The practical half: how to make a settings conversation actually go somewhere.

• Bring weeks, not nights. One bad night is a story; four weeks of the same pattern is evidence.
• Bring the pattern in one sentence. "Pressure sits at the top of its range for the last two hours and my events cluster there" is something to work with. "I think my pressure is too low" is an opinion, and it will be met with one.
• Say what you feel, not only what you measured. Bloating, waking gasping, a mask you fight — those are clinical information too, and they are frequently the thing that changes a decision.
• Ask what would need to be true for a change to be considered. It converts a no into a plan.
• One change at a time, then wait. Whoever makes the change, changing several things at once and judging it after two nights teaches you nothing. This is the oldest rule in this hobby and it is the most ignored.

And if you genuinely cannot get anyone to look at your data — which happens, and is a real problem rather than a personal failing — that itself is worth saying out loud to your GP or family doctor, who can refer.

Nothing on SomniTalk is medical advice. Talk to your own clinician before changing therapy.

Worth adding, because this is one of the leading reasons people quietly give up on therapy and one of the least discussed at appointments.

Swallowing air is not a sign you are doing it wrong. When the pressure needed to hold your airway open is close to what it takes to open the top of your oesophagus, some of the air goes to the wrong place. The result is bloating, belching, wind and abdominal discomfort — sometimes bad enough to wake you, and frequently bad enough to be blamed on food.

Things people report helping, none of which are settings changes:

• Sleeping position. Raising the head of the bed, or side-sleeping rather than flat on the back, changes it for a lot of people.
• Not eating late, and being careful with fizzy drinks and alcohol in the evening.
• Fixing a mouth leak, if there is one — swallowing tends to be worse when air is moving through the mouth.
• Time. For a fair number of people the first weeks are the worst and it settles as they adapt.

Three questions arrive under this heading constantly, so here is the experience half of each — not a recommendation, because which machine is right for you is a prescription question.

"Is the newer model worth waiting for?" Between generations of the same manufacturer's machine the therapy is usually much the same; what changes is the interface, the connectivity, the humidifier design and how you get at your own data. If the data route matters to you, that is the specification to compare, and it is the one the marketing pages say least about.

"Should I push for a bilevel or an ASV?" Those are different treatments, not upgrades. They are prescribed for specific findings — pressure that cannot be tolerated, ventilation problems, or central events that do not settle — and the route to one is a clinician looking at your data, not a purchase decision. What you can usefully do is arrive with the data that would prompt the question.

"Can I buy used, or without a prescription?" The rules differ by country, and a second-hand machine carries two practical risks that people underestimate: hours of use you cannot verify, and a device whose settings someone else chose. If you go that way, the hours counter and the settings are the first two things to check.

The comparison question this article gets asked most often, answered from the data side rather than the sales side.

Between a machine generation and its successor, the therapy algorithms are usually close relatives and the detailed data written to the card is broadly the same shape — so if you already know how to read one, you are not starting again. What tends to change is everything around it: the screen, the setup flow, the humidifier and tube fittings, the cellular reporting, and sometimes whether a card slot is present or populated at all.

Two practical notes for anyone buying, new or used:

• Check that the card slot is there and that a card is in it. Some units ship with the slot empty, and a machine that is not writing to a card is not keeping the detailed record you will want later. A card that has never been in the machine holds nothing retrospectively.
• On a used machine, check the hours counter and the current settings before anything else. The hours tell you what you actually bought; the settings tell you what somebody else was prescribed, which is not what you were.

Nothing on SomniTalk is medical advice. Talk to your own clinician before changing therapy.

Power. Almost every current machine takes 100–240 V and switches automatically, so a plug adapter is usually all you need abroad — check the label on your own power supply rather than trusting that. A voltage converter is generally not needed and is one more thing to fail. Take the original power supply; the machine is fussier about it than a phone is.

Off-grid. Battery packs exist for this and are sized in watt-hours; the humidifier and the heated tube are what drain them, so many people run the machine dry on battery nights and accept one dry morning. Work out the numbers before the trip, not at the campsite.

Altitude. Machines compensate automatically within their stated range. If you are going somewhere genuinely high, that is worth checking against your own machine's documentation.

The card. A trip is the classic case for uploading the whole card when you get home — only the new nights import.

There is, and it is usually the tube rather than the humidifier.

In rough order of how often it works:

1. Use the heated tube if you have one, and raise the tube temperature rather than lowering the humidity. Warming the tube keeps the air above the point where it sheds water, so you keep the moisture and lose the condensation. This is the fix most of the time.
2. Insulate the tube if you do not have a heated one — a fabric tube cover costs very little and does the same job passively.
3. Get the tube off the cold floor and out of the draught. Running it under the duvet, or over the headboard rather than down the side of the bed, changes more than people expect.
4. Only then lower the humidity, one step at a time, and give each step a couple of nights. Dropping straight to zero is what produces the desert.
5. Warm the room slightly, if that is an option. Rainout is far more common in winter and in cold bedrooms for exactly this reason.

Two extras worth knowing:

• Water level does not change the humidity output — the machine humidifies to a target; a fuller chamber just lasts longer. Filling it to the top does not make it worse.
• Dry mouth that survives all of this is often a leak story, not a humidity story. Air escaping through the mouth all night dries it out no matter what the humidifier does.
• Use distilled water where you can, and empty and dry the chamber daily — it is the least glamorous and most useful maintenance habit there is.

The four options, honestly, with what each one is actually good at.

1. Fix the nose first. If your nose is blocked, your mouth opens because it has to. Congestion, allergy, rebound congestion from long-term decongestant sprays, a structural problem — all worth dealing with before buying anything, and all a clinician's question.
2. A chin strap supports the jaw so it does not drop. It does not seal lips: if you breathe out through parted lips with your jaw closed, a strap changes nothing. Cheap, no risk, worth a try, not a guarantee.
3. Mouth tape does seal, and for some people it is what finally works. The caveats are not decoration: never tape if your nose is blocked, if you have been drinking, if you are nauseated or prone to reflux or vomiting, if you take anything sedating, or if you cannot comfortably breathe through your nose while awake. If you have any condition affecting breathing, swallowing or consciousness, ask your clinician first. Nobody should be talked into this by a thread.
4. A full-face mask removes the problem instead of managing it, by pressurising the mouth too. Bulkier, its own leak profile, and for a determined mouth-breather often the shortest route to a good night.

What this board will not do is tell you which of those to pick, or tell you to change a pressure setting to compensate for mouth leak. Comfort hardware is yours; prescribed settings are not.

This is a standing argument on every board, it generates strong opinions, and it has a genuine safety dimension that gets skipped when it is argued rather than explained. So: the even-handed version, with the caveats attached rather than in a footnote.

The problem itself is simple. With a nasal mask or pillows, if your mouth opens in the night the pressurised air takes the shortest way out. You lose some of the therapy, you dry out, and the leak can be steady enough that the nightly average barely notices.

Rainout is the gurgling, the spitting, the cold splash on the face at 4 a.m., and the puddle in the tube in the morning. It is not a fault. It is condensation: warm humid air from the humidifier meets a cooler tube or a cooler bedroom and gives up its water on the way.

That one sentence contains every fix there is. You either put less water in the air, or you stop the air cooling before it reaches your face.

Happily — the answer is usually drift, which is this desk's whole subject: something that changed too slowly for anybody to decide it.

Pick a month when you felt well and the most recent month, and compare these, in this order:

1. Usage. Total hours, but more importantly the shape. Nights that end at 4 a.m., or start an hour later than they used to, change how you feel far more than a tenth of a point of AHI.
2. Leak. The trend, not the average. A cushion that is slowly giving up produces a leak line that creeps rather than jumps, and the creep is invisible night to night and obvious across months.
3. Pressure. If your machine adjusts itself, compare the pressure it settles at across the two months. Rising demand is a real signal — one for the person who prescribed it to interpret, not for you to act on.
4. The event mix, not the event count. The same AHI made of different things is a different night. A month where the obstructive share shrank and the clear-airway share grew is a genuinely different situation.
5. Everything outside the machine. Weight, a new medication, alcohol timing, seasonal congestion, a colder bedroom, more back-sleeping. Winter is no coincidence above.
6. The parts — where most "nothing changed" cases land. A cushion softened past the point where it seals at your pressure; stretched headgear; a grey filter; a film in the chamber; a hose split at the cuff. All of it has a working life, none of it announces itself, and nobody notices the day it ends.

Then take the comparison, not the complaint, to your clinic. "These two months, this is what moved" is a five-minute appointment; "I feel worse" is a twenty-minute one that often ends where it started.

Answer these six and the choice usually makes itself.

1. Can you breathe through your nose all night, every night? If not — chronic congestion, a structural problem, bad seasonal allergies — a nasal mask or pillows will fight you, and a full face is the pragmatic answer until the nose is sorted.
2. Does your mouth fall open in your sleep? If it does and you would rather not use a strap or tape, a full face sidesteps the argument entirely.
3. Do you sleep on your side or move a lot? Pillows and small nasal masks have the least to catch on the pillow; a big full face has the most. This is the biggest comfort difference for side sleepers.
4. Beard? Not disqualifying, but a full face has a chin seal running through it. Nasal masks and pillows dodge that entirely; some people trim a channel, some use a liner.
5. Do you read, or use a phone, or wear glasses in bed? Masks that sit low on the face or route the tube over the forehead leave your field of view clear; a big full face does not.
6. Does the idea of a mask on your face make you tense? Claustrophobia is common and it is a real factor. Pillows and minimal-contact nasal masks are usually the gentlest starting point.

Then the practical bit: ask about the fitting or exchange window before it runs out. Most suppliers will swap a mask that does not work, and most people never ask. Fit it lying down in your own sleeping position, and give a new mask three or four nights.

Adding the hardware half, since point 2 is where the first month is actually spent.

The things that most often turn out to be the problem, roughly in order of how often:

• Size, not tightness. Overtightening a mask that is the wrong size is the single commonest mistake — it deforms the cushion and makes the leak worse, not better.
• Where the tube sits. A tube pulling sideways off the pillow drags the seal open every time you turn over. A hose lift, a clip, or just running it over the headboard fixes more leaks than a new mask does.
• Mouth leaking with a nasal mask. Air escaping through the mouth is its own problem and has its own solutions — worth diagnosing before buying different hardware.
• Dry mouth and a blocked nose, which are often the same story as the leak above rather than separate complaints.

Suppliers will usually let you try a different mask within a fitting period. Ask.

A full-face mask stops air escaping around your mouth. It does not stop air escaping past a seal that is not sealing, and it does not stop the room drying you out. In this order:

1. Leak at the mask, especially at the mouth end. Full-face masks leak most at the chin and the corners of the mouth, and worse if the cushion is old or the mask is a size too big. The tell is that it is fine when you fall asleep and not fine at 3 a.m.
2. Use the heated tube instead of more humidity. Raising the tube temperature breaks the desert-or-swamp trade-off: the air arrives warm enough to hold its moisture rather than dropping it in the hose. Raising humidity alone is what causes the splash.
3. If you have no heated tube, insulate the hose — a fabric cover costs very little — and get it off a cold floor and out of the draught. Rainout is a temperature problem before it is a humidity problem.
4. Then look at the room. A dry bedroom, especially in winter with the heating on, will dry you out whatever the machine does. Some people fix this with a room humidifier and never touch a machine setting again.
5. Then look at your medications. A long list of very ordinary drugs cause dry mouth — some blood-pressure medicines, antidepressants, antihistamines, diuretics, and more recently the appetite drugs. If the dryness started when a prescription did, that is a question for the prescriber.
6. And the ordinary things. Water through the day rather than a glass at bedtime, alcohol close to sleep, and daytime mouth-breathing out of habit all show up in the morning.

If you have worked through all of that and it is still bad enough to damage your teeth or wake you, say so to your clinician and to a dentist. Persistent dry mouth is a treatable complaint in its own right.

The second flavour, and the more annoying one, because the usual advice has already been taken. Anonymised:

"I switched to a full face mask specifically because of dry mouth. It is better than it was, but I still wake up dry, and if I push the humidity any higher I get water in the hose and a cold splash in the face. I seem to be choosing between a desert and a swamp."

That sentence — desert or swamp — is the actual problem, and there is a way out of it that does not involve picking one.

The follow-up question this always produces, so let us pre-empt it: after the import succeeds, that machine shows up as a BMC device rather than under the name on the box.

That is correct and it is your machine. The Luna G3X is manufactured by BMC Medical and sold under a different brand; internally the device identifies itself with the manufacturer's model name, and the app reports what the machine says about itself rather than what the retail box says. Your data, charts and settings are complete — only the label differs.

Worth knowing for the same family of machines: their cards use a newer format that some older desktop tools, including released OSCAR versions, cannot read at all. If you have been told your machine's data is unreadable, that may be why.

Work through these in order. It is close to the order of how often each one turns out to be the answer.

