DMS Delays Too Short?

I flew two AT DMS motors for the first time yesterday at ROC, an I140-14 in a Rocketry Warehouse 38 Special modified for dual-deploy and a J425-14 in a Giant Leap Escape Velocity. I left the delays as-is and put the BP in the well for use as a backup for the electronic drogue. In both cases, the DMS's ejected at just about 12 seconds rather than the rated 14 seconds. In the Escape Velocity it wasn't a big deal because it was probably only about 1 second from apogee anyway, but the 38 Special was still moving at nearly 150 ft/sec when it popped. Fortunately it was drogueless so it didn't break anything, but it could have been nasty. Has anyone else had experiences with the DMS's delays being short? Thanks!

Since I started getting on-board video (Mobius), making it easier to tell just when the deployment event occurs (frame-by-frame analysis, also correlated with an A2 for most of the flights below), my observation has been that AT delays (DMS or RMS) can be fast, and CTI delays are more likely to be late.

Going through all my data so far, I have (these were all SD motor-eject flights):
AT DMS I280DM-14A (-2s drill): Fired @ 13.43s post-ignition (11.8s after burn-out @ 1.61s)
AT DMS I500T-14A (no drill): Fired @ 13.73s post-ignition (12.6s after burn-out @ 1.15s)
AT RMS-38/360 I218R-M (no drill): Fired @ 9.03s post-ignition (7.6s after burn-out @ 1.46s)
AT DMS I280DM-14A (-4s drill): Fired @ 12.30s post-ignition (10.5s after burn-out @ 1.77s)
CTI Pro38-4G I287SS-15A (-3s drill): Fired @ 14.67s post-ignition (13.1s after burn-out @ 1.58s)
CTI Pro38-5G I297SK-15A (-5s drill): Fired @ 11.53s post-ignition (9.8s after burn-out @ 1.70s)
CTI Pro38-4G I255RL-16A (-3s drill): Fired @ 16.40s post-ignition (14.8s after burn-out @ 1.56s)

So two of the AT's were about right (interestingly the two I drilled, also both I280DMs), two were very short (the two I didn't drill, one DMS and one an RMS "M" delay). One of the CTIs was on, the others were a second or more longer than what I had supposedly drilled them to, and that was with multiple turns of the drill, blowing-out the hole, cleaning the bit, repeating until no more loose delay chunks in the hole or on the bit.

I flew an I140W-14A (-2s drill) in December, but I didn't get video for that flight (the SD Card was still plugged into my computer ). The A2 data there suggests an early deployment, reporting 2.21s burn time, 9.7s coast-apogee and -1.4s apogee-eject (i.e. 8.3s burn-out to eject, 10.5s ignition to eject). That flight came in 230' less than the 2720' OR predicted, when my flights generally go about 10% over what OR predicts, so that would also seem to agree with an early ejection (about 500' lower than the flight probably should have gone with the proper delay).

A 14 second delay is allowed to have a variation of +/- 20% = +/- 2.8 seconds by NFPA 1125. The range is 11.2 seconds =< delay <=16.8 seconds so a 12 second ejection is within spec. If your delay was 2 seconds early, the rocket velocity was 64 fps and the rocket was 48' below it's planned apogee.

If you need/want more precision and accuracy, you need to use electronic deployment.

Bob

bobkrech said:

A 14 second delay is allowed to have a variation of +/- 20% = +/- 2.8 seconds by NFPA 1125. The range is 11.2 seconds =< delay <=16.8 seconds so a 12 second ejection is within spec. If your delay was 2 seconds early, the rocket velocity was 64 fps and the rocket was 48' below it's planned apogee.

If you need/want more precision and accuracy, you need to use electronic deployment.

Bob

Click to expand...

I'm pretty sure there is a 2 sec maximum delay time variation, but either way 12 sec on a 14 sec specified delay is within NFPA 1125 accuracy window.

John

Good point on the ±20%, though some of my observations seem to be even beyond that, unless AT delays actually start at ignition instead of burn-out. 7.6s is more than 20% below 10s (though some drilling tolerance probably explains it there), and 8.3s more than 30% below 12s (though I don't have video to back-up that 8.3s measurement, only A2 data for that flight, noting that I had good correlation between A2 and video on other flights). My CTIs were all within ±20% when measuring from burn-out. All of my ATs would be within ±20% if the time starts at ignition instead of burn-out (well, the one drilled DMS would have been 3% over, again I'd take that as drilling tolerance).

So for the CTI (Pro38 at least, I haven't used other sizes) and AT DMS + RMS-EZ reloads, a pre-assembled delay closure is included that presumably takes the burn-time of the motor into account. I haven't used any AT RDKs, it appears they come in a bunch of flavors that are meant to account for the burn-time of each motor (even the propellant type), there isn't simply one delay length for each of "S", "M" and "L" (how I thought it worked until doing some research today). So it's interesting that this all seems to try to maintain the time-from-burn-out for the delay, yet from what I've seen that doesn't work so well for AT motors, it really feels more like time-from-ignition.

And yes, as suggested my ultimate goal is electronic deployment so that this whole thing doesn't matter (if only I could find an e-match that works, hopefully a bunch of Wildman ejection lighters will show up on my door this week, MJG said they'll ship me a replacement box for my 80 bad initiators by the end of the month). But I would still prefer to keep the motor eject backup at least for now, and when using AT it seems like you'd want to set a VERY long delay to ensure it doesn't pop first.

I had one more flight (I161W-14A, RMS-EZ drilled -4s) that didn't have on-board video but did have a ground camera shot clearly seeing burn-out and ejection, again correlates perfectly between the on-board A2 data and video camera frame-by-frame analysis. That one was 8.9s burn-out to eject, 11.47s ignition to eject, so it's within ±20% either way, but again shorter than I had desired by over a second. Common thread here seems to be if I want a specific delay time on an AT motor I should probably shoot for 1-2s longer than what I really want. And for CTI if anything it seems like I want to drill out a bit more than I have so far, even my "desired" delays were generally rounded-up from what OR recommended, so in some cases I was already starting out with a second longer delay than I really wanted, having the charge go another second or two later than intended made things even scarier watching from the ground. I should make a +1s shim for my Pro-DAT like qquake2k did on his thread to get a bit more resolution.

I run any motor eject backup where dual deployment is the main, a second or 2 long let the altimeter do the work. The backup is just to help, sometimes if it's to late, the best you can hope for is a non ballistic recovery.


TA

Johnly said:

I'm pretty sure there is a 2 sec maximum delay time variation, but either way 12 sec on a 14 sec specified delay is within NFPA 1125 accuracy window.

John

Click to expand...

Nope. Not unless it has change since I wrote the S&T testing manual which is based on NFPA 1125.

The current requiremen5 is:

minimum variation below 7.5 seconds = +/-1.5 seconds; or

from 7.5 seconds to 15 seconds = +/- 20%; or

longer than 15 seconds = +/-3 seconds.


The old requirement was:

minimum variation below 5 seconds = +/-1.0 seconds; or

from 5 seconds to 15 seconds = +/- 20%; or

longer than 15 seconds = +/-3 seconds.

The change occurred because one of the manufacturers request NFPA to make the change, and they did.

Bob

I've had the gut feeling on AT's that the delay gets burning while waiting on motor to ignite , I'm sure its just a gut ache.

michigander said:

I've had the gut feeling on AT's that the delay gets burning while waiting on motor to ignite , I'm sure its just a gut ache.

Click to expand...

On any core burner, the igniter starts the propellant and the delay charge at the same time. If the motor chufs the delay won't burn as quickly as when the casing pressure comes up with a proper burn. The delay total burn time is supposed to be the stated delay time plus the propellant burn time.

I saw a USLI team do a delay burn check by igniting only the delay grain on a CTI motor, I believe. The stated 14 or so second delay smoldered for over a minute before blowing the ejection charge. With no pressure in the casing, it took its own sweet time.

bobkrech said:

Nope. Not unless it has change since I wrote the S&T testing manual which is based on NFPA 1125.

Click to expand...

Anyway, its always entertaining when someone argues S&T standards with someone from S&T.

Seriously, thanks Bob for bringing the education.

It sure seems like a possibility. I thought I've heard that the Loki instructions (I live in CA so I've never used one myself) says to pull the ignitor a bit back from the top so as to not light the delay immediately, but both the AT and CTI instructions I've seen say to put the ignitor at the very top. So it seems like one vendor considered this while the others perhaps didn't. Though the longest motor ignition I've had so far was the CTI I255, you can hear the relay click almost 2 seconds before the first sign of smoke, and it smoked for a full 2 seconds before it finally showed any flame, yet the delay was still almost 2 seconds longer than the drilling target. So that flight/motor didn't seem to be affected by the long time to ignition. I'd have to look more closely at the AT flights to see if any extra time spent on the ground correlates to the shorter observed delay.

