Harrison Neta 12 curtain rail

[hvxl]

Has anyone managed to modify the way a Harrison Neta 12 curtain rail is controlled? The mechanics work fine, but frequently the device doesn't react to any RF signals anymore. I have to cycle the power before it will work again. It sometimes works correctly for a week, but at other times it fails day after day.

When I first bought the system, it worked perfectly. But since then I have acquired more equipment that uses RF. My theory is that the device receives some RF signals that somehow lock up the receiver. Especially in the vicinity of the curtain rail, my 433MHz wireless headphones also frequently pick up lots of noise (also when the rail is unplugged, so it's not coming from the Neta 12 itself).

I have contacted Harrison to ask if this is a known problem and if they could suggest any solution. But they could not even be bothered to respond.

Obviously this is not a good situation in an automated house, especially when nobody is home for a while. So I was thinking about replacing the RF receiver with some other type of controller. Maybe a Qubino DC shutter module, or an esp8266. However, the housing doesn't appear to be designed to be opened. So I was wondering if anyone has already tried this and has some tips to share.

[johnz]

I have not changed the way it is controlled, but I was definitely able to open the housing. Unless something has changed in the last two years it is just some small screws. I took down the rail, but can’t remember if that was necessary for opening the housing.

[hvxl]

You are right. I thought I remembered looking at opening it a long time ago, so I didn't look too closely this time before posting. There are two screws on the back (so the rail needs to be taken down to be able to get at those) and four on top. The two bigger screws hold the motor in place. They don't need to be unscrewed. There is also a seal that must be broken to open the housing. I guess that's what stopped me last time. But since Harrison doesn't respond and the warranty has expired long ago, I had no trepidation to break it this time.

Next step: Figure out how it works, and how to change that.

[hvxl]

I think I have pretty much figured out the schematic:

neta12.pdf

Harrison Neta12 controller schematic

(16.43 KiB) Downloaded 142 times

It all makes more sense now. There are two voltage regulators. One for 12V and one for 5V. So the power supply of 12V is a bit misleading. The device will need 16V for the regulator to be able to do its job.

Then there are two relays that switch the polarity of the power to the motor (connected to T2). The current to the motor passes through a 0.5 Ohm resistor (R1). The voltage across that resistor goes to one side of a comparator in the PIC12F675. The output of the potentiometer goes to the other side of the comparator. So, the potentiometer doesn't directly control the travel distance of the curtains. It just sets a threshold for the motor current. When the curtains are nearly closed, they encounter increasing resistance, resulting in increased current drawn by the motor.

Next I checked the signal coming from the RF receiver. I could not detect any pattern in the signal coming from the RF receiver (RF2). It was definitely not steady when the remote was not activated. So, it does look like it is almost constantly picking up RF noise.

Finally I tried reading the contents of the PIC. But as you may have expected, memory protection had been switched on by Harrison. So the only thing I could read was the configuration word. It was set to 0x3044, meaning:

Code: Select all

Field     Option               Category                                Setting
FOSC      INTRCIO              Oscillator Selection bits               INTOSC oscillator: I/O function on GP4/OSC2/CLKOUT pin, I/O function on GP5/OSC1/CLKIN 
WDTE      OFF                  Watchdog Timer Enable bit               WDT disabled
PWRTE     ON                   Power-Up Timer Enable bit               PWRT enabled
MCLRE     OFF                  GP3/MCLR pin function select            GP3/MCLR pin function is digital I/O, MCLR internally tied to VDD
BOREN     ON                   Brown-out Detect Enable bit             BOD enabled
CP        ON                   Code Protection bit                     Program Memory code protection is enabled
CPD       ON                   Data Code Protection bit                Data memory code protection is enabled


This shows that the watchdog timer is disabled. So, if the RF noise ever causes the program to go down a bad path, it won't recover automatically.

Having been able to read the configuration word, I am convinced I can load my own program into the PIC12F675. The part for driving the motor should be straight forward. But finding the remote control signal in all the RF noise is much harder. Converting to an ESP8266 requires more hardware changes, but has the advantage that the device can report back whether the curtains are open or closed. So, I'm not yet sure which option to choose.

[hvxl]

I decided to first try to write my own firmware for the PIC12F675. I could always switch to an ESP8266 as plan B. I'm happy to report that plan A worked out. The curtains have been operating flawlessly with my own firmware for over a month now.

[Bwired]

great project Schelte!
i have the Neta running for a very long time! and still working great!
Would like to control it over the esp8266 as well

[hvxl]

An esp8266 based system would be nice as it could also report back state information. I even already have the necessary parts. But now that simply loading my own firmware works so well, I have little incentive to completely rebuild the controller circuitry.

[Bwired]

I can understand