Redundant dual deploy, when do you consider it 'necessary'?

[Grog6]

You guys would freak over my junk box of electronics, samples, leftovers, dumpster contents...

 

[Buckeye]

While this does provide some coverage for HW faults - but they don't really occur UNLESS YOU ABUSED the electronics in some what like leaving them outside (lost) for months.

My altimeters (PerfectFlights and Ravens) have fared well after some abuse - hard landings, water landings, hanging on a power line for a couple rainy weeks - and they still seem fine. After such events, I dry them out, clean them up with alcohol, run all the checks on the computer interface, and ground test. These are tough little devices.

 

[jderimig]

You guys would freak over my junk box of electronics, samples, leftovers, dumpster contents...

Any vintage vacuum tubes?

 

[Orien]

My altimeters (PerfectFlights and Ravens) have fared well after some abuse - hard landings, water landings, hanging on a power line for a couple rainy weeks - and they still seem fine. After such events, I dry them out, clean them up with alcohol, run all the checks on the computer interface, and ground test. These are tough little devices.

I've been impressed with the robustness of Missileworks as well. My RRC2L survived what had to be a 1000 g right turn at almost Mach one and the ensuing freefall from 2000 feet. When I found the av bay, it was dutifully beeping out the altitude.

 

[Buckeye]

View attachment 493940View attachment 493941View attachment 493942

As for me, I'm for redundancy. The cost is, what, $50 for an RRC2L? I invest a lot of thought, time, care, and cash into these things. I don't like the idea of my rocket's survival ultimately depending on one little cheap ematch firing. I'm really surprised there's so much opposition to it, honestly.

Wow, that is quite the contraption! Looks nice, but heavy.

Back to page 1 of this thread, if your concern is an ematch failure, then you can add another one for $0.30 and rest easy.

My opposition to RDD is that it is often treated like a religion. "The way it must be done."

I have had a few less than ideal main deployments, but never ever a drogue failure and ballistic return to earth. That's my priority. I make sure drogue/drogueless is loosely packed, 2 ematches for drogue event, motor backup if possible, and no shear pins in the booster (drag separation is rarely a thing).

If guys prepping 2 computers, 2 batteries, 4 ejection charges, 8 ematches, and yards of wiring keeps them at their table and not in the launch queue, then that's more flying time for me!

 

[Michael L]

Any vintage vacuum tubes?

Two large boxes.

 

[Michael L]

Wow, that is quite the contraption! Looks nice, but heavy.
I'll bet it weighs less than a wood sled. 3D prints are usually mostly hollow. 15% infill is pretty common.
If guys prepping 2 computers, 2 batteries, 4 ejection charges, 8 ematches, and yards of wiring keeps them at their table and not in the launch queue, then that's more flying time for me!
All I have to do is charge batteries (not "rocket science" and done all at the same time), do a quick link test with the computer to make sure the trackers is still alive, weigh charges, dump, test the ematch with an ohmmeter, add an ematch (I only use one) to the charge well, seal it up (tape), connect, assemble the av bay and rocket. It's not much trouble at all and it gives me the time to check all of the details. Plus I can't fly HP if I don't do it that way.

 

[Grog6]

Two large boxes.

I still have a tube tester.

 

[jderimig]

Two large boxes.

[/Thread derail] Any sweep tubes? [/]

 

[boomtube-mk2]

All those systems share a single point of failure, which is the system integrator.

I never fail, I just sometimes don't succeed . . . badly.


When I look at this, I can't help but wonder if having dual-deploy with back-up, thus two of everything and the resulting complexity thereof, doesn't invite trouble in itself.
That's a lot of stuff to keep track of.

 

[Michael L]

I still have a tube tester.

Yeah... me too. I have my Hickok 539C but it's got an issue. I think its simple but I haven't been motivated to open it up. Plus calibration and all that is a bit intimidating (probably for no reason). I have some simpler testers at home. Restoring test equipment for fun can be a mighty deep hole. I bought the eTracer V3.0 kit. It's decent but the designer disappeared for a while. So I bought the "other" eTracer kit. It's much nicer, better supported, and works at higher voltage. Plus I can print test data stickers to put on the box once I've tested the tube. I rarely do that btw. It's handy if you're tube seller.

[/Thread derail] Any sweep tubes? [/]

lol yeah... I keep them in the stall with my Unicorns I wish... There are some restorations I could do if only...

I never fail, I just sometimes don't succeed . . . badly.

View attachment 494058
When I look at this, I can't help but wonder if having dual-deploy with back-up, thus two of everything and the resulting complexity thereof, doesn't invite trouble in itself.
That's a lot of stuff to keep track of.

It's actually not that hard to keep track of. It's a bit pricey so having a sled that can be moved from rocket to rocket would be nice.

 

[boomtube-mk2]

It's actually not that hard to keep track of. It's a bit pricey so having a sled that can be moved from rocket to rocket would be nice.

How difficult would it be to have a "Universal Sled" that can be used in rockets of differing diameters?
Perhaps using some form of adaptor like those used for motors.