Of course Bob is correct but here's the exact wording from the copy of 1125 that I have:


Production lots shall be corrected, destroyed or retested by the manufacturer under any of the following conditions:


(4)The time delay of any test item varies more than 1.5 seconds or 20 percent, whichever is greater, from the established mean time delay value of the rocket motor or motor-reloading kit. In no case shall this variation exceed 3 seconds.

Well, I guess -20% was a bit too much. Next time I'll just leave out the powder and rely on the electronics.

cerving said:

Well, I guess -20% was a bit too much. Next time I'll just leave out the powder and rely on the electronics.

Click to expand...

My TAP once said you either trust your electronics or you dont...

I have had a save from motor backup and one failure .

Now only use when flying new altimeters with no redundancy.. which is rare.

Kenny

KenRico said:

My TAP once said you either trust your electronics or you dont...

Click to expand...

In my case I have complete trust in my electronics, but absolutely no trust in the matches. My first (and so far only) attempt at DD resulted in the drogue not firing. Thankfully the main deployed at 700' (after a 4477' flight, so over 300 mph at deployment), which at least saved the rocket and electronics though the airframe wound up ~18" shorter once it was repaired (lower section zippered all the way to the hardpoint mount). But the bad e-match (both the ones I used in that flight were made by another rocketeer, two matches from the same batch, one that worked and one that didn't) went from ~1&#937; to being open without ever igniting, so no separation at apogee. Having a motor backup would have been great here, except I was flying an I65W-PS DMS, so no option for motor eject. And now I'm sitting on 80 purchased ignitors that seem to have the exact same problem as the home-made one that failed in action. So once my replacement/alternative e-matches arrive (sadly looks like neither will show up in time for my next launch tomorrow), I'll need to see several successful ground tests before I'll be comfortable attempting DD again, and you can bet I'll have the motor backup for a while also. Most likely (after a few successful ground firings) I'll probably try a few SD flights first (electronic main w/motor backup for the main instead of a drogue), perhaps with e-matches only (no BP) in the second bay just to make sure that they actually burned. A waste of matches, but at this point I need the confidence-building.

michigander said:

I've had the gut feeling on AT's that the delay gets burning while waiting on motor to ignite , I'm sure its just a gut ache.

Click to expand...

That is a fact. The igniter should be right against the delay grain hence it starts immediately. At least that is how it should be and re same for any other motor using motor delay/ejection.

michigander said:

I've had the gut feeling on AT's that the delay gets burning while waiting on motor to ignite , I'm sure its just a gut ache.

Click to expand...

Oh yeah, sure thing. Big trouble for say tarc flights that need an accurate delay, if the motor chuffs you just lost some time. And altitude too, the chuffs didn't push the rocket, just wasted energy.

Well, I guess I jinxed myself by speaking bad of AT's early delays. I flew two motors (both AT) yesterday:

AT RMS-38/360 I245G-L (RDK-13PLUS) (-2s drill, should have been 12s): Fired @ 16.23s post-ignition (14.8s after burn-out @ 1.4s)
AT RMS-38/720 J350W-M (no drill): Never fired (rocket lawn-dart)

The first flight the rocket reached apogee ~12.6s, OR predicted 12.8. It recommended 11.4s delay, so I bought the -L delay for this motor and drilled -2s from it using the RMS DDT. But since the charge fired almost 3 seconds late and 3.6s post-apogee, the rocket was moving fast enough that it made a 2.5" zipper. At least this one will fly again.

The second flight just never fired. After an hour and a half of searching the alfalfa field I finally found it, the BP was still sitting in the well. Disassembling the forward closure the top half of the BP poured out, further down it was packed-solid and came out as a single chunk. Below that the opening into the motor chamber was also plugged with solid BP. I'm not sure if that happened as a part of the impact or what, and if it did why it never fired, but that rocket and all the electronics that was in it are a loss (as is my second attempt at my L2 cert. :rant. The bits (forward delay spacer, O-ring and insulator) were all in their proper places and I could see up to the closure hole plugged with BP so the delay charge had burned.

So now not only do I not trust e-matches, but I've lost all faith in motor eject as well. Sadly, the Wildman ejection ignitors showed up Friday, so I didn't have a chance to ground-test before making this flight, I should have rigged the main charge well even without any confidence in it, it just might have saved the rocket and cert.

Will

A L2 rocket with electronics is worth at least $300 sitting on the launch pad. If you can take a match and (3) $100 bills from your wallet, strike the match and burn the 3 bills, then read no further, otherwise....

