at high speed, control and smooth flight was not for the ill at heart and real touchy!

 

[Stomper]

If this worked.?.? Apogee ejection speed could be near zero, Correct? That would be cool and easy on rocket parts and chutes and such!

Correct. Straight up would give almost zero velocity at apogee each time. Probably need to start thinking seriously about the possibility of backslides at apogee..

It works, sort of. It stabilises to a vector quite nicely. Unfortunately the vector is not straight up! It is about 30deg off vertical, and due to the auto-tune on the internal model of the inertial system. That's what I get for trying to use COTS parts without tweaking firmware. The next incarnation is likely to be based on a CleanFlight platform and will need the firmware "adjusted" to ensure the correct nominal path.

My biggest take-away from the flights so far is how little system gain is actually needed. I will be printing some more fins with smaller area to get the gain down further. Ideally what I fly it on should have strakes to get the airflow over the rear fins a bit more predictable too.

What does this have to do with my RocketFlightComputer?!

 

[BLH]

You should add some other hooks, inputs and features that allow users to do stuff like this. That would separate your flight computer from all the rest making it more marketable...

 

[OverTheTop]

What does this have to do with my RocketFlightComputer?!

Apologies for any perceived thread hijacking. Somebody asked me a question.

 

[BLH]

What does this have to do with my RocketFlightComputer?!

Yes, I ask the question and apologize also. But it sounds like your making an open source-ish flight computer and to be honest, you have allot to learn and a long ways to go.... First you need to get the hardware reliable, and then come up with Mach filters, launch detection and things like that that aren't so simple that others have already solved. I encourage you to proceed but there are allot of flight computers that have been fully tested and have a history of "working". Just competing in that market, while it will be a learning experience, will probably never cover your development cost. People tend to go with what they perceive works, which may" be correct or not" and there are some really good flight computers out there that have proven themselves over many years and are used widely, hence people using them are not going to switch for no real reason. Even if it cost a few $10's of dollars less, how much is my time and rocket worth to try something unproven and new with no real benefit or advantages?

A wise man once told me if you want something useful, "Find a need and fill it". I think if you had something different from all the others, that might be a need. OverTheTop is working on such a thing as is Joe Banard and his vectored thrust stuff.... No current flight computers will run these along with other flight stuff. Dream BIG and look ahead. Building and proving another "standard flight computer" doesn't seem to fit the need and will take you many years before the market excepts it. And allot more money.

If you had, for example, a real cool operating system and/or open source stuff (Although Bdales flight computer has that) going above and beyond what others do, you might have something. IE, Have all the drivers and hardware for digital IO, analog IO, events, altitudes and velocities and such along with advanced stuff that no else does like 9-axis IMU's, you would fill the need for advanced flight computers. As rocketry advances there will be a need for more advanced flight computers... Why not be the first to start something like that and then once people see you can make an advanced computer work really cool, you can come out with a basic simple one and everyone will go with it because they know you know what the heck you are doing and "trust" you designs... Thoughts to ponder.

Best of luck always.

 

[Stomper]

Yes, I ask the question and apologize also. But it sounds like your making an open source-ish flight computer and to be honest, you have allot to learn and a long ways to go.... First you need to get the hardware reliable, and then come up with Mach filters, launch detection and things like that that aren't so simple that others have already solved. I encourage you to proceed but there are allot of flight computers that have been fully tested and have a history of "working". Just competing in that market, while it will be a learning experience, will probably never cover your development cost. People tend to go with what they perceive works, which may" be correct or not" and there are some really good flight computers out there that have proven themselves over many years and are used widely, hence people using them are not going to switch for no real reason. Even if it cost a few $10's of dollars less, how much is my time and rocket worth to try something unproven and new with no real benefit or advantages?

A wise man once told me if you want something useful, "Find a need and fill it". I think if you had something different from all the others, that might be a need. OverTheTop is working on such a thing as is Joe Banard and his vectored thrust stuff.... No current flight computers will run these along with other flight stuff. Dream BIG and look ahead. Building and proving another "standard flight computer" doesn't seem to fit the need and will take you many years before the market excepts it. And allot more money.

