Electronics/ Honeywell Galaxy alarm help

Jan 16, 2026 Last reply: 5 months ago 22 Replies

Hi All,



I have a Honeywell Galaxy 2 house alarm and am dabbling with IoT development using an ESP32. The software side of it I am fine with but not confident on the electronics side. The project I am working on is to connect the ESP to the alarm so that when it goes off it will text, email, SMS me. Rather than have a separate power supply for the ESP I was hoping to power it from the alarm panel itself.



The ESP I have is a lilygo t-call v1.3 but am struggling to find a definitive spec on what the power supply requirements are. Some say it can peak at 2A others seem to say it is generally 70mA !!



The alarm has a general aux power supply which seems to have a max of



500mA (although suspect this is shared with the keypads) and the trigger header (where I connect to to determine the alarm status) seems to have a max of 100mA.

So my questions are....



  1. is there a way I can reliably check what the consumption actually is? At the moment I am powering it via the onboard USB connector attached to my laptop for development.
  2. What is the best way for me to power it from the alarm panel/ what are my options. If I don't know the consumption I was wondering if I could power it from the backup battery (via a fuse)?

Many thanks in advance.



Lee.


Microcontrollers tend to have very variable power demands - can be tens of mA when active and uA when sleeping. The T-Call uses the ESP32-WROVER-E module which has a table in the datasheet (page 29):

formatting link
Table 16: Current Consumption for Wi-Fi (2.4 GHz) in Active Mode Work Mode RF Condition Description Peak (mA) Active (RF working) TX

802.11b, 1 Mbps, DSSS @ 19.5 dBm 350mA 802.11g, 54 Mbps, OFDM @ 14.0 dBm 243mA 802.11n, HT20, MCS7 @ 13.0 dBm 233mA 802.11n, HT40, MCS7 @ 13.0 dBm 229mA RX 802.11b/g/n, HT20 108mA 802.11n, HT40 113mA

It uses an A7670 LTE module:

formatting link
says its power consumption is under 4mA.

If you have a non-ancient network you should in on 802.11n at least, so the peak power of the MCU would be ~230mA in transmit and ~110mA in receive. Add on 4mA for the LTE mobule.

Also, the T-Call doesn't say what its power supply regulator is. If it's a linear regulator, the current would be (almost) the same, at a higher voltage. So if your alarm power was 5V it would be taking 234mA at 5V.

But I see an inductor on the board, and if it's a switching regulator then the power would be (almost) the same, so that's

234mA @ 3.3V = 772mW = 140mA @ 5V. (plus a bit for switching losses)

I think your 500mA supply is likely to be ample.

Do you know the voltage of the alarm power output - is it 5V or something more like lead acid battery voltage (12-14V)?

One option is to use a switch mode power supply board, commonly called a 'buck converter' when stepping down voltage. That will take your 12V down to a fixed or variable output voltage, eg the 3.3V of the T-call. For example:

formatting link
(many many listings, also on ebay, aliexpress, etc).

Many of these are clones (the genuine LM2596 chip is from Texas Instruments but these likely have a Chinese knockoff), but the idea is you feed in (say)

12V and adjust the blue potentiometer with a screwdriver to set your chosen output voltage. As a switching regulator it's more efficient and runs cooler than the normal linear type.

The clones tend to over-egg their supported output current so I'd not be using this at the stated 3A limit, but below 1A you'll likely be fine.

You can get versions with screw terminals, LED voltage displays, rotary knobs, higher currents/voltages, all kinds of stuff. But the kind above are about a quid each and fine for a basic job.

Theo

Hi Theo,

Firstly many many thanks for your fantastic response it is much appreciated. Has definitely given me a good way to think about it.

I th- page 30 - Chip itself max 68mA (could reduce with different modes)

- page 53 - WiFi max 240mA

  1. SIM800L for the mobile connection. The datasheet
    formatting link
    is not so helpful other than it seems to say it needs a supply that can give it up to 2A which is massive compared to the one you found.

The alarm supply is nominally 12v although I measured it as 13.7v. I have a couple of the variable buck converters - although I had assumed if the input voltage varied (e.g. switch to battery backup) then the output would also? So I have on order some fixed 3.3v ones of these

formatting link
which says it will go up to 5A

Any ideas on how to power this from the alarm system? I was wondering whether any of this would work

  1. Adding caps to deal with power spikes (assuming it doesn't consume 2A all the time? Also, how do I protect the ESP/ alarm from damage?
  2. Connect it to the 12v backup battery
  3. Could I somehow add a separate battery for the ESP and charge it from the alarm aux power so the battery deals with spikes.

Also, any chance you could send me a link to the T-CALL module you found as that might be a simpler/ safer option.

Many thanks once again

Lee.

No. designed to produce a stable output from a variety of inputs.

You can even get 'buck/boost' converters that will produce a steady output when the input is either below or above the required output voltage.

I'm using one myself to provide a steady 12 volts from a lead acid battery which is contonuously charged (and thus tends to be around 13.5 volts most of the time). The buck/boost converter produces 12 volts out from an input voltage range of 8 volts to 30 volts or so.

