Help with Ejection Charge Issue

Hi—

I am doing a dual deployment rocket for the first time using an Entacore AIM Altimeter. I got everything set up for an ejection deployment test, loaded my black powder, taped it up, and then went to turn on the altimeter. And the black powder immediately was ignited and the ejection occurred unexpectedly.

The only reason I can think this would happen is because the Entacore altimeter manual says that it performs a continuity test on startup. But I was under the impression that continuity tests don’t set off an igniter, and furthermore it seems dangerous to me to load up the black powder while the electronics are switched on.

If anyone can provide me advice regarding why this happened, I’d really appreciate it. I don’t want to have anymore unexpected ignitions when messing with black powder, and I’m also trying to finish this test before I drive back up to school tomorrow. Thanks in advance for your help!

Here is my wiring... unless I’m totally missing something, it looks alright to me. I’m really baffled right now and scared to proceed before I figure out what made it ignite unexpectedly like that :/

Update #3: I just put a plain old firewire igniter in the port without any black powder, turned on the altimeter, and it didn’t go off... perhaps I accidentally only secured one wire of the igniter to the screw terminal, and that made it go off during continuity test?? I’m not versed enough in electronics to know for sure. If anyone has any insight, that’d be much appreciated.

Update #4: did it again. Didn’t ignite when I turned the altimeter on and off, I then loaded black powder and put the igniter in the powder well, and it went off the second I turned the switch on again.

.....

I use the XTRA, never used the Entacore altimeter, but I've never heard of the power up issue you had. I did have the same thing happen thing happen to me me (charges blow on power up) with my Raven 4 when I reversed the battery polarity

Demosthenes said:

Update #4: did it again. Didn’t ignite when I turned the altimeter on and off, I then loaded black powder and put the igniter in the powder well, and it went off the second I turned the switch on again.

.....

Click to expand...

My first Level one attempt was an abort due to the same issue using Entacore AIM USB altimeters and ematches. I had a redundant configuration, went to the pad (yes, I had ground tested) and when I turned on one side - kaboom. Felt like a complete doofus draped in shock cord and parachutes. I found that more often than not, power-on caused one or both ematches to fire. I worked with Entacore technical support it was deemed to be an intermittent issue with both altimeters. Seemed odd to me, but whatever.

For my next L1 attempt not quite a month ago, I built an Eggtimer Quantum and launched using it for a flawless certification flight. Thanks, @cerving!

I wish I could explain with confidence why we both had issues. I have a hunch it is a turn-on transient that allows sufficient current to fire the ematch, but it’s just that - a hunch.

g.pitts said:

My first Level one attempt was an abort due to the same issue using Entacore AIM USB altimeters and ematches. I had a redundant configuration, went to the pad (yes, I had ground tested) and when I turned on one side - kaboom. Felt like a complete doofus draped in shock cord and parachutes. I found that more often than not, power-on caused one or both ematches to fire. I worked with Entacore technical support it was deemed to be an intermittent issue with both altimeters. Seemed odd to me, but whatever.

For my next L1 attempt not quite a month ago, I built an Eggtimer Quantum and launched using it for a flawless certification flight. Thanks, @cerving!

I wish I could explain with confidence why we both had issues. I have a hunch it is a turn-on transient that allows sufficient current to fire the ematch, but it’s just that - a hunch.

Click to expand...

wow... Thank you so much for sharing this. I thought I was doing something wrong and was going crazy trying to figure it out. I guess I just have a defective altimeter.. of course I will try some last ditch attempts to see if maybe it’s a me problem, but if I don’t find anything, I’ll just hold off on my launch and order a new altimeter. Thank you again for sharing this, I have yet to read about anyone else having this issue online so you are a real life saver.

Demosthenes said:

wow... Thank you so much for sharing this. I thought I was doing something wrong and was going crazy trying to figure it out. I guess I just have a defective altimeter.. of course I will try some last ditch attempts to see if maybe it’s a me problem, but if I don’t find anything, I’ll just hold off on my launch and order a new altimeter. Thank you again for sharing this, I have yet to read about anyone else having this issue online so you are a real life saver.

Click to expand...

Hey, no worries! I spent countless hours going over my PCB design to see if there was something I missed, but never found any issues with it... Working with David from Entacore (who was really helpful and responsive), we ended up with the “transient issue” diagnosis.

When is your scheduled launch date?

I am looking at your wiring and I noticed you have another set of wires connecting to the C on the side where the +/- V connections are at.

There really should not be any wires connected there. It is possible that current is flowing to the igniters as soon as you power it up.

g.pitts said:

Hey, no worries! I spent countless hours going over my PCB design to see if there was something I missed, but never found any issues with it... Working with David from Entacore (who was really helpful and responsive), we ended up with the “transient issue” diagnosis.