If you had, for example, a real cool operating system and/or open source stuff (Although Bdales flight computer has that) going above and beyond what others do, you might have something. IE, Have all the drivers and hardware for digital IO, analog IO, events, altitudes and velocities and such along with advanced stuff that no else does like 9-axis IMU's, you would fill the need for advanced flight computers. As rocketry advances there will be a need for more advanced flight computers... Why not be the first to start something like that and then once people see you can make an advanced computer work really cool, you can come out with a basic simple one and everyone will go with it because they know you know what the heck you are doing and "trust" you designs... Thoughts to ponder.

Best of luck always.

Hello BLH,

sorry if this sounds a little bit rude... but I learned to start with the basics and that when you learned the basics, proceed to do the advanced stuff...

In therms of rocketry... yes there is much to learn for me.

I opened this thread to ask you experienced guys if there is a market or if someone could be interested in this and if there is... what should be the features of a new product.
I tried to implement most of the stuff people told me.

I don´t really care if there is a huge marked for my flight computer... This is just hobby...I want to learn from this hobby stuff... because my other company is doing something similar and I can use the gained experience in the projects there. I don't need and don't want to make profit with this one.

My offer is... join my project if you need this kind of flight computer I offer here... or tell me what I should change and get a cheap flight computer for your own rocket project and make it work like you want it.

What are the flight computers out there ? I only found some lets say aged projects. Could you provide any links or names?

cheers
Stomper

 

[BLH]

Well, I'm sure this is not an exhaustive list, but manufactures that make either a computer or good altimeter now would be Altus Metrum, Perfect Flight, Egg Finder, Marsa, Missleworks,... Some of these are just dual deployment altimeters but any flight computer would have to perform that function really well to start with.

I'm not sure people want a "CHEAP' flight computer. Maybe better to call it "Low Cost". One must remember that the rockets that warrant a computer are going to be higher end and costlier air frames and HP guy's buy what works and has proven to work. I know most of the manufactures listed above spent several years getting their devices up and running and reliable!

At the least, you are probably going to have to spend allot more $$$ for development, burn up allot of rocket fuel and complete the processes of crashing numerous rockets before you get up to speed.

I'm sure you could build a great one but if your going to put in all that time, money and effort, maybe buy one of each kind and test them out to see what they can and can't do. Find advantages and disadvantages of each and come up with a "modern" design. I think you may find that the computers themselves are not the poopy part, but the user interfaces, which are typically cell phone based are what needs more work....

This is why building altimeter/computer boards are like building drills... Most people don't want to build one because of the cost and difficulty in the details and user interfaces, they just want the holes.

We could always use new and better stuff and German products are historically high quality, and higher cost and work really really well. That's way they cost more. I am personally not interested in some unproven "Cheap Chinese" flight computer in my nice rocket.

I think if you do the market research you will come up with something great!

Good luck always.

 

[Stomper]

Thanks...

I think my rocket computer is a all in one solution for an entry level rocket engineer.
I know that it won´t cover all the needs of a high performance rocket.

And of course... there is something on the market that do things better.

I try to cover the most important functions... log your data, ignite, deploy, recover.

All in one and reliably.

 

[BLH]

I'm sure your computer will be really neat and once you get it going and proven, who knows a "fine low cost German design" might become the quality standard! Then you'll sell a buck load.

 

[OverTheTop]

Hi Stomper. When coding up your flight computer I suggest you do a code read through of any other FC programs. Not to steal their code, but to find out what you don't know. There are some peculiar things that happen during flights, like fluctuating pressures around Mach and other oddities. Other non-nominal flight conditions, say a chuff on the pad, also need to be allowed for. Other FCs have been through development and found out these things the hard way. You will have a better starting point if you can leverage their history rather than recreating their learning curve.

I don't think many people get into rocketry for profit, but learning is a good substitute if you have something else putting bread on the table. Enjoy the project and its challenges .

 

[Stomper]

Thanks for your hint... I was thinking about the problem of breaking the sonic barrier... but couldn´t figure out with my brain, what would happen. To have a look at the algorithm of other projects is a good idea.

I wanted to do a simulation on such things with a CFD Simulation in the future.

I think the first software will not cover flights above mach 1

Best regards
Stomper