 

[Michael L]

It seems like it wouldn't be too hard. The hardest thing that I haven't finished yet is putting redundant dual deploy and tracking in an Aspire (29mm rocket). Why would I do that? I'm going to launch that puppy with an I205 in it lol. That and of the 3 Aspires I've built, deployment has failed (my fault) twice. Once I used the wrong ignitors and once I reused a TeleMini that spent a few months outdoors due to the first failure. The two failures wouldn't have been failures if I had used motor ejection BUT I'd have had to reduce the ejection charge in the motor that I used.

 

[cautery]

How difficult would it be to have a "Universal Sled" that can be used in rockets of differing diameters?
Perhaps using some form of adaptor like those used for motors.

I am designing an av-bay internal module that will be entirely mobile to multiple airframes and airframe diameters. One signal/power connection each end, and av-bay/payload bulk fastener patterns are identical.... template located on-center. Should be pretty cool if it works. (That's always the thing, huh?)

Additionally, the mass in the module is laterally, and longitudinally stationed about the air-frame center-line such that it remains a neutral and effectively a single mass unit falling ON the center-line of the airframe.

 

[UhClem]

uman-Body Model ESD protection is standard on every pin of every IC made in the last 20 years with the exception of super high-frequency inputs....which we don't have in our cheapy FC's.

Absolutely wrong. They do have diodes to clamp the inputs to Vcc and ground. (Usually. Parts designed to run at 3.3V that have 5V compatible inputs will skip the Vcc diode.) These will still be killed by ESD.

ESD protection is more complicated than that and has to be designed in. For example I prefer to use the MAXIM RS485 chips that include ESD protection.

 

[UhClem]

Those of you that launch when it's cold and dry out could do a service for your club by buying a box of the very cheap wrist straps, and donate them to the club. attach the grounds to the launcher platform, if it's metal.

Of very limited utility. The key thing to do is maintain equal potential and that can be done (carefully) without using an ESD wrist strap. Which is just part of an overall ESD control system.

 

[Grog6]

I agree, you can just stand there with your hands on a grounded object.I'd rather wear a heel strap personally.

 

[FredA]

Absolutely wrong. They do have diodes to clamp the inputs to Vcc and ground. (Usually. Parts designed to run at 3.3V that have 5V compatible inputs will skip the Vcc diode.) These will still be killed by ESD.

Sorry - not correct. HAVE YOU DONE IC DESIGN?
Inputs are protected and checked to withstand the HBM at most major IC manufacturers.
It's their liability if the part dies before being soldered into a circuit.
Once in circuit, they are more robust as other dissipation paths exist.
Clamp diodes to the supply DO NOT CARE if the part is 3V or 5V.
Clamp diodes are not the only input protection.

I' know there are exceptions based on the needs of the circuit.
But every IC manufacturer I know starts with the assumption that ALL pads are protected till proven they can't work correctly with the protection network. The protection network is built into the pad structure that they copy & paste around the part - automatic and they have to do work to remove it.
Our dirt-cheap MCU's from the likes of STMicro and ATMEL are all protected.

 

[Andrew Brown]

I haven't seen all of the FC's on the market, but as FredA said, the ones with common components all have ESD protection between the pads and the guard rings on the IC. These are not small geometry devices, and they are qualified against a JEDEC ESD standard at the package level, often in the range of 5 - 10 kV HBM (enough to be a decent sized spark from your finger to light switch, which also requires a path to ground). Note that this doesn't apply to small geometry devices such as dense memory, GPU's, RF connections, etc, and you do need ESD protection for those. These are cheap, large device geometry microcontrollers that are designed to be commodities. Once soldered on a board, their ESD tolerance is generally improved by adding regulators, decoupling caps, and pullup/down resistors here and there. The FET drivers for deployment charges are discretes and pretty robust, and isolate the GPIO's of the micro from anything that you would be handling anyway. The general exception would be if there are any RF pads where there are reduced ESD tolerant pads, like a GPS receiver or telemetry module. I'm not saying to be careless with ESD and start touching everything on the board on dry winter days after you've put wool socks on, but these are fairly tough modules. The only thing you are handling anyway are battery connections (hopefully not directly touching terminals anyway, besides, the micro is protected by the regulator), and the FET outputs where the ematches are connected.

I'd be interested to see if there actually have been ESD fails, or if this concern is academic.

 

[Grog6]

We used to kill chips all the time in the winter, until We finally put in place mitigation strategies.
The one that finally worked was conductive, grounded floor wax, Anti-static, grounded mats on all the tables, heel straps, and a mandate for all- cotton clothes.
The tsop8 CML output gates would pop like tiny popcorn, when they plugged in cables. They got series resistors with bav99 diodes in a later revision, but the static voltage hits faster than the diodes can turn on.

 

[Andrew Brown]

>>The tsop8 CML output gates would pop like tiny popcorn, when they plugged in cables.

