Guiding - What causes this guide graph?

[swag72]

I've had an issue of late where I have lost 1 sub a night due to guiding errors. Tonight I stayed and watched the guiding graph in Maxim as I have tried to pinpoint when the issue happens. Anyway, I have the following guiding graph, which as you can see has settled and will guide perfectly now for the rest of the night.

I've seen this type of spike before in some posts and stuff, but can not remember or find what it may be a symptom of. Would appreciate you having a quick nosey of the graph and suggesting what it may look like. As I said, will now guide perfectly all night. I'm guessing that it's not drift otherwise it would continue.

I'm using a QHY5 and 10x60 finder guider all atop of my Pentax 75.

[swag72]

Any thoughts on this graph? I can confirm that the guiding for the rest of the night (6 hours worth) was absolutely fine, this just seems like a one off, but always *I think* at the start of the run, perhaps about 10 minutes in.

[FraserClarke]

Not sure of the cause, but can you set the Y tolerance (or minimum move) lower, so that the software corrects the error before it gets so large?

Assuming Y is declination, I guess it might be caused by changing pre-load on the dec gears -- it slowly slips until the error gets big enough to correct, and then the guider moves the gear in the other direction, which pre-loads it properly and keeps it stable for the rest of the night?? (I'm assuming, when you say "rest of night" you mean "rest of night, as long as there isn't a pier-flip" ?)

[James]

I'm guessing here but could it be backlash?

James

[swag72]

Not sure of the cause, but can you set the Y tolerance (or minimum move) lower, so that the software corrects the error before it gets so large?

Assuming Y is declination, I guess it might be caused by changing pre-load on the dec gears -- it slowly slips until the error gets big enough to correct, and then the guider moves the gear in the other direction, which pre-loads it properly and keeps it stable for the rest of the night?? (I'm assuming, when you say "rest of night" you mean "rest of night, as long as there isn't a pier-flip" ?)

The Y minimum move is already set to zero, as is the X axis. Y is declination. I've not tried a flip for a while, so yes, it's all imaging up to the zenith, then it puts itself to bed!! If it is as you suggest, how could this be dealt with?

I'm guessing here but could it be backlash?

James

I don't know James - If it was backlash would it not be a regular issue? It will go on happily for another 6 hours.

[James]

It miht be that initially the guiding has to be 'taken up', once taken up the commands go through and the graph goes back to where it should. Like you though I'd suspect that it would happen again though.. sorry!

James

[FraserClarke]

Backlash and pre-load are interlinked (backlash is basically caused by a loss of pre-load on the gears). As it's the dec axis, the guiding corrections are probably always in the same direction (i.e. you have a slight bit of polar alignment error), so once the backlash is taken up, it won't happen again in that direction. Do you dither around on the targets during a sequence? If so, i'd expect you'd see the effect more often. if not, then you'd only see it once per session.

You could try forcing the correction before you start the imaging run;

1) Set-up as normal

2) Click on the guide star

3) By hand, change the Y position of the guide star by a few pixels (need to experiement with which direction!)

4) Start guiding and wait for the guider to pull the star into the box and the trace to settle...

[Jessun]

Not much help perhaps, but I see a similar thing on the iOptron. In my case it's DEC that keeps moving - it gets 'stuck' at a small rate and wanders off. The only cure is a rapid movement in the other direction that forces it to settle. And it's random too.

My HEQ5 never did though.

Hope you get it sorted!

[RikM]

I have seen a similar thing happen with PHD. I don't know the cause but I just bailed out of the sub, stopped guiding, waited a few seconds and started guiding again and it was fine ever after. Same thing has happened a few times but not every time. I normally off balance the scope slightly, so the gears are always pulling up-hill.

[Shibby]

Not much help perhaps, but I see a similar thing on the iOptron. In my case it's DEC that keeps moving - it gets 'stuck' at a small rate and wanders off. The only cure is a rapid movement in the other direction that forces it to settle. And it's random too.

I get this with my Vixen GP. Usually after slewing or centering a target manually, it wanders for a while until i give it a nudge or two in the opposite direction. Such an effect would show up like this on a guide graph; I doubt it's backlash because that would look like a sudden, large movement (something I also suffer from!)

[swag72]

Interesting comments - Thanks guys - Same thing happened last night, for about 5 minutes after the tracking starts it's OK, then it starts to rise, slowly really, just moving slowly above the central line. It didn't move upwards for more than 1.5 to 2 pixels again, then dropped quickly. 1st 20 minute sub was fine, 2nd one had to be binned as it dropped quickly. Don't know that I'll ever get to the bottom of it!

Surely by nudging it, then I will upset my sub alignment from the previous night that I have just spent ages to get right?

[ollypenrice]

Could it be a poor connection on the ST4? I've had this. It all became clear when I did a flip and could no longer guide in Dec. This was because, after the flip, the dead connection was the one I needed in Dec. As Fraser says, you only guide one way in Dec as a rule. I also like his suggestion to push the dec off target manually so as to force it to take up the backlash.

Bad connexions are a beggar because they come and go, making them hard to pinpoint.

Olly

[swag72]

I'm not using an ST4 now Olly, I'm ASCOM pulse guiding, so all going through the EQDIR cable. I do wonder if it is the mount taking up the backlash that is there initially, then happily guiding all night. I don't want to move anything though once I am guiding, otherwise it will shift my target. Is there another way to do it? Is there a way to perhaps get the backlash taken up quickly at the start with a guiding setting then set everything back to normal once it's done the business? That would be useful as I could get that done in a short space of time at the start and not lose any subs at all.