Quick update. I thought that 2A+ current from the USB port on my old laptop (running Ubuntu) seemed a bit high. So, I run "usb-devices" both with and without the device connected. When connected only one USB device is added with the following output

T: Bus=02 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#= 11 Spd=12 MxCh= 0 D: Ver= 2.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs= 1 P: Vendor=10c4 ProdID=ea60 Rev=01.00 S: Manufacturer=Silicon Labs S: Product=CP2104 USB to UART Bridge Controller S: SerialNumber=01D9FD0D C: #Ifs= 1 Cfg#= 1 Atr=80 MxPwr=100mA I: If#= 0 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=00 Prot=00 Driver=cp210x E: Ad=01(O) Atr=02(Bulk) MxPS= 64 Ivl=0ms E: Ad=82(I) Atr=02(Bulk) MxPS= 64 Ivl=0ms

Which seems to be saying it is only providing a max power of 100mA and the device is working fine. So, I was wondering if there is a way of ensuring the device doesn't take more than say 400mA (I believe the 2x keypads are also powered from aux and it is fused at 500mA on the board according to the spec.

Thanks very much both. Looks like I can use the variable ones I have for now as I have ordered 3.3v ones and now found out the T-CALL board needs

5v so will need to order some more.

That datasheet (or the non-captcha version I found):

formatting link
is quite helpful - see page 22.

It shows that the 2A bursts are 577us wide with a 4615-577 = 4038us recovery time until the next one. VBAT must drop no more than 350mV.

From that we can do some calculations...

Amount of charge needed to cover a burst Q = It = 2A * 577e-6 seconds = 1.154e-3 coulombs

Looking for a suitable capacitor to hold up the supply, Q=0.5 CV^2 (C is the capacitance) = 0.5 * C * 0.350*0.350 = 0.06125 C

Equating charges and finding for C:

1.154e-3 = 0.06125 C C = 1.154e-3 / 0.06125 = 18800 microfarad

which is a chunky capacitor but not impossible, especially at lower voltages. Of course, this calculation assumes the supply has gone away completely rather than staying there providing its normal current.

During the recovery, we need it to charge in 4038us. Assuming a constant current supply:

Q = It

1.154e-3 = I * 4038e-6 I = 1.154e-3 / 4.038e-3 = 286mA

That would be what you'd need if a constant current. I suspect it can do it faster (via a higher current) assuming something doesn't trip, but I suspect the alarm PSU is dumb enough that it won't. This kind of bursty current isn't going to overheat anything, and most trips (fuses etc) are ultimately based on a thermal event.

Anyone want to check my envelope?

TBH the claim of 4mA for the other one seems a bit surprising given it needs to talk to the LTE network which is likely to cause more current pulses.

Unless this one is doing 2G which is more power hungry perhaps?

(those current pulses might be the same as the buzzes that were picked up on microphones just before somebody's 2G phone rang?)

I think you will be fine running it from the alarm PSU. The current spikes are short and you're unlikely to damage anything. The problem is more likely that the mobile module undervolts itself during spikes and doesn't work, rather than blowing any fuses. Some good supply capacitors should deal with the spikes.

This one:

formatting link
I assumed T-Call was a product and not a generic name, BICBW.

Theo

Firstly thank you very much for the Theo. Those diagrams on page 22 are way beyond my knowledge and how you gleaned all that from the data sheet is genius thanks :). I can't validate your numbers unfortunately but maybe someone else who understands this stuff can.

Yes the SIM800L only supports 2G so it will be using that. Also, I have tried looking in the data sheet for under what circumstances it spikes with a view of avoiding those scenarios via code but couldn't find anything. My only fear is killing the alarm system and also I believe the keypads are also powered from aux. If the worst case is I blow a fuse then hey that can easily be replaced so might as well give it a go and see what happens.

No idea about the power supply out, but Honeywell do (used to do?) a separate little box for outgoing calls/texts that was powered by the main box.

There are various comms cards and expansion modules that can fit in the main box of my Galaxy, I'd be surprised if there isn't sufficient capacity to power an ESP32, I'll see if I can dig out my folder of documentation ...

Thanks very much both. I think the ESP32 is fine it is the SIM800L module that is also on the dev board that is the potential issue. Trouble is that I need it to call/ text me when the alarm goes off so need the module.

The alarm is a Galaxy 2-44+ if that helps and I found this manual online for it

formatting link
There appears to be a number of things wired into the aux pins already although there seem to be more +wires connected that -wires - is this normal? I have taken a few photos and uploaded them here which might also help

formatting link
Not sure what all the aux connections are but at a guess....

- 2 x keypads

- I think the board bottom right is an expansion board to have extra zones so maybe that is connected?

- I assume the red box is a landline module of some sort?

Another option might be to get a 240v to 5v/ 3.3v transformer although not sure what the transformer looking thing is top left which seems to be between inbound mains wire and the board.

Any help greatly appreciated.

Many thanks

Lee.

Need to dig out the old laptop, the docs must be on there ...

Thanks Andy. Much appreciated :).

Hi Andy,

I was wondering if you were able to dig out the docs by any chance?

Many thanks

Lee.

I've just been to the IT graveyard, and pulled out three M.2 NVMes and one external HD containing Acronis image backups, filtering through them now for Honeywell/Galaxy stuff ...

You are a star. Thanks very much Andy much appreciated.

OK, I found them, will upload somewhere ...

formatting link

It's basically all the Honeywell/Galaxy/Flex documents I could lay my hands on while I was preparing to install ...

How far have you got with your ESP32 interface?

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required