Hard off switch on the Hammer Head

Genesis

After the first reset its not a "might" any more.

How hard is this, really? Assuming there is some battery-backed persistent storage in this thing, you set it just before hitting the solenoid, and when you safely return from that action, you clear the flag in battery backed memory. If you find the flag set on boot, you have crashed due to the solenoid draw!

Only a crazy person hits themselves in the head twice with a hammer
Once. After the first reset atempting to fire the solenoid, a second attempt is known futile and foolish. IF you have other, lower-power options (e.g. the buzzer) then hit THAT instead after you get the first unsolicited reset during the attempted solenoid firing.
For how long? Another 30 seconds? If the situation has gotten that bad then there's an argument that the only chance the diver has is if he comes to his senses (assuming he can!) and takes action on his own. He can't do that with a dead handset, because he doesn't know what he has to do.

Odds are it won't matter, but continually rebooting leaves one with a blank or "locked" screen when you could at least be trying to get someone's attention.

If, for example, the solenoid requires 100ma to fire but the beeper only requires 20ma to make noise, then hitting the solenoid a second time after the first unsolicited reset may actually kill, as the beeper might be noticed and might not cause a reset.

Yes, I know we're talking about actions at the margins here, but IMHO it is indefensible under any circumstances to fire the solenoid after coming back from a reset caused by the solenoid being fired! In that case you already know that its not going to work out....

__________________
"A venturesome minority will always be eager to get off on their own, and no obstacles should be placed in their path; let them take risks for Godsake, let them get lost, sunburnt, stranded, drowned, eaten by bears, buried alive under avalanches - that is the right and privilege of any free American."
http://www.denninger.net
http://www.diversunion.org/liability.htm - Fix the Diving Cert racket

cys

Apart from the issue of solenoid O2 injection on start-up, it can be argued that the left handset is doing exactly what it's supposed to be doing in this situation -- trying to keep the diver alive in a low O2 condition by actuating the solenoid.

There are TWO handsets. The diver only has to look at the right handset to see the pO2. Further, the HUD is controlled by the unaffected right handset, so the diver is getting visible HUD warnings, and, if things are really dire, a tactile warning as well via the vibrating HUD alarm.

Like others who have posted to this thread, I really like my HH a lot. It's a beautifully designed system in so many ways. It's worth reading the manual even if you don't own one...

silent running

Joe, the first part of your point seems based on the assumption that the proper way to fly an ECCR is manual with the electronics as a back-up, a position I don't agree with. If I wanted an MCCR, I would've bought one. And I'd bet that the KISS units with their CMF will do better job of maintaining SP than somebody trying to do so by pushing the man inject on their ECCR once in a while, not to mention that any decent ECCR should be able to maintain SP better in all conditions than an MCCR, but I digress...

My main point is that if you are able to read your PO2 on the 2dry, you are concious, thus able to manually inject and the solenoid is mostly a convenience. So in that situation, I'd rather have my deco info for as long as I could instead of the solenoid, not to mention the problem of soleniod firing causing more reboots as it draws power down. If you are unconcious and in a low power situation, whether from poor battery contact or a dying bat, it's probably too late anyway, and if you're unconcious due to an O2 tox, the unit continuously rebooting/injecting is certainly not going to be helpful in bringing you around. I can see some merit in the unit trying to maintain at least say .2 PO2 if there's enough power to fire the solenoid for an unconcious diver scenerio, but surely the solenoid need not fire if the loop has more than .2, even when rebooting. If I had to choose btw solenoid only and no solenoid in low volt conditions, I would take the no solenoid and display/deco option as I think the low volt/concious diver condition is much more likely than unconcious diver/low voltage. No matter what, the firing of the solenoid when rebooting underwater before the controller sees the PO2, is not good and needs to be fixed...

jradomski

no your not seeing things corretly.. My comment wasn't during a normal condition my comment to genesis is once the abnormal condition has been entered.. at a point where the voltage is too low to properly control the solenoid, and has progressed from a low condition (below the target) to a threatening condition.

I am not disagreeing about the firing during reboot issue either.. once its been rebooted I dont see a problem having the solenoid try to fire, but it should only fire if the po2 is too low not as part of powerup sequence.. My feeling (knowing the hardware) is that the measured po2 is not valid when it gets to the point to see if the solenoid needs to be fired for the po2 being too low.... one of these days I will probably dump the code out of the handset and disassemble itnand see what exactly is going on..

__________________
Joe Radomski
CCR Trimix Instructor Trainer
ANDI Instructor Trainer Director #10

All posts are personal opinions and DO NOT reflect any affiliated agency unless specifically stated.

Genesis

Cys: I've read the manual and there is a lot to like about the HH.

However, I still disagree with some of the design decisions, particularly when put up against the generalized position that the designer has taken of "covering the ways you can kill yourself with this thing to the best of its ability."

The "parachute" idea has merit BUT the inability to shut the unit off when it sees water (but not depth), thereby leading to batter issues, which can then lead to a reboot/injection loop, is IMHO a fairly clear violation of the claimed principles behind the design.

Thre is no particular reason not to allow the diver to select "off" on the handsets (or even DEFAULTING to that status) if the unit is not under pressure. Waking up every half-minute or so just long enough to sample the depth sensor and O2 status - and if less than some modest depth under water (e.g. 3-5') AND PO2 > .20 immediately going back to sleep - even if the wet contacts are closed - doesn't do harm and (mostly) avoids the genesis of this problem. You need only be powered up for a couple hundred milliseconds maximum to allow the ADCs to settle and take your readings. If you sample every 30 seconds, say, then you've cut power consumption in "standby" to 1/150th of its otherwise "idle" amount! This is, basically how computers like the VR3 do it. Indeed, they take the view that you must FORCE them on for an extended period of time, otherwise on the surface they will attempt to conserve as much power as possible.

If you do THAT and then implement both "do not fire the solenoid on power up if submerged" AND "do not fire again IF you rebooted while trying to do so" then the problem goes away. All of this is easily implemented in software, with the only potentially difficult part being knowing if you reset on the last firing (you can'd use NVRAM for this, you need SRAM - something battery backed - because NVRAM has a limited number of write cycles possible - or an accessible processor flag to indicate if a reset occurred due to a brownout.)

I understand that the HUD and secondary should have alerted you to the problem already but here's the deal - if that's the case then why would you care if the solenoid tries to fire in this situation, since by definition you almost have to be unconscious (yet still on the loop!) AND in a low-power situation you're unlikely to actually be fixing anything (you might get away with it once or twice, but when you run out of power you do!)

An automated parachute actuator that fires when you're 20' above the ground doesn't change the outcome - you still bounce.

Joe - you may be right - the problem may be that the "fire solenoid" portion of the code gets entered before a valid PO2 is present, and if RAM is intentionally wiped to zeros on boot, the system would think that the PO2 was in fact zero - which would lead to the solenoid firing..... if so that (failing to insure that things initialize in the proper order on system start) is a fairly serious logic error. If that is the case then IMHO a recall is in order to update the code and fix it.

__________________
"A venturesome minority will always be eager to get off on their own, and no obstacles should be placed in their path; let them take risks for Godsake, let them get lost, sunburnt, stranded, drowned, eaten by bears, buried alive under avalanches - that is the right and privilege of any free American."
http://www.denninger.net
http://www.diversunion.org/liability.htm - Fix the Diving Cert racket

jradomski

There is a good reason.. surface swims... If you are in the water on the loop doing a surface swim you dont want the unit switching off.. I definately dont want the po2 dropping because the unit shut off, since the loop volume will decrease may become insufficient, also if it doesnt (get to the pont you cant take a breathllike with an ADv) you run the risk if the po2 dropping too far (especially with an ADV and hypoxic diluent) before the next "wake up" interval is reached..

__________________
Joe Radomski
CCR Trimix Instructor Trainer
ANDI Instructor Trainer Director #10

All posts are personal opinions and DO NOT reflect any affiliated agency unless specifically stated.

Genesis

Why not make it selectable?

You still have the parachute..... how long does it take to "wake it up"? The ADC I'm using is "alive" within tens of milliseconds (30ms or so from a cold start) so if you were to sample every 10 seconds, for example, you would have a duty cycle of 0.3%, which extends battery life by a factor of 333X. Call it 300x since there's likely some overhead in getting things "woke up" - but you don't have to turn on things like the LCD unless you're going to actually go "active" (assuming he's got soft control of all the various other items individually.)

Further, if you DO detect a hypoxic loop you then inject and stay on for some period of time to validate whether someone really IS breathing the loop (that is, is the PO2 changing or was this a "one time" deal?)

I gotta put the mouthpiece in my mouth - as soon as I do the first interval that fails the check causes the unit to wake up, inject and STAY ON so long as a changing loop PO2 continues to be detected.

One of the original commenters on this thread noted that he had run into excessive battery drain due to the unit staying "on" when on boat rides due to the handsets turning themselves on from ambient spray on board. Sounds like a legit issue - and one that can be addressed - to me.

__________________
"A venturesome minority will always be eager to get off on their own, and no obstacles should be placed in their path; let them take risks for Godsake, let them get lost, sunburnt, stranded, drowned, eaten by bears, buried alive under avalanches - that is the right and privilege of any free American."
http://www.denninger.net
http://www.diversunion.org/liability.htm - Fix the Diving Cert racket

cys

Genesis, I really can't find fault with your comments about the design in your last two posts. At the same time I appreciate the design goal behind the wet sensor and the lack of a hard off. If the system could be made to only partially wake up as per your last post and still detect a hypoxic loop with an actively breathing diver, that would be cool.

decoweenie

Wasn't that the cause of the Cis fatality in UK ?

__________________
"...after a while you get bored offering advice to a bull that like to keep butting the fence with its head rather than walking through the open gate..."

- Rebreather World PM

silent running

Then we agree about that part, that's why at the end of my last post I said that the controller firing the solenoid as part of the reboot, before seeing the PO2, is no good...