eXO Support

[flz]

Creating a discussion here to track the testing of the support for eXO devices.

Preliminary support has landed in the iaqualink library (https://github.com/flz/iaqualink-py, exo branch) and corresponding changes in home-assistant (https://github.com/flz/home-assistant, iaqualink-http2 branch).

Testing is a bit tricky, you'll need to do two things:

• Override the homeassistant/components/iaqualink directory in your home-assistant installation (or run a new docker installation straight from my git repository).
• Override the iaqualink directory that's installed (move it out of the way and put the new one in place).

Once you've got this running, you might want to set the HTTPX_LOG_LEVEL=trace environment variable for home-assistant which will give you some extra logging.

Support was added based on the work by @martinfrench92 in mk-french@6c7ac06.

I would be grateful if people could send me some "shadow" responses. I've worked from the one that was posted in the community forum but it's missing "vsp_speed" and "heating" for example and I'd like more samples of auxiliary devices and sensors.

[mk-french]

Nice one @flz,

Here's the full response with a few things redacted.
example_shadow.txt

Note I do not have the PH or ORP sensors.

[mk-french]

Maybe a few screenshots of the iaqualink app would help so I know what is used/actionable.

I've emailed a few your way.

That's odd, your sns_1 sensor seems to indicate a reading?

I assume it's just the default value given it is '7'.

I've got your new code running in dev against my eXO and it's throwing a few exceptions on some of the sensors. When I get a chance, I'll do some debugging and push the fixes.

[flz]

That's great, thanks!

I had a look at the screenshots. There's more that could be integrated in iaqualink to support setting schedules and attributes but I'm not sure it helps with automation.

[mk-french]

I agree. I couldn't think of much use for it so I never bothered...

[BadgerBadgerAndFox]

Just got my hydroxinator back online, zodiac swapped out my original exo. It’s got the orp, ph, temp sensors with a variable speed pump and color lights connected (lights are on aux due to requirement of Australian version which controls an external relay)
Would love to help but may need some assistance to get it working in docker, I’m running docker on synology, any chance this could get a docker hub build?

[ozrex75]

I'd love some guidance in testing this out. I've updated my HA docker to the latest (2021.12) but it's running on iaqualink-py 0.3.90, and when I try and overwrite the contents of /site-packages/iaqualink, I get:

Setup failed for iaqualink: Unable to import component: cannot import name 'AqualinkBinarySensor' from 'iaqualink' (/usr/local/lib/python3.9/site-packages/iaqualink/init.py)

[ozrex75]

Please see my attached shadow file
shadow.txt

I see a bunch of sensors created within HA, but no attributes or values. Here is part of my HA log:

2022-02-05 13:58:50 ERROR (MainThread) [homeassistant.util.logging] Exception in async_write_ha_state when dispatching 'iaqualink': ()
Traceback (most recent call last):
File "/config/custom_components/iaqualink/sensor.py", line 49, in native_value
state = int(self.dev.state)
ValueError: invalid literal for int() with base 10: 'V85R67'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "/usr/src/homeassistant/homeassistant/helpers/entity.py", line 530, in async_write_ha_state
self._async_write_ha_state()
File "/usr/src/homeassistant/homeassistant/helpers/entity.py", line 563, in _async_write_ha_state
state = self._stringify_state()
File "/usr/src/homeassistant/homeassistant/helpers/entity.py", line 536, in _stringify_state
if (state := self.state) is None:
File "/usr/src/homeassistant/homeassistant/components/sensor/init.py", line 371, in state
value = self.native_value
File "/config/custom_components/iaqualink/sensor.py", line 51, in native_value
state = float(self.dev.state)
ValueError: could not convert string to float: 'V85R67'

2022-02-05 13:58:50 ERROR (MainThread) [homeassistant.util.logging] Exception in async_write_ha_state when dispatching 'iaqualink': ()
Traceback (most recent call last):
File "/config/custom_components/iaqualink/sensor.py", line 49, in native_value
state = int(self.dev.state)
ValueError: invalid literal for int() with base 10: '24:00'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "/usr/src/homeassistant/homeassistant/helpers/entity.py", line 530, in async_write_ha_state
self._async_write_ha_state()
File "/usr/src/homeassistant/homeassistant/helpers/entity.py", line 563, in _async_write_ha_state
state = self._stringify_state()
File "/usr/src/homeassistant/homeassistant/helpers/entity.py", line 536, in _stringify_state
if (state := self.state) is None:
File "/usr/src/homeassistant/homeassistant/components/sensor/init.py", line 371, in state
value = self.native_value
File "/config/custom_components/iaqualink/sensor.py", line 51, in native_value
state = float(self.dev.state)
ValueError: could not convert string to float: '24:00'

[flz]

Thanks for the report! The code has an assumption that all sensors are numbers, that shouldn't be too hard to fix.

Do the rest of the devices work?

[ozrex75]

When I first checked, I couldn't see anything within the discovered devices, but I've re-checked and I do see some entities!

Note: It appears to create an individual device per entity. Should it be grouping them into one "device" with multiple entities within?

I can see switch entities for Aux1, Aux2, Boost, Low and Production. I tried turning Production off and on via the switch in HA and it appeared to work! Note if I toggle the switch, it reverts to it's previous state for a few seconds while it's updating on the device, then it appears to refresh to the new state.

It has created a device called "water temp" but has no entities within. There is another device with an entity named "temp" but only has a value of 1.

[ozrex75]

This appears to stop working after exactly 8 hours.

I get the following in the HA log:

2022-02-07 21:58:17 DEBUG (MainThread) [iaqualink] <- 401 Unauthorized - https://prod.zodiac-io.com/devices/v1/[REDACTED]/shadow

Perhaps some form of token is expiring? The Aqualink android app is pretty crappy and seems to suffer a similar issue. I found a workaround with the android app where you clear the app storage then re-sign in.

[mk-french]

This appears to stop working after exactly 8 hours.

I get the following in the HA log:

2022-02-07 21:58:17 DEBUG (MainThread) [iaqualink] <- 401 Unauthorized - https://prod.zodiac-io.com/devices/v1/[REDACTED]/shadow

Perhaps some form of token is expiring? The Aqualink android app is pretty crappy and seems to suffer a similar issue. I found a workaround with the android app where you clear the app storage then re-sign in.

