433MHz long range device for alarm sensors

[okw]

I got a Staniot SecPanel 3 alarm system for Christmas and so far it works surprisingly well. Except for the range I need for a magnetic door switch. I tried to install it in a storage locker 50 meters from the alarm central, but I only have reception right outside the storage unit, once I enter the room it can't reach.
As far as I see it, I can:
-add an external high gain antenna - might do the trick.
-replace CR2032 (3V) with li-ion battery (3.7V) - might wear ICs out faster, or burn something right away.
-design my own - more work, but more fun.

Many years ago I've made a few 433MHz sensors based on cheap Aliexpress modules and Arduino. Worked fine, but these only sent temperature and humidity to a home automation central.
I'm sure these alarm sensors have more security(?), some rolling/learning code, etc? The manual says it supports ev1527.
There are transceivers made specifically for ev1527, with 4 button inputs, which I assume is the basis for these cheap Ali garage door openers.
Anyone with experience and can recommend front-end ICs or SoCs for loooong range on a budget? One option for CR2032 battery and one power hungry for use with a 3000mAh li-ion (I'll probably use the li-ion in the storage unit).
Or is my best bet to use one of those ev1527 front-ends and add an amplifier on the output?

[floobydust]

Reverse-engineering a secure sensor protocol is a lot of work.
I would look at the transmitter circuit design. Sometimes they bungle it as a bad design, or intentionally weaken TX power for FCC approvals. U2 is a F4455B ASK transmitter in SOT23-6, your circuit is missing C4 maybe check the circuit a bit.
First I'd try put a decent antenna on it. Or look for what's blocking /interfering with the RF signal.
You could tap into the keying input and then just add your own high power TX.

IC datasheet:
12.1 Use the power amplitude control of external resistance
The R7 is used to adjust the RF amplitude output level, which may be required to meet consistency adjustment. For example, the following table lists a 50 ohm test
The conductive RF output level of the test board and the typical value of the corresponding R7 resistance value, as shown in Figure 2. F4455B demonstration board using ring antenna R7 can adjust the radiation field allowed by FCC or ETSI compliance as needed. Please contact Sinota to get the recommended R0 value, so as to Comply with FCC and ETSI standards."

Oh well, there is no R7 or TX power table in the datasheet. Must be on the eval board.

[okw]

Thanks. Yes a higher power TX would certainly be an option.

I just studied the PCB now. PAOUT to the antenna is identical to the datasheet, except an additional series inductor between C3 and C4. C4 is missing, but also NC in the datasheet.
If there was a R7 to increase power, it would be great

[okw]

I just bought two 433MHz antennas, one 4cm long (-10db@433.92MHz) and one longer 16cm (-21dB@433.92MHz), so they turned out to be quite decent. I used a PS200 to measure.
I cut the IPEX connector off of two SMA cables and soldered them to the PCBs. I put the long antenna on the alarm central (RX) and short on the sensor (TX). Or should I switch them?
Mounted in the storage unit, I jsut barely get signal between them now. If I move the sensor a bit, the central doesn't receive anymore.
As I replaced the antennas, I'd like to tweak the matching networks on both units. I attached the TX circuit for the sensor and RX circuit for the central, as well as LiteVNA measurements of the stock sensor, and the external antenna sensor. Nothing for the central yet. I set the span for 1.8MHz, not sure if that's too narrow, but since the 433.MHz "band" don't have any channels, it should be dead on(?).
Stock sensor is quite 50 ohms (47-53, depending on how I place the cable and where I sit in relation to the DUT).
The external antenna one is another story. How do I go about matching better?

[Stringwinder]

Rod antennas can benefit from putting them in the corner of two plates (~ 90 deg) and
surprisingly this is called a corner antenna.

https://www.microwavetools.com/corner_reflector/

Replacing a CR2032 with CR123A makes changes less frequent.
Also makes function in cold weather better.

Tested CR123A for a sensor (Tapo T110) in my outdoor mailbox 15 meters away.
Installed last summer and battery (CR2032) died a few weeks ago.
No more unnecessary walks in cold weather/rain to find that the mailbox is empty.