TX37u -IT LaCrosse Temerature Sensor Repair

[Martin Riddle]

Anyone here repair one of these things?

I get a Radio signal indicator on the receiver, but it appears the
thermister maybe bad.

It was reading blank before I opend it.
It was reading fluctuating temperature after I cleaned it.
And now reads Blank after I reflowed the solder around the thermister.

I suspect a bad TC, but figured I ask here is someone had seen
something similar.

I'll probably unsolder the TC and hook up a decade box and see what
happens.

Cheers

[Jeff Liebermann]

On Wed, 04 Jan 2017 13:45:45 -0500, Martin Riddle
<marti...@verizon.net> wrote:

>Anyone here repair one of these things?

Yes, but with lousy results. The plastic senor cases are glued or
solvent welded, making disassembly rather difficult.

LaCurse uses "chip on board" technology in their sensors and consumer
weather station products. The chip is glued to the PC board, wire
bonded to pads on the PCB, and then covered with a blob of epoxy. I
cracked one open to see if the put on any junction coating. Nope. If
you push on the epoxy in any direction, you break the wire bonding,
and say goodbye to your sensor or weather station.

I found a TX6U sensor in my junk pile and cracked open the case.
Here's what's inside:
<http://www.11junk.com/jeffl/pics/weather%20stations/La%20Crosse/index.html>
The "chip on board" is the black glob. The green corrosion on the
contacts is typical after a battery leak. The white stuff on the PCB
is where La Crosse forgot to clean the PCB after hand soldering.

>I get a Radio signal indicator on the receiver, but it appears the
>thermister maybe bad.

Model number of your sensor? Numbers are amazingly useful.
These pages might help:
<http://www.lacrossetechnology.com/sensor-finder/>
<https://www.youtube.com/watch?v=V4WbJ5jzKiA>

>It was reading blank before I opend it.
>It was reading fluctuating temperature after I cleaned it.
>And now reads Blank after I reflowed the solder around the thermister.

Let the thermistor cool down.

What did you use to clean the board?

Did you touch the black "chip on board"? If yes, you may have killed
it.

Did you pair or sync the sensor with your unspecified model weather
station base unit? Some version require you to reset the weather
station in order to get it to sync:
<https://www.youtube.com/watch?v=TP_MwvWNA78>

>I suspect a bad TC, but figured I ask here is someone had seen
>something similar.

Yep. When you remove the batteries, you lose the connection between
your sensor and the weather station. Start over and will produce a
reading when they synchronize.

>I'll probably unsolder the TC and hook up a decade box and see what
>happens.

Good plan but a trim pot might be easier.

--
Jeff Liebermann je...@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

[Jeff Liebermann]

On Wed, 04 Jan 2017 22:25:47 -0800, Jeff Liebermann <je...@cruzio.com>
wrote:

>Model number of your sensor? Numbers are amazingly useful.
>These pages might help:
><http://www.lacrossetechnology.com/sensor-finder/>
><https://www.youtube.com/watch?v=V4WbJ5jzKiA>

Sorry. I didn't see the model number in the Subject line:
<http://www.lacrossetechnology.com/tx37u-it>

I don't have one in my junk box and couldn't find a photo of the
insides with Google Images.

[Martin Riddle]

On Thu, 05 Jan 2017 00:01:34 -0800, Jeff Liebermann <je...@cruzio.com>

wrote:

>On Wed, 04 Jan 2017 22:25:47 -0800, Jeff Liebermann <je...@cruzio.com>
>wrote:
>
>>Model number of your sensor? Numbers are amazingly useful.
>>These pages might help:
>><http://www.lacrossetechnology.com/sensor-finder/>
>><https://www.youtube.com/watch?v=V4WbJ5jzKiA>
>
>Sorry. I didn't see the model number in the Subject line:
><http://www.lacrossetechnology.com/tx37u-it>
>
>I don't have one in my junk box and couldn't find a photo of the
>insides with Google Images.

The TX37U is similar to the TX6U, but with a simplified RF section,
and a 0603 TC.
It pairs with the reciever, so I'm hoping its just a bad TC.