1. Cushion age. Silicone softens and stops holding shape, gradually enough that nobody notices the day it stops working. If your cushion is months old, replace it before doing anything else — that is the "nothing has changed" case, and the commonest one.
2. Size, not tightness. Over-tightening the wrong size makes leaks worse: the cushion deforms and folds. If you are cranking the straps to get a seal, the size is wrong. Fit it lying down, not sitting up in a shop.
3. Tube routing. A hose that pulls sideways off the pillow drags the mask open every time you turn over — that is the roll-over case. Running it over the headboard, or using a hose lift or a clip, fixes more leaks than a new mask does.
4. Pressure late in the night. If the machine adjusts itself, it is often at its highest when you are deepest asleep and your face has relaxed. Blowing off at higher pressure is a seal problem, not a reason to go looking at settings.
5. Skin and hair. Moisturiser or oil at bedtime, a beard growing out, and hair caught under the strap all break seals.
6. Mask-off periods. If the chart shows the mask coming off entirely rather than leaking, that is a different thread — usually comfort, congestion, or an arm doing it in your sleep.

If none of that moves it, the mask may simply be the wrong shape for your face. Most suppliers have an exchange window; the sooner you ask, the more likely it applies.

Leak is the most visible way therapy fails, and it blocks everything else — so it is worth a thread of its own. Three anonymised versions of the same message, because the differences matter:

"My leak has been fine for months and this week it has doubled. Nothing has changed."

"It blows off my face when the pressure goes up, always in the second half of the night."

"It only leaks when I roll onto my side, and then it wakes me up."

Those are three different faults with three different fixes, and telling them apart is most of the work.

With a nasal mask or pillows, waking up parched is usually not a humidity problem at all. It is your mouth falling open in the night: air goes in through the mask and straight out through your mouth, taking the moisture with it. A humidifier cannot win against a one-way breeze — which is why turning it up did nothing.

Worth working through in this order:

1. Rule out the nose first. If you cannot breathe through your nose comfortably, your mouth will open — that is reflex, not willpower. Congestion, allergies, a deviated septum and rebound congestion from decongestant sprays all live in this box, and several are treatable. Fixing the nose fixes the mouth surprisingly often.
2. Then humidification, properly. With a heated tube, raise the tube temperature rather than only the humidity — warmer air carries moisture to your face instead of shedding it in the hose. One step at a time, two or three nights each.
3. Then the mouth itself. A chin strap supports the jaw; it does not seal lips, so people who breathe out through parted lips often find it does nothing. Mouth tape does seal, works for some people, and carries real safety caveats — it has its own thread on this board.
4. Or change the mask class. A full-face mask sidesteps the argument by pressurising the mouth too. Bulkier, its own leak problems, but often the shortest route for a determined mouth-breather.
5. Protect your teeth while you sort it out. Chronic overnight dry mouth is a dental problem, not only a comfort one — saliva is what protects enamel and gums overnight. A dentist is the right person for that half.

Two things not to do: do not stop using the machine to escape the dryness, and do not chase pressure settings for it. Dry mouth is a leak-and-humidity problem; pressure belongs to your prescriber.

From the kit side, because two items on that list are ours — and because of what the same people write a month later.

When somebody at week three does have a fixable problem, it is nearly always one of two: a mouth that falls open once the room gets dry, or a cushion half a size out that seals on your back at 11 p.m. and gives up at 3 a.m. after you have rolled over. Neither announces itself as a dramatic number.

The follow-ups agree on something else. Improvement gets noticed sideways — you stop falling asleep in front of the television, or an afternoon passes without the wall. Very few say "I woke up transformed"; a lot say "I did not notice until I looked back".

One hardware footnote, because it costs people an evening.

You need a card reader that actually reads the card, and on most machines the card is a full-size SD card sitting behind a flap. Take it out with the machine switched off, and put it back the same way. Removing the card does not erase anything and does not harm the machine — the machine writes to it again the next night.

Do not rename, tidy, move or "clean up" anything on the card before uploading. The folder names are the importer's map. If you have already tidied a card, no harm done — put the files back where they were, or upload the card again after the next night's recording.

For the brand-by-brand background, the shelf has an overview topic: CPAP Machines & Devices, and a walkthrough of getting the data off the card: How to Download CPAP Data from Your SD Card (ResMed, Philips, Löwenstein, BMC, Hoffrichter).