Spa temperature has occasional 'spikes' and 'dips' S100103 Honolulu

[NZMR2]

Hi all,

I'll start by stating I'm pretty green when it comes to the tech details behind these units. I guess I know enough to be dangerous!! I've recently purchased a complete 6 wire module from Metaljay and had it delivered to me in New Zealand. I managed to get it set up using Hardware: 6 wire 2021 (air) and DSP: 6 wire 2021 (air), PCB: V2B. It connects and I can control via the web interface. It's not super responsive and trying to control via the spa itself is a bit hit and miss. The display sometimes flicks off / does a few strange behaviours. I also can't use the timer buttons anymore.

I managed to get it hooked up to my Home Assistant via MQTT. I've got it showing in my main dashboard.

For now, I'm just setting some timers to see if it's working and it seems to be. My intent is to use some smarts to turn it on when my solar is generating excess. However, when I reviewed the history on the 'Spa temperature', I can see that it's getting spikes (up to about 92 degrees) every few minutes. This also shows up in the logs where it says the heater has turned off then turned on a few seconds later. Watching the spa display it looks like this is actually happening on the spa itself too.

Has anyone seen this behaviour before? Any thoughts as to what to do to sort it out?

Thanks,
Danny

[visualapproach]

The timer button may be disabled in config. The spikes is probably bit errors. Probably due to poor connection. Not necessarily bad solders but it can be too long wires or mismatched filters, faulty esp and so on.

[Rodrigocicci]

Same issue with the same pump. Tried with my module, everything works but i had this same blinking and spikes in graphic. Though it was my module and built a second one. Same issue. Got one from metaljay and same issue. Will have a week off next week and i will try re flashing the module and reconfigure HA.
I will post here if anything changes.

[visualapproach]

If using the LP filters, try without. And vice versa.

[NZMR2]

Rodrigocicci, just wondering if you've had any luck resolving this?

[alexapps1]

I have exactly the same issue with S100102 pump. I tried a D1 Mini the first time and figured it was because I was out of spec, so I thought the problems were due to 3.3v vs 5v logic.

Bought the kit from metaljay, just tried it, and still get exactly the same issues.

-- Temperature up and down on control panel not working, although sounds do play.
-- When changing temperature in Web UI it seems to hunt for the temperature set.
-- Assume still getting temperature reading spikes as even the display said >90c.

I've soldered my cables on the H1B and H2B side (as per metaljay pics although he is using the old U3 rather than U2 level shifter in those pics. I assumed H1B and H2B are post resistor and capacitor which I figured are there to help stabilize things.

There is someone selling one on eBay with cables soldered on the H1A and H2A side. I don't know if this is the issue, maybe I should have done this?

[visualapproach]

The signals from cio seems to be bad. It occasionally gets proper messages (incl temp) but then it has already passed the target so it sends the opposite button (up or down) and only settles when it by chance gets the right value.
I reckon you don't want to do soldering since you bought it pre made. Otherwise I would recommend trying as short cables as possible and try with the A ports.

[alexapps1]

Thanks @visualapproach.

Sorry, I should have been clear, I bought the parts but soldered myself. I will try short cables on the A ports now, and shorten the cables on the B ports if that still doesn't work.

[alexapps1]

Shortened cables to 1 inch, using A ports and no change.

Very strange that whatever I change I seem to consistently get the same behaviour.

Being in the UK with 50hz electricity wouldn't change anything clock related would it?

I can try shorter cables on B port unless you can think of anything else to try first to save me unsoldering those!

[metaljay]

Sorry I missed this, have you configured your pump and board revision correctly? Have you tried experimenting with other options? I ask as the fact the display isn’t working is either the red LLC or pump config

[alexapps1]

Yes I think so

It's all working, it's just like there is noise on the data line to the pump, just misses messages or sends false messages back.

Only thing I haven't tried is shortening the cables on the 'B' side as I needed to put it back together to use it.

Shortening cables on A side made no difference, moving from D1_Mini to full recommended setup made no difference.

Cables being nowhere near anything else, and pump/heater being off made no difference, which is why I'll be surprised if it is anything to do with cable length but I've been wrong before...!