PEC training confusion

[Anthonyexmouth]

It is a permanent setup, just built a pier and trying to get everything dialled in

[Demonperformer]

Sorry, realise yours is permanent and wasn't trying to hijack the thread (much!), but the process obviously takes a significant amount of time and was wondering if it was still a worthwhile option when all the gear is put back together each session in not necessarily exactly the same relative positions. If PEC is working out of step it would be likely to cause more problems than it solves?

[Anthonyexmouth]

3 minutes ago, Demonperformer said:

Sorry, realise yours is permanent and wasn't trying to hijack the thread (much!), but the process obviously takes a significant amount of time and was wondering if it was still a worthwhile option when all the gear is put back together each session in not necessarily exactly the same relative positions. If PEC is working out of step it would be likely to cause more problems than it solves?

it's all staying together, hopefully the days of setups are over. 

[Demonperformer]

Yes, I realise yours is, but not everyone on this forum has that luxury ...

[Anthonyexmouth]

Belt mod is waiting to be collected from FLO, will grab it Monday and try the PEC training again when that's done. Field flattener arriving Monday too. Xmas in February. 

[Anthonyexmouth]

I should probably leave well alone until i do the belt mod, but i just downloaded phd2 log viewer and it says my polar alignment error last night was over 5 arc mins and yet the guide assistant said it was 0.2 when i ran it last night. 

[Anthonyexmouth]

On 21/02/2019 at 23:05, vlaiv said:

I marked out bad:

P2P error is large - you want that to be smaller - with guiding you want your p2p to be less than 2".

Second thing marked is max drift rate - you want it to be lower, so that drift limiting exposure is something like 8-10s. You want to guide at about half to full drift limited exposure. Meaning that your guide exposure should be anywhere from half to full max drift rate. Longer guide exposure helps with seeing, but you can't have it be larger than drift limited exposure (about half is max really when you have "smooth" mount).

Belt mod should sort both of these issues, and a bit of backlash as well.

Getting the belt mod tomorrow so all of this might be redundant, but it would be nice to understand this a little better. Just doing another guide assistant and right Ascension P2P after is now 32.22 arc seconds and drift limiting exposure is 0.6 sec. I take it this is worse than the other night. What causes this? 

[michael8554]

1 hour ago, Anthonyexmouth said:

right Ascension P2P after is now 32.22 arc seconds

What's your RA Periodic Error like ?

Michael

[vlaiv]

Above two items that I've pointed out are related.

Stock mount can have P2P Ra error as much as 40-50". This is "leading and trailing" error combined - meaning mount is a bit faster than it should be for some part of worm revolution and then it is slower than it should be for other part of worm revolution. This happens because mechanical parts are not perfect circles, but rather "egg shaped". In practice it means that if you leave your mount unguided, during the course of one worm cycle stars will drift in RA that much - P2P value. It also means that if you do one full worm period single exposure - star streaks will be that much arc seconds long.

With PEC applied this P2P error should be significantly reduced - order of 5" and that is because of fact that PEC is compensating for part of that "egginess", but it can't do it perfectly. You should also note that there won't be two exactly same worm periods with respect to star position deviation - P2P error will change from period to period and that is caused by longer wavelength errors of the mount - like RA gear being slightly out of shape. RA period is 24h so you won't be able to see that period unless you record for couple of days - which is not possible.

Rate of change is "slope" of PE curve and has direct impact on how "guidable" your mount is. It is usually reported as maximum rate of change - where the curve is the steepest. If you want to guide out all periodic changes in your mount - you should choose guide period such that it can guide out even when the position error is changing the fastest - to be able to react to that change.

P2P error is not very important in it self if slope of it is small - or rate of change is small - there is low speed change in star position - this guides out easily. Problem is of course that worm period is fixed, and higher P2P implies faster rate of change with even perfect sine wave, and period is never perfect sine wave as it consists of different harmonics. This is why you want both of these to be as small as possible.

[Anthonyexmouth]

1 hour ago, vlaiv said:

Above two items that I've pointed out are related.

Stock mount can have P2P Ra error as much as 40-50". This is "leading and trailing" error combined - meaning mount is a bit faster than it should be for some part of worm revolution and then it is slower than it should be for other part of worm revolution. This happens because mechanical parts are not perfect circles, but rather "egg shaped". In practice it means that if you leave your mount unguided, during the course of one worm cycle stars will drift in RA that much - P2P value. It also means that if you do one full worm period single exposure - star streaks will be that much arc seconds long.

With PEC applied this P2P error should be significantly reduced - order of 5" and that is because of fact that PEC is compensating for part of that "egginess", but it can't do it perfectly. You should also note that there won't be two exactly same worm periods with respect to star position deviation - P2P error will change from period to period and that is caused by longer wavelength errors of the mount - like RA gear being slightly out of shape. RA period is 24h so you won't be able to see that period unless you record for couple of days - which is not possible.

Rate of change is "slope" of PE curve and has direct impact on how "guidable" your mount is. It is usually reported as maximum rate of change - where the curve is the steepest. If you want to guide out all periodic changes in your mount - you should choose guide period such that it can guide out even when the position error is changing the fastest - to be able to react to that change.

P2P error is not very important in it self if slope of it is small - or rate of change is small - there is low speed change in star position - this guides out easily. Problem is of course that worm period is fixed, and higher P2P implies faster rate of change with even perfect sine wave, and period is never perfect sine wave as it consists of different harmonics. This is why you want both of these to be as small as possible.

thanks, still a bit confusing but starting to get my head around it. cycle up to FLO tomorrow and collect my belt kit and baader click lock. Reducer is now fitted. 

Forecast is clear all tomorrow night so fingers crossed. 

[Anthonyexmouth]

Results after the belt mod and a first attempt at adjusting the RA worm gear

Backlash chart is closer to what it should be but its gone all jumpy. 

This doesn't look much better

[Anthonyexmouth]

Graph is worse than before the belts added. I was better off before building this pier and doing the mod. 

[vlaiv]

24 minutes ago, Anthonyexmouth said:

Results after the belt mod and a first attempt at adjusting the RA worm gear

Backlash chart is closer to what it should be but its gone all jumpy. 

This doesn't look much better

Did you redo pec training, or at least disabled last one?

You stripped down your mount and there is no way that old PEC is in sync now. If you have not done so already - disable old PEC (and get new one when you get the time for it).

[Anthonyexmouth]

Its disabled, I'll do a new one right now. 

Do I disable guide output in PHD2? 

Edited February 25, 2019 by Anthonyexmouth

[vlaiv]

Looking at above graph, which is all over the place, I can't help but wonder if its really windy tonight, or seeing is particularly poor?

If none of above is out of usual, then you probably over did worm/worm gear adjustment - it's no longer smooth, or end floats, it can also cause jumpy movement.

As for PEC - do the same thing you did last time - go for quick recording, so leave the guide output and record via EQMod control panel. But I would do this last - once you are happy with smoothness of your mount. If you happen to open up mount again for another round of adjustment - you'll need to redo PEC - so save it for last.

[Anthonyexmouth]

Just checked EQMOD, should my RA / Dec rates be .50ish? Sure thats what I changed the to, if so should they be the same in PHD" or is that something different?

[vlaiv]

I'm using x0.4 or x0.3, but yes, change it down to at least x0.5.

When you do, you need to recalibrate PHD2

[Anthonyexmouth]

Should the setting in PHD match this? 

Also, how do you estimate your "seeing" 

Sky looks crystal clear and cloudless. 

[michael8554]

Doesn't the "High-frequency Star Motion in GA tell you the seeing conditions ?

Michael

[vlaiv]

2 minutes ago, Anthonyexmouth said:

Should the setting in PHD match this? 

Also, how do you estimate your "seeing" 

Sky looks crystal clear and cloudless. 

After you do your calibration and review calibration data - both DEC and RA measured guide speeds should be close to ones you set. PHD2 will report both of these values in "/s, but will indicate each, I can't remember now what is the label for them. Let me see if I can find it online. Here it is:

You don't set guide speed in PHD2 - it reads it off mount driver (EQMod in this case). If you review your mount calibration data - you will see both guide speed and Expected RA rate - which is just 15"/s times guide speed. You will also see RA and DEC rate - these two should be very close to Expected values.

I judge seeing in following ways - by eye - see if stars are twinkling. If there is no twinkle in almost any of the stars - seeing is probably great. You can also look at the star thru telescope and see what airy pattern of the star looks like. You don't have to look thru a telescope - you can use your camera - just get very bright star in FOV and do very short exposures and look at star shape change from one exposure to next. You can also judge seeing by FWHM of two second exposure. Two seconds is usually short enough so that your mount won't cause distortion of the star - just measure FWHM and you will get "relative" seeing that way - values that you can compare night to night.

[Anthonyexmouth]

1 minute ago, michael8554 said:

Doesn't the "High-frequency Star Motion in GA tell you the seeing conditions ?

Michael

Hope that wasn't directed at me. At this point i'm doubting I know what a telescope looks like. 

[vlaiv]

2 minutes ago, michael8554 said:

Doesn't the "High-frequency Star Motion in GA tell you the seeing conditions ?

Michael

I guess it would for very short guide exposure, but not for something like 2.5s. With such exposures FWHM is better measure.

High frequency star motion can be susceptible to all kinds of other factors than seeing alone - wind shake, roughness of mount, ...

[Anthonyexmouth]

Are these 2 almost right angle steps in the PEC gragh places it's sticking? 

[vlaiv]

Yes that would be perfect example of "sticky" graph - mount nearly stops (slope goes almost vertical) - and then it continues "like nothing happened" at perfect speed - slope horizontal.

[Anthonyexmouth]

1 minute ago, vlaiv said:

Yes that would be perfect example of "sticky" graph - mount nearly stops (slope goes almost vertical) - and then it continues "like nothing happened" at perfect speed - slope horizontal.

Is that likely to be caused by my blundering around with the worm gear tension? or maybe tight belts?