EEVblog #1165 - Cypres Parachute Safety AAD Teardown

DejaVu. Dave revisits the new generation Cypres 2 parachute Automatic Activation Device and tears it down.
A safety device that automatically activates your parachute at a pre-set altitude. It has saved over 4000 lives to date.

Bitcoin Donations: 38y7DE8HEHNj8fGDtUr4PkCn9nWxiorvvy
Litecoin: ML7oQokTwB38bgzzjLDbRV97HKAHuwRfHA
Ethereum: 0x11AceA38DCA9DbFfB4F35f3F746af65F9dED28ce

Support the EEVblog through Patreon!

Buy anything through that link and Dave gets a commission at no cost to you.

Stuff I recommend:

Donate With Bitcoin & Other Crypto Currencies!

[JuanMurphy]

A few things that you might find interesting about the CYPRES. Correct on its "self-calibration" and sensing. It will, as you stated, sample every 30 seconds. Once it senses a pressure change it will lock itself to its last "unchanged" pressure. It will use this pressure as "ground". So the CYPRES is (by just turning on) effective when the take-off location is close to the jump location. If there are vertical obstacles or the jump area is at a different altitude than the take-off altitude then the CYPRES must be manually adjusted to work properly. Some funny things that happen that most users of this product dont know. It only arms itself once it gets 1, 500' above its "activation" altitude which is about 750' (I think)...so in the "old days' of skydiving when 2, 000; exits were routine the CYPRES may not have been armed. So it would sense "take-off", lock onto an altitude, then begin sampling every few seconds until it realizes that its falling, then begins to sample continuously until it gets to its starting altitude and pressure remains constant (on the ground). For all "normal" jump operations this is fine. There have been times where user error has confused the CYPRES. One incident in California years ago was a skydiving fatality where a jumper had checked her CYPRES prior to getting to the jump area. There was, apparently, a significant hill between where she started and the DZ. The CYPRES thought it was in an aircraft climbing as the pressure change from driving was similar to an aircraft's initial ascent. She did not reset the CYPRES at the DZ and ended up having a malfunction/accident. The CYPRES did not fire as it set itself to an altitude that was either 1000' below the DZ or several hundred feet above it.

Also, the expert model was set to activate when a jumper was between 125ish feet and 850ish feet and reaching a speed of 35m/s or 78mph. So if both of those parameters were met (or more importantly if the pressure readings were +/- equal to those parameters) it would fire. This was developed in the 80s when parachutes were much larger and the rates of descent were much slower. They have developed another model that is for smaller high-performance parachutes where the speeds required to cause activation are much higher and another model for students and tandem parachutes were the activation altitudes are considerably higher.

The CYPRES also records data, so when sent back to manufacturer (as is the case for fatalities and some malfunctions) they can pull all the pressure/time data from the unit. They actually withhold much of the data as it is proprietary but they will release the data related to the vertical speed and altitude of activation.

CYPRES are also made for pilot parachutes where they will essentially activate on a delay from bailout.

[DVSProductions]

Where did the chip in the sensor go?
At 20:30 it was there
At 24:18 it is gone

[DoRC]

Pretty sure the plastic piece was a cover for the cutter that got pushed out of thee way not melted rope.

[user]

The summer I turned 16 (1970) my older brother had joined a local jump club made up of mostly active military personnel. Anyway, I decided I wanted to try jumping too so the old man signed the release form (probably thinking if anything happened one less mouth to feed, lol). Anyway, after the first five mandatory static line jumps the day came for my first free fall. They had an aluminum gadget charged with a co2 cartridge and a pin one would pull when we got over a certain altitude. Well, like an idiot, after I got the parachute strapped on (a Korean War surplus military parachute, a double L, by the way) I accidentally pulled that damm pin. I hollered to the jump master and she said back up against the refueling shed. This I did and a second later that thing went off and my rip cord went flying into the air. But since I had my back pressed against the shed they were able to restring the rip cord and not have to repack the parachute. Anyway, the plane we used had the side door removed. They told me if that thing went off again they would toss me out of the plane, no matter what the altitude. For the record I scared the ground master because I followed directions and counted out the full 1000, 2000, ... 5000 before pulling the rip cord for a full 5 second delay. From what I'm told, that is rare for first time free fallers.

[PhiTheProducer]

The wet replaceable filter is in case of a water landing or in case of heavy humidity, you’ll want to replace it. Any moisture can cause a difference of atmospheric pressure on the sensor.

[grantrennie]

Great video, the chord cutter is a standard quarry daisy chain detonator, they use these in quarries and mines to set off the charges and they are secondary detonators that lie on top of the ground, the detonator cables are threaded through them and when they go off, they cut through the plastic insulator and let's the spark shoot through to the next row of charge detonators.. Common in mining and have been repurposed to parachute line cutters..

