Measured motor delay thread

I'd like to see any significantly short or long out-of-specification measured delay times posted here from people using altimeters capable of measuring them. I realize they won't be perfectly accurate due to variables like motor temperature, chuffing, altimeters which use G forces to measure motor burnout (as I suspect they do; Gs can go negative before motor burnout), inadequate vent holes to accurately sense apogee, etc., but they'll be in the ballpark. That's why I'm asking for just the radically long or short ones.

I'll start:

Aerotech SU F32-8T - 11.0s measured (8s was already well past predicted apogee for this rocket; luckily the recovery system was robust)

For NAR cerified motors you can go here and click on the link for a particular motor. On the second page of the data sheet for most motors you'll see the actual measured times from when the motors were certified.

Per NFPA 1125, delay is measured from the point where the thrust drops to 5% of the peak until ejection.

billspad said:

For NAR cerified motors you can go here and click on the link for a particular motor. On the second page of the data sheet for most motors you'll see the actual measured times from when the motors were certified.

Per NFPA 1125, delay is measured from the point where the thrust drops to 5% of the peak until ejection.

Click to expand...

I forgot to mention that I already know about NAR motor cert sheets; I have a directory full of them.

I want field data since it's there and easy to get to for anyone with an Altimeter II. Others should, too. Why? The cert motors are a very limited sample and in the case of the reloadable ones tested are assembled by people who are almost certainly very proficient in that task. Besides keeping track of CATOs, long or short delays are probably even more of a factor in rocket destruction because of the relative rarity of CATOs. There have been many threads about Aerotech motor delay issues. Whether those problems are caused by incorrect assembly procedures or other problems, it would be nice to get measured data from actual flights of all MPR and HPR motors of all brands since it's there to be easily had and easily posted.

Composite MPR and HPR motor delay data would be what I personally am most interested in.

It would be interesting to see measurements. Delays have a reputation of being inaccurate, especially for core-burning motors.

The problem is that anyone who would need accurate delay time should be using electronics anyway. (I.e., if your rocket is too heavy to handle a 20% early/late deployment, you should not be using motor ejection.)

Winston said:

The cert motors are a very limited sample and in the case of the reloadable ones tested are assembled by people who are almost certainly very proficient in that task.

Click to expand...

You'd be surprised. :wink:
Checking the procedures in the instructions with someone who isn't proficient can provide some good feedback to the manufacturers.

JohnCoker said:

It would be interesting to see measurements. Delays have a reputation of being inaccurate, especially for core-burning motors.

The problem is that anyone who would need accurate delay time should be using electronics anyway. (I.e., if your rocket is too heavy to handle a 20% early/late deployment, you should not be using motor ejection.)

Click to expand...

I remember reading a post from a big gun flier on the board of a big gun rocketry organization who stated he "burst the bubble" of a flier who obsessed about delays, used multiple simulators on rockets with motor delays and also prided himself on drilling delays. He said to forget it all and just use electronics!
You know.................. He's right. One just points it straight up, arms it and flies. No need to worry about it. Of course if the rocket goes half cocked sideways, it'll get zapped whether using motor or electronic delay. Fly a stable rocket with electronics, good matches and apogee will go just right. Kurt

KennB said:

You'd be surprised. :wink:
Checking the procedures in the instructions with someone who isn't proficient can provide some good feedback to the manufacturers.

Click to expand...

Kenn is the resident S & T non proficient person. He voluntarily limits his flying so he doesn't get too proficient.

I usually read the instructions when I assemble a reload because I can never remember if the fat O-ring goes in the forward or aft end. It doesn't help that it changes in different casing sizes. Kenn assembled the motors for some recertification testing we did last week. He read every word of the directions and found a thing about igniter location in that particular motor that I wasn't aware of. My point is that the proficient builder is just as likely to screw up as the non proficient builder.

And I agree with John that electronics should be used when an accurate delay is necessary. Actually, I think electronics should be used whenever possible.

billspad said:

Kenn is the resident S & T non proficient person. He voluntarily limits his flying so he doesn't get too proficient.

I usually read the instructions when I assemble a reload because I can never remember if the fat O-ring goes in the forward or aft end. It doesn't help that it changes in different casing sizes. Kenn assembled the motors for some recertification testing we did last week. He read every word of the directions and found a thing about igniter location in that particular motor that I wasn't aware of. My point is that the proficient builder is just as likely to screw up as the non proficient builder.

And I agree with John that electronics should be used when an accurate delay is necessary. Actually, I think electronics should be used whenever possible.

Click to expand...

Any new motors??

5x7 said:

Any new motors??

Click to expand...

No, just routine recertification firings.

ksaves2 said:

He said to forget it all and just use electronics!

Click to expand...

Virtually all MPR and, I'd guess, the majority of HPR flights rely on motor delay. The point of listing measured delay times here from those who have onboard electronics that only measure and do not control (like the Altimeter II) is to gather field data on a large variety and large number of motors with little effort on the part of those with such electronics - simply read it off the display and post it here along with the motor type and delay chosen.

From that data, we'll be able to determine, for instance, if some reloadable motor types/brands are more prone to long or short delays than others, perhaps due to a more complex assembly process and user error or some flaw in the design. We won't know the cause for certain, but we'll know whether it does or doesn't have an issue.

Winston said:

Virtually all MPR and, I'd guess, the majority of HPR flights rely on motor delay. The point of listing measured delay times here from those who have onboard electronics that only measure and do not control (like the Altimeter II) is to gather field data on a large variety and large number of motors with little effort on the part of those with such electronics - simply read it off the display and post it here along with the motor type and delay chosen.

From that data, we'll be able to determine, for instance, if some reloadable motor types/brands are more prone to long or short delays than others, perhaps due to a more complex assembly process and user error or some flaw in the design. We won't know the cause for certain, but we'll know whether it does or doesn't have an issue.

Click to expand...

There are a couple of small MAD units out there, the Transolve P7 (https://www.transolve.com/Transolve/Files/Products/P7/P7.html)
and the Pico AD4 that could actually be used on a modroc (https://www.picoalt.com/)

Yeah they have a cost but I caution about obsessing over the motor delay grains. The delays may vary from batch to batch due to changes
in the reagents that the motor maker can't control. Two seconds either way wouldn't be surprising. What set of data is obtained this week might be different 6 months from now and what is true 6 months from now might be off a year from now. Any statement that you make of "who has long or short delays" may not be valid in a short period of time. I suspect that the large motor providers test some representative samples from their different batches and trying to police that could be futile. You could perhaps pass your findings on to the motor maker in question if you find a consistent discrepancy, to make amends, but to publicly draw conclusions from a small dataset is
fraught with trouble. Even if you are on a Standards and Testing Committee I would suggest you use your findings discretely between yourself and the motor providers.

With MPR there is nothing wrong with using some sort of shock absorbing device in or parallel to the shockcord to try to compensate for delay discrepancy. Kurt (I am not involved commercially in amateur rocketry in any way)

Winston said:

I'd like to see measured delay times from people using altimeters capable of measuring them. I realize they won't be perfectly accurate due to variables like motor temperature, chuffing, altimeters which use G forces to measure motor burnout (as I suspect they do; Gs can go negative before motor burnout), inadequate vent holes to accurately sense apogee, etc., but they'll be in the ballpark.

I recently cleared the memory on my Altimeter II, so I only have two values to list:

Aerotech SU E20-7W - 6.1s measured
Aerotech SU F32-8T - 11.0s measured (8s was already long for this rocket; luckily the recovery system was robust)

Click to expand...

The burn time is just as interesting to me, I have seen quite a few reports on TRF of motors burning much longer or shorter than advertised based on altimeter data.

billspad said:

For NAR cerified motors you can go here and click on the link for a particular motor. On the second page of the data sheet for most motors you'll see the actual measured times from when the motors were certified.

Per NFPA 1125, delay is measured from the point where the thrust drops to 5% of the peak until ejection.

Click to expand...

I know S+T is adressing this on recertification, but everone should be aware that the data sheets on many of the motors posted on NAR site are very old and do not have data from individual firings.

ksaves2 said:

Yeah they have a cost but I caution about obsessing over the motor delay grains. The delays may vary from batch to batch due to changes
in the reagents that the motor maker can't control.

Click to expand...

No one is "obsessing," I just want easily obtainable field data. So should everyone else. At least with the Altimeter II, that data is being thrown away every time owners choose "erase."

5x7 said:

The burn time is just as interesting to me, I have seen quite a few reports on TRF of motors burning much longer or shorter than advertised based on altimeter data.

Click to expand...

That would be nice, too. Anything that could hamper successful deployment of recovery gear using motor delay is important because it's relevant to rocket survival and overall safety.

Winston said:

No one is "obsessing," I just want easily obtainable field data. So should everyone else. At least with the Altimeter II, that data is being thrown away every time owners choose "erase."

Click to expand...