When is your scheduled launch date?

Click to expand...

I think I’ll give them a call tomorrow, but most likely the same conclusion will be reached since I’ve quadruple checked the wiring now.

My launch is scheduled for November 16th, so in an ideal world I’d have more than enough time to order a new altimeter and get ready in time. Problem is that my school break ends tomorrow and between classes/homework/reserving a field for testing safely on University property, I’m out of luck. No worries, I will use my next winter break to get things going again, test at home, and then launch in the spring!

SinfulDarkLord said:

I am looking at your wiring and I noticed you have another set of wires connecting to the C on the side where the +/- V connections are at.

There really should not be any wires connected there. It is possible that current is flowing to the igniters as soon as you power it up.

Click to expand...

C is a no-connect on the altimeter, and it is documented that this terminal can be used as a junction terminal as he is doing. I didn’t trace out all the wiring, but I’m a 40+ year career EE and had the same exact issue as she did.

SinfulDarkLord said:

I am looking at your wiring and I noticed you have another set of wires connecting to the C on the side where the +/- V connections are at.

There really should not be any wires connected there. It is possible that current is flowing to the igniters as soon as you power it up.

Click to expand...

I thought it might have to do with that too, but all that’s going on there is connecting my battery to a switch. The switch wires deceivingly go underneath the board, but it’s wired exactly according to the Entacore manual: positive battery lead in V+, negative battery lead and one end of switch in C, other end of switch in V-. Also, the firewire igniters wouldn’t fully light up when I turn on the altimeter during a standalone test. But as soon as the igniter is inside of black powder and I turn on the altimeter, immediately on start up the black powder ignites, making the “transient” theory catch my eye, because otherwise how could my black powder be exploding on startup without the actual igniters even being used up??

Who knows though... certainly not me lol. I’m just frustrated at this point. Thank you for trying to help me out with this issue I’m having!! I really appreciate it

Demosthenes said:

I thought it might have to do with that too, but all that’s going on there is connecting my battery to a switch. The switch wires deceivingly go underneath the board, but it’s wired exactly according to the Entacore manual: positive battery lead in V+, negative battery lead and one end of switch in C, other end of switch in V-. Also, the firewire igniters wouldn’t fully light up when I turn on the altimeter during a standalone test. But as soon as the igniter is inside of black powder and I turn on the altimeter, immediately on start up the black powder ignites, making the “transient” theory catch my eye, because otherwise how could my black powder be exploding on startup without the actual igniters even being used up??

Who knows though... certainly not me lol. I’m just frustrated at this point. Thank you for trying to help me out with this issue I’m having!! I really appreciate it

Click to expand...

Ah yes I see, strange that should have worked, how are you wiring the igniters? Can you show us a picture of that please.

I would recommend wiring it without the C port connection.

Very frustrating. Back in competition days, my school team used the Entacore AIM-3 and didn't have these issues.

I'd recommend the Missleworks RRC2 or RRC3, or the Perfectflite Stratologger CF.
(if you want to solder, eggtimer kits are good too)

Demosthenes said:

I thought it might have to do with that too, but all that’s going on there is connecting my battery to a switch. The switch wires deceivingly go underneath the board, but it’s wired exactly according to the Entacore manual: positive battery lead in V+, negative battery lead and one end of switch in C, other end of switch in V-. Also, the firewire igniters wouldn’t fully light up when I turn on the altimeter during a standalone test. But as soon as the igniter is inside of black powder and I turn on the altimeter, immediately on start up the black powder ignites, making the “transient” theory catch my eye, because otherwise how could my black powder be exploding on startup without the actual igniters even being used up??

Who knows though... certainly not me lol. I’m just frustrated at this point. Thank you for trying to help me out with this issue I’m having!! I really appreciate it

Click to expand...

FWIW, I was able to reproduce my ematches firing during power-on without any back powder involved. Not every time, but most times I would power on and the AIM USBs would set off the ematches. I’m using a LiPO 2S which has fast and high instantaneous current capability, which may be the key difference in what we are seeing between our respective setups.

Cycling power to an altimeter with charges attached is a bad idea. The pins of the micro-controller that is running the show are undefined in that grey period between the time power is applied and it completes its reset process. Sure they may end up as inputs (aka open circuit) but you have no idea what they will do before that.

An arming switch that interrupts the circuit to the charges is always a good idea.

Looking at that photo it appears that this uses the typical N channel MOSFET output and includes a gate to source pull down. That protects you while the control signal is floating. I am curious just what resistance was chosen but the photo doesn't have enough detail to make out any markings.