I noticed this same thing when I originally extended the plug-in. I fixed it by attempting a re-login when this error is thrown. I agree, the app is crap, and I actually have better reliability with the plug-in. XD

[ozrex75]

Not sure if I have done this correctly, but see this pull request: #19

[benjycov]

Hi All,

I'm struggling to get this to work - I have an Exo IQ LS.

I have overridden the iaqualink component by deploying the one from (https://github.com/flz/home-assistant, iaqualink-http2 branch)), and putting it in custom_components, set the version in the manifest, re-added the integration and it looks like this in HA:

i.e. HA thinks it is using the custom component version.

But in the logs I get the same error:

Am I missing something with the override?

Regards... Ben

[h0eferm]

Hi @cloudbr34k84

I got this running using NodeRed:

https://discourse.nodered.org/t/iaqualink-for-pool-equipment-strike-again/63321/5?u=h0eferm

Best,
h0eferm

[cloudbr34k84]

Yeah ok, thanks but i dont know anything about Nodered..

[Veuchez]

HI @benjycov!
could you share your nodered flow please?
I can read the states of the various sensors of the eXO chlorinator but I cannot send the on/off commands, PH value setting etc.
Thanks!

[benjycov]

@Veuchez - happy to share again, but I have never actually made mine do any control, I simply use the NR to get pH, ORP, temp, etc.

[Veuchez]

@benjycov Ok, in the interface I saw control buttons and I thought I was able to send commands from NodeRed. How do you control the switching on and off of the pump, chlorinator, etc.?

[h0eferm]

Hi @flz and @martinfrench92,

sorry for bothering you, but could you please explain how to integrate the exo support? Would appreciate your support.

I'm getting following error after overwriting the directory:

Setup failed for iaqualink: Unable to import component: cannot import name 'AqualinkToggle' from 'iaqualink.device' (/srv/homeassistant/lib/python3.9/site-packages/iaqualink/device.py)

Many thanks and best,
Marcus

[mk-french]

I guess if you have AUX2 set away from 'Heater' then it doesn't appear in the data - makes sense...

Just pushed a patch.
Set the required version to 0.5.2 and see how you go!
"requirements": ["git+https://github.com/martinfrench92/iaqualink-py.git@zs500#iaqualink==0.5.2"]

-Martin

[BadgerBadgerAndFox]

Thanks, that fixed it. I'll start playing around with it and grab the shadow responses.

[andres-canello]

Hi @martinfrench92 , I've tried this branch and hit this error, any ideas?

This error originated from a custom integration.

Logger: homeassistant.config_entries
Source: custom_components/iaqualink/init.py:121
Integration: Jandy iAqualink
First occurred: 17:00:32 (1 occurrences)
Last logged: 17:00:32

Error setting up entry [redacted] for iaqualink
Traceback (most recent call last):
File "/usr/src/homeassistant/homeassistant/config_entries.py", line 357, in async_setup
result = await component.async_setup_entry(hass, self)
File "/config/custom_components/iaqualink/init.py", line 121, in async_setup_entry
devices = await systems[0].get_devices()
File "/usr/local/lib/python3.10/site-packages/iaqualink/system.py", line 75, in get_devices
await self.update()
File "/usr/local/lib/python3.10/site-packages/iaqualink/system.py", line 221, in update
await self._parse_shadow_response(r1)
File "/usr/local/lib/python3.10/site-packages/iaqualink/system.py", line 247, in _parse_shadow_response
devices.pop("vsp_speed") # temp remove until can handle dictionary
KeyError: 'vsp_speed'

[mk-french]

@andres-canello, I have pushed a patch to patch the patch. ;)

Somehow the default return argument of the pop function was dropped in the last fix. You can see the exception is showing that it can't find the 'vsp_speed' key, (I assume) you don't have a variable speed pump so the data is not there...

Grab 0.5.3 and see how you go!

-Martin

[andres-canello]

@martinfrench92 , worked like a charm. I have a regular pump, not a variable speed one. Thanks a lot!

[benjycov]

Hi Cloud, I'm not sure why that is happening, mine is definitely the right way around. It may be that my equipment is slightly different to yours (I have an Exo LS with ORP & PH control and a Zodiac VSP pump). What you can do is look at the whole object returned by the controller and see what fields are useful to you (this also answers your earlier question) and then alter the sensors NR creates in HA accordingly. If you add a debug node. [image: image.png] If you then view the debug window (once the flow has run again), you will see the whole message structure which you can explore, there is lots and lots of data available. I am using this field (object -> state -> equipment -> swc_0 -> filter pump -> state. [image: image.png] I hope that helps. Regards... Ben

…

On Mon, 18 Jul 2022 at 10:24, cloudbr34k84 ***@***.***> wrote: Hi Cloud, I had the same problem as you, i just couldn't for the life of me get the exo version of the iAqualink to work in HA (even with Martins excellent). So I used the code to write my own Node Red flow, which queries the API and populates some HA sensors for PH, ORP, Temp, etc. I have attached the flow so you should be able to import it into NR, put your own username and password in (search the text for ***@***.*** and YourPassword). I hope that helps until we can get the exo version into HA properly. Regards... Ben exo-flows.txt <https://github.com/flz/iaqualink-py/files/9120391/exo-flows.txt> Hi ben So for some reason the time- in which my Pump turns on seems to be opposite. See image. Any thoughts?? [image: image] <https://user-images.githubusercontent.com/58960644/179482337-5aa865af-6ccd-4161-a295-4380d7f80058.png> — Reply to this email directly, view it on GitHub <#16 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AQZVQ2ENLLP6KACNV34MMSTVUUPEPANCNFSM5LCZRZ7A> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[cloudbr34k84]

Hey Ben, this was the first using NR lol so bare with me. I cant see any of your pictures btw. Could you resend them please?

[benjycov]

Apologies for the delay! Here is the screenshot with the debug node added, and then in the debug frame you can see the full object outputted and expanded. There is a lot of fun data to play with!

[cloudbr34k84]

No worries at all.
Ok so i have added Debig as per picture, and re rerun the node. I then click on the debug green node and then click the Bug icon, bt i dont see anything. what am i missing. Can you export yours again please.

[benjycov]

Hi Brad, This is an easy one, you have attached the debug node to the wrong output on the 'Has token expired?' switch, reattach it to the bottom output (the same one that feeds the HA entity nodes), and you should get an output in the debug. Regards... Ben

…

On Sun, 24 Jul 2022 at 07:42, cloudbr34k84 ***@***.***> wrote: No worries at all. Ok so i have added Debig as per picture, and re rerun the node. I then click on the debug green node and then click the Bug icon, bt i dont see anything. what am i missing. Can you export yours again please. [image: image] <https://user-images.githubusercontent.com/58960644/180635682-171f0393-4f43-4747-bedc-b1e746a484ef.png> — Reply to this email directly, view it on GitHub <#16 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AQZVQ2GTTUZIWQHFNHBGWOLVVTQWBANCNFSM5LCZRZ7A> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[cloudbr34k84]

ok great! thank you..

I have a Variable speed pump but its not connected to Zodiac system
I dont have PH sensors or ORP, but I do have temperature

how to do i modify the NR so i can get more entities etc??

[benjycov]

No worries. If you look at the properties of the ORP, PH HA nodes you should be able to copy them, change the name and path of the property you want exposed as a HA entity and go from there. Out of interest what other info are you interested in?

…

On Sun, 24 Jul 2022 at 14:14, cloudbr34k84 ***@***.***> wrote: ok great! thank you.. I have a Variable speed pump but its not connected to Zodiac system I dont have PH sensors or ORP, but I do have temperature how to do i modify the NR so i can get more entities etc?? — Reply to this email directly, view it on GitHub <#16 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AQZVQ2FWEYQMELAXWBWYN63VVUXUBANCNFSM5LCZRZ7A> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[cloudbr34k84]

I figured it out!

Under Set Headers you have specified your Device ID, so i changed it to mine! now
https://prod.zodiac-io.com/devices/v1/JT20002421/shadow

[benjycov]

Hi Brad,

Wow! Just Wow... Basically we (well you really) just discovered a security flaw in iAqualink. I just set up a new flow and put in your serial number and I am indeed able to query your system.

In a nutshell what they have done is trusted that anyone authenticated on the first phase who gains a valid serial number can then query any system.

I will put something together and report this to Zodiac/iAqualink tomorrow. (its late here).

In the meantime please can you edit your post above and obfuscate both of our serial numbers!!

Thanks... Ben

[benjycov]

Hi Brad,

Just to add, I knocked up another NR flow to sequentially check serial numbers, and sure enough I quickly found lots of different system IDs that I can query.

Presumably we could also change settings and stop start these systems.

Ben

[cloudbr34k84]

yep removed the ID's, apologies!
damn! they are lazy!

[benjycov]

It happens, they just failed to check that the authenticated user has that system ID registered to their user.

I've dropped them an email with the problem, lets see what they come back with!

[ChutneyMary]

Hi Ben, Just be sure to know that this has always been an unofficial API. The API key used in Node-Red is generic for all users and came about through someone sniffing the eXO using Wireshark. Zodiac has never published an open API and doesn't look likely to despite many requests from users. The consequences of you reporting this is that all users here will possibly lose their ability to integrate 3rd party devices for monitoring their pool (by the removal of the generic API key or otherwise). The onus is on users to protect their Device ID's. Are Zodiac completely lazy in not providing an official API? - Yes! Are Zodiac completely useless in not securing their systems? - Yes! Do I care if someone else can read my pool pH or ORP? - No. Do I care if I can't integrate my home automation monitoring system with my pool? - Yes.

…

On Mon, 25 Jul 2022 at 08:06, benjycov ***@***.***> wrote: Hi Brad, Wow! Just Wow... Basically we (well you really) just discovered a security flaw in iAqualink. I just set up a new flow and put in your serial number and I am indeed able to query your system. In a nutshell what they have done is trusted that anyone authenticated on the first phase who gains a valid serial number can then query any system. I will put something together and report this to Zodiac/iAqualink tomorrow. (its late here). In the meantime please can you edit your post above and obfuscate both of our serial numbers!! Thanks... Ben — Reply to this email directly, view it on GitHub <#16 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AH3MQTAPF62PZVUWRCEKF6TVVXK7BANCNFSM5LCZRZ7A> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[cloudbr34k84]

How do you protect your user ID when someone can just query them like ben mentioned?

[ChutneyMary]

Zodiac has no motivation or intention to provide 3rd party access to their systems. If you have used their useless iAqualink App for awhile, you will know that they can also force firmware updates on 'their' equipment (with no user ability to decline the update).
So the ability to close off 3rd party access is real and an unfortunate consequence of their backward philosophy in IoT.

If you wish to keep access with HA, I would be inclined to not post Node-Red flows demonstrating security vulnerabilities.

[Harrisdamon]

@cloudbr34k84 Thank You Brad! Very happy.

[cloudbr34k84]

too easy! hope it helps. Its only sensors, but still good to know when things are on

[Harrisdamon]

I see all the entities now in HA, do you know if they can be automated?

[cloudbr34k84]

what do you mean automated?? They are only sensors, or binary sensors

[Harrisdamon]

Just need to be able to turn the filter pump on or off based on solar export

[cloudbr34k84]

@benjycov I was wondering if you could use the same method to access my electricity account details??
https://www.agl.com.au/
There is an app as well

[hybrixxx]

Ciao, since last night I'm unable to switch ON/OFF the EXO IQ from the API. It returns an INTERNAL SERVER ERROR. Am I the only one? Authentication is working, querying status is working, but updating production status is not working:

Login=$(curl  -X POST -H "Host:prod.zodiac-io.com" -H "accept:application/json" -H "content-type:application/json" -H "accept-encoding:gzip" -H "user-agent:okhttp/3.12.0" -d '{"api_key":"xxxxxxx", "email":"xxxxxxx", "password":"xxxxxx"}' "https://prod.zodiac-io.com/users/v1/login")

AuthenticationToken=`echo ${Login} | jq -r '.authentication_token'`

Authorization=`echo ${Login} | jq -r '.userPoolOAuth.IdToken'`

curl -X POST -H "Host:prod.zodiac-io.com" -H "accept:application/json" -H "authorization:"$Authorization"" -H "content-type:application/json; charset=UTF-8" -H "content-length:62" -H "accept-encoding:gzip" -H "user-agent:okhttp/3.12.0" -d '{"state":{"desired":{"equipment":{"swc_0":{"production":1}}}}}' "https://prod.zodiac-io.com/devices/v1/xxxxxxxx/shadow"

[BadgerBadgerAndFox]

Have you tried updating the app to current version?
You could reset the exo (hidden service menu is long press on the ok button from memory) and delete in the app and pair again.
I’ve reset mine a few times without issue.

[mk-french]

I also had the issue with the app not being able to load my eXO device for a long time. I didn’t really care to fix it as I just used HA. I fixed it recently by removing and re-pairing with my account and also re installing the app (not sure if that was required…).

Regarding the API issue, I spent a small about of time snooping to see if something obvious had changed but wasn’t able to pick up any of the eXO related exchanges through the proxy. Interestingly the commands weren’t being sent to the device in that state. I had to remove the MITM for it it start working again. Requires more investigating!

[BadgerBadgerAndFox]

Thanks Martin, hopefully it’s not something insurmountable.

[BadgerBadgerAndFox]

This warning is on the website:
iAquaLink Update: Be sure to upgrade to the latest version of the iAquaLink app (v6.4.2) for important feature and security updates. With this new update, previous versions of the app will no longer be supported and will have limited functionality. If you already have version v6.4.2, you’re all set--no further action needed!

https://au.iaqualink.com/en-au/manuals

[hybrixxx]

I already have the updated APP since some time ago. I'll reset the EXO IQ to try to recover access to the APP and later I will try the sniffer again.

[hybrixxx]

Error, It was fully compatible with old exo IQ un 2019 until It stop being compatible. I have chlorinator and robot from 2019 and now only robot works un the web. El jue, 29 sept 2022 8:05, BadgerBadgerAndFox ***@***.***> escribió:

…

The iAquaLink web portal is not compatible and has never been compatible with the new exo/hydroxinator models. It’s only used by the older iAquaLink devices. — Reply to this email directly, view it on GitHub <#16 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ALIVYA7PEP2JFKXJ5UM2YJLWAUWRFANCNFSM5LCZRZ7A> . You are receiving this because you commented.Message ID: ***@***.***>

[BadgerBadgerAndFox]

Interesting I got my exo late 2019 and it was never compatible, confirmed by zodiac support in the US and locally that the web interface is not for them.

[BadgerBadgerAndFox]

The app is currently presenting a warning that if you are using a version older than the current 6.4.2 it will not function, I assume this means they have changed the API authentication token

[hybrixxx]

Ciao, the API authentication process is the same. With CURL commands i'm still able to use same process I had to get the token. The problem is only when I request the API to update any parameter to a new desired value.

[benjycov]

I’ll just throw my 10p in. I also got my Exo in 2019 and I swear that when I first got it it worked on the web interface. It is still listed on the web interface but fails when you try and interrogate it.

…

On Thu, 29 Sep 2022 at 08:51, BadgerBadgerAndFox ***@***.***> wrote: Interesting I got my exo late 2019 and it was never compatible, confirmed by zodiac support in the US and locally that the web interface is not for them. — Reply to this email directly, view it on GitHub <#16 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AQZVQ2ARQPDCNGJD4S6KPNTWAVC7BANCNFSM5LCZRZ7A> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[hybrixxx]

I have asked FLUIDRA and theyconfirmed it will never work in the web again as they web is not compatible anymore with the EXO IQ. So lets cross fingers that they fix the API...

[hybrixxx]

I've reseted the EXO IQ and repaired with Aqualink succesffully. My issue now is that HTTPCanary is not working as it did in the past, I'm not able to sniff the communications :(

[flz]

Apologies for the radio silence, I've updated the exo branch in iaqualink-py as well as the iaqualink-http2 branch in home-assistant. If you check out the latter and update the manifest to include @exo, it should be all you need to test. I personally open it in VSCode and do "Reopen in Container".

I don't have the hardware to test so I'll rely on people to tell me what's working and what isn't. I understand there might have been changes on the eXO API side, the recent iAqualink API changes should be fixed (e.g. errors caused by "absent" sensors).

[hybrixxx]

Ciao, do you have a sample of the new curl format required by the API to turn on/off the exo IQ? Thanks

[GizzGool]

Hello

First of all, a big thank you to everyone who helped us connect our eXo to Home Assistant, as the latter isn't officially supported (yet?), especially @benjycov for the YAML files. 👍

I just have one question. When we look at the return of my exo_data variable, I see the following data:
MQTT connection: 2
MQTT disconnection total: 1
OTA State: 0
Number of reboots from the MQTT issue: 15631

So, it looks like the device would be able to connect directly to an MQTT broker?
Or is it simply used on the iAqualink server side, which uses an MQTT broker, given the current MQTT connection count = 2 (and why is the total MQTT disconnection = 1?)

It would indeed be much simpler if we could locally point the eXo to our broker...

Thank you in advance for your clarification.

[benjycov]

Yes I have wondered this myself, could we intercept the connection from the exo to Zodiac with DNS poisoning or IP natting... maybe... but I assume it uses some kind of authentication/VPN layer that we would also have to imitate... I'll have a quick look at the data leaving my exo and see if there is anything obvious we can intercept.

[GizzGool]

Thank you for your super quick response @benjycov, I really appreciate your help and therefore I have some hope with your research 😃

[benjycov]

Okay I decided to spend a couple of hours sat by the pool trying to do this! :) Alas I failed. :(

After having a poke around its using the Amazon AWS IOT service (mqtt with some buzzers and bells), it uses TLS on port 8883. I got the traffic flowing to my mitmproxy, but while I saw TLS connection - the attempts fail, which confirms enforcement.

Basically the next step would be actually interrogating the firmware on the controller board to try and extract the certificate... as it happens I do have a spare board for me exo - but its a pretty deep job and as the cloud API method works for now - I think I'll leave that one for another day :)

[GizzGool]

Thank you very much, Ben, for taking the time.👍
It's a real shame that major manufacturers are still spending time locking everything down. They could at least give the end consumer a choice (with the worst-case scenario of losing support).
I'm also experiencing these difficulties with Somfy's proprietary 🤬 IO protocol and my roller shutters.
Because, of course, I think there's a lot of pressure. Somfy has provided a local API, but it still forces us to use their gateway, which unfortunately, in my case, causes problems due to the distance of certain devices from it, which regularly go offline.
I would so love to be able to control the roller shutters locally (with a small ESP box, for example) without having to hack a remote control!
So I hope some tinkerers will break the eXo's firmware, in particular...

[benjycov]

Hi, Yes so if it’s not stable on your aquiline app it might point to wider probs… for ref I find the app quite stable (it’s all backed off against Amazon)… so it should ‘just work’. Even with that, expect you to get something, I’d double check your login/password and the system ID starting ‘JT….’. If you can’t get it working, I’ll have a look at how you can give us a bit more debugging info to get you sorted :) Regards… Ben

…

On Fri, 8 Aug 2025 at 00:12, riccidan ***@***.***> wrote: Hi Ben, Thanks for the file Mate, well I dont have errors now, but I dont have any Zodiac sensors either. That said my Zodiac is often showing as "offline" on the webapp so I'm guessing that's more of a cloud issue. The exo_data sensor is showing as 'unavailable', would that be right? — Reply to this email directly, view it on GitHub <#16 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AQZVQ2D5A2N3VOUQGKQBAJL3MPFNXAVCNFSM6AAAAABZQ3YT5KVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTIMBTHE3TGMY> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[riccidan]

Hi Ben, Thanks for the offer. I dont think it's a cloud issue, rather a WiFi connectivity issue with the hydroxinator. I have an open ticket with Zodiac/Fluidra but I dont hold much hope as it's out of warranty and they're just walking me through the usual re-connection steps. My robot cleaner is online, but the pool unit consistently goes offline. Have even tried creating a dedicated 2.4ghz guest network for it exclusively, to no avail. Regards, Ricci

…

On Fri, 8 Aug 2025 at 08:21, benjycov ***@***.***> wrote: Hi, Yes so if it’s not stable on your aquiline app it might point to wider probs… for ref I find the app quite stable (it’s all backed off against Amazon)… so it should ‘just work’. Even with that, expect you to get something, I’d double check your login/password and the system ID starting ‘JT….’. If you can’t get it working, I’ll have a look at how you can give us a bit more debugging info to get you sorted :) Regards… Ben On Fri, 8 Aug 2025 at 00:12, riccidan ***@***.***> wrote: > Hi Ben, > > Thanks for the file Mate, well I dont have errors now, but I dont have any > Zodiac sensors either. > > That said my Zodiac is often showing as "offline" on the webapp so I'm > guessing that's more of a cloud issue. The exo_data sensor is showing as > 'unavailable', would that be right? > > — > Reply to this email directly, view it on GitHub > < #16 (reply in thread)>, > or unsubscribe > < https://github.com/notifications/unsubscribe-auth/AQZVQ2D5A2N3VOUQGKQBAJL3MPFNXAVCNFSM6AAAAABZQ3YT5KVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTIMBTHE3TGMY> > . > You are receiving this because you were mentioned.Message ID: > ***@***.***> > — Reply to this email directly, view it on GitHub <#16 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AQU24FTBQVPAMMAIFO4ANR33MPGONAVCNFSM6AAAAABZQ3YT5KVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTIMBTHE3TMNA> . You are receiving this because you commented.Message ID: ***@***.***>

[benjycov]

Actually, I went through a whole support thing with them over WiFi, the exo would only sometimes connect. I tried deploying a new access point in the pool house right next to the exo, still same issue. In the end they sent me a whole new main board… turns out that wasn’t the issue. It’s really sensitive about connection quality/frame size. For example I found that if forced the through the VPN to my other house it connected fine. Then I ditched the crappy copper connection got Starlink and now fibre and never had the problem since. Check if your router has a setting for MSS and lower it, it’s probably something like 1420, try dropping it to 1200, if that improves things iteratively increase it until you find the sweet spot. I hope that helps…

…

On Fri, 8 Aug 2025 at 00:39, riccidan ***@***.***> wrote: Hi Ben, Thanks for the offer. I dont think it's a cloud issue, rather a WiFi connectivity issue with the hydroxinator. I have an open ticket with Zodiac/Fluidra but I dont hold much hope as it's out of warranty and they're just walking me through the usual re-connection steps. My robot cleaner is online, but the pool unit consistently goes offline. Have even tried creating a dedicated 2.4ghz guest network for it exclusively, to no avail. Regards, Ricci On Fri, 8 Aug 2025 at 08:21, benjycov ***@***.***> wrote: > Hi, > > Yes so if it’s not stable on your aquiline app it might point to wider > probs… for ref I find the app quite stable (it’s all backed off against > Amazon)… so it should ‘just work’. > > Even with that, expect you to get something, I’d double check your > login/password and the system ID starting ‘JT….’. > > If you can’t get it working, I’ll have a look at how you can give us a bit > more debugging info to get you sorted :) > > Regards… Ben > > On Fri, 8 Aug 2025 at 00:12, riccidan ***@***.***> wrote: > > > Hi Ben, > > > > Thanks for the file Mate, well I dont have errors now, but I dont have > any > > Zodiac sensors either. > > > > That said my Zodiac is often showing as "offline" on the webapp so I'm > > guessing that's more of a cloud issue. The exo_data sensor is showing as > > 'unavailable', would that be right? > > > > — > > Reply to this email directly, view it on GitHub > > < > #16 (reply in thread)>, > > > or unsubscribe > > < > https://github.com/notifications/unsubscribe-auth/AQZVQ2D5A2N3VOUQGKQBAJL3MPFNXAVCNFSM6AAAAABZQ3YT5KVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTIMBTHE3TGMY> > > > . > > You are receiving this because you were mentioned.Message ID: > > ***@***.***> > > > > — > Reply to this email directly, view it on GitHub > < #16 (comment)>, > or unsubscribe > < https://github.com/notifications/unsubscribe-auth/AQU24FTBQVPAMMAIFO4ANR33MPGONAVCNFSM6AAAAABZQ3YT5KVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTIMBTHE3TMNA> > . > You are receiving this because you commented.Message ID: > ***@***.***> > — Reply to this email directly, view it on GitHub <#16 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AQZVQ2E7VEY3G6W65RF3CIL3MPIRDAVCNFSM6AAAAABZQ3YT5KVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTIMBTHE4TINI> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[riccidan]

That's actually really interesting, thanks for the tip Mate. To be honest the Guest 2.4ghz network hasn't improved connectivity so I suspect Im dealing with the same issue you did. My (Telstra Smart Modem Gen3) router doesn't have MSS settings. Does the board have a USB port by chance? Wondering if it might be possible to add an external wifi dongle.