If one wants to do this for their own mortification, o.k. But I don't see what there is to gain with a small sample size. Plus one needs to be concerned about
the accuracy of the electronics device used for measuring. As far as motor burn is concerned, the most accurate measurement means is a loadcell test stand. There may be differences between burn times obtained from various altimeters that can record that data and a test stand that can graphically record
the burn. Periodic testing by S & T is nice on appropriate test equipment but I caution against making public accusations using consumer altimeters.

Concerning:

"Aerotech SU E20-7W - 6.1s measured
Aerotech SU F32-8T - 11.0s measured (8s was already long for this rocket; luckily the recovery system was robust)"

You have more motors from that same batch to test? That would be interesting. Is the 11 seconds a fluke? Is it consistently high from other motors in the same batch? 6.1 is pretty close to stated but again how is the absolute accuracy of the device. We can only accept what Jolly Logic says on faith. Kurt

Winston said:

No one is "obsessing," I just want easily obtainable field data. So should everyone else. At least with the Altimeter II, that data is being thrown away every time owners choose "erase."

Click to expand...

Winston

When we certify motors, we actually measure the thrust of the motor verses time. The thrust stand will also recorded the force of the motor ejection charge so we have a direct electronic record of the ejection time. We also video the motor firing and have both a visual and audio reference on the motor ejection timing, and since we have plot of data versus time, we can observe any abnormalities during a motor firing. Unless I see view a recorded accelerometer curve and a recorded barometric curve, I'm not going consider any inflight data to be quantitative as I do not know what the actual data looks like.

Bob

bobkrech said:

Winston

When we certify motors, we actually measure the thrust of the motor verses time. The thrust stand will also recorded the force of the motor ejection charge so we have a direct electronic record of the ejection time. We also video the motor firing and have both a visual and audio reference on the motor ejection timing, and since we have plot of data versus time, we can observe any abnormalities during a motor firing. Unless I see view a recorded accelerometer curve and a recorded barometric curve, I'm not going consider any inflight data to be quantitative as I do not know what the actual data looks like.

Bob

Click to expand...

Holy guacamole Bob. I didn't realize that the ejection charge was monitored during S & T tests but indeed that would be the most accurate.
Follow the thrust curve and then get the spike from the ejection. Since it's all plotted against time one has a very accurate representation of the burn events including the exact delay. My point is I would be very skeptical with indirect measuring of said events using an altimeter/accelerometer. Especially when trying o draw conclusions that might be erroneous and unjustified. Kurt

ksaves2 said:

Holy guacamole Bob. I didn't realize that the ejection charge was monitored during S & T tests but indeed that would be the most accurate.
Follow the thrust curve and then get the spike from the ejection. Since it's all plotted against time one has a very accurate representation of the burn events including the exact delay. My point is I would be very skeptical with indirect measuring of said events using an altimeter/accelerometer. Especially when trying o draw conclusions that might be erroneous and unjustified. Kurt

Click to expand...

Measuring and reporting the ejection delay time is a motor certification requirement under NFPA 1125, and NAR S&T certifications are 100% compliant with it.

https://www.nar.org/wp-content/uploads/2014/08/ST-MotorTestingManual.pdf

Here is the recent AT F44W certification. https://www.nar.org/wp-content/uploads/2014/08/F44W.pdf Note the burn time and ejection delay data on page 2.

Bob

Bob, I really do appreciate those that work with the NAR S&T Team to give us good data sets to work with.

Related to the S&T procedures Bob, do you have a choice set of "lighters" that you use during the tests or do you use whatever is factory supplied?

Also, what if a motor chuffs? Do you toss that data point out of the set?

Just wondering.

Greg

I'm with you on that. A Large Rocket coming down almost as fast as it went up, with a crowd of cub scouts isn't fun, and I have seen it happen way to many times.

Andrew

GregGleason said:

Bob, I really do appreciate those that work with the NAR S&T Team to give us good data sets to work with.

Related to the S&T procedures Bob, do you have a choice set of "lighters" that you use during the tests or do you use whatever is factory supplied?

Also, what if a motor chuffs? Do you toss that data point out of the set?

Just wondering.

Greg

Click to expand...

We assemble and fire the motors according to the manufacturers instructions. If the manufacturer supplies igniters, we use the supplied igniters. If the manufacturer does not supply the igniters, we follow their recommendations and bill the for them.

I can't remember ever having a chuffing motor in a certification test, however I do remember having a motor that did not ignite using the manufacturer's recommended method. We figured out how to ignite the motor and the manufacturer changed his instructions to conform to the way we go it to work.

Bob

bobkrech said:

We assemble and fire the motors according to the manufacturers instructions. If the manufacturer supplies igniters, we use the supplied igniters. If the manufacturer does not supply the igniters, we follow their recommendations and bill the for them.

