Rebreather Guardian HUD on Classic Kiss

The "KISS Version” of Uri’s Rebreather Guardian HUD
By Neil Sutcliffe


Uri Baran's HUD is available at www.underwaterexperiences.com.



When I was researching RBs with a view to picking one to buy, I thought how nice it would be to have the KISS with a HUD.

Having dived my (unmodified) KISS for a few months following my Mod1 course I was (and am) very happy with it – I like manual control, like the triple redundancy on the displays, and the whole KISS concept, I wouldn’t want to change the philosophy.
I haven’t had many problems with the unit and the few I’ve had have been minor.

I have had a couple of instances where my PO2 got much lower than planned. Entirely my fault and I’m normally very PO2 vigilant, but it can happen and I suspect most KISS divers will have had a similar experience sometime.

A combination of being colder, higher work load, higher task load, being above minimum loop volume, basal flow lower than ideal, and a little narked (air dil) – then you check the displays after a bit longer than usual, and there it is – low PO2! In my case, the lowest I’ve had was still above 0.8 – not the end of the world but a lapse in discipline, and I’ve tried to make damned certain not to let it happen again!

Having got reasonably comfortable with the unit, I still felt a HUD would be useful. So started a series of emails between Uri Baran and myself about his HUD, the Rebreather Guardian (RBG) – Uri very patiently answered all my questions and concerns about the unit so I bit the bullet and committed to it.

There were a number of reasons I favoured Uri’s unit:

1. Three independent readouts of PO2 – 1 for each cell
2. Circuits each independent except for the battery
3. Ability to run the unit without reference to handsets in normal use
4. Even shorting out the HUD connections the KISS displays work OK
5. User-specifiable signal sequence
6. Ability to have a buffered cell 1 output to a VR3 and keep 3 handsets
7. Virtually no additional loading on the cells

For me 1 and 2 were the main deciding factors, very much compliant with the way I am used to the KISS working, independent readouts, with me getting all the information and using it exactly as I would with the KISS handsets, just with an immediately visible and attention-grabbing light signal.

The next step was to work out how to connect the unit into the KISS, ideally without having to drill holes in my KISS. I had heard from Uri that Dave Thompson had had the RBG connected to his KISS to check the concept, and that he had connected using the spare O2 add port on the left of the scrubber head.


Kim at Jetsam confirmed that the thread for the port was 1/8 – 27 npt and having searched unsuccessfully for a cable gland with a 90 degree elbow Uri and I decided on a ¼” Swagelok elbow intended for hose connections – basically the same sort of connection as the O2 connection into the scrubber head on the other side. (Swagelok part number SS-400-2-2 from www.swagelok.com.

I sent this off to Uri who connected a cable with a Fischer connector into the elbow, he used a stainless washer and grommet to fashion a gland – and then potted the whole thing for good measure.

Uri’s unit allows the customer to specify a signal sequence with a high degree of flexibility. Each cell has its own display consisting of a green and a red LED (see Mark Chase’s report on the RBG).

My signal sequence was as follows:

• PO2 < 0.05 All off
• 0.05 – 0.30 Red fast flash
• 0.30 – 0.60 Red slow flash
• 0.60 – 1.10 Red & Green slow flash together
• 1.10 – 1.25 Green slow flash
• 1.25 – 1.35 Green solid
• 1.35 – 1.45 Green solid with off flicker
• 1.45 – 1.60 Green and Red solid with off flicker
• > 1.60 Red solid with off flicker

The rational for this was that when all is OK it shows solid green, if a bit high or low a green-only signal, if significantly away from desired PO2 red with green, if needs immediate attention then red-only signal. Flash means low, flicker means high. We adjusted the on/off ratio for the flicker to make it distinctly different from flash and the flicker is more attention getting than e.g. a solid red signal. These ranges also mean I could elect to run the unit at 1.35-1.45 for deco with a not too distracting signal.

The next thing was a trip down to see Uri for us to fit the Swagelok connector into the head and fashion a wiring loom to connect the RBG to the cells, this was remarkably straightforward once we had worked out a good way to do it and produced a neat wiring loom (See photos). The cables from the Swagelok fitting (ribbon cable) divide into cable pairs which route through the small holes next to the O2 cell apertures without blocking them (these allow equalisation of pressure above and below the cells). Once the cables are routed to above the cells, they connect in with a simple T-piece which connects both the KISS display and the RBG but could easily allow the RBG or KISS displays to be disconnected if required.