Cheers

[Martin Riddle]

On Thu, 05 Jan 2017 00:01:34 -0800, Jeff Liebermann <je...@cruzio.com>

wrote:

>On Wed, 04 Jan 2017 22:25:47 -0800, Jeff Liebermann <je...@cruzio.com>
>wrote:
>
>>Model number of your sensor? Numbers are amazingly useful.
>>These pages might help:
>><http://www.lacrossetechnology.com/sensor-finder/>
>><https://www.youtube.com/watch?v=V4WbJ5jzKiA>
>
>Sorry. I didn't see the model number in the Subject line:
><http://www.lacrossetechnology.com/tx37u-it>
>
>I don't have one in my junk box and couldn't find a photo of the
>insides with Google Images.

SO I finally found the TC on the TX34, it's located right next to the
only 10k resistor on the board. At first I thought it was a cap, since
it read open. But looking at the voltage across it, I can see they
pulse and sample to avoid heating the TC.
After hooking up the decade box, and trying all the typical NTC
values, It turns out to be a 10K tc with a B25/85=3486.
The reciever will only display 139f(59c)@ 2810ohms, using
R2 = R1 / (exp( B*(1/ T1 - 1/ T2) )), I determined the B25/60=3462
which should be close to the only Vishay part with a B25/75=3477 and
B25/85=3486. (NTHS0603N02N1202JP)
I'll order some and see how accurate it turns out to be.

Cheers

[Martin Riddle]

Update, I see some NTC specs include B25/50=3380, which turns out to
fit nicely plus their stocked at mouser and digikey.

Cheers

[Jeff Liebermann]

On Sat, 07 Jan 2017 10:01:02 -0500, Martin Riddle

<marti...@verizon.net> wrote:

>On Sat, 07 Jan 2017 00:42:59 -0500, Martin Riddle
><marti...@verizon.net> wrote:
>
>>On Thu, 05 Jan 2017 00:01:34 -0800, Jeff Liebermann <je...@cruzio.com>
>>wrote:
>>
>>>On Wed, 04 Jan 2017 22:25:47 -0800, Jeff Liebermann <je...@cruzio.com>
>>>wrote:
>>>
>>>>Model number of your sensor? Numbers are amazingly useful.
>>>>These pages might help:
>>>><http://www.lacrossetechnology.com/sensor-finder/>
>>>><https://www.youtube.com/watch?v=V4WbJ5jzKiA>
>>>
>>>Sorry. I didn't see the model number in the Subject line:
>>><http://www.lacrossetechnology.com/tx37u-it>
>>>
>>>I don't have one in my junk box and couldn't find a photo of the
>>>insides with Google Images.

>>SO I finally found the TC on the TX34, it's located right next to the
>>only 10k resistor on the board. At first I thought it was a cap, since
>>it read open. But looking at the voltage across it, I can see they
>>pulse and sample to avoid heating the TC.

Nice troubleshooting job.

Methinks the main reason they sample is to reduce power drain and thus
extend battery life.

>>After hooking up the decade box, and trying all the typical NTC
>>values, It turns out to be a 10K tc with a B25/85=3486.
>>The reciever will only display 139f(59c)@ 2810ohms, using
>>R2 = R1 / (exp( B*(1/ T1 - 1/ T2) )), I determined the B25/60=3462
>>which should be close to the only Vishay part with a B25/75=3477 and
>>B25/85=3486. (NTHS0603N02N1202JP)
>>I'll order some and see how accurate it turns out to be.
>

>Update, I see some NTC specs include B25/50=3380, which turns out to
>fit nicely plus their stocked at mouser and digikey.

Most of the stuff I see has a thermistor with thin leads (like the one
in the photo I posted). The one's stuck to the PCB tend to use the
PCB as a heat spreader, resulting in long response times. The leaded
variety are also easier to insert in the convective air stream. I've
compared the chip type with a leaded type inside a proper pagoda
(Stevenson radiation shield):
<http://802.11junk.com/jeffl/pics/wx/slides/radiation-shield-01.html>
(Well, maybe not so proper). The PCB mounted variety shows about a
5-10 minute additional delay for abrupt changes in air temperature,
such as sunrise and sundown. However, it doesn't matter much since
temperature is updated every 5 or 10 minutes (I forgot which).

[vicfl...@gmail.com]

I had same problem.
All I did was touch soldering iron the the thermistor and it came right back.