…

On Fri, 8 Aug 2025, 8:51 am benjycov, ***@***.***> wrote: Actually, I went through a whole support thing with them over WiFi, the exo would only sometimes connect. I tried deploying a new access point in the pool house right next to the exo, still same issue. In the end they sent me a whole new main board… turns out that wasn’t the issue. It’s really sensitive about connection quality/frame size. For example I found that if forced the through the VPN to my other house it connected fine. Then I ditched the crappy copper connection got Starlink and now fibre and never had the problem since. Check if your router has a setting for MSS and lower it, it’s probably something like 1420, try dropping it to 1200, if that improves things iteratively increase it until you find the sweet spot. I hope that helps… On Fri, 8 Aug 2025 at 00:39, riccidan ***@***.***> wrote: > Hi Ben, > > Thanks for the offer. I dont think it's a cloud issue, rather a WiFi > connectivity issue with the hydroxinator. > > I have an open ticket with Zodiac/Fluidra but I dont hold much hope as > it's > out of warranty and they're just walking me through the usual > re-connection > steps. > > My robot cleaner is online, but the pool unit consistently goes offline. > > Have even tried creating a dedicated 2.4ghz guest network for it > exclusively, to no avail. > > Regards, > > Ricci > > On Fri, 8 Aug 2025 at 08:21, benjycov ***@***.***> wrote: > > > Hi, > > > > Yes so if it’s not stable on your aquiline app it might point to wider > > probs… for ref I find the app quite stable (it’s all backed off against > > Amazon)… so it should ‘just work’. > > > > Even with that, expect you to get something, I’d double check your > > login/password and the system ID starting ‘JT….’. > > > > If you can’t get it working, I’ll have a look at how you can give us a > bit > > more debugging info to get you sorted :) > > > > Regards… Ben > > > > On Fri, 8 Aug 2025 at 00:12, riccidan ***@***.***> wrote: > > > > > Hi Ben, > > > > > > Thanks for the file Mate, well I dont have errors now, but I dont have > > any > > > Zodiac sensors either. > > > > > > That said my Zodiac is often showing as "offline" on the webapp so I'm > > > guessing that's more of a cloud issue. The exo_data sensor is showing > as > > > 'unavailable', would that be right? > > > > > > — > > > Reply to this email directly, view it on GitHub > > > < > > > #16 (reply in thread)>, > > > > > > or unsubscribe > > > < > > > https://github.com/notifications/unsubscribe-auth/AQZVQ2D5A2N3VOUQGKQBAJL3MPFNXAVCNFSM6AAAAABZQ3YT5KVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTIMBTHE3TGMY> > > > > > > . > > > You are receiving this because you were mentioned.Message ID: > > > ***@***.***> > > > > > > > — > > Reply to this email directly, view it on GitHub > > < > #16 (comment)>, > > > or unsubscribe > > < > https://github.com/notifications/unsubscribe-auth/AQU24FTBQVPAMMAIFO4ANR33MPGONAVCNFSM6AAAAABZQ3YT5KVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTIMBTHE3TMNA> > > > . > > You are receiving this because you commented.Message ID: > > ***@***.***> > > > > — > Reply to this email directly, view it on GitHub > < #16 (comment)>, > or unsubscribe > < https://github.com/notifications/unsubscribe-auth/AQZVQ2E7VEY3G6W65RF3CIL3MPIRDAVCNFSM6AAAAABZQ3YT5KVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTIMBTHE4TINI> > . > You are receiving this because you were mentioned.Message ID: > ***@***.***> > — Reply to this email directly, view it on GitHub <#16 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AQU24FV6I2PUHFGYWAOANHD3MPJ6JAVCNFSM6AAAAABZQ3YT5KVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTIMBUGAYTANY> . You are receiving this because you commented.Message ID: ***@***.***>

[GizzGool]

Hi Ricci,
For your information and to perhaps help you, I also had problems with WiFi disconnection due to the poolhouse being too far from the house router (the eXo does not seem to be very well equipped in terms of antennas!). I had to install an additional Asus RP-AX58 access point (in order to take advantage of the Mesh of an Asus router) which is located just a few meters from the eXo and since then everything has been ok.

[riccidan]

Ignore that question, just realised it's firmware driven.

…

On Fri, 8 Aug 2025, 2:04 pm Ricci Danieletto, ***@***.***> wrote: That's actually really interesting, thanks for the tip Mate. To be honest the Guest 2.4ghz network hasn't improved connectivity so I suspect Im dealing with the same issue you did. My (Telstra Smart Modem Gen3) router doesn't have MSS settings. Does the board have a USB port by chance? Wondering if it might be possible to add an external wifi dongle. On Fri, 8 Aug 2025, 8:51 am benjycov, ***@***.***> wrote: > Actually, I went through a whole support thing with them over WiFi, the > exo > would only sometimes connect. I tried deploying a new access point in the > pool house right next to the exo, still same issue. > > In the end they sent me a whole new main board… turns out that wasn’t the > issue. > > It’s really sensitive about connection quality/frame size. For example I > found that if forced the through the VPN to my other house it connected > fine. Then I ditched the crappy copper connection got Starlink and now > fibre and never had the problem since. > > Check if your router has a setting for MSS and lower it, it’s probably > something like 1420, try dropping it to 1200, if that improves things > iteratively increase it until you find the sweet spot. > > I hope that helps… > > > > > On Fri, 8 Aug 2025 at 00:39, riccidan ***@***.***> wrote: > > > Hi Ben, > > > > Thanks for the offer. I dont think it's a cloud issue, rather a WiFi > > connectivity issue with the hydroxinator. > > > > I have an open ticket with Zodiac/Fluidra but I dont hold much hope as > > it's > > out of warranty and they're just walking me through the usual > > re-connection > > steps. > > > > My robot cleaner is online, but the pool unit consistently goes > offline. > > > > Have even tried creating a dedicated 2.4ghz guest network for it > > exclusively, to no avail. > > > > Regards, > > > > Ricci > > > > On Fri, 8 Aug 2025 at 08:21, benjycov ***@***.***> wrote: > > > > > Hi, > > > > > > Yes so if it’s not stable on your aquiline app it might point to > wider > > > probs… for ref I find the app quite stable (it’s all backed off > against > > > Amazon)… so it should ‘just work’. > > > > > > Even with that, expect you to get something, I’d double check your > > > login/password and the system ID starting ‘JT….’. > > > > > > If you can’t get it working, I’ll have a look at how you can give us > a > > bit > > > more debugging info to get you sorted :) > > > > > > Regards… Ben > > > > > > On Fri, 8 Aug 2025 at 00:12, riccidan ***@***.***> wrote: > > > > > > > Hi Ben, > > > > > > > > Thanks for the file Mate, well I dont have errors now, but I dont > have > > > any > > > > Zodiac sensors either. > > > > > > > > That said my Zodiac is often showing as "offline" on the webapp so > I'm > > > > guessing that's more of a cloud issue. The exo_data sensor is > showing > > as > > > > 'unavailable', would that be right? > > > > > > > > — > > > > Reply to this email directly, view it on GitHub > > > > < > > > > > > #16 (reply in thread)>, > > > > > > > > > > or unsubscribe > > > > < > > > > > > https://github.com/notifications/unsubscribe-auth/AQZVQ2D5A2N3VOUQGKQBAJL3MPFNXAVCNFSM6AAAAABZQ3YT5KVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTIMBTHE3TGMY> > > > > > > > > > > . > > > > You are receiving this because you were mentioned.Message ID: > > > > ***@***.***> > > > > > > > > > > — > > > Reply to this email directly, view it on GitHub > > > < > > > #16 (comment)>, > > > > > > or unsubscribe > > > < > > > https://github.com/notifications/unsubscribe-auth/AQU24FTBQVPAMMAIFO4ANR33MPGONAVCNFSM6AAAAABZQ3YT5KVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTIMBTHE3TMNA> > > > > > > . > > > You are receiving this because you commented.Message ID: > > > ***@***.***> > > > > > > > — > > Reply to this email directly, view it on GitHub > > < > #16 (comment)>, > > > or unsubscribe > > < > https://github.com/notifications/unsubscribe-auth/AQZVQ2E7VEY3G6W65RF3CIL3MPIRDAVCNFSM6AAAAABZQ3YT5KVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTIMBTHE4TINI> > > > . > > You are receiving this because you were mentioned.Message ID: > > ***@***.***> > > > > — > Reply to this email directly, view it on GitHub > <#16 (comment)>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/AQU24FV6I2PUHFGYWAOANHD3MPJ6JAVCNFSM6AAAAABZQ3YT5KVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTIMBUGAYTANY> > . > You are receiving this because you commented.Message ID: > ***@***.***> >

