PEC Training...how do you do it?

JohnH · #1 20-05-2007, 07:03 PM

Had some good news from Vixen recently they finally made PEC available for their Stabook controllers (used on the Sphinx and GP mounts). I did a very quick training routine and it seems to work well enough but it made me think - what is the best way to train PEC - using an autoguider, manaually? Where in the sky should the star used be? If autoguiding how fast should the corrections be applied? Etc etc...

Robby · #2 20-05-2007, 07:12 PM

Hi John,
With my G-11 mount, I PEC train at 0.5x speed, using a webcam autoguider, and on a star close to the meridian (where RA movement is greatest).
Good luck
Cheers

Dennis · #3 20-05-2007, 07:39 PM

Hi John

Manually, with my Vixen GPDX and Skysensor 2000 combo, I get comfortable, insert a high power eyepiece (with illuminated cross hairs) for a magnification of between x300 and x400, and locate a bright star at high altitude near the equator, not the SCP. Centre the star.

Then I set the SS2K Menu command to “PEC Mode = Record” and sit there and watch the star. When it drifts, I press the appropriate RA+ or RA- motor movement key to bring the star back to centre. The SS2K correction speed can be set from x0.1 to x0.9 and IIRC I used x0.5.

I do this for one revolution of the worm which IIRC was 8 minutes (of time) on the GPDX. Having completