StuartT

46 minutes ago, Starflyer said:

The calibration works over 25 pixels and afaik that's set and can't be changed.  I aim for about 12 steps, you're getting four or five steps at 1050ms calibration step size, try dropping it to 350ms, that should sort it out.

Thanks. I find PHD2 to be somewhat impenetrable to be honest. I've tried reading the documentation but it makes very little sense to me. I didn't even realise you could change the calibration process! 

I'll give this a try 

I am finding that when I calibrate (first pointing to the equator) I get an error saying 'not enough calibration steps - results may be inaccurate'.

This is the log viewer diagram. Surely if there aren't enough steps PHD2 could just do more? What's stopping it?

19 hours ago, geoflewis said:

Do these links to PHD2 documentation about calibration help...?

https://openphdguiding.org/man/Basic_use.htm#Automatic_Calibration

https://openphdguiding.org/man/Tools.htm#Calibration_Details

Thank you! Yes they do. 

19 hours ago, knobby said:

@StuartTAre you using an off axis guider ? as your focal length in PHD2 looks about correct for a 150ED / Reducer.

Focal length in PHD2 should be of the guide scope.

Yes, I am using an OAG. Hence I used the FL of the scope (and reducer - which is spelled incorrectly in the profile 🤣)

6 hours ago, michael8554 said:

Hi Stuart

1. Those dozen or so lines in the GuideLog are setup info - which items would you like explained  ?

The graph shows outward steps as blue W1, W2 etc, and red N1, N2 etc.

These should ideally be evenly spaced, groups of pairs show backlash and/or stiction.

Bunching at the start of the N run would show you hadn't cleared Dec backlash before staring the Cal.

And the faster return steps are colourless E4 to E0 , and N4 to N0.

2. But for info on how effective the Cal was, you should look at the report at the start of the next item in the GuideLog,  "Guiding begins at......... "

Michael

Apologies. I shouldn't have included the text. I was meaning the diagram. Thanks for explaining. Does it matter that the red and blue lines are not aligned with N and W? does that rotation just indicate that my guide cam is rotated relative to the meridian and equator?

Can anyone explain what this is telling me? I can't seem to find any documentation on this (Andy Galasso's Analysing PHD2 guiding results doesn't include calibration)

Thanks

1 hour ago, vlaiv said:

I haven't tried that one (that I remember), but in essence - it should be close in performance as it relies on same things.

Same way you can hit "auto" in PecPrep to get the curve (it tries to remove noise and insignificant harmonics and isolates important stuff) - it can be done in PHD2, after all - they have access to same raw data (PHD2 can record position prior to issuing correction).

There are only two things that are different:

- predictive pec needs some time to learn and it won't give you best performance from the start of the guiding sessing

- PHD2 relies on corrections being followed to the letter - and that might not be the case. In each new round of correction, in order to get "uncorrected" behavior of the mount - PHD2 needs to mathematically reverse last correction and figure out where the mount would be without correction issued (so it can get accurate reading on periodic error). Problem is - corrections are not perfect. Sometimes they overshoot, sometimes they undershoot - no way of knowing how much of a correction was taken up by the mount (backlash and other mechanical issues).

For this reason I would give slight edge to the PEC recording and preparation as it can use multiple PE periods to establish good mean periodic error and correction will be applied from the start regardless if one is guiding or not.

as usual a well-informed and helpful answer. Thank you! 

So in order to get the best raw data for analysis in PECPrep should I just run the guiding asst for a while? Presumably that will just be the mount behaviour without guide pulses?

23 minutes ago, vlaiv said:

Point of PEC is to produce correction of the mount as mechanically is - without any external aids - like guiding.

This will make mount run smoother and will put less strain on guider system once you start using it.

Guide corrections will mask true underlying shape of PE curve and produced PEC will be sub optimal (even if you apply it with guiding only).

Thanks for this. Can I ask what you think of the alternative approach - using the 'predictive PEC' algorithm in PHD2? I hear it's very effective

On 26/06/2018 at 12:37, vlaiv said:

I turn off guiding and use PHD2 to create log file (there is option in PHD2 to disable guide commands). I usually create about hour / hour and a half of log file. While doing that - at some point I press time stamp button in EQMod. EQMod auto PEC is also turned off. Next I load PHD2 log data into PecPrep and create PEC curve.

 

Hi @vlaiv sorry to be late to the party here, but I am just getting to grips with PEC. I am curious as to why you say guide commands have to be switched off? I have just analysed a PHD2 log file and analysed in PECPrep with only the 'auto filter' added. Here is what I got. The PEC curve is created is not a pure sine wave, but has a higher frequency oscillation (period 11, mag 33.4) added to the main worm gear period (period 198.7 mag 100). Is it ok to load a curve like this into EQMOD? Thanks!

(or is it just easier to let PHD2 do everything by selecting the 'predictive PEC' algorithm as Cuiv recommends?)

12 hours ago, symmetal said:

Signal to noise ratio  = signal power / noise power. As the ratio can be large numbers, the dB scale is used where SNR(dB) = 10 log ( signal power / noise power)

Note the use of the term power. In electrical circuits power(W) = volts(V) x amps(I). Using ohms law where V = I x R or voltage = current x resistance you can rewrite it as

power = V² / R.

The signal and noise levels we talk about in our images are actually voltages expressed in ADUs. The resistance is the same in all cases and so isn't relevant and can be considered as having a value of 1 so it can be rewritten as

power = V²  and  therefore

SNR(dB) = 10 log ( signal voltage² / noise voltage² )

Working in squares of voltages is awkward so it's rewritten as 

SNR(dB) = 20 log ( signal voltage / noise voltage )

A 40dB SNR therefore corresponds to a power ratio of 10,000 : 1 or a voltage ratio of 100 : 1

As mentioned above we are only concerned with voltage ratios and not power ratios.

If we are referring to AC signals then the RMS values of the signal and noise voltages are used.

Still confused? 🙂

Alan 

 

Thanks. for explaining. Makes sense now. 

On 20/01/2023 at 20:46, Seelive said:

I assume they are measuring SNR in terms of the ratio of straight ADU units rather than the power ratio so 40dB would be just 1:100, much more believeable.

not sure what you mean here. The dB scale is logarithmic. A Bel is a power of ten. So a decibel is a tenth of that. Hence, 40 dB is 10^4 (or 10,000)

12 hours ago, Seelive said:

I assume they are measuring SNR in terms of the ratio of straight ADU units rather than the power ratio so 40dB would be just 1:100, much more believeable.

Err..

9 hours ago, Sunshine said:

Wow this is beautiful, I know nothing about processing but I can tell you certainly do! the colours seem to be so nicely exposed and saturated.

1 hour ago, newbie alert said:

Yes nice Stuart, must be pleased with that..

Thank you both. Yes, I am very happy with this one. I am pleased to have so much detail visible.

I shall be adding some more data from my new Askar D2 filter for some Sii signal (if the Rosette has any)

21 hours ago, Adam J said:

Depends how its measuring SNR. But a SNR of 1:10000 seems unlikey as thats almost what i would expect from a daylight image lol.

Adam

Yes, I agree. I think the tool must be flawed.

I even made the measurement on a preview containing only nebulosity (because stars would have a much higher SNR). 1:10 000 sounds wildly improbable

8h 40m, Esprit 150, ASI2600MC, L Extreme

3h 50m, Esprit 150, ASI2600MC, L Extreme

I guess no one knows... 🤷‍♂️

I've just discovered the SNR script in PI and so naturally I was curious about my images. A recent colour sub measured at 28-30 dB (depending on channel). An integrated result measured 38-40 dB

What should I be considering a good SNR? (40 dB is 1:10 000 which sounds like it's quite good for SNR?)

Thanks 

5 minutes ago, simmo39 said:

Thank you that is a brilliant walk through, I may be getting the D2 filter my self v soon and this will help me a lot with my luddite understanding of these things. 

I found I needed double the exposure time for the D2 as Oiii and Sii are fainter than Ha. So I used 300s for the L Extreme and 600s for the D2

5 minutes ago, simmo39 said:

When merging the 2 Oiii channels did you combine them post stack or did you combine as a separate stack? Sorry for the interrogation.

Here is what I did.

Initially from the frequency response curve for my OSC I estimated the proportions of RGB that make up the three emission lines.
Hα = 80%R + 15%G + 5%B
Oiii = 95%G + 45%B
Sii = 75%R + 15%G + 5%B

Method
1 integrate data from each filter separately

2 SPCC each image

3 Register the two images

4 For L Extreme filter
split RGB
create Ha = 0.8*R + 0.15*G + 0.05*B
create Oiii_LExt = 0.95*G + 0.45*B

5 For Askar Oiii / Sii filter
Split RGB
create Oiii_A = 0.95*G + 0.45*B
create Sii = 0.75*R + 0.15*G + 0.05*B

6 Add the two Oiii images with simple addition
Oiii = Oiii_LExt+Oiii_A

7stretch the three images (Ha, Oiii, Sii)

8 Use LRGB combination
L blank
R Sii
G Ha
B Oiii

9 Postprocess - as usual

5 minutes ago, Stuart1971 said:

I think you could go a bit more on the processing too, hope you don’t mind, but had a little tweak….not sure if it’s better or not…but the first looked a bit dark on my screen, but maybe it’s just my screen….🤔

 

I agree. In fact, I just posted a slightly more stretched version to Astrobin. (Original now replaced). Thanks

2 minutes ago, simmo39 said:

V nice, was the combination of the data straight forward? 

I created the Ha, Oiii and Sii channels from the RGB of each filter (using the response curves of my camera), then merged the two Oiii channels (as both filters pass that). Then did an LRGB comb (black L and SHO for RGB). 

This image is a combination of 3.5h with an L Extreme filter (Ha and Oiii) and 3.5h with an Askar Sii / Oiii 6nm filter to produce an SHO result.

Overjoyed to report that my 'first light' with the new mini PC went like a dream! Everything connected up perfectly (and stayed connected up), sky was crystal clear and my guiding was the best I have ever had at 0.38" (about one third of my imaging scale!). I was in astro heaven basically. 😍

So I got over three hours on the Soul with the L Extreme and here it is (ready to add the Sii data to it with my new Askar filter)

1 minute ago, scotty38 said:

I just looked and I can see all the Mele drive. Let me check on what sharing etc I may have set up on there

Edit - ok great if it's all working...

Yes, it turns out I had to explicitly turn on sharing and give permission to "Everyone" - even though that wasn't necessary when connecting drives in the opposite direction.

Ah. Yes. That fixed it. Thanks. 

For some reason I didn't have to do that the other way around. I guess RDP did it for me.