[sugpro] Re: Verifying Motor Performance Through Flight Tests
- From: Ben Brockert <wikkit@xxxxxxxxx>
- To: sugpro@xxxxxxxxxxxxx
- Date: Wed, 18 Feb 2015 18:00:31 -0500
I work with Adrian at our day jobs, so let me know if you have any
specific questions for him.
On Wed, Feb 18, 2015 at 2:56 PM, Steve Peterson
<steve_peterson@xxxxxxxxxxxxx> wrote:
> Thanks, Richard; I had a further thought on the subject regarding Cd.
>
> You can use "coast down" information gleaned from the flight data to
> calculate CdA. Adrian (IIRC) did this with fairly good success. Basically
> the data from burnout to apogee (or thereabouts) can be used as I described
> for the thrust data to determine net force on the rocket, except the
> coasting data is the result of only gravitational and aerodynamic forces.
> Weighing the rocket post-flight will allow you to easily deduct the (now
> constant) gravitational force (assuming no parts are lost :-)) leaving only
> the aerodynamic force.
>
> That force will be the result of the Cd and the cross-sectional area, both
> of which are not easy to measure, although various programs can estimate it,
> as you point out. That leaves only the air density to be estimated and, of
> course, flight angle (and angle of attack) to be judged/assumed--although
> one might argue that with enough test flights (and not-too-dispersed data)
> the aerodynamic force would include the effects of "typical" flight angle
> and angle of attack.
>
> Seems like it's a good excuse for repeated flights :-)
>
> --Steve
>
>
> On 02/18/2015 11:35 AM, Richard Nakka wrote:
>>
>> You've described the process and challenges nicely, Steve.
>>
>> It's been a while since I last harvested motor performance from flight
>> computer data, but it can be done with, I expect, reasonable
>> "engineering" accuracy. Timely discussion, as I plan to do this soon
>> with my latest batch of flight test data.
>>
>> I recommend using the barometric, rather than accelerometer data,
>> integrations (and all the other calcs) can be done relatively easily
>> using a spreadsheet software. Adrian seems to have greater faith in
>> the barometric data compared to accelerometer ( I fly the Raven3).
>>
>> Propellant mass as a function of time can be extracted from SRM.xls.
>>
>> With regard to drag coefficient (Cd), I use AeroLab to obtain this.
>> AeroLab gives Cd as a function of mach number, which can be handy for
>> higher velocity flights.
>>
>> Richard
>>
>>
>>
>>
>>
>> On Sun, Feb 15, 2015 at 6:11 PM, Steve Peterson
>> <steve_peterson@xxxxxxxxxxxxx> wrote:
>>>
>>> Michael,
>>>
>>> The basics: if you have position (altitude) with respect to time, then
>>> the
>>> change in position over time is the velocity. The change in velocity with
>>> respect to time is acceleration. If you have the mass (at the same moment
>>> in
>>> time that you've calculated the acceleration for, then rearrange Mr.
>>> Newton's formula (F=ma) to get the net force. Any decent altimeter will
>>> give
>>> you altitude (to some precision/accuracy) vs. elapsed time (to some
>>> precision/accuracy). After that is when the gremlins get you....
>>>
>>> Altitude: change in altitude may not represent a true change in position
>>> (that is, distance) because the rocket may be headed off at an angle. You
>>> will have to either assume a certain angle of flight and calculate the
>>> true
>>> distance, or assume that it flew vertically (in which case the change in
>>> altitude is the change in distance).
>>>
>>> Mass: it isn't constant, so you'll have to calculate it based on the
>>> grain
>>> geometry and your static tests, etc. I don't know if any of Richard's
>>> spreadsheets list mass burned vs. time, but if they do, that would give
>>> you
>>> a good start, assuming your manufacturing is under tight enough control.
>>>
>>> OK, so you've calculated F--but hang on, because that's *net* F. That is,
>>> thrust minus the force of gravity and minus the force of drag. The force
>>> due
>>> to gravity is just g*mass and we've already dealt with mass.
