Dehumidifing LNAs

[joe...@alpine-club.dk]

Keeping your eletronics dry and especially cooled electronics can be a little tricky as we heard from Steve in SARA November meeting. There are a lot of various teories of how it is done, some work and some dosent. Furthermore, it can take a long time to find out wether it was a good idea or not. Bad teories can be destructive for our delicate electronics.

There is the posibility to use some sort of dehumidifier where Silicagel can be effective and can keep the electronics dry but it requires maintenance to keep it active depending on how good your mechanics are.

But there are also a modern solution, a micro dehumidifier from Rosal. Have a look at these inspiring videos:

https://www.youtube.com/watch?v=n7EWexck8NE

https://www.youtube.com/watch?v=Vabq-s62IVM

https://micro-dehumidifier.com/product/vr2x-rsm-5-litre-dehumidifier/?v=0ecbf9426bcf

Think they sounds promising!

Joergen

[Alex P]

Use a small resistive heater to keep the hardware ~ 5 dgC above the environment ?

[joe...@alpine-club.dk]

That is very effective Alex but Steve has cooled the LNA. This is a good idea for the noise figure but also a challenge for humidity. Basically LNA has to be hermetic and filled with nitrogen or Silicagel.

[Marcus D. Leech]

On 2025-11-23 19:01, joe...@alpine-club.dk wrote:

That is very effective Alex but Steve has cooled the LNA. This is a good idea for the noise figure but also a challenge for humidity. Basically LNA has to be hermetic and filled with nitrogen or Silicagel.

So, the problem is that unless you very aggressively cool the LNA (not impossible, just unpleasant), then the effect on noise figure is modest.   Let's say the LNA has a room-temp
  noise temperature of 50K @ 297K, dropping the temperature of the LNA to 267K (a 30K drop), will likely drop the Tlna by perhaps 5K.   That 5K may or may not be a good
  improvement, depending on what your other noise sources are.

The main advantage of LNA temperature control for the amateur is in *gain* stabilization, which can have a serious effect on measurements.  So, controlling its temperature
  in the "positive" direction is easier, and achieves the goal of gain stabilization.

mandag den 24. november 2025 kl. 00.50.25 UTC+1 skrev Alex P:

Use a small resistive heater to keep the hardware ~ 5 dgC above the environment ?

--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/sara-list/85860721-c197-4162-ab69-46b5a2b2e4b7n%40googlegroups.com.

[joe...@alpine-club.dk]

Steve have allready integrated cooling so my input is best understood as a help and inspiration.

Most MMIC LNA amplifiers will drop noise figure by 0.3 - 0.4 dB when operated at -40 compared to ambient. This is atractive because it creates acces to otherwise unaccesable performance. On the other hand, there are challenges keeping your electronics dry. Some complications are you properly need a machineshop with lathe and milling machine etc. It requires knowhow and a deep understanding of the physics involved to produce boxes nessecary for the job.

The dehumidifier shown in the youtube videos may be a shortcut to be succesfull hanling cooled LNAs.

[Andrew Thornett]

I've made up a set of kit to regas my Atik 16200M cooled astro camera - only cost me about £30 to put something together that was effective.

....so if the LNA can be put in a sealed box with option to open hole one side and hole another side then this is highly practical. With regards to that, put it in a metal box and wrap in emulsifying tape to seal from gas escape. The traditional method of providing the two holes is two small screw holes and sealed by a screw +/- rubber washer. Must make sure holes are big enough for whatever tubing you use from gas cylinder. 

NB I also can't see a reason why the metal box can't be attached to large heatsink one side before emulsifying tape added.

To regas open both holes, start gas cylinder, leave few seconds to clear tubing and start gas going through box and clear out air from box, close hole on far side.

Best wishes

Andy

Sent from Outlook for Android

---

From: sara...@googlegroups.com <sara...@googlegroups.com> on behalf of joe...@alpine-club.dk <joe...@alpine-club.dk>
Sent: Monday, November 24, 2025 8:22:04 AM
To: Society of Amateur Radio Astronomers <sara...@googlegroups.com>
Subject: Re: [SARA] Re: Dehumidifing LNAs

 

To view this discussion visit https://groups.google.com/d/msgid/sara-list/2ac9617e-3b36-4ece-a600-742d2a2ae5f4n%40googlegroups.com.

[joe...@alpine-club.dk]

One method is to fill compartment with a dry gas where nitrogen is the most accesable I guess. However, this requires the box remains sealed or there will be humidity injected via pressure changes caused by the weather temperature cycling. In profesional electronics with sufficient mechanics, no problem and will last forever or close to. 

For amateurs this is more challenging with mechanics, sealed RF connectors, sealed power input etc. If we could simply pump humidity out constantly! Simpler and cheaper mechanics.

Then there is the question Marcus rises, is it worth doing it? Properly not but temperature stabilisation is a good thing, at least for RTL-SDRs.

I like the idea with a constant dehumidifying LNA and for 60£ it could be an idea for a cristmas pressent

?

[b alex pettit jr]

It is practically Impossible to perfectly seal a system for long term use.

 ( I have seen commercial electronic boxes fill with water after being in the afternoon desert heat and then quickly chilled and drenched during a rain storm )

The least complex would be to devise a constant low pressure feed of nitrogen or argon .. a few inches of water pressure would work ...

A BBQ grill regulator drops the pressure to 10"  H2O &  you actually  "could" use propane .... Just don't smoke near your LNA   

Alex

[Stephen Arbogast]

I live  in South West  Colorado,    very  dry here .  The   Colorado  River is  drying  up..  I have been using  the  Kracken RF   Discovery Dish    with  Hydrogen  Line  feed  sealed  in plastic  for a year now...  no problems...   I am  lucky.

[Andrew Thornett]

Does humidity affect day to day performance of RF LNAs and filters or just lifespan?

Sent from Outlook for Android

---

From: 'Stephen Arbogast' via Society of Amateur Radio Astronomers <sara...@googlegroups.com>
Sent: Monday, November 24, 2025 10:19:14 AM

To: Society of Amateur Radio Astronomers <sara...@googlegroups.com>
Subject: Re: [SARA] Re: Dehumidifing LNAs

--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/sara-list/91123196-a6e2-48ae-8fa3-da38fcc3cd42n%40googlegroups.com.

[b alex pettit jr]

Probably not the Humidity, but the Condensation Thereof on cooled components that can potentially corrode them.

[b alex pettit jr]

AND, components like the nooelec SAWbird consume ~ 1 watt of power so it inherently will be condensation free ..( if Not Cooled )

=============================================================================================

--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---

You received this message because you are subscribed to a topic in the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/sara-list/yi_BGlxoGqY/unsubscribe.
To unsubscribe from this group and all its topics, send an email to sara-list+...@googlegroups.com.

To view this discussion visit

https://groups.google.com/d/msgid/sara-list/311914086.341148.1763989242362%40mail.yahoo.com

.

[joe...@alpine-club.dk]

I think it is relatively simple to make an airtight container, all vacuum guys do all the time. You can buy the stuff but it can be expensive and not practical to build electronics into. So we must steal some ideas from these guys. A few guidelines:

Cheap o-rings can be used down to 1e-5 milibar.

A flat rubber gasket will never work because it  not stabilised in x and y directions unless there are a dedicated groove milled so gasket is locked in x and y direction. Z direction is compressed by the lid screwed down to metal-metal and groove is about 70% of gasket height. A flat gasket has a low pressure per area wich is a bad thing but a round gasket will have a high pressure per area.

A round gasket is prefered and in a properly designed square groove allmos perfect.

Gasket contact surfaces must be half mirror finish.

Use metal not acrylic boxes. Water vapor can migrate trough acrylic, havent we all seen caravans with misty windows?

A cable is not air tight, it is a hose with some wires trough it.

You cant seal a screw and thread. All holes must be placed outside the gasket line.

The lid must be thick (>>3mm)  and have very small overhang between screws, it is not enough with a screw in each corner.

There are properly other advices but you get the idea!

And of course as Andy writes, fill it with dry air not random humid air.

This is properly far beyond what we as amateurs can handle so wouldnt i be if we just pump humidity out out of a simple box? Look at the videos, there is a posibility!

Apart from profesionally working in microwaves and outdoor microwave systems I have also made vacuum test systems for satellite testing. Please ask if there are any questions on this subject.

Best regards to you all

Joergen

[b alex pettit jr]

Another idea, not too painful to implement,  would be attaching a 'breathable' 

desiccant canister to the electronics box via a hose .. w/ visible humidity

indicator on it .. all air sucked in or out then passing through the drying material. .

Alex

[joe...@alpine-club.dk]

That was my next suggestion Alex!

In Radar waveguide instalations two systems are common, one with a controlled leak in the far end and a pump with a build-in dehumifier in the other end. The pump will deliver a slight overpressure in the other end. All waveguide connections needs to be good but not ideal to keep the system dry. This works and is used in a lot of larger systems around the world. Expensive, makes noise and uses power. Not really amateur friendly!

The other is a container filled with Silicagel plus an input and output hose. One hose is connected to the waveguide and the other open in free air. Also used a lot in Radar instalations.

I will try to build an amateur version and show it here so we can get a good discussion over that solution Alex. Think it can be a very good solution to keep our gear dry.

Andy, Alex is absolutely corect, humidity in a LNA will not really cost NF but condensation is poison for NF and will create galvanic corrosion.

So let us get dry!

Joergen

[Andrew Thornett]

Given the relatively low price of SAWBird at around £50, is it worth the purchase of a £60 dehumidifier? I notice we also have the potential problem of humidity causing corrosion on the power board that comes with the dehumidifier.

If the humidity caused direct problems with the data, I can see it might be worth it.

For more expensive LNAs and designer LNBs, the cost effectiveness might be otherwise.....?

Having said that, this tiny dehumidifier is amazing and looks like it would be perfect for expensive cooled astro cameras.

Andy

Sent from Outlook for Android

---

From: sara...@googlegroups.com <sara...@googlegroups.com> on behalf of joe...@alpine-club.dk <joe...@alpine-club.dk>

Sent: Monday, November 24, 2025 2:25:38 PM

To: Society of Amateur Radio Astronomers <sara...@googlegroups.com>
Subject: Re: [SARA] Re: Dehumidifing LNAs

--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/sara-list/0b9b817f-bb72-4fb3-9c23-3114f2ab347bn%40googlegroups.com.

[joe...@alpine-club.dk]

Spot on Andy, it depends.

I started diskusion on the subject "cooled LNA" because Steve in australia got a defect LNA, defect because of condensation creating some sort of shorts on his PCB. For Steves remote location, continous dehumidification at low costs could be ideal.

For most of us, this is not nessecary but it will not reduce performance to keep our electronics dry.

[Andrew Thornett]

On the other hand.....

I could see potential benefit for having LNA at constant temperature. The easiest way to achieve that is to cool it below any likely outside temperature - avoiding need for heater at well. Plus heating adds noise.

But cooling below ambient leads to condensation hence need to dehumidify. 

You can tell I am just thinking on my feet here....probably should think it out properly then post a single post....

Nevertheless, Alex gave me some advice - he didn't use these words but I interpreted his advice as follows - apologies, Alex, if I have misunderstood you:

A radio telescope system is only as good as its weakest link. Therefore, there is no point is having a wonderful cooled LNA when the dish/coax/waveguide/SDR/etc. Is not optimised to the same level.

Even then, a poor radio environment, trees, or other sources of obstruction or reflection of radio signals, or other sources of RFI, unstable power supplies, even cloud and poor weather/too many airplanes/etc. may limit how far a system can be optimised.

Is above a correct understanding?

Andy

Sent from Outlook for Android

---

From: sara...@googlegroups.com <sara...@googlegroups.com> on behalf of joe...@alpine-club.dk <joe...@alpine-club.dk>

Sent: Monday, November 24, 2025 3:43:22 PM

To view this discussion visit https://groups.google.com/d/msgid/sara-list/01af1e13-1b9a-4691-aba3-325a1ad853bcn%40googlegroups.com.

[b alex pettit jr]

AND, as the Thermal Gain Coefficient of the  nooelec SAWbird H1  LNA is only

 -0.1 dB gain decrease per + 10 dgC change in temp 

its a minor part of system drift sources 

Exec Summary : There Are Much Bigger Fish To Fry !

[Adrian]

Hi

 Although the idea for the Micro Dehumidifier solution to keep the interior of an enclosure containing an LNA dry sounds like a great solution, but then it could be pointed out that one might be exchanging one problem for another.

If you look at how the proposed dehumidifier works you will find that it accomplishes the dehumidification process through a very clever way of electrolysis of water.

 If you look carefully at the reaction taking place the water is electrolyzed to H+ and O2. with the H+ ions transported away from the interior to reform H2O with outside atmospheric oxygen but the interior is left with an excess of O2.