[benjycov]

As it happens I also have Somfy shutters (on my Poolhouse)… I use RFLink to control them locally - have you looked at that as an option? I also have their cloud hub thing in a drawer somewhere - but I found it too laggy.

…

On Fri, 8 Aug 2025 at 09:36, GizzGool ***@***.***> wrote: Thank you very much, Ben, for taking the time.👍 It's a real shame that major manufacturers are still spending time locking everything down. They could at least give the end consumer a choice (with the worst-case scenario of losing support). I'm also experiencing these difficulties with Somfy's proprietary 🤬 IO protocol and my roller shutters. Because, of course, I think there's a lot of pressure. Somfy has provided a local API, but it still forces us to use their gateway, which unfortunately, in my case, causes problems due to the distance of certain devices from it, which regularly go offline. I would so love to be able to control the roller shutters locally (with a small ESP box, for example) without having to hack a remote control! So I hope some tinkerers will break the eXo's firmware, in particular... — Reply to this email directly, view it on GitHub <#16 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AQZVQ2AFCEVQOO54Y3J6JYT3MRHQXAVCNFSM6AAAAABZQ3YT5KVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTIMBUGM2TIOA> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[GizzGool]

Thank you very much Ben for your response.
Unfortunately, I cannot use this solution because SOMFY now uses the proprietary IO protocol and with an RFLink you can only control the old roller shutters which used the RTS radio protocol 🤨

[benjycov]

Fellow eXo owners - grab a beer, sit down next to your pool and relax, for I have news, and it is very good news.

I've written a brand new integration for the eXo, no node red, no merging code, no templates, no rest sensors, just one integration, and you can get it right now, here!!!

Go easy on me, because this is my first every integration. Let me know how you get on and if you can let me know (in the discussion or issues) which Exo you are using with what extras that would be great. :)

Enjoy!

Ben

[GizzGool]

You rocks 🍻

Addition: works perfectly the first time! Again, thanks so much for the development

[riccidan]

Legend! Thanks for giving this a crack. Now all I have to do is fix my WiFi connectivity to my EXO. :(

…

On Thu, 7 Aug 2025 at 23:16, GizzGool ***@***.***> wrote: Hello First of all, a big thank you to everyone who helped us connect our eXo to Home Assistant, as the latter isn't officially supported (yet?), especially @benjycov <https://github.com/benjycov> for the YAML files. 👍 I just have one question. When we look at the return of my exo_data variable, I see the following data: MQTT connection: 2 MQTT disconnection total: 1 OTA State: 0 Number of reboots from the MQTT issue: 15631 So, it looks like the device would be able to connect directly to an MQTT broker? Or is it simply used on the iAqualink server side, which uses an MQTT broker, given the current MQTT connection count = 2 (and why is the total MQTT disconnection = 1?) It would indeed be much simpler if we could locally point the eXo to our broker... Thank you in advance for your clarification. — Reply to this email directly, view it on GitHub <#16 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AQU24FQUTGDZFEU436HNNNL3MNGUXAVCNFSM6AAAAABZQ3YT5KVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTIMBTGQ4TQMQ> . You are receiving this because you commented.Message ID: ***@***.***>

[delbz]

Wow. Just replaced the rest sensors with this integration. Very impressive. Thank you for your dedication to solving this challenge.

[GizzGool]

Hello Ben,
Your eXo integration is reporting an error code "3" to me, knowing that the device does not display one locally and also seems to be working correctly.
So, could you please provide a table of correspondence between the error numbers and their meaning, because in the documentation, Zodiac does not mention a number for errors?
Thank you in advance and thank you again for your integration which now greatly simplifies our lives!

[benjycov]

Hi Gizz, So first of all if you look at the attributes for the error it should tell you (if we know) what the text relates to - as it happens 3 is 'Low Conductivity'. There are some error codes I don't know - discussion here: benjycov/exo_pool#1 Maybe I need to expose the text as an entity in its own right? Your exo controller should have a flashing flight and be showing low conductivity on the screen. Low conductivity tends to be quite transient, and can mean the salt level is low or in my case it meant I needed a new set of blades in the chlorinator. Are you finding that the chlorinator is struggling to keep up with demand per chance? Regards... Ben

…

On Sun, 10 Aug 2025 at 15:19, GizzGool ***@***.***> wrote: Hello Ben, Your eXo integration is reporting an error code "3" to me, knowing that the device does not display one locally and also seems to be working correctly. So, could you please provide a table of correspondence between the error numbers and their meaning, because in the documentation, Zodiac does not mention a number for errors? Thank you in advance and thank you again for your integration which now greatly simplifies our lives! — Reply to this email directly, view it on GitHub <#16 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AQZVQ2CDSQSHNKNSHXFPEW33M5IGBAVCNFSM6AAAAABZQ3YT5KVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTIMBWGQZTGOA> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[GizzGool]

Ben, you're really too fast, if all the supports in the world could be like you 🤩
New version installed, it's perfect!
Indeed, it seems the message is intermittent on my end, knowing that I'm well above the salt recommendation. Great for the finances if I have to change the blades in the middle of summer!
Finally, I don't know if I'm the best person to tell you if the chlorinator is having trouble responding to the request? In any case, the upward update is done fairly quickly (as example once I've cleared the error). I'll run some tests nearby with a change in pH or ORP value to let you know in this way.
Or you can also ask me more specifically about the tests you would like me to do on my side, it will be with pleasure.
Thank you again for your responsiveness.