I can't remember ever having a chuffing motor in a certification test, however I do remember having a motor that did not ignite using the manufacturer's recommended method. We figured out how to ignite the motor and the manufacturer changed his instructions to conform to the way we go it to work.

Bob

Click to expand...

Well Bob, I sure am glad someone is watching our backside. Most if not all problems with motors that I have seen were caused by Human Error.

Andrew

bobkrech said:

Winston

When we certify motors, we actually measure the thrust of the motor verses time. The thrust stand will also recorded the force of the motor ejection charge so we have a direct electronic record of the ejection time. We also video the motor firing and have both a visual and audio reference on the motor ejection timing, and since we have plot of data versus time, we can observe any abnormalities during a motor firing. Unless I see view a recorded accelerometer curve and a recorded barometric curve, I'm not going consider any inflight data to be quantitative as I do not know what the actual data looks like.

Bob

Click to expand...

This is not meant to be data of certification accuracy or anything for certifiers to even be concerned about. This is BALLPARK stuff meant to find obviously long and short delays, like my 8 second Aerotech MPR motor delay that was 11 seconds long. The data's there, so lets use it, warts and all.

From my initial post here:

I realize they won't be perfectly accurate due to variables like motor temperature, chuffing, altimeters which use G forces to measure motor burnout (as I suspect they do; Gs can go negative before motor burnout), inadequate vent holes to accurately sense apogee, etc., but they'll be in the ballpark.

Winston said:

This is not meant to be data of certification accuracy or anything for certifiers to even be concerned about. This is BALLPARK stuff meant to find obviously long and short delays, like my 8 second Aerotech MPR motor delay that was 11 seconds long. The data's there, so lets use it, warts and all.

From my initial post here:

I realize they won't be perfectly accurate due to variables like motor temperature, chuffing, altimeters which use G forces to measure motor burnout (as I suspect they do; Gs can go negative before motor burnout), inadequate vent holes to accurately sense apogee, etc., but they'll be in the ballpark.

Click to expand...

11 seconds based on indirect measurement of a one time event without any idea of the accuracy of your Jolly Rogers III (or whatever it is) device.
For all we know there could be differences between devices due to the electronic vagaries of each individual. You are treating your device as if it were as accurate as a static test stand. You re-inforce that by this statement, "like my 8 second Aerotech MPR motor delay that was 11 seconds long. The data's there, so lets use it, warts and all."

First of all we don't know that your delay was "actually" 11 seconds long. Secondly, you imply that inaccurate data is o.k. to draw conclusions from with,
"The data's there, so let's use it, warts and all." Certainly not a very scientific approach. I'd use your device for what it is ultimately meant for. It's to give one an idea of the altitude the rocket went. Even then, with the changes in weather and barometric pressure, the actual "true" altitude might be off a bit from the altitude that the device reads. It is also known that accelerometer derived altitude is a lot more inaccurate than barometric. If the rocket is a little off vertical, accel only is more prone to be inaccurate.

Finally your comment concerning ballpark data can't be supported because you don't know for certain the degree of introduced errors. It's obvious you haven't seen an unfiltered datastream off a deployment device.

Leave the testing to the motor testing committees of NAR and Tripoli. If one knows of a consistent anomaly, report it through the regular channels. Cato's are easy. There have been wrong delays packed with reloads in the past by mistake that the motor makers made good on when it was pointed out. It's rare but could occur. Kurt

ksaves2 said:

11 seconds based on indirect measurement of a one time event without any idea of the accuracy of your Jolly Rogers III (or whatever it is) device.

For all we know there could be differences between devices due to the electronic vagaries of each individual. You are treating your device as if it were as accurate as a static test stand. You re-inforce that by this statement, "like my 8 second Aerotech MPR motor delay that was 11 seconds long. The data's there, so lets use it, warts and all."

Click to expand...

NO, I'm NOT treating it as accurately as a test stand.

First of all we don't know that your delay was "actually" 11 seconds long.

Click to expand...

The accelerometer, I assume, measures the beginning of delay time the only way it can, from the point where the rocket is no longer accelerating which it would assume is motor burnout. From the acceleration curve in Openrocket, that should have occurred about 0.3s before specified motor burnout. Other predicted values were amazingly close to predicted (for instance, measured max-v was only 4% lower than predicted), so I don't think that was wildly off. I subtracted 0.3s from "cost to apogee" + "apogee to eject" to get 11s.

"The data's there, so let's use it, warts and all." Certainly not a very scientific approach.