So what happens now to say the copper traces in the LNA circuit board. Copper reacts directly with oxygen to form cuprous oxide (Cu₂O) at lower temperatures, and cupric oxide (CuO) at higher temperatures or with prolonged exposure to oxygen and no water is needed for these reactions to occur. While water vapor promotes hydroxide and carbonate formation from CO2 (leading to patina), the basic oxide layers (Cu₂O → CuO) arise purely from oxygen exposure. Now condensation of humidity from atmospheric water vapor or rain water significantly accelerates copper corrosion because dissolved gases (CO₂, NOₓ, SOₓ) that lower the pH, create acidic films that attack the protective Cu oxide layer.

 Now in addition Micro Dehumidifier claims;  "When operating, the density of oxygen in the container increases as the water molecules dissociate into hydrogen ions and oxygen, and the oxygen remains in the container. Due to normal leakage and the occasional opening of the container, the increase in oxygen density is typically 2 to 3%."  However if the container is vented sufficiently to allow for the excess O2 to escape wouldn't that defeat the purpose of both retaining the cooling function as well as keeping the excess humidity from entering to the container.  The tug of war between O2 increase, or allowing venting and introducing heat and humidity seems like an effort of diminishing returns. 

 Likely then other methods as suggested such as drying agents like silica gel which can be regenerated by heating (≈ 120–150 °C),  might be of simpler utilization to reduce the potential corrosion effects of humidity condensation inside the cooled LNA environment.  Even this approach has it's drawback as silica gel only has a moderate water uptake (- 40% of its weight) and is good for moderate humidity control but it still leaves residual moisture that hopefully is reduced sufficiently below the dew point concentration. An even better solution would be a hermetically sealed inert gas space also excluding oxygen as a better solution to the corrosion problem.

Adrian

[Adrian]

Thanks Joergen

  On looking at other dehumidification products that Micro Dehumidifier carry  they have one called Drykeeper that seems to be a passive material that they claim work much better than silica gel..

https://micro-dehumidifier.com/drykeeper/

Joergen 

 This could be a  better option for the cooled LNA application purpose,.

Adrian

[Andrew Thornett]

I note that drykeeper data does not tell us how much of product was used just how much water it absorbed.

Sent from Outlook for Android

---

From: sara...@googlegroups.com <sara...@googlegroups.com> on behalf of Adrian <kjan...@gmail.com>
Sent: Tuesday, November 25, 2025 5:16:55 PM

To: Society of Amateur Radio Astronomers <sara...@googlegroups.com>
Subject: Re: [SARA] Re: Dehumidifing LNAs

--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/sara-list/967840e7-c0b5-46d1-b555-7030443e7634n%40googlegroups.com.

[Dave Typinski]

If corrosion due to condensation is the issue, why not flood the inside of the
enclosure with liquid coolant and connect it to a liquid cooling loop? Bury the
heat exchanger (radiator) deep enough in the ground for gain stability despite
outside air temperature fluctuations.
--
Dave

[A. C.]

Frst, what liquid coolant? Like 50/50 Glycol? Have you any idea what the dielectric constant of that stuff would do to the tuned striplines of an LNA at 1.4 GHz?

Well try it on your LNA first and tell us how it worked out.

--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---

You received this message because you are subscribed to a topic in the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/sara-list/yi_BGlxoGqY/unsubscribe.
To unsubscribe from this group and all its topics, send an email to sara-list+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/sara-list/6925FD4A.9070907%40typnet.net.

--

Adrian

"My God, it's full of stars!"
 Dave Bowman

[b alex pettit jr]

It will take a LOT of COOLING to noticeably improve the Noise Figure of a nooelec SAWbird .. 

https://www.mouser.com/datasheet/2/412/TQP3M9037_Data_Sheet-1500717.pdf

Gain Stability 

Alex P

=============================================================

[Dave Typinski]

I thought striplines acted like coax cable, not caring what's exterior to them.
Or are we talking about microstrip traces? Those might care about the
coolant, depending on the thickness of the PCB material surrounding the
conductor(s).