[benjycov]

I've uploaded a new release which now has a sensor that shows the current error code translation.

…

On Sun, 10 Aug 2025 at 16:42, Ben King ***@***.***> wrote: Hi Gizz, So first of all if you look at the attributes for the error it should tell you (if we know) what the text relates to - as it happens 3 is 'Low Conductivity'. There are some error codes I don't know - discussion here: benjycov/exo_pool#1 Maybe I need to expose the text as an entity in its own right? Your exo controller should have a flashing flight and be showing low conductivity on the screen. Low conductivity tends to be quite transient, and can mean the salt level is low or in my case it meant I needed a new set of blades in the chlorinator. Are you finding that the chlorinator is struggling to keep up with demand per chance? Regards... Ben On Sun, 10 Aug 2025 at 15:19, GizzGool ***@***.***> wrote: > Hello Ben, > Your eXo integration is reporting an error code "3" to me, knowing that > the device does not display one locally and also seems to be working > correctly. > So, could you please provide a table of correspondence between the error > numbers and their meaning, because in the documentation, Zodiac does not > mention a number for errors? > Thank you in advance and thank you again for your integration which now > greatly simplifies our lives! > > — > Reply to this email directly, view it on GitHub > <#16 (comment)>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/AQZVQ2CDSQSHNKNSHXFPEW33M5IGBAVCNFSM6AAAAABZQ3YT5KVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTIMBWGQZTGOA> > . > You are receiving this because you were mentioned.Message ID: > ***@***.***> >

[benjycov]

No worries. If you can do me a favor and put future comments/changes/ideas against the actual github project that would be great: https://github.com/benjycov/exo_pool/issues In terms of the responsiveness of switches, we might be able to tune it a bit more, how it works at the moment is that we tell the chlorinator (well the API) to make the change, we then change the switch optimistically, we then check back 10s later and ask the chlorinator 'have you done it yet'. Mostly it has.. but i figure none of us make changes that often... so we can probably live with it this way. The new set of blades problem got me last summer - it made a massive difference though, I hadn't realised how bad they had got (5 seasons done), I was constantly shocking the pool to keep up, this year it rarely drops below 750.

…

On Sun, 10 Aug 2025 at 18:23, GizzGool ***@***.***> wrote: Ben, you're really too fast, if all the supports in the world could be like you 🤩 New version installed, it's perfect! Indeed, it seems the message is intermittent on my end, knowing that I'm well above the salt recommendation. Great for the finances if I have to change the blades in the middle of summer! Finally, I don't know if I'm the best person to tell you if the chlorinator is having trouble responding to the request? In any case, the upward update is done fairly quickly (as example once I've cleared the error). I'll run some tests nearby with a change in pH or ORP value to let you know in this way. Or you can also ask me more specifically about the tests you would like me to do on my side, it will be with pleasure. Thank you again for your responsiveness. — Reply to this email directly, view it on GitHub <#16 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AQZVQ2HAA42KUT4NELSVFI33M552XAVCNFSM6AAAAABZQ3YT5KVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTIMBWGUZTINI> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[GizzGool]

If you can do me a favor and put future comments/changes/ideas against the actual github project that would be great: https://github.com/benjycov/exo_pool/issues

No worries.

In terms of the responsiveness of switches, we might be able to tune it a bit more, how it works at the moment is that we tell the chlorinator (well the API) to make the change, we then change the switch optimistically, we then check back 10s later and ask the chlorinator 'have you done it yet'. Mostly it has.. but i figure none of us make changes that often... so we can probably live with it this way.

I also think that once it's settled, you normally don't touch it much anymore...

The new set of blades problem got me last summer - it made a massive difference though, I hadn't realised how bad they had got (5 seasons done), I was constantly shocking the pool to keep up, this year it rarely drops below 750.

This must certainly be the case for me.
In fact, since we see a significant amount of bubbles around the cartridge, we're thinking it must be okay for enough chrlorine production? But maybe that doesn't mean anything.
Did you buy an original cartridge or an equivalent (because the original ones are really not cheap!)?
Do you perhaps have a link if it's online?

[benjycov]

So I paid €549 for the TRI-L (so the large official one), I’m in France and I found it on easypiscine.fr, then got my local pool shop to price match. I did though seriously consider one off AliExpress at about 1/3 of the price! Maybe I will next time… I rationalised that over 5/6 years it wasn’t too bad. Ours was still ‘fizzing’ a bit but in hindsight not as much as the new one.

…

On Sun, 10 Aug 2025 at 20:59, GizzGool ***@***.***> wrote: If you can do me a favor and put future comments/changes/ideas against the actual github project that would be great: https://github.com/benjycov/exo_pool/issues No worries. In terms of the responsiveness of switches, we might be able to tune it a bit more, how it works at the moment is that we tell the chlorinator (well the API) to make the change, we then change the switch optimistically, we then check back 10s later and ask the chlorinator 'have you done it yet'. Mostly it has.. but i figure none of us make changes that often... so we can probably live with it this way. I also think that once it's settled, you normally don't touch it much anymore... The new set of blades problem got me last summer - it made a massive difference though, I hadn't realised how bad they had got (5 seasons done), I was constantly shocking the pool to keep up, this year it rarely drops below 750. This must certainly be the case for me. In fact, since we see a significant amount of bubbles around the cartridge, we're thinking it must be okay for enough chrlorine production? But maybe that doesn't mean anything. Did you buy an original cartridge or an equivalent (because the original ones are really not cheap!)? Do you perhaps have a link if it's online? — Reply to this email directly, view it on GitHub <#16 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AQZVQ2BEO5LDL46WIPDZNHT3M6I7PAVCNFSM6AAAAABZQ3YT5KVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTIMBWGU4DKNY> . You are receiving this because you were mentioned.Message ID: ***@***.***>