Motor Performance Consistency

Over the course of this year, my TARC team discovered pretty large inconsistencies between various motors of the same type and batch. When keeping the rocket in the exact same configuration and just throwing a new motor in, the altitudes would vary wildly (more than would be reasonably attributed to condition differences within the same hour). When we reduced the mass of the rocket (and keeping the CG the same), the altitude would at times decrease.

At this point, we seem to have narrowed these discrepancies down to inconsistencies within motors. The motors we've been using are primarily single-use Aerotech F42. Without spending a lot of money to test various motors between vendors, reloads, and single-use, we can't easily verify the theory that some motors are less consistent than others. Have you found there to be such variations in motors, specifically in the MPR F range?

kavel_r said:

Over the course of this year, my TARC team discovered pretty large inconsistencies between various motors of the same type and batch. When keeping the rocket in the exact same configuration and just throwing a new motor in, the altitudes would vary wildly (more than would be reasonably attributed to condition differences within the same hour). When we reduced the mass of the rocket (and keeping the CG the same), the altitude would at times decrease.

At this point, we seem to have narrowed these discrepancies down to inconsistencies within motors. The motors we've been using are primarily single-use Aerotech F42. Without spending a lot of money to test various motors between vendors, reloads, and single-use, we can't easily verify the theory that some motors are less consistent than others. Have you found there to be such variations in motors, specifically in the MPR F range?

Click to expand...

Composite propellant motors can perform differently if ignited differently. If the motor does not ignite properly (entire internal propellant face igniting from the top down), it will not have the correct pressure and the thrust will be lower and the burn longer with less total impulse because of a loss of efficiency.

if the ignition is really bad, there will be chuffing where some of the propellant burns, but the motor does not come up to pressure and it can sit there 'sputtering' for a while until it finally pressurizes. This burns up propellant without producing usable thrust and you lose lots of total impulse.

If there is any actual production variation, it would be detectable in weighing the motors to see if there is different weight, which would usually be the propellant. This is more common in Estes black powder motors. Composite motors should be very consistent in weight.

White lightning propellant gets a crust on the exposed inner surface of the propellant. This crust can affect ignition. Using a very hot igniter will burn through the crust and help ignite fast and correctly.

You can also use the Blue Thunder propellant which does not get a crust.

Our TARC team found pretty significant differences in propellant weight in CTI motors where you can actually take things apart and weigh individual grains. Significant in this case meaning a couple of tenths of a gram in a grain weighing 10g or so. The students mixed and matched grains to get a consistent propellant weight, which helped noticeably. With a single-use, all you could do is weigh the overall motor.

That said, our teams also had significant variability in flights that was poorly explained by other factors (weight, weather, etc.). It's so frustrating to launch 3 times in an hour, get two flights at 3 and 5 points, then call a scored flight and get 20 points.

Those thoughts about ignition are interesting and make sense.

I know some of the "expert" TARC teams/schools in my area do weigh motors to match them for more consistency. This year they were using Aerotech Blue Thunder F39s (this is a 24/40 reload), other years they have used Cesaroni motors. It must help - they won it all a few years ago. One of their two teams was eighth this year.

The F42, according to this: https://www.nar.org/SandT/pdf/Aerotech/F42.pdf is Blue Thunder and has a very consistent total impulse (within the half-dozen samples fired during certification at least).

Hmmmmmmm.......

Another issue is the delay. One of the teams I was helping was using motors that used drilled delays. They found altitude variance based on early ejection before apogee. Changing to a longer delay got a more consistent apogee altitude but of course if it went too long it created other issues. Smaller motors are also more temperature sensitive due to the smaller propellant mass.


Tony

I will just tack on that reducing the mass of a rocket can actually cause it to have a lower apogee. Think of throwing a baseball vs a ping-pong ball. The lighter rocket might move more laterally and less vertically. So you can't count that test as adding any data.

Bat-mite said:

I will just tack on that reducing the mass of a rocket can actually cause it to have a lower apogee. Think of throwing a baseball vs a ping-pong ball. The lighter rocket might move more laterally and less vertically. So you can't count that test as adding any data.

Click to expand...

That's technically true, but it's also pretty atypical for TARC rockets (BT-70 and larger, F motors and smaller) to be in the range where more weight might mean more altitude. Normally, less weight on the pad will result in more acceleration and less weathercocking as well.

To illustrate the point of mass:

54mm CF Mongoose clone

Long burn motors like light rockets, fast burn motors like heavier ones. I always thought weather-cocking was a function of over-stability, not total weight. Interesting thought.