[bernharddrescher]

I once have been to a factory tour for pressure sensor packaging. The gel inside the sensor is most likely a protective layer for the die and bond-wires as the package / sensor requires "direct" contact to the environment.

[dorbie]

Cool, glad you finally got around to it. It's a CYPRES 2. Most moisture concerns come from submersion during water landings e.g. swoop ponds. The devices remain inside a closed parachute container almost all their life and are inspected by a rigger regularly. The U.S. regulations primarily say that if such a device is used it must be used in compliance with the manufacturer's guidelines, they don't really dictate much else although industry organizations like PIA, parachute manufacturers (via approvals) and clubs have acted to maintain quality.

[johncuzzourt]

My last jump was in the 1990s. I had a good canopy over my head, and at about 1500 ft the AAD went off and my main canopy folded in front of me!
Fortunately, my reserve was open and not involved with my main. I pulled my cutaway cord and landed the reserve with only minor injury to my big toe. I vowed never to jump again until I could buy, test, and maintain my own equipment. The owner of the DZ was angry at ME because I threw away the $50 cord and handle as I'd been trained for when pulling the reserve cord!

[grocuttt]

I've been jumping with various Cypres models for years, it's nice to finally see the inside of one (Thanks Dave). There's a few things I've heard over the years that people might be interested in. This is all from memory so I might have miss remembered some of the details.

Depending on the model they have to be serviced every 4 or 5 years. As well as replacing the battery they also perform any updates that are required. These updates can include both software and hardware, so expect the hacked on components were probably added during one of the services and were probably not there when the unit was originally manufactured. This is also the reason for the reenterable potting compound.

They do a LOT to eliminate failures. Obviously they test the hell out of it during production and limit the service life to remove the front and end parts of the failure rate bath tub curve. During each service they also retest them and compare the data to that units previous test results. The idea is that even if a unit is currently within spec it may be rejected because a measurement has drifted to much from its previous tests, which might indicate it could go out of spec before the next service is due.

I can't remember if it was a Cypres or a competing model, but I remember a case where there was a batch of pressure sensors that were faulty in a fairly subtle way. Because of all the testing that goes on they faulty sensors were detected before the sensor manufacture even found out.

There have been fatalities due to units firing when the user did have a good canopy, but the cases I know of weren't actually down to a failure of the Cypres. The way they work is they fire if you're travelling faster than a critical speed when you're below a critical altitude (78mph / 750ft for the expert model I think). The snag is that people started jumping really small canopies and putting them into a high speed dive just before landing. As they pushed the limits further and further some people got to the point where they exceeded the firing thresholds even though they had a good canopy. To address this there's now a "Speed" variant of Cypress that has different firing criteria.

[whatthefunction]

On my 1st jump I locked up mentally. They had to punch me to snap me out of it.

[brencostigan]

The foam will be there to reduce the exchangeable air volume to ensure the highest speed sampling of external air pressure. If they didn’t take this measure to exclude air by filling the voids then the air exchange through the pressure port orifice would render the pressure measurement too slow and inaccurate.

[johnfrancisdoe]

The UI box may be a standard Hitachi / clone LCD, a button and maybe some caps/filtering. All within the extended Faraday cage formed by the two metal cans and the shielding braid. If the red wire in the cable and the red wire soldered to the can are the same red wire, they might not trust the braid to remain connected for 12.5 years of packing parachutes, bumping around and jumping out of airplanes.

[TheSharkey]

User manual says unit is waterproof for 24 hours at 8ft depth. The unit here looks identical to the current model. Possibly differs internally though. Product is designed and made in Germany.

[eliotmansfield]

Wiring and assembly doesn’t ooze quality imo.

[whatthefunction]

I am an aficionado aficionado. I love aficionados

[TheRealSasquatch]

i'd love to sit through the code review for something like this.

[xXRocketDukeXx]

Dave, is it Possible for you to make video comparisson for different AAD (Vigil/m2)? Im sure there are many past service life and worthless lying around 🤔

[kylegreen]

The filter is typically only replaced at service time or by a certified rigger who can service it (this depends on the manufacture date nowadays, but I think it is typically every 4 years although I'd have to ask my buddy who owns SSK, the USA maintenance house for Cypress units). I keep a Vigil 2 in my rig, so I'm not 100% sure. I think it is usually replaced when the device is submerged in water (i.e. the skydiver biffs it and goes for a swim on a failed swoop over the pond or something of the like).

[christopherguy]

It would have been interesting to hook up a power supply and see it working first.