1.) Commercial e-matches cost about $1 each. For dual deployment you need 2 so spend the $2 for commercial e-matches with a failure rate of ~<=1% for your cert flight. Consider it like a 0.7% value insurance premium. (Commercial launch insurance premiums are 10% of value, e.g. the policy cost for a $1B rocket is $100M.) It's a small price to pay of a successful cert flight.

2.) Never use a home made e-match that you did not make and/or never launch an electronic deployment rocket without doing a ground test with the flight e-matches first to make sure you get prompt ignition of the BP charge and have the proper amount of BP to get a good, but not overly energetic ejection. That's another 1 time 0.7% value insurance premium expense that's well worth it.

3.) If you use a motor ejection apogee backup you need to be aware and design for a deployment that could be +/-5 seconds early or late. Why? Because you could be off by +/-2 seconds in your calculation of the time to apogee, and the maximum allowable error in an ejection delay is +/- 3 seconds. That's +/- 5 seconds.

4.) This means you should design your rocket so it will survive a 160 fps (100 mph) deployment event without damage. That can be done for 100 G main deployment at apogee with some forethought, but it is a piece of cake for a 40 G drogue deployment since the delta V (and the g-loading) is due to a typical delta-v of ~64 fps which is equivalent main deployment of +/-2 seconds of apogee (delta-v = 40 mph or 64 fps). If you rocket suffers any damage under a drogue deployment under that non-optimum condition it was designed to lightly.

If you do these 4 items before your first flight, the probably of an in-flight failure of the deployment system becomes extremely low. And your cost per flight is much less than 100% of the rocket value.

Bob

woferry said:

The bits (forward delay spacer, O-ring and insulator) were all in their proper places and I could see up to the closure hole plugged with BP so the delay charge had burned.

Click to expand...

I wonder what could cause the BP not to light if the delay element burned completely? I've seen long delays, short delays, failure of delays to light (typically very old motors), but never intact BP that didn't light. Speculation?

dpower said:

I wonder what could cause the BP not to light if the delay element burned completely? I've seen long delays, short delays, failure of delays to light (typically very old motors), but never intact BP that didn't light. Speculation?

Click to expand...

I f you pour nice loose BP into the well and some on the bottom of the well is "chunked" can't see a cause other than grease.

Okay, I've had it with AT motor ejection. So after my two failed flights two weeks ago, I decided to go with an I357T (RMS-EZ) Saturday, as it sim'ed to a lower altitude than the reloads I already had and the winds were projected to be high so being only single-deploy motor eject I didn't want to have to walk as far for the recovery (and I already had high-flying motors built into my CTI cases that I didn't get to fly the last launch, so I didn't want to pull the installed reloads out to load something different in those cases so I had only my AT case empty). The sims suggested an 11.5s optimal delay, so I used my UDT to remove 2 seconds from the 14s-default EZ delay assembly. This time:

AT RMS-EZ-38/360 I357T-14A (-2s drill): Fired @ 4.7s post-ignition (3.8s after burn-out @ 0.92s)

So, this charge went off less than 5 seconds from ignition (no ignition issues, less than a half-second from first smoke to flame), 8s earlier than it should have fired. I have the whole thing on video as well as the on-board altimeter logs. The rocket was about half-way towards its projected altitude, so the shock of the early ejection snapped the airframe in half (well, first it zippered the entire top half even though I had taken some anti-zipper methods this flight, then snapped it off just above the shock cord mount). That's 3 out of 3 flights in the past 2 weeks where I've had a rocket damaged or destroyed by a motor eject-related issue in an AT motor, and my rocket fleet has gone from 2 to 0. So I won't fly another AT motor without electronic eject (given this failure even using the motor as a backup is out), but of course I won't be flying any motors at all until I build some new rockets.

woferry said:

Okay, I've had it with AT motor ejection. So after my two failed flights two weeks ago, I decided to go with an I357T (RMS-EZ) Saturday, as it sim'ed to a lower altitude than the reloads I already had and the winds were projected to be high so being only single-deploy motor eject I didn't want to have to walk as far for the recovery (and I already had high-flying motors built into my CTI cases that I didn't get to fly the last launch, so I didn't want to pull the installed reloads out to load something different in those cases so I had only my AT case empty). The sims suggested an 11.5s optimal delay, so I used my UDT to remove 2 seconds from the 14s-default EZ delay assembly. This time:

AT RMS-EZ-38/360 I357T-14A (-2s drill): Fired @ 4.7s post-ignition (3.8s after burn-out @ 0.92s)

So, this charge went off less than 5 seconds from ignition (no ignition issues, less than a half-second from first smoke to flame), 8s earlier than it should have fired. I have the whole thing on video as well as the on-board altimeter logs. The rocket was about half-way towards its projected altitude, so the shock of the early ejection snapped the airframe in half (well, first it zippered the entire top half even though I had taken some anti-zipper methods this flight, then snapped it off just above the shock cord mount). That's 3 out of 3 flights in the past 2 weeks where I've had a rocket damaged or destroyed by a motor eject-related issue in an AT motor, and my rocket fleet has gone from 2 to 0. So I won't fly another AT motor without electronic eject (given this failure even using the motor as a backup is out), but of course I won't be flying any motors at all until I build some new rockets.