Tony

A TARC team I work with actually took 20 Identical Cesaroni motors and tested them at a local lab. They said the results didn't vary that much with the thrust, it was the delay times that really varied. None of the delays were altered. For example a 12 second delay would actually give them anywhere from 9 sec. to 16 sec. Probably has to do with how the motor ignited.

This is my biggest dislike of TARC. I think the rules as written are great for 95% of the entrants but for the top 10-20 at nationals, I would suspect there is more random variation at play than actual differences in skill. Even if you knew the exact impulse of the motor and had a perfect sim, you can easily get a couple of percent variation for winds or minor launch rod angle errors.

It would be nice to see a "masters" category with a lower impulse limit or some other restrictions that would press the design & building skills a little more. As it stands, I think you could argue the results are being decided by factors outside of the students control. Take away the two sided spec and let them go for it.

Thanks for the replies everyone! This is all good info. I never thought of weighing and swapping grains to get a consistent propellant weight.

We also saw quite a bit of variations in the delays.

Another issue we were concerned about is off-axial thrust, like the motor exhaust vectors around and doesn't stay centered in the nozzle. I saw a video (can't find the link anymore) of a slow-mo Estes motor burn with significant off-axial thrust, do you know if this is a problem with composite motors too?

It can be,but rarely in my experience.
Usually from abnormal nozzle erosion causing the vectored thrust.This can happen due to molding issues. [air bubbles in phenolic for one]
Only composite propellent I see, vectoring thrust on regular basis are metal storms, but I don't think they are ever used in competition.

Vectoring on Estes motors is almost always caused by a bad nozzle. They are essentially compressed clay and occasionally they fail or have a manufacturing defect that creates an issue. As Jim mentions, any propellant with metal can create slag in the exit cone which can create vectoring. It's much more likely with smaller motors that don't fully burn the metal before it exits, and of course any sparky motor that intentionally spews metal out with the exhaust.


Tony

I've had major thrust vectoring on Aerotech E15s when the igniter stayed stuck in the motor from overzealous taping.

It’s normal for rocket performance to vary from flight to flight, and not all such variation is motor-related. Wind, wind shear, tip-off and other factors can make big differences.


The fact that you are seeing much more variation with F motors (or any particular impulse range) than with other motors, suggests to me that ejection spikes may be implicated. By this I mean a spike in *recorded* altitude caused by shock to the instrument (e.g. forces on the diaphragm) during ejection. Although these spikes can be randomly positive or negative, the instrument delivers the highest recorded altitude, so negative spikes don’t interfere with you readings, but positive shocks can, and usually do. (OK. I acknowledge that spikes can arise from other causes, and other spikes have different characteristics. Out of scope, just now.)


The placement of an ejection spike in the curve figures into its impact on accuracy. If, for example, you have an over-long delay and ejection takes place well below apogee, the spike may not exceed the real apogee, and you’re fine. If the delay is near apogee, you may get a spuriously high reading. Of course, launch weight can figure into the placement of the spike.


If at least some of your altitude readings are much higher than the simulated predictions, I’d suggest a couple of unofficial flights with recording altimeters aboard.

gtg738w said:

This is my biggest dislike of TARC. I think the rules as written are great for 95% of the entrants but for the top 10-20 at nationals, I would suspect there is more random variation at play than actual differences in skill. Even if you knew the exact impulse of the motor and had a perfect sim, you can easily get a couple of percent variation for winds or minor launch rod angle errors.

It would be nice to see a "masters" category with a lower impulse limit or some other restrictions that would press the design & building skills a little more. As it stands, I think you could argue the results are being decided by factors outside of the students control. Take away the two sided spec and let them go for it.

Click to expand...

This would open up interesting possibilities - one that comes to mind is making points a function of the "ready to launch" rocket mass. Low mass = good ;-) This would then stress optimizing the vehicle design for efficiency - this is very much a real world consideration. As things stand, people just throw more motor in to overcome overbuilding of the airframe.

doug_kosty said:

This would open up interesting possibilities - one that comes to mind is making points a function of the "ready to launch" rocket mass. Low mass = good ;-) This would then stress optimizing the vehicle design for efficiency - this is very much a real world consideration. As things stand, people just throw more motor in to overcome overbuilding of the airframe.

Click to expand...

Back in the most popular age of NAR competition, most of the individual events had mass limits. It did make things more challenging. At my first contest, I had to butcher my Estes Orbital Transport on the field to get it down to the required Swift (B) B/G mass. When I flew Dual Payload, I opted to fly with less than max total impulse so that I could carry more tracking powder. For TARC, I think I would consider leaving mass alone, but penalize total impulse used.