jiberjaber

Mark - you can access the Indi panel from the astroberry.local webpage, there's a bunch of icons on the left but I can't recall which one is the indipanel as I seldom (never) use it... I use the PI to do it all and just VNC in to it to check progress / setup stuff

Sounds like you might need to investigate that first

Wow such an odd problem that! Have you tried it with a none modified camera to isolate it to lens / camera ?

There's a few threads on here about the Todmorden Pier - it uses bolts and glue basically to hold it all together, it's pretty rock solid - the blocks are very heavy by themselves, so there is a surprising amount of weight in the structure. Mine is slightly different in the holding down arrangement is actually just the bottom block concreted into the hole - it would appear the usual approach is to concrete 'j-bolts' in to a base then bolt down on to that. I'm quite happy with it, not finished yet, need to run some conduit out to it and backfill around the base with some gravel or similar to level it off. One of those jobs I'll get round to eventually! The good thing is if we move I can just knock it down. If you go for it, don't forget to factor in a mount - pier adaptor of some description

Rogelio's book, Mastering PixInsight goes to some length to underline that whilst DBE can do normalisation, it is not it's primary objective. One thing I did find useful is less is more with DBE, I reduced the sample points from 10 to 7 and got better gradient removal as a result. I always run DBE after a suitable crop and before any channel combination. I find the Photometric colour calibration to be excellent! If it's narrow band, then DNALinearfit script rocks prior to combining! The problem I've concluded with PI is it's built from a scientific perspective whereas PS is from an artistic perspective - these differing approaches suit different people I think - so whatever works I guess PI is very garbage in = garbage out when it comes to workflow, I think PS (plugins etc) sound perhaps more forgiving by limiting complexity and choice to give an easier experience? Whilst I've got PS, the most I've ever done in it (other than composite widefields) has been messing to clean up a mask for reuse in PI

Good call on the blink upload, I too have used that in the past to get going but if you can't get it in to a state to upload the code it might be a struggle to get it on there. Certainly look at the 'old fashioned way' of getting the code on by shorting the GPIO_00 to ground before power up and holding it there. https://circuits4you.com/2016/12/14/reset-programming-esp8266/

You won't get any response to SSH or ping unless you coded to provide those services. If you can still see the web page then the device is still working but sounds like it's some odd conflict somewhere, timing or some such.

OTA = Over The Air via IP (I think you experimented with that earlier when I mentioned it? Are the ESP's still running the code that was on them, if so, they are not dead I think - It's been a while since I used these devices (the ones I have don't have a separate programme button) - Are you sure you are holding the right button down for the right sequence, I think it is programme button whilst power up isn't it?

I get this sometimes when there is a conflict with the serial port on the programming PC, a reboot usually solves it but it is one of the reasons I move to OTA as soon as I can. Have you tried a reboot, or even another USB port? On Windows, Cura will grab the USB ports when it loads up resulting in the same issue, but you are on Linux, so I expect that to not be the problem

Yep - that's how I did mine initially - there's a wind rose for Graffina but it's a little bit flaky to set up (ok once running though.)

If you have seen the NodeRed -> ESP MQTT 'ON / OFF' message in the MQTT Explorer then this points to the ESP not acting on the message. Have you subscribed to the topic and set up the call-handler on the ESP ? If you haven't seen the NodeRed message to the ESP, then it is a NodeRed problem in the first instance. Have you got a debug attached to the action for triggering the output ?

MQTT has a structure, that is a TOPIC and a PAYLOAD, The payload is usually data and not used like say a print statement. You also need to be careful of overloading the payload in Arduino as it is a buffer in the pubsub client so can only handle so much at a time (though this can be increased but need to keep an eye on the ration of memory used). My initial implementation was just to send a single value (eg: TOPIC = /home/hall/temp PAYLOAD = 27.01) but I've since been investigating using a JSON string I think it's called as I can then send more info in one message which helps with any other clients subscribing (just forward thinking, there's a bit of an issue with the pubsub library when subscribing to too many topics, and you don't want the software interrupt being triggered really often!) I can't remember the exact notation (see my git I guess for details) but using the JSON approach I could send TOPIC = /hom/hall PAYLOAD = {27.01,56.00,1011.11,falling} (or something like that) and that would allow me to deliver in one message temperature, humidity, pressure and trend, as an example. If you get the JSON format right the monitoring software will also be able to decode it too, but it's late and I have probably written it wrong in my example 😳

RE Workflow, Well for the bit you do in PI as listed above yes I think if DSS is working for you and you are happy with the results then it leaves more time for the more artistic bits then keep on trucking.... Do you know what DSS does in the background? Might be worth finding out to see if there is similarity or differences in how it might be done with PI. It might also effect what you get with PI compared with DSS if there is a difference.. Myself, as I paid for P,I I like to get max use out of it I use the weighted batch pre-processing script to pre-process my files and sometimes let it go all the way through to integration to see what the result looks like but I am usually using data over more than one session / filter / binning so I usually delete the integration and registered files after the look-see anyway. I apply cosmetic corrections to the pre-processing and my files are usually darks, flat-darks, flats and lights. (Probably the same seeing as you are using a ASI1600MM also?) I then use the sub frame selection process to weight my images (all sessions by filter), pick the best reference, align, local normalisation and then a weighted integration. After which I can then apply Mure denoise (script) and dynamic crop, dbe etc... So the process is quite similar to the lightvortext tutorials but I also make use of the process and image containers where I need to which allows me to batch up processes (eg: I can batch the registration for all filters, then the same for local normalisation and finally for integration) which allows me to leave stuff running in between batches and do something else... You've nothing to lose giving it a try if you already own PI (other than losing time of course )

You could try doing it all in Pixinsight on a couple of ones you've split between the two previously and see how it looks?

You can set the wifi and SSH settings using a config file in the /boot directory so you could get it up and running headless? https://desertbot.io/blog/setup-pi-zero-w-headless-wifi Basically create a blank file called SSH and add your wifi details to the supplicant file as shown...

If the aim is a quick setup in the garden, why not just add a todmorden pier and leave the mount installed and covered. It's just concrete blocks that can easily be dismantled if you are worried about moving house for example and with a coat of paint or a bit of wood cladding can blend in well. Probably looking at less than £100 inc painting and a cover. From my experience, it's not so much the assembly that takes time, it's the PA etc. With my current setup I just head out, take the cover off the mount (HEQ5Pro) and attach the scope (ED80) and connect three cables (power, ethernet and mount control) (Pi running EKOS, Dew Controller, focuser all mounted on the scope) Of course I don't want to get in the way of teh business case for new toys either You certainly need a small rig to play with, I've got the iOptron Sky Guider Pro for that but that isn't ASCOM controlled.

I have about 5-6 ESP8266 devices around the house running this which might be of use (the good thing about this project of yours is it is finally making me put my code into a repository rather than at teh mercy of onedrive on my home PC) I use a variant of this code for the house thermostat (with some node.red in the background) (covers DHT22, BME280 and SHT30 sensors) https://github.com/oldjiberjaber/ESP_MQTT_Temperature

The broker is quite lightweight, any (SB)PC which you have running permanently will be quite happy running it in the background. Some people even run them on their routers!

I've just tried that application, great stuff. Previously I was using MQTT.fx - a good improvement!

And in other news, I finally bit the bullet and put my WXStn code on git hub. Took a while to go through and check I hadn't left any personal details in there (api keys, passcodes etc) but given you've moved towards weatherstation and MQTT etc I thought it might be useful for your consideration / examples of how to (not?) do it https://github.com/oldjiberjaber/wxstation (I've got no idea of how github works mostly but seemed like the easiest place to share the code with you from)

I think I mentioned this chap before, but certainly worth looking at. I have used his "The Script" before a long time ago when I was first starting out on the MQTT and NodeRed journey and it looks like there is more good stuff included in it, his blog is good for new gadgets, I've just noticed there is an ESP based board with MOSFET on his blog which I intend to take a look at. https://tech.scargill.net/the-script/

Mine topped out at 81C today!

Have you combined kstars with the all-sky script from Pete?

Welcome back Matt!

I'll be interested in your results, I tried making a mask for my lens and it's hard to get enough light in to see the focusing spikes in live view with teh ISO cranked all the way up (works in a photo though - but that's a lot harder to get the focus right using it that way)