It also looks like the stimulus for the continuity test is provided via a couple of resistors per channel which means that the test current is very likely always applied.

UhClem said:

Cycling power to an altimeter with charges attached is a bad idea. The pins of the micro-controller that is running the show are undefined in that grey period between the time power is applied and it completes its reset process. Sure they may end up as inputs (aka open circuit) but you have no idea what they will do before that.

An arming switch that interrupts the circuit to the charges is always a good idea.

Looking at that photo it appears that this uses the typical N channel MOSFET output and includes a gate to source pull down. That protects you while the control signal is floating. I am curious just what resistance was chosen but the photo doesn't have enough detail to make out any markings.

It also looks like the stimulus for the continuity test is provided via a couple of resistors per channel which means that the test current is very likely always applied.

Click to expand...

Certain microcontrollers are designed to be “glitchless” during power-on - either cold power-on or warm-reset. They will come up via circuit design in the micro to a known “safe” state and not glitch active. And of course some micros do not have that feature. So it’s hard to say what may happen during that power-on window without knowing which micro is used in the AIM USB design. I didn’t get that far in debug of the issue, as knew that I was going the route of the Eggtimer Quantum.

<edit> I should add that I went through my design as a part of the issue investigation with Entacore. David de Bruyn had zero concern with switching the power to the AIM USB.

Not sure if the same conditions would apply to the AIM altimeter verses the XTRA....but just in case...

I had a pair of AIM XTRA units that gave me fits several years back, and there was a very important firmware update needed after the very thing you describe happened to me. It prevented me from flying any rockets on a trip to Balls 22 (2013). Happened to me twice as well. What you describe sounds exactly like what I experienced.

Long story short, upon power up, if the GPS "altitude" was far enough off of the barometric "altitude", it fired the charges immediately upon power up. The firmware update fixes the incongruity between the two.

David de Bruyn, owner of Entacore, explains the situation in post #135 of my old (and still depressing) thread here: https://www.rocketryforum.com/threads/balls-22-projects.56834/page-5

ECayemberg said:

Not sure if the same conditions would apply to the AIM altimeter verses the XTRA....but just in case...

I had a pair of AIM XTRA units that gave me fits several years back, and there was a very important firmware update needed after the very thing you describe happened to me. It prevented me from flying any rockets on a trip to Balls 22 (2013). Happened to me twice as well. What you describe sounds exactly like what I experienced.

Long story short, upon power up, if the GPS "altitude" was far enough off of the barometric "altitude", it fired the charges immediately upon power up. The firmware update fixes the incongruity between the two.

David de Bruyn, owner of Entacore, explains the situation in post #135 of my old (and still depressing) thread here: https://www.rocketryforum.com/threads/balls-22-projects.56834/page-5

Click to expand...

Thanks, Eric, I'm sorry to hear about your scrubbed launch! This does indeed sound similar to the issues I had with my AIM USBs. My AIM USBs were the latest hardware design running the latest FW 4 / SW 4 that was released in November 2017, so I believe it is fair to say that our issues are different despite their similarities. I suppose it's water under the bridge at this point, but personally I'd like to understand the root-cause of this issue.

g.pitts said:

Certain microcontrollers are designed to be “glitchless” during power-on - either cold power-on or warm-reset. They will come up via circuit design in the micro to a known “safe” state and not glitch active.

Click to expand...

Which micro-controllers? I have read a lot of data sheets and have never found one that claimed to be glitchless during power on/off.

UhClem said:

Which micro-controllers? I have read a lot of data sheets and have never found one that claimed to be glitchless during power on/off.

Click to expand...

My most recent experience is with the TI family. Here's one as an example, but it's a thing these days.

https://www.ti.com/lit/pdf/sprs742&usg=AOvVaw3MibNX-NR9STL3_I-7vsfj

That is a pretty weak version of glitch free: "GPIO pins as input (state depends on internal PU/PD)". Figure 5-1

So the pins are not actively driven but the internal pull up/down resistors do come into play. Which the code hasn't had a chance to configure yet.

UhClem said:

That is a pretty weak version of glitch free: "GPIO pins as input (state depends on internal PU/PD)". Figure 5-1

So the pins are not actively driven but the internal pull up/down resistors do come into play. Which the code hasn't had a chance to configure yet.

Click to expand...

It's common practice to add an external pull-up or pull-down to support the need of what the GPIO touches. I've done it with countless designs. The alternative is to have the microcontroller vendor impose a decision on you, which is undesirable. I'm not sure why this is such a big deal to you, but the bottom line is that the GPIOs won't glitch during power-up or reset.