>>> However, the force due to drag is more problematic. As you know, it
>>> consists
>>> of the Cd of the rocket (which will vary with velocity), the angle of
>>> attack, atmospheric conditions (launch pad altitude, altitude of the
>>> rocket
>>> at any instant in time, temperature, barometric pressure at launch, etc.)
>>> and, of course, the square of the velocity.
>>>
>>> Your question then becomes, will you know all that stuff with sufficient
>>> accuracy to give you a meaningful result? And will your altitude be
>>> accurate/precise enough to allow you to do all the math on it to get the
>>> acceleration with any kind of accuracy/precision?
>>>
>>> From what I recall, the Featherweight altimeters are about the most
>>> accurate/precise out there (although I would also check with the altus
>>> metrum guys because I've read that their stuff is pretty darned good,
>>> too).
>>> Both will record fast enough to get you data with short enough time
>>> intervals. I know Adrian Adamson (Featherweight) has done a lot of study
>>> on
>>> this--you might check the Featherweight forum and also over on TRF.
>>>
>>> I should also mention that the Featherweight altimeters (or at least the
>>> Raven), and possibly the altus metrum products, will also provide
>>> acceleration data so that you don't have to do the double differentiation
>>> to
>>> calculate acceleration from altitude. I haven't looked into how
>>> accurate/precise it is, however. But you still have to know the
>>> atmospheric
>>> info and the aerodynamics of your rocket--and those two are usually the
>>> killers.
>>>
>>> A lot of people have looked into doing this and, as I recall, very few
>>> have
>>> managed to come up with anything that was very persuasive (and they were
>>> using commercial motors), although I am certainly no expert on this
>>> stuff.
>>> It's pretty easy (especially if you can program) to simulate a few data
>>> points and do the calcs to see what you come up with. Munge the altitude
>>> data a bit to simulate inaccuracies and see how much it throws off your
>>> answer. Vary the Cd by .1, .2, .3 or so and see what happens. Etc etc.
>>> You'll soon get a feel for just how hard this is.
>>>
>>> --Steve
>>>
>>> On 02/15/2015 10:17 AM, Michael Monteith (Redacted sender
>>> michael_r_monteith@xxxxxxxxx for DMARC) wrote:
>>>>
>>>> I hope this isn't off topic as it has to do with verifying motor
>>>> performance really. I was thinking on what I would need to verify
>>>> rocket
>>>> motor performance during a flight test. So I was thinking of what would
>>>> be
>>>> the requirements to gather the data in flight. There is so many
>>>> altimeters
>>>> and ranges of price. Some show they output thrust time. But not sure
>>>> exactly if it's what I'm thinking it is or I'd be better off getting one
>>>> cheaper and calculating it.
>>>> http://data.rocketsetc.com/altimeter_data.html
>>>>
>>>> So to my question. What data is required and how fast? I see all
>>>> the
>>>> thrust curves for static testing but trying to figure out how you
>>>> backtrack
>>>> and figure from a flight test what the thrust curve is for comparison?
>>>> This
>>>> is what I want to arrive at, a thrust curve for flight test vs thrust
>>>> curve
>>>> on static testing.
>>>>
>>>> My initial guess is at least having time and altitude and having rocket
>>>> mass etc. From there you can calculate acceleration etc and arrive at
>>>> thrust. I don't want to think of the formula right now for this. It
>>>> might
>>>> be in my pile of books but those are boxed up in Missouri and won't see
>>>> them
>>>> for about a month now. But don't recall anything like that.
>>>>
>>>> I figured I might as well buy the right recording altimeter to begin
>>>> with. I don't mind spending the money but only if I do it preferably
>>>> once
>>>> and right. Specifically the right data, accurate, and the right speed.
>>>> I
>>>> think the more time I spent on it the more confused I was with all the
>>>> options on them all. At least until I know the bare minimum. I don't
>>>> know
>>>> if anyone has gone down this road or not. I saw Richard made mention on
>>>> one
>>>> of his pages that it was something for a future page.
>>>>
>>>> If we need to take it offline feel free to email me.
>>>>
>>>> Michael Monteith
>>>>
>>>>
>>>
>>
>
>
Other related posts:
