@klimbot (sorry for the delay in responding to your response, I've been in the middle of essays.) Thank you, that's reassuring and a helpful clarification.

I've been doing a lot more reading about the product line since placing the order a few days ago. I can see there was a history of minor issues like needing a capacitor between ground and reset (I think I got that right). But it appears those were early iterations of the product.

I gotta say, I don't usually get impatient for things I've ordered to arrive, but I really am excited by this one.

As an aside, after being missing for three weeks, a clever guy in the Australia post lost items office did some lateral thinking and tracked down my missing stuff. He is officially my hero. So I have all my picaxe stuff back, but, that's SO YESTERDAY. :^)

@klimbot , thank you very much for the heads up on stability...

Just to clarify, was the stability an issue at the coding phase or did you find problems with the devices running over a longer period of time?

Each of my projects that I need to build will run 24 x 7, so stability over the long term is important, particularly with the autopilot, I REALLY don't need it to go toes up while I am sleeping.

@ajb2k3 Thanks for the welcome.

The autopilot project is pretty simple, I made the previous version using one of the lower spec picaxe chips, and it really does not get much simpler than that. The autopilot idea has been tackled in a number of different ways in the past. The Raspberry Pi is a popular candidate and there's a big public project on that one that has been around for a while now.

But I am one of those guys who really likes simple, self contained systems, and the M5Stack really appeals for that reason. I mean, it has the screen and buttons already included, and that may sound trivial to some, but to me it's practically the deal maker on its own.

The autopilot motor current is about 10 to 15 amps in rough seas, so needs a nice big H-bridge to manage it. The original unit uses a pair of relays, and they've been going for 20 years, so that's an option.

I have been thinking some more about the compass and it slowly dawned on me that I really should be using an externally mounted compass anyway. Installation logic says the compass should be low down, near the middle of the boat, but the wheel house on my boat is right up top and a bit closer to the stern. So I think I will stay with the basic units and just buy an external compass module, maybe a 9 axis accelerometer version based on the MPU-9250 chip.

Still getting my head around the different models of M5Stack. I THINK I only need the most basic version. All I need is I2C support.

I am not really into networking or integrating this stuff in any way. There's no real advantage. Maybe later, if I have time, I will look into it, but right now I need to keep the boat warm, the wind generator charging (without it fighting the solar regulators) and make sure the boat doesn't sail in circles while I am making the tea.

So... I THINK three of the ESP32 Basic Core IoT Development Kits will do the job? I mean, the thing is about 1000 times more powerful than my old picaxe chips. Maybe more. Any reasons they might not cope?

Hello all,

I am a new member who has just signed up.

Long story short, I have been working with Picaxe chips for about 15 years and have solved a handful of little project problems with them over that time. However, last month my ENTIRE electronics toolkit was lost in the post, tools, components and all, so I found myself in the strange position of starting, literally, from scratch.

Since I was starting from scratch, I figured it was a good time to ask around and see if maybe there was something more suited to my needs than the picaxe, particularly as I have three pretty urgent projects on the go (which is why my stuff ended up in the post, I was travelling around with it and left it behind on an interstate visit.).

One of the suggestions that was made to me was the M5Stack system, and I have to say, at first glance it looks absolutely perfect for my needs.

All three of my projects require a simple display, some buttons to control settings and the ability to switch on or off a few circuits. And of course, that is all built in with the M5Stack. Perfect!

One project requires interfacing with a handful of temperature probes and yes, I've read the article on interfacing more than 3 DS18B20 probes which answered all my questions there. Tick! (I only need four)

One project required interfacing with an opto-isolated hall-effect current sensor, and I really cannot see any problems there. Tick!

But one project requires interfacing with a GPS and a compass. Now, I've found the commonly used GPS module (M8N) that is in the M5Stack format and it looks perfect. The external GPS antenna is a real bonus, and the whole integration looks straight forward and suited to my needs.

However I cannot decide if the built in MPU9250 in the M5Stack Gray kit will be suited to my needs.

I am building an autopilot for my boat. I already have the steering motor (actually I already have the whole autopilot, but the brains of the thing are going potty) so all I need to do is control the motor with a simple reversible 12 volt DC circuit, which I had done successfully with some big MOSFETS using the picaxe. (Yes, they were lost in the post too.)

The code logic is very simple and the processing power required is minuscule. But I cannot figure out if the onboard MPU9250 will work as a compass given that:

1). It is inside the case and might therefore be affected by power flow in the unit itself.
2.) Whether it is actually an ok compass in itself. I don't need more than two or three degrees accuracy, but if it drifts more than that, it will affect how long I take to get from A to B.

Can anyone shed some light on the compass and how it actually behaves in the real world?

For the record, it won't be a show stopper if it is no good, as I can buy plenty of cheap and simple I2C compass chips that are ok, but I like the idea of using the inbuilt version if I can.

Matt