Click to expand...

I'm going to state this one more time and that's it. This is BALLPARK data from results obtained in the field intended to find OBVIOUS delay outliers that I'm certain many here have personally experienced, whether due to manufacturing variability in single use motors or assembly problems in a particular type of reloadable motor casing due to user error or whatever. The 8 sec delay motor that was 11 sec was obviously, visibly long. The altimeter simply confirmed that.

It's to give one an idea of the altitude the rocket went. Even then, with the changes in weather and barometric pressure, the actual "true" altitude might be off a bit from the altitude that the device reads. It is also known that accelerometer derived altitude is a lot more inaccurate than barometric.

Click to expand...

Accelerometer II is barometric and I always use the 1/4" dia hole area per 100 cu.in. fuselage volume rule.

Finally your comment concerning ballpark data can't be supported because you don't know for certain the degree of introduced errors. It's obvious you haven't seen an unfiltered datastream off a deployment device.

Click to expand...

As, I've said, all other predicted values were measured to be amazingly close to simulation predictions. I even compensated for the slight amount of error due to a probable misreading of motor burnout through the sensing of zero acceleration.

If one knows of a consistent anomaly, report it through the regular channels.

Click to expand...

That's the kind of problem this field data is intended to detect!

This is interesting. Out of necessity due to the various agency regulations pertinent to their use and shipment, motors to be certified are sent from the manufacturer requesting certification. They can't be production samples bought off the shelves like the reviewers of other products chose to do for some reason:

https://www.nar.org/wp-content/uploads/2014/08/ST-MotorTestingManual.pdf

2.3 Scheduling Test Sessions

Upon receipt and approval of a written application for testing, the S&T Chairperson will, in cooperation
with the manufacturer and committee members, assign a testing location and schedule a test session of
sufficient duration at the earliest possible date. The S&T Chairperson will advise the manufacturer in
writing of the following:

Fees associated with the certification testing
The quantity of each motor type required
The address for shipment of the test articles
Proposed test date and location

2.4 Shipping of Motors for Testing

Motors must be legally shipped to the designated test site in a manner that will ensure arrival a minimum of
seven days before the scheduled test date. Exceptions for later arrival may be made by agreement with the
S&T Chairperson. Upon receipt of the test motor shipment the supervisor of the receiving test site (S&T
Chairperson or Regional Subcommittee Chairperson, as appropriate) will advise the manufacturer
promptly. Any damage, losses or omissions will be reported immediately to the manufacturer for remedy.
The manufacturer will include with the shipment a complete list of all items included, and will clearly label
or mark all items for identification.

I understand that for first certification, but I think recertification should use off the shelf (perhaps literally) motors.

Good point.

Winston said:

This is interesting. Out of necessity due to the various agency regulations pertinent to their use and shipment, motors to be certified are sent from the manufacturer requesting certification. They can't be production samples bought off the shelves like the reviewers of other products chose to do for some reason:....

Click to expand...

CarVac said:

I understand that for first certification, but I think recertification should use off the shelf (perhaps literally) motors.

Good point.

Click to expand...

In the recertification mentioned above, Bill did purchase, with his own money, the same reloads that were submitted just as a check. One had a different batch number than the sample sent by the vendor and the other had the same batch number. Depending on the source, we may have to search far and wide to ensure we are getting a batch that is fresh enough to qualify for recertification. With the vendor sending samples from the production line, we know we have the latest to compare with the testing done five years earlier.

This would also make a lot more work for the S&T volunteers. I understand that outfits like Consumer Reports buy all the items they test to maintain their editorial independence. In our small community, we don't need to keep that kind of distance. We also don't have the income sources to have that type of relationship.

I hope no one is intimating any deviousness on the part of the vendors. That would be shortsighted as performance that varies from the published thrust curves and delay times would surely be reported on MESS forms and a motor would accumulate enough dings to bring it to S&T's attention. :wink:

KennB said:

I hope no one is intimating any deviousness on the part of the vendors. That would be shortsighted as performance that varies from the published thrust curves and delay times would surely be reported on MESS forms and a motor would accumulate enough dings to bring it to S&T's attention. :wink:

Click to expand...

No deviousness, just that manufacturers certainly wouldn't take any LESS care in the manufacture of motors they send in for evaluation, right? On CATOs, I suspect that probably a minority are actually reported, especially in the case of single use motors, but that's just a hunch.

I'll change my original thread post to better reflect what I'm after. I suspect that even fewer long/short delays are reported than CATOs because people can often get away with them without damage to the rocket, as I did.