A nice bonus was that we had not needed to drill or modify the original structures of the KISS so I could remove the unit tomorrow and restore the KISS to its original state. I am very unlikely to do this (unless I ever buy or build a different rebreather and want to sell the KISS).

All that then remained was to mount the RBG box to the KISS somewhere and route the HUD cable with the display on the end to the DSV – Uri’s standard mounting for the display fits the KISS DSV perfectly. I’ve not finalised the mounting position for the RBG box yet but currently it is tucked into the bottom of the counterlung case just above the O2 1st stage (I have a 2l lung on the exhale side so there is plenty of room).


A try dive was the next step so I took to the balmy waters of Dorothea Quarry only to discover an O-ring was missing from one of the Fischer connectors – it was pretty obvious that all was not well as the HUD displays varied wildly and made no sense. This did however confirm one of the great design features of the RBG – even if the cable from the cells to the RBG unit is shorted out it does NOT affect the reading on the KISS displays as high value resistors in the wiring inside the head protects the KISS displays from the short, so at least I got a dive!

Once I’d inserted the missing O-ring in the Fischer connector all was well and a return trip to Dorothea made me a happy man. Even after just one dive with the HUD I was making the transition from diving using the KISS displays and watching what the HUD was doing, to diving using the HUD and checking it against the KISS displays every so often.

Calibrating the unit is a doddle and just involves pressing a button when the cells are flushed with 100%O2 when you are calibrating the handsets, confirmed by a signal from the HUD display.. The process was made a bit more awkward for me by the way I had the unit inside the counterlung case making it harder to reach the calibration button, but I should be able to improve on this as it was only a temporary lash-up. It also signals on power-up if the battery is getting low, the unit is supposed to be OK for about 48hrs operation on a full charge.

The really good news at this stage was that I was about to go to the Red Sea on a liveaboard (MY Juliet through Oonasdivers) for a week, so the system would get a really good check-out.

After a week in the Red Sea, I can confirm that the RBG worked perfectly and I would now be reluctant to go back to HUD-less diving. It is so easy to monitor PO2 without having to look at your wrist displays frequently. The HUD alerts you to dropping PO2 and the need to add O2 and you can easily add to bring PO2 back into desired range. No need to check handsets, just a quick squirt of O2 and check it settles back in solid green.

I’d be reluctant to say you don’t have to consciously check it, especially in bright light (6m on a Red Sea reef in full midday sunshine), but in normal or dim light conditions it would be very hard not to notice the LED flashing green, and almost impossible to miss green & red together! The red LED being above the green make it even more obvious as visually the signal moves up and really does get your attention fast. I can’t imagine UK waters will ever be bright enough for the HUD to ever be anything less than obvious!

I suspect, though haven’t tried it yet, that I would be able to read the signals even if I had lost my mask and had no spare.

On several occasions I noticed 1 LED on slow green flash (1.10 to 1.25bar) and wondered how long it had been flashing but checking the handsets showed that it must have only just started and in fact I’d noticed it immediately, the flashing cell reading marginally under 1.25 and the other 2 just over. The HUD display is a little bright on night dives but this could be easily sorted with a sleeve with a grey glass/plastic lens over the HUD to dim it.

Diving with a HUD is fantastic, particularly if your hands are occupied with other things, I was doing a little photography and it is great to be able to check your PO2 with just a minor shift of gaze, while holding a camera in position waiting for that once-in-a-lifetime shot, same would apply for things like reeling up an SMB line, laying line, scootering, etc. Would also be the case for busy stressful events like rescuing another diver.

On most dives I now only check the handsets at the start, after descent and raising to working PO2 (1.3), and once in a while during the dive to confirm. In fact a couple of days before my trip finished, I managed to flood a KISS handset and had no soldering iron to replace the defunct electronics. I sealed the display housing up again (flooded handset could affect the HUD display by shorting the cell) and ran the unit perfectly happily with 2 handsets and the 3 HUD displays. I’d be fairly happy to dive the unit with just the RBG and no handsets.

Uri has been excellent throughout the whole process, putting up with my inundating him with emails, and being at pains to ensure that the whole system was right for me – I would like to thank him wholeheartedly for his commitment and customer service.

You can probably tell I’m pretty happy with the RBG. It is very much in keeping with the original concept for the KISS – manual control, independent displays for each cell (the 3 channels are independent except for the battery), with the user making decisions based on the display rather than a computer with “voting logic”. It’s just that the display readout is a light sequence rather than a numerical display, and tends to grab your attention when out of range, rather than waiting for you to check.

