RockSIM 6.x bugs and work-arounds

RockSIM 6.x bugs and work-arounds

Post by Jerry Irvin » Sat, 31 May 2003 03:56:34





Quote:




> >>>I noticed a funny problem ? (or feautre I don't understand)

> >>>I will fly a loc-iv on a g40-4. it will say optimum delay is 4.65
> >>>seconds.

> >>>I will then fly it on a g-40-10. it will say optimum delay is 7.65
> >>>seconds.

> >>>this anomily is the same for jsut about everthing I fly, a short
> >>>delay, a short optimum time. A long delay  , a longer optimum time.

> >>>Anyone know what I'm doing wrong ?

> >>I think that, if you select a specific delay, it *stops* simulating
> >>when the delay finishes.  So, if you select a delay that is shorter
> >>than the coast-time-to-apogee, the rocket is still moving fast, but
> >>slowing down, at the end of the simulation.

> >>"Best" delay time seems to be picked by looking for the time when the
> >>rocket is moving the slowest.  For a sim that stops before apogee,
> >>this will always be right at the end of the delay.

> >>I haven't looked, but I'll bet you find that the G40-4 has a delay
> >>time of 4.65 seconds in the .eng file you are using.

> >>It's simply picking the "best" delay time *out of the values it
> >>calculated*.

> >> - Rick "But, yes, I'd still consider it a bug"***inson

> > Again, I don't know nuttin 'bout RockSim, but a LOC-IV will reach apogee
> > LONG before 7.65 seconds. The 4.65 number sounds close: I calculated about
> > 4.4, but it depends on what you pick for the mass and CD which the original
> > poster didn't specify.

> RockSim calculates a pretty low Cd for a LOC-IV, around 0.29. wRASP uses
> 0.60 by default. The real Cd is probably somewhere in the middle,
> depending upon building techniques, finish, etc.

Real CD is probably at 0.60.  RocSim gets kudos for that default value.  
GHS suggested 0.85 and he would be right more of the time if the
alphabet still stopped at F.

Jerry

--
Jerry Irvine, Box 1242, Claremont, California 91711 USA

Please bring common sense back to rocketry administration.
Produce then publish.  http://www.FoundCollection.com/

 
 
 

RockSIM 6.x bugs and work-arounds

Post by Jerry Irvin » Sat, 31 May 2003 04:25:02




Quote:
> True of ANY simulation tool. If you don't know about what result to expect,
> you don't know when it's telling you garbage.

> GIGO!

This applies to so many aspects of rocketry, administration included.

Jerry

--
Jerry Irvine, Box 1242, Claremont, California 91711 USA

Please bring common sense back to rocketry administration.
Produce then publish.  http://www.usrockets.com

 
 
 

RockSIM 6.x bugs and work-arounds

Post by Eric Ellingto » Sat, 31 May 2003 05:08:14


Quote:





>>>> I noticed a funny problem ? (or feautre I don't understand)

>>>> I will fly a loc-iv on a g40-4. it will say optimum delay is 4.65
>>>> seconds.

>>>> I will then fly it on a g-40-10. it will say optimum delay is 7.65
>>>> seconds.

>>>> this anomily is the same for jsut about everthing I fly, a short
>>>> delay, a short optimum time. A long delay  , a longer optimum time.

>>>> Anyone know what I'm doing wrong ?

>>> I think that, if you select a specific delay, it *stops* simulating
>>> when the delay finishes.  So, if you select a delay that is shorter
>>> than the coast-time-to-apogee, the rocket is still moving fast, but
>>> slowing down, at the end of the simulation.

>>> "Best" delay time seems to be picked by looking for the time when the
>>> rocket is moving the slowest.  For a sim that stops before apogee,
>>> this will always be right at the end of the delay.

>>> I haven't looked, but I'll bet you find that the G40-4 has a delay
>>> time of 4.65 seconds in the .eng file you are using.

>>> It's simply picking the "best" delay time *out of the values it
>>> calculated*.

>>> - Rick "But, yes, I'd still consider it a bug"***inson

>> Again, I don't know nuttin 'bout RockSim, but a LOC-IV will reach apogee
>> LONG before 7.65 seconds. The 4.65 number sounds close: I calculated
>> about
>> 4.4, but it depends on what you pick for the mass and CD which the
>> original
>> poster didn't specify.

> RockSim calculates a pretty low Cd for a LOC-IV, around 0.29. wRASP uses
> 0.60 by default. The real Cd is probably somewhere in the middle,
> depending upon building techniques, finish, etc.

I should add that I was using the demo version of RockSim 5.x when I was
playing around with the LOC-IV, and the calculated Cd isn't actually
displayed in the demo version. To determine the Cd value RockSim
calculated, I first ran a test flight, specifying that the Cd should be
calculated in the design screen. Then I plugged different values in for
static Cd (again, on the design screen) until the manually entered
flight matched the calculated flight.

I'd really love to try Apogee's AeroCFD to see what Cd it comes up with
for the LOC-IV. I'd also like to do a few flights with a recording
altimeter to derive a Cd. Alas, neither is in my budget this month.

 
 
 

RockSIM 6.x bugs and work-arounds

Post by Bob Kapl » Sat, 31 May 2003 05:12:42



Quote:
> Real CD is probably at 0.60.  RocSim gets kudos for that default value.  
> GHS suggested 0.85 and he would be right more of the time if the
> alphabet still stopped at F.

Actually GHS said 0.75. Not bad for 3FNC with an A8-3.

I HAVE backtracked some of my models to CDA under 0.30: egglofters with
massive boatails such as I flew in ELD at NARAM-41 (still up there!) and
NARAM-42 (US record).

        Bob Kaplow      NAR # 18L       TRA # "Impeach the TRA BoD"
                >>> To reply, remove the TRABoD! <<<
Kaplow Klips & Baffle:      http://www.pleimling.org/le/Phantom4000.pdf
    www.encompasserve.org/~kaplow_r/    www.nira-rocketry.org    www.nar.org

 Save Model Rocketry from the HSA!   http://www.space-rockets.com/congress.html

 
 
 

RockSIM 6.x bugs and work-arounds

Post by Mark Johnso » Sat, 31 May 2003 05:50:12



Quote:


> > RockSim calculates a pretty low Cd for a LOC-IV, around 0.29. wRASP uses
> > 0.60 by default. The real Cd is probably somewhere in the middle,
> > depending upon building techniques, finish, etc.

> That's just absurd for this rocket! I'd go with something in the 0.6-0.8
> range depending on construction and finish.

Yea, verily. My experience with backtracked Cd calculations indicates
that for most of us builders (Trip Barber and some Internats guys
excepted) any drag coefficient calculation below 0.5 is severely
suspect. [My own personal best backtracked value came on an
F Altitude rocket about 15 years ago...the Cd came out to 0.57]
 
 
 

RockSIM 6.x bugs and work-arounds

Post by Bob Kapl » Sat, 31 May 2003 08:14:34


Quote:

> Yea, verily. My experience with backtracked Cd calculations indicates
> that for most of us builders (Trip Barber and some Internats guys
> excepted) any drag coefficient calculation below 0.5 is severely
> suspect. [My own personal best backtracked value came on an
> F Altitude rocket about 15 years ago...the Cd came out to 0.57]

Quite true for 3FNC like SD, Altitude, etc. VERY not true for boat-tailed
egglofters like I've been flying for the past couple decades.

        Bob Kaplow      NAR # 18L       TRA # "Impeach the TRA BoD"
                >>> To reply, remove the TRABoD! <<<
Kaplow Klips & Baffle:      http://www.pleimling.org/le/Phantom4000.pdf
    www.encompasserve.org/~kaplow_r/    www.nira-rocketry.org    www.nar.org

 Save Model Rocketry from the HSA!   http://www.space-rockets.com/congress.html

 
 
 

RockSIM 6.x bugs and work-arounds

Post by John DeMa » Sat, 31 May 2003 08:49:13


Quote:

> "Bob Kaplow" wrote
> > Eric Ellington writes:
> > > RockSim calculates a pretty low Cd for a LOC-IV, around 0.29. wRASP uses
> > > 0.60 by default. The real Cd is probably somewhere in the middle,
> > > depending upon building techniques, finish, etc.

> > That's just absurd for this rocket! I'd go with something in the 0.6-0.8
> > range depending on construction and finish.

> Yea, verily. My experience with backtracked Cd calculations indicates
> that for most of us builders (Trip Barber and some Internats guys
> excepted) any drag coefficient calculation below 0.5 is severely
> suspect. [My own personal best backtracked value came on an
> F Altitude rocket about 15 years ago...the Cd came out to 0.57]

  I'll chime in with another skeptical eye.  Something like an average
build of a LOC-IV is probably around 0.7 Cd.  Keep in mind also that the
larger (longer) rockets are at a higher Reynolds #, so they typically have
less drag than smaller modrocs.  Another affect is that the small mass of
a smaller rocket will give it a higher natural frequency, causing pitch
excursions (dynamics in angle-of-attack)... hence higher Cd than expected.

  For some drag numbers I ran in a windtunnel on small rockets, see my
NARAM R&D report from 1995:
   http://web.syr.edu/~jsdemar/rocketdrag.html

  Now that I have access to a windtunnel with a 6x3x3ft test section,
I'm thinking about comparing larger rockets.

  -John

 
 
 

RockSIM 6.x bugs and work-arounds

Post by John DeMa » Sat, 31 May 2003 08:55:24


Quote:

> "Larry W. Hardin" writes:
> > Strange... I ran the LOC-4 (don't know where I picked up the
> > file) and got an optimum delay of 6.10 seconds when I ran both
> > G40-7 and G40-10.  There was no G40-4 in the list.

> Even 6.something sounds long to me. More like 4.something...

   I just flew a LOC-4 today on a G80-4.  It was just about perfect, slightly
into the wind.  A demo for a high school AP Physics class.  It wrapped itself
around a lighting tower on the ballfield, about 30ft (half-way) up.  Recovered
by one of the teachers.

 -John

 
 
 

RockSIM 6.x bugs and work-arounds

Post by EMR » Sun, 01 Jun 2003 04:46:11


I think many are making the assumption that the LOC-IV that was simmed
matches what in reality the rocket it built to.  In other words, what
I find often with RockSIM is that I build the model in the SIM and I
always have to add a MASS to adjust for Weight and CG.  I do this by
measuring my fully finished rocket with recovery system installed.

What I have found is that the amount of epoxy I use for fin fillets,
the filler for spirals, the number of coats of primer and paint,
always makes the rocket heavier than a straight build SIM model.

So, if this fellow just used a straight SIM model the longer delays
may be necessary.  Or even perhaps if he built is per the instructions
but did not use a lot of epoxy or filler or paint...

... perhaps those with field experience indicating that the 4-5 second
delay is perfect have that experience on a fully (tough) built model
with full finish.

I have had great success with adding the MASS to finalize my weight
and CG.  Also, don't forget to set the launch conditions up to where
you are launching it.  The default give very different results than
when set up to the altitude, humidity, temperature, wind and
barametric pressure.  (Note the latitude doesn't do anything currently
from my understanding).

Just food for additional thought.

Regards,
Nick

 
 
 

RockSIM 6.x bugs and work-arounds

Post by Leonard Fehsken » Sun, 01 Jun 2003 05:26:40




Quote:
> I think many are making the assumption that the LOC-IV that was simmed
> matches what in reality the rocket it built to.  In other words, what
> I find often with RockSIM is that I build the model in the SIM and I
> always have to add a MASS to adjust for Weight and CG.  I do this by
> measuring my fully finished rocket with recovery system installed.

There's a mass and CG override feature that's intended specifically for
this purpose.  You measure the mass and CG location of the model as built,
enter these values, and you don't have to play around with a mass object.
You can turn the feature on and off with a single checkbox.

len.

 
 
 

RockSIM 6.x bugs and work-arounds

Post by AlMax7 » Sun, 01 Jun 2003 06:56:57



Quote:
>>There was no G40-4 in the list.

no, you can type over the list to type in the exact dely you have.
Quote:
> Strange... I ran the LOC-4 (don't know where I picked up the
> file) and got an optimum delay of 6.10 seconds when I ran both
> G40-7 and G40-10.  There was no G40-4 in the list.

> Larry


> > I noticed a funny problem ? (or feautre I don't understand)

> > I will fly a loc-iv on a g40-4. it will say optimum delay is 4.65
> > seconds.

> > I will then fly it on a g-40-10. it will say optimum delay is 7.65
> > seconds.

> > this anomily is the same for jsut about everthing I fly, a short
> > delay, a short optimum time. A long delay  , a longer optimum time.

> > Anyone know what I'm doing wrong ?

 
 
 

RockSIM 6.x bugs and work-arounds

Post by AlMax7 » Sun, 01 Jun 2003 07:03:36


Quote:
>>>    I just flew a LOC-4 today on a G80-4.  It was just about

perfect
wow, interesting. Mine has 4 d-12 tubes as well.

I've flown it with an f60-9 that ejected 20 feet off the field.
The nose cone was runied, but the rocket and 12" payload were fine.

I thought I had "Cut" it down to 6, but since I timed the flight, the
delay still was 9 seconds after shut off. Ross told me since the moter
was 12 years old, the delay had gone long and the peak thurst lower.

I've flown it also on a g80-10 to perfect apogee, but alass, that also
had 4 d-12 kabooms going on the cluster at the same time, so added an
f48 to the group, all bets off.  But, that had a 12 " payload section
added.
They called that version a Heavy Duty Beauty. A loc-iv with 4 24mm
tubes for cluster and a 12" payload section.

I've flown it on an aerotech h70-10  single use to a perfect apogee as
well.

I'll try my g40-4 and see how close you and Bob are. I thhink you'r on
track.

Quote:


> > "Larry W. Hardin" writes:
> > > Strange... I ran the LOC-4 (don't know where I picked up the
> > > file) and got an optimum delay of 6.10 seconds when I ran both
> > > G40-7 and G40-10.  There was no G40-4 in the list.

> > Even 6.something sounds long to me. More like 4.something...

>    I just flew a LOC-4 today on a G80-4.  It was just about perfect, slightly
> into the wind.  A demo for a high school AP Physics class.  It wrapped itself
> around a lighting tower on the ballfield, about 30ft (half-way) up.  Recovered
> by one of the teachers.

>  -John

 
 
 

RockSIM 6.x bugs and work-arounds

Post by Bob Kapl » Sun, 01 Jun 2003 07:36:45


Quote:

> I think many are making the assumption that the LOC-IV that was simmed
> matches what in reality the rocket it built to.  In other words, what
> I find often with RockSIM is that I build the model in the SIM and I
> always have to add a MASS to adjust for Weight and CG.  I do this by
> measuring my fully finished rocket with recovery system installed.

> What I have found is that the amount of epoxy I use for fin fillets,
> the filler for spirals, the number of coats of primer and paint,
> always makes the rocket heavier than a straight build SIM model.

> So, if this fellow just used a straight SIM model the longer delays
> may be necessary.  Or even perhaps if he built is per the instructions
> but did not use a lot of epoxy or filler or paint...

Actually, with an under weight sim, you usually will end up with a SHORTER
coast time...

Quote:
> ... perhaps those with field experience indicating that the 4-5 second
> delay is perfect have that experience on a fully (tough) built model
> with full finish.

With a 4" tube, the LOC-IV is totally drag limited. The variation in coast
time due to weight is VERY small across a 2:1 weight range.

        Bob Kaplow      NAR # 18L       TRA # "Impeach the TRA BoD"
                >>> To reply, remove the TRABoD! <<<
Kaplow Klips & Baffle:      http://www.pleimling.org/le/Phantom4000.pdf
    www.encompasserve.org/~kaplow_r/    www.nira-rocketry.org    www.nar.org

 Save Model Rocketry from the HSA!   http://www.space-rockets.com/congress.html

 
 
 

RockSIM 6.x bugs and work-arounds

Post by John DeMa » Sun, 01 Jun 2003 09:00:27


  I'd expect your right about needing a longer delay with the 4 D12's in there
at the same time.  And the full H should need the 10 seconds (or maybe 8).

  I forgot to point out that this LOC-4 has an added payload section. A little
heavier, but not that signficant.

  -John

Quote:

> >>>    I just flew a LOC-4 today on a G80-4.  It was just about
> perfect
> wow, interesting. Mine has 4 d-12 tubes as well.

> I've flown it with an f60-9 that ejected 20 feet off the field.
> The nose cone was runied, but the rocket and 12" payload were fine.

> I thought I had "Cut" it down to 6, but since I timed the flight, the
> delay still was 9 seconds after shut off. Ross told me since the moter
> was 12 years old, the delay had gone long and the peak thurst lower.

> I've flown it also on a g80-10 to perfect apogee, but alass, that also
> had 4 d-12 kabooms going on the cluster at the same time, so added an
> f48 to the group, all bets off.  But, that had a 12 " payload section
> added.
> They called that version a Heavy Duty Beauty. A loc-iv with 4 24mm
> tubes for cluster and a 12" payload section.

> I've flown it on an aerotech h70-10  single use to a perfect apogee as
> well.

> I'll try my g40-4 and see how close you and Bob are. I thhink you'r on
> track.



> > > "Larry W. Hardin" writes:
> > > > Strange... I ran the LOC-4 (don't know where I picked up the
> > > > file) and got an optimum delay of 6.10 seconds when I ran both
> > > > G40-7 and G40-10.  There was no G40-4 in the list.

> > > Even 6.something sounds long to me. More like 4.something...

> >    I just flew a LOC-4 today on a G80-4.  It was just about perfect, slightly
> > into the wind.  A demo for a high school AP Physics class.  It wrapped itself
> > around a lighting tower on the ballfield, about 30ft (half-way) up.  Recovered
> > by one of the teachers.

> >  -John

 
 
 

RockSIM 6.x bugs and work-arounds

Post by EMR » Mon, 02 Jun 2003 01:53:36


Quote:



> > I think many are making the assumption that the LOC-IV that was simmed
> > matches what in reality the rocket it built to.  In other words, what
> > I find often with RockSIM is that I build the model in the SIM and I
> > always have to add a MASS to adjust for Weight and CG.  I do this by
> > measuring my fully finished rocket with recovery system installed.

> There's a mass and CG override feature that's intended specifically for
> this purpose.  You measure the mass and CG location of the model as built,
> enter these values, and you don't have to play around with a mass object.
> You can turn the feature on and off with a single checkbox.

> len.

Very true, in older versions this mass override was VERY fixed and as
you changed motors it kept the rocket the same mass.  I don't know
when this was fixed, but in version 6 the mass changes with the
different motors.  My weigh is a left over old habit, but you made me
check it again, so I will probably start doing it with the over ride
now.  Cool.  Thanks.

Nick