AMS2 Pro humidity consistently measures too low

But then, do we really need a more accurate measurement for this application? Will it actually provide any practical benefit that saves the day?

What I was trying to explain is why relative humidity (RH) is a poor indicator of how much water vapor is present in the air—because RH is always relative to the ambient temperature at the point of measurement.

If our goal is to understand how dry the air inside a container truly is, it would be far more useful to display absolute humidity (AH) instead of RH.

Let me give a practical example from my own equipment:

One of my dryers shows 10% RH at 40°C, while the other shows 13% RH at 35°C. Which one would you consider to be drier?

If you’ve followed the reasoning above, it should be clear that the unit showing 13% RH at 35°C is actually slightly drier than the one showing 10% RH at 40°C—despite the higher RH reading.

One final point to add to this thread is that all sensors are consumables, and RH sensors are no exception. When exposed to elevated temperatures over extended periods, they tend to lose accuracy relatively quickly.

A well-known example is the SHT3x series, which has a recommended maximum operating temperature below 85°C for prolonged measurement periods. One of my non–Bambu Lab dryers even disables humidity readings above 60°C specifically to preserve the sensor’s lifespan.

If Bambu Lab does the same, I don’t know—but they really should implement similar protections.

So now you can decide for yourself whether we truly need higher accuracy for this type of RH measurement, and whether that added precision is actually worth the cost in this context.

So, to get an accurate reading, the firmware would have to factor in the internal temp of the ams unit? Would be cool if bambu could impliment such a thing, since they have all the data needed.

Yes, exactly—relative humidity (RH) is inherently tied to temperature, so to get a meaningful or accurate representation of how much water vapor is actually in the air (i.e., dryness), the firmware would need to factor in the internal temperature of the AMS unit.

Bambu Lab already collects both RH and temperature data, so in theory, they could easily calculate absolute humidity (AH) in g/m³ and display that instead—or at least provide it as an advanced option. That would make it far easier to compare how “dry” the air actually is inside the AMS, especially across different temperatures.

Would definitely be a great firmware addition, especially for users drying sensitive filaments like PA, PC, or PVA where every gram of moisture matters.

1 Like

I completely agree on the difference between relative and absolute humidity. But I think, relative humidity is the value that is relevant for filament storage. In my understanding, it determines, if the air will take up moisture from the filament or if the filament takes up moisture from the air.

1 Like

Please before posting anything do some research into Humidity and how it works as you are completely wrong

Relative humidity is exactly as it’s worded it’s relative and in this case directly proportional to the temperature.

All you are doing relatively is measuring the maximum moisture content of the air at that given temperature so as temperature increases relative humidity will automatically lower as the air can contain more moisture

Absolute humidity is what is required in order to measure exact readings.

The groups on Facebook are full of people commenting about humidity and meter readings that have absolutely no clue what they are talking about

It’s a very complicated question. You’re right that RH is the figure that we want to know whether the filament takes up moisture or not. However that depends on the types of filaments. For some filaments, even in a very low RH (<15%), these filaments, are still prone to draw water, essentially becoming a dessicant, because they’re either reacting with water or having a micro structure to capture water, instead of just attracting water. So for these filaments the absolute humidity also matters because for their main driving force of absorbing water is the micro structure.

RH works great with some lower absorption rate filaments, but AH determines whether some of the Nylons or PVAs would last long enough.

To remove the water from those filaments even a 1% RH might not be enough, you need heat to give energy to the water molecules to escape the micro prisons.

@hotellonely What is a practical example where you are relying on an AH rather than RH number for filament storage and/or drying?

When it comes to drying using very low humidity air, I start to see reference to dew point. For instance, Vision Miner says you should dry certain filament in a blast oven fed by air with a dewpoint of -50C (if I remember correctly) or less. In part this may be due to the way really low humidity gets measured, by literally detecting at what temperature a mirror gets fogged. It’s no joke; that’s how some of them work.

No, I’m just saying that even we have a really low RH reading, some filaments would still die slowly unless we revive it later on with heat

1 Like

You’re right that relative humidity (RH) is a relative measure, and that it depends on temperature—no disagreement there. However, there are a couple of incorrect assumptions in what you’ve written:
1. RH is not directly proportional to temperature.
The relationship is exponential, not linear. As temperature increases, the air’s capacity to hold moisture increases rapidly—not in a straight line—so RH can swing dramatically with small temperature changes, even if no additional water is added.
2. RH doesn’t “automatically” decrease with higher temperature unless absolute humidity remains constant.
If water vapor is added (like in a closed system), RH may stay the same or even increase. RH is a function of both temperature and water vapor content—not just temperature alone.
3. It’s true that absolute humidity (AH) gives a more stable and comparable metric across temperatures, which is why professionals use it when precision matters—such as in drying systems, climate control, or aviation.

Do your own research before jumping on others :grin:
As a pilot, I use humidity data regularly for everything from density altitude to dew point spread—
but of course, I’m the one who’s completely wrong and everyone else is perfectly right. :+1:

2 Likes

ah! I found this old video from stefan, it says better than me

in short,
a ton of silica gel beads (very very low RH) is still not good enough to keep CF Nylon or PVA dry

but it should be good enough for everything else that we print daily

You’re absolutely right that dew point becomes the standard reference once we deal with very low humidity environments, especially in professional drying systems—and yes, chilled mirror hygrometers do literally detect fogging on a surface to determine the dew point. That method is still among the most precise ways to measure ultra-low moisture content.

But here’s where absolute humidity (AH) becomes practical in filament storage and drying:

Example 1: Comparing two dry boxes at different temperatures

Let’s say one dry box shows 10% RH at 40 °C, and another shows 13% RH at 35 °C.

At first glance, 10% RH might sound “drier,” but when converted to absolute humidity, the 13% at 35 °C box is actually dryer in terms of real water vapor content (g/m³).

Without AH (or at least temp-compensated RH), we’d make the wrong decision based on misleading RH alone.

Example 2: Storing filament near dew point thresholds

For moisture-sensitive materials like PA, PVA, or PEI, you want to ensure the air stays well below the critical moisture absorption threshold. Using RH alone won’t help you if the ambient temperature shifts.

AH or dew point tells you how much water vapor is actually available to be absorbed—independently of temperature swings.

Example 3: Designing or calibrating heated drying cycles

If you set a target of “RH < 10%” without accounting for the chamber temperature, you may over- or under-dry the filament. But if you define your goal as “AH < 1.0 g/m³”, it’s unambiguous regardless of temp.

That’s essential when building drying profiles for hygroscopic or advanced materials.

So yes—dew point is key when you’re using desiccant-fed or regenerative dryers with specs like “-40 °C dew point or better.”

But absolute humidity is often more intuitive and more practical for daily filament storage and monitoring, especially when your equipment reports only RH and temperature.

2 Likes

@hotellonely By the way, something of a tangent, but I’m pretty sure his vacuum was nowhere near hard enough for even a FAFO. According to his own video, he pulled only about 28 inches of mercury of vacuum:


which is just nowhere near a hard vacuum:

For contrast, in HVAC, inability to maintain a vacuum below 500 micron indicates the presence of moisture that needs to be pumped out, and Stefan never got below 40,000 micron.

2 Likes

Yeah, I think he wasn’t doing vacuum good justice there. In reality the vacuum dryers are much stronger, and also, heated :stuck_out_tongue:

1 Like

PVA will absorb moisture just from being touched. When I work with PVA, I always use gloves and only print with thoroughly dried filament. I dry it both before and after printing, and store it vacuum-sealed between prints to prevent any moisture uptake.

1 Like

Here’s what grok 4 (Elon’s latest!) has to say about vacuum drying PVA:

To dry PVA (Polyvinyl Alcohol) filament effectively using vacuum, a strong vacuum is required to lower the boiling point of absorbed moisture and facilitate evaporation without high heat, preventing filament degradation. Based on expert guidance and experiments, the optimal vacuum strength is typically below 500 microns (or approximately -29.5 inHg / 0.67 mbar), with deeper vacuums (e.g., 300 microns or lower) accelerating the process.reddit.com+2 more This is similar to nylon, as PVA is also highly hygroscopic (absorbing up to 10% moisture by weight) and benefits from low-pressure drying to avoid hydrolysis.wevolver.compvamed.net

Why This Vacuum Strength?

  • Mechanism: Vacuum reduces air pressure, dropping water’s boiling point (e.g., at 300 microns, water boils at ~20-30°C).reddit.comcnckitchen.com For PVA, which is sensitive to high temperatures (drying above 60°C risks softening or clumping), this enables efficient moisture removal at mild heat (40-50°C).wevolver.comforum.prusa3d.com
  • Threshold for Effectiveness: Below 500 microns is considered “deep vacuum” for moisture extraction; shallower levels (e.g., 1000 microns) are less efficient and may require longer times or higher heat.cnckitchen.com Tests show vacuum drying outperforms dehydrators for hygroscopic filaments like PVA.youtube.com

Drying Time and Temperature Guidelines

Use a vacuum chamber or dryer with heat. Typical parameters for PVA:

Vacuum Strength Temperature (°C) Drying Time (Hours) Notes
Below 500 microns 40-50 4-6 Basic drying for mildly damp PVA; combine with desiccant for best results.wevolver.compvamed.net
300 microns or lower 40-50 2-4 Faster moisture boil-off; ideal for heavily absorbed water; monitor to avoid overheating.reddit.comforum.prusa3d.com
Below 100 microns 50-60 6-12 (or overnight) Deep drying for severe cases; use with low heat to prevent clumping.cnckitchen.comwevolver.com

Equipment Recommendations

Vacuum drying is effective for PVA, but combine with heat and monitor to avoid under-drying. wevolver.compvamed.net

1 Like

Interestingly, it seems even more aggressive in its advice (below 10 microns!) for drying nylon:

Drying Time and Temperature Guidelines

Use a vacuum chamber or dryer with heat. Typical parameters for nylon:

Vacuum Strength Temperature (°C) Drying Time (Hours) Notes
Below 500 microns 40-50 4-6 Surface moisture removal; suitable for mildly damp filament.sovol3d.com
15-100 microns 60-70 8-12 Deep drying for heavily absorbed moisture; use with desiccant for best results.forum.bambulab.comforum.raise3d.com
Below 10 microns 70-90 Overnight (12-24) Optimal for severe cases; monitor to avoid overheating.matterhackers.comyoutube.com
  • Verification: Hold vacuum for 30-60 minutes; if pressure rises slowly, moisture is present—continue until stable.cnckitchen.com

In which case, Stefan was maybe off by a factor of about 4000x, or more, in his FAFO attempt, not to mention he wasn’t heating his (just ambient) but only vacuum.

1 Like

I think , we have become very scientific and went far away from the original question :joy:
Very interesting though, I learned quite a bit!

Anyway, I just looked up the reported humidity levels reported by my X1C + 2x AMS1:
First AMS has dessicants, which apparently are very saturated :joy: I probably missed or ignored the alarm. Anyway that are the readings:

  • the printer says level 1/5,
  • the App reports 20%
  • the sensor reads 19% RH at 27°C

The second AMS for PLA has no dessicant.

  • the printer says 4/5,
  • the app reports 80%,
  • the sensor says 40% RH at 27°C

I think this clearly proves that the percentage reported by BambuLab has nothing to do with humidity but instead is some arbitrary level of humidity.

I think they mapped RH from 0% … 50% to levels 0…5. In the app, they map those levels to some % value. that would match the suspicious % values that seem to increase in steps of 20%
so I think it works like:

0…9% RH = lvl 0 = 0% (app)
10…19% RH = lvl 1 = 20% (app)
20…29% RH = lvl 2 = 40% (app)
30…39% RH = lvl 3 = 60% (app)
40…49% RH = lvl 4 = 80% (app)
50…100% RH = lvl 5 = 100% (app)

2 Likes

Oh, @NeverDie actually hat intermediate values, then my mapping is not correct. Yet I think some similar mapping is used by BambuLab. I still think, the value just is some indication of “much or little humidity” without physical meaning.

Not that it will change anyone’s mind, but just closing the loop: I checked the JLCPCB price for the chinese TH sensor whose datasheet I posted earlier in the thread, and at quantity 1000 their price would be $0.5835:

and even at quantity 14, the minimum order, it’s $0.64.

Since JLCPCB is a middleman, I’m sure Bambulab could get a better price by ordering direct from the manufacturer.

Unfortunately, none of this gets us any closer to what the accuracy is for the Bambulab black box humidity sensor. In the best case, maybe there’s a fixed calibration offset to get a more accurate number.