Refinements Made
Added 7 February 2006

After using Uri’s HUD for a while as described in the article, I realised that a few improvements could be made. Uri and I had anticipated this and so the unit’s programming port had not been finally sealed, allowing us to make a few “tweaks”.

Firstly I opted to expand the solid green/all OK PO2 range very slightly down to 1.22, here is the current ranges (all in bar) and corresponding signals:

> 1.60 Red solid with off flicker
1.45 – 1.60 Green and Red solid with off flicker
1.35 – 1.45 Green solid with off flicker
1.22 – 1.35 Green solid
1.10 – 1.22 Green slow flash
0.60 – 1.10 Red & Green slow flash together
0.30 – 0.60 Red slow flash
0.05 – 0.30 Red fast flash
PO2 < 0.05 All off

The most commonly occurring niggle related to O2 addition, when O2 is added manually to the Kiss there is a brief peak in the PO2 as the O2 rich gas passes the sensors before it has mixed thoroughly. Kiss users are accustomed to seeing the PO2 on the handsets climb quickly when they add O2, and then settle back down to a steady value.

Although these PO2 “spikes” are expected and do not seem to be a problem they do briefly put the PO2>1.35, and sometimes >1.45 for a second or two (note that the real PO2 blip may be a bit longer as the cells take a few seconds to respond).

These high PO2 “blips” were reflected in the HUD’s display, so when adding O2 I would get a brief “high PO2” signal, i.e. solid green with off flicker, or green and red solid with off flicker.

I was concerned that I would get used to seeing these signals when adding O2 and would be less likely to notice/react to warnings of true high PO2. However, I obviously still wanted to be warned about high PO2 levels promptly.

After discussing a few ideas between Uri and myself, we agreed a nice way to solve the problem – here’s how it works…

When the unit detects a rising PO2, going through the 1.35 threshold it starts a delay period of 3.5 seconds during which it does not change the LED display. This allows a brief spike of PO2 above 1.35 or above 1.45 to not be signalled by the unit, however after the 3.5s period the PO2 is displayed again (usually it has fallen back down to <1.35 by then). As a safety feature if the sensed PO2 rises to >1.6 then the delay is aborted and the actual PO2 displayed immediately, this means that a rapidly rising PO2 (e.g. due to O2 free-flow) would not be significantly delayed in being displayed.

This has worked very well in real-world use. I no longer get an alerting display from the HUD when adding O2 unless I overdo it significantly, but it signals to me promptly if PO2 goes high.

The second area we addressed was similar but at the low PO2 end of the normal range. As I try to dive my rebreather at minimum loop volume I occasionally manage to crack the ADV and add a tiny bit of diluent (air in my case) as I breathe down the loop volume (usually as I am just about to add O2). As the diluent is added on the inhale side of the loop and very close to the cells it is immediately sensed by the cells and therefore the RBG HUD, and results in a “low PO2” type signal e.g. green or red and green slow flash (see above).

We therefore devised a similar system to the high O2 blip ignoring system; a down-going signal passing through the 1.22 threshold is ignored for 4 seconds, unless the PO2 signal drops below 0.6 or rises back above 1.22 in which case the delay aborts and the actual PO2 is displayed. As a result if I add a tiny bit of air diluent with the ADV I don’t get an alert unless PO2 stays low, generally the sensed PO2 will be back up as the next inhale brings higher PO2 gas over the cells. I still get a low display (red flash) if I add diluent shallower than about 18m as the PO2 of the air diluent is then <0.6, hardly a problem in the real world – I’m usually on ascent then anyway and dumping loop gas and adding O2 rather than adding diluent.

The features described are of course implemented independently in each of the 3 channels of the RBG HUD unit (3 separate circuits).

I have dived the unit a fair bit since these changes have been implemented (though we’ve only just changed the up-going “ignore period” to 3.5 seconds, it was 3s initially) and they work very well, I have done my best to catch the unit out e.g. by dropping PO2 to 1.0 and then raising it very rapidly to >1.6, or raising PO2 to 1.5 and then adding diluent from the ADV, and the display has always been appropriate.

The lack of “false alarms” due to expected brief deviations of PO2 from the set point avoids them being distracting and more importantly means that I am unlikely to ignore real alarms.

I’m even more pleased with the unit than ever!

Neil Sutcliffe



Uri Baran's HUD is available at www.underwaterexperiences.com.

Discuss this article HERE