I'm not sure this is a fair comparison to touching an ematch output. High speed CML outputs are notoriously sensitive, and plugging them up to a cable of indeterminate length that may or may not have been discharged certainly could blow the gate oxide of a driver FET. This is also far from HBM, which is inherently a slow process. You should continue doing what works for you though, and if using a wrist strap and connecting it to a metal stake in the ground makes you confident, by all means, carry on. I'm simply saying that I don't believe it's necessary based on the BoM and designs I've seen on my limited altimeter experience. I am curious if anyone actually has blown an altimeter out from ESD.

 

[jderimig]

I am curious if anyone actually has blown an altimeter out from ESD.

I have never seen one in 14 years.

 

[Grog6]

The output is likely to be a low-side switching transistor, NPN bjt or n-channel mosfet; those are fairly robust.

 

[UhClem]

I have taken VLSI design courses at the graduate level so I have a better than average understanding of the problem.

The clamp diodes on a 3.3V part will not allow 5V tolerant inputs so yes, they do care. They are often the only input protection unless the part has special protection structures added on.

See something like the TI document SLAA530 on system level ESD protection. (I pulled up a random, and fairly recent, MSP430 data sheet and it says nothing about ESD protection. HBM or otherwise.) Protection during manufacture is the responsibility of the manufacturer. The IC vendor certainly is not going to replace parts damaged by ESD.

SLAA530 reminds me a lot of the final chapter in Henry Ott's book.

One of my favorite articles that touches on ESD is "The Art and Science of RS485" from Circuit Cellar magazine. (A Google search will usually turn up a copy somewhere.) It discusses a plant where 10% of their RS485 receivers (4,000 bipolar 75175s) were dying every day.

 

[Andrew Brown]

This thread has digressed from the original post, so this will be my last post on this topic. The MSP430 describes the ESD threshold against the JEDEC standard in section 8.2. It's +/- 1 kV HBM, which is actually pretty wimpy for a product in that class. The design of the ESD diodes is very specific for meeting HBM and CDM requirements. It's not a simple clamp diode. We generally only use proven IP blocks for our ESD pads, and they are worth every penny.

The original post asked the simple question of when dual deploy redundancy is necessary. A lot of discussion then centered around possible failure from ESD. My response is that the electronics are quite robust to ESD, especially once implemented on the altimeter board with regulators and high current drivers. The overall sensitivity of the electronics is small in comparison to a low current e-match. I am far more concern of a discharge setting off an ematch than damaging the electronics. If one were to create a list of failure modes of dual deploy altimeters, I would imagine ESD would be pretty far down the list.

 

[kramer714]

With a good designed 3d printed av bay, it is easy to fit (2) altimeters, (2) 2s lipos, and (2) microswitches in an avby that is less than 3 inches long in a 3 inch rocket. Using an inernal bulkhead means that you only use 3" of total length for the av bay even in a 6" long coupler. I dont use 'end bulkheads that much anymore. Too much wasted space in the coupler.

 

[Buckeye]

With a good designed 3d printed av bay, it is easy to fit (2) altimeters, (2) 2s lipos, and (2) microswitches in an avby that is less than 3 inches long in a 3 inch rocket. Using an inernal bulkhead means that you only use 3" of total length for the av bay even in a 6" long coupler. I dont use 'end bulkheads that much anymore. Too much wasted space in the coupler.

Nice compact av-bay. I like it. Care to show the layout of your electronics? Looks like everything is mounted horizontally on the internal bulkheads?

 

[kramer714]

Yup sits sideways

for the 3 inch I have a strattologer version and a RRC2 version

The red 'inner tube' is a piece of coupler that I cut the long way so it fits inside a coupler. It is 1-1/2 inch long I bond it in place iinside the coupler. The bulkheads are smaller than a normal ones and fit inside (not agasint the ends but inside) the coupler against the 'inner tube'.

The mount is screwed onto one of the bulkheads with (3) #6 countersunk screws. and holds a 400mAh 2s lipo with a JST connector, plus a micro switch that is operated by a screw that goes through the switchband.

I really dont like eye bolts or U bolts, they take up too much room and are right in the center. For this av bay I use (2) #10 screws that run through the av bay connected to an aluminum saddle, the shock cord is held by the saddle no u-bolt or eye bolt needed. There is a 1/4-20 screw that goes in the middle, I started putting a threaded coupler in the center so there are (2) screws flush to the bulkheads. The hex in the center keeps the coupler from turning (the picture shows the older 'round center design' the CAD shows the hex.

I can get a 38-600 Aerotech in a 12" long body tube below the lower bulkhead. There is less than 1/2 inch clearance from the top of the motor to the lower bulkhead in the av bay. The motor is well inside the coupler tube. Plus there is room for a chute and shock cord in the nose for head end deploy. The pictures show 3 inch and a 4 inch goblin, both use this kind of design.

 

[Banzai88]

Good discussion, and I appreciate everyone's pro/anti dual redundant positions. Thanks!

So, of the current crop of available flight computers that have 'redundant' channels, looks like the Featherweight Raven and the Marsa 33LHD both have 4 channels, and for the DIY folks there's the Eggtimer Proton with 6.

Anything else out there?

Those of you who use a single computer, which one and why?

 

[Voyager1]

Altus Metrum EasyMega and TeleMega have 6 channels.