Click to expand...

So we can keep track of these issues, please fill out a MESS report for this malfunction. Check off the "Delay inaccurate (estimated actual delay)" and put in the information about the length of the delay you've shared above.
Thanks.

Thanks, I wasn't aware that MESS covered delays, report filed.

You might double check your universal drilling tool. My friend has one that he had problems with and it was set very different from mine. It can be adjusted via the set screw. I am not sure what the best way to calibrate it is.

woferry said:

Thanks, I wasn't aware that MESS covered delays, report filed.

Click to expand...

I'm not sure too many people know that delay issues can/should be reported, either. Video or altimeter data is useful in checking how well the published data matches with the field performance.
Thanks for filing.

terryg said:

You might double check your universal drilling tool. My friend has one that he had problems with and it was set very different from mine. It can be adjusted via the set screw. I am not sure what the best way to calibrate it is.

Click to expand...

I don't think any misadjustment could result in -2s becoming -10.5s (I missed a digit above, it was actually 4.37s after ignition, 3.45s after burn-out), and I had last used this UDT to do a -4s adjustment on an I280DM DMS motor (post #2 above), where that delay was just a tad longer than it should have been (10.5s vs. 10s target). So I haven't measured the bit yet but it's clearly not that far out of whack. Plus of course having used the drill before on several DMS, EZ and CTI reloads, I knew how much material to expect to be removed from a -2s drill, and that's what I got when I drilled this motor, there weren't any large chunks that came out with the UDT, it was more of a shave.

Yeah, when I drilled -3s out of an I140-14 I was surprised how little of the delay grain actually came out. That one ended up being about a -5 too. When the drills first came out, people would think that the numbers were the delay time and drill out 9 seconds thinking it would be a 9 second delay... big surprise when it pops halfway up. Reading instructions is always a good first step..

When I flew my first I140 DMS six months ago, I used the delay tool to remove -8 seconds, to result in leaving 6. The motor pretty much ejected either at burnout or even slightly before, and I wondered how that could possibly be, since the UDDT is maxed at -8 seconds, so you'd never be able to leave less than 6 seconds of real delay left. Now that you've mentioned it, could my unit be 'off' and require recalibration? How do you definitively tell? Need to ensure short delays when flying gliders, but not *blowthrough short*.

AstronMike said:

When I flew my first I140 DMS six months ago, I used the delay tool to remove -8 seconds, to result in leaving 6. The motor pretty much ejected either at burnout or even slightly before, and I wondered how that could possibly be, since the UDDT is maxed at -8 seconds, so you'd never be able to leave less than 6 seconds of real delay left. Now that you've mentioned it, could my unit be 'off' and require recalibration? How do you definitively tell? Need to ensure short delays when flying gliders, but not *blowthrough short*.

Click to expand...

I don't have the UDDT myself yet but at -8 seconds there should be a 1/4" gap between the drill knob and UDDT body when set up on the motor to drill. With the DMS, and RMS-EZ charges going off way early, set right or not isn't going to help I guess. I never did like the sound of molded plastic bulkheads, to me plastic with pressure and heat don't mix well. Something is giving somewhere, that or the QC on the delay grains sucks.

I recently flew an I284-14A RMS-EZ, drilled at the "-8" setting, which should have given a 6s delay +/- 1.5s. The delay burned through right at motor burnout, same failure mode as AstronMike. I've flown some DMS motors OK, but haven't needed the "-8" setting before. I checked this UDDT against my other UDDT, and they measure identical. Interestingly, I've never had such a total delay failure like this on old-style non-EZ aerotech motors, so wasn't concerned about motor reliability. Of course, this occurred on my most expensive, most complex build to date, which suffered significant damage.

Yes a MESS form has been submitted, as well as a warranty claim to AT.

I'd sure like to know how this failed, and I'll also need to decide if I want to use the same motor once the rebuild is complete, or use either pure electronic recovery (difficult in this particular rocket), or use a different motor.