One wonders how much of an impact the dielectric constant of the coolant will
actually have. PCs operating at multiple GHz can have the entire case flooded
with coolant and they work fine. Thermaltake calls it "immersion cooling."
Certainly not a perfect analogy to an LNA, but all the same there are a lot of
traces in a PC seeing rather high speed signals.

https://www.youtube.com/watch?v=Myq3JCfXLgI

--
Dave

On 11/25/25 15:00, A. C. wrote:
> Frst, what liquid coolant? Like 50/50 Glycol? Have you any idea what the
> dielectric constant of that stuff would do to the tuned striplines of an LNA at
> 1.4 GHz?
> Well try it on your LNA first and tell us how it worked out.
>
> On Tue, Nov 25, 2025 at 11:02 AM Dave Typinski <dav...@typnet.net
> <mailto:dav...@typnet.net>> wrote:
>
> If corrosion due to condensation is the issue, why not flood the inside of the
> enclosure with liquid coolant and connect it to a liquid cooling loop? Bury
> the
> heat exchanger (radiator) deep enough in the ground for gain stability despite
> outside air temperature fluctuations.
> --
> Dave
>
> --
> --
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To post to this group, send email to sara...@googlegroups.com

> <mailto:sara...@googlegroups.com>

> To unsubscribe from this group, send email to
> sara-list-...@googlegroups.com

> <mailto:sara-list-...@googlegroups.com>

> For more options, visit this group at
> http://groups.google.com/group/sara-list?hl=en
> ---
> You received this message because you are subscribed to a topic in the
> Google Groups "Society of Amateur Radio Astronomers" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/sara-list/yi_BGlxoGqY/unsubscribe.
> To unsubscribe from this group and all its topics, send an email to
> sara-list+...@googlegroups.com

> <mailto:sara-list%2Bunsu...@googlegroups.com>.

> To view this discussion visit
> https://groups.google.com/d/msgid/sara-list/6925FD4A.9070907%40typnet.net.
>
>
>
> --
>
> Adrian
>
> "My God, it's full of stars!"

> *Dave Bowman*

>
> --
> --
> You received this message because you are subscribed to the Google
> Groups "Society of Amateur Radio Astronomers" group.
> To post to this group, send email to sara...@googlegroups.com
> To unsubscribe from this group, send email to
> sara-list-...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com/group/sara-list?hl=en
> ---

> You received this message because you are subscribed to the Google Groups

> "Society of Amateur Radio Astronomers" group.

> To unsubscribe from this group and stop receiving emails from it, send an email
> to sara-list+...@googlegroups.com

> <mailto:sara-list+...@googlegroups.com>.

> To view this discussion visit

> https://groups.google.com/d/msgid/sara-list/CACs6ujbDPqNmicWWV4z-2WSwJZtRgkaSZBeG%2BEwiQNRs%3DgFCYw%40mail.gmail.com
> <https://groups.google.com/d/msgid/sara-list/CACs6ujbDPqNmicWWV4z-2WSwJZtRgkaSZBeG%2BEwiQNRs%3DgFCYw%40mail.gmail.com?utm_medium=email&utm_source=footer>.

[joe...@alpine-club.dk]

Thank You Adrian

A slightly different aspect of the Micro dehumidifyer, build-up of oxygen with its risk of oxydizing our nice LNA micro strips.

Bare cupper surfaces can be atacked. Most of the microstrip magic is happening between groundlayer, laminate material and bottom of top Cu layer. Top of outer Cu layer is not THAT important so as long as it dosent grow down to bottom of top Cu layer damage i think we can live with that. It will not improve noise figure but I think costs will be small. What if PCB gets a thin layer of PCB laquer, will this remove the risk Adrian? It could be a an easy fix.

I like the idea with floating in a preserving and cooling liquid. Prety sure it will be really damaging to our good noise figure!

In my opinion silicagel is a good path to go and I will show my take on an implementation in my 2nd generation LNA with a barebone Sawbird. Just need a little work yet! And Alex, it is not cooled but has at heater + thermistor so it can be heated and maybe even thermostatically regulated.

Joergen

[b alex pettit jr]

Hello Joergen,

This is what I designed to thermally stabilize a remote SDR .... The AirSpy Mini changes its band pass shape over temp.

( foam enclosure not installed n this photo )

Alex

==============================================

[joe...@alpine-club.dk]

I will also have a separate thermostat on the RTL-SDR plus accurate PT100 sensors on both LNA and RTL-SDR so there wil be a lot of bottoms to play with.

Joergen

[Stephen Arbogast]

My very  low  tech  way of   cooling my AirSpy  SDR....  probably  not necessary  since it  is located  inside my house.  A  brass bowl of  pennies  which  are not 100% copper  and no longer  in  circulation..  see   attached picture.

But  my discovery  dish  HL-feed LNA gain does  change  with temperature.  

[b alex pettit jr]

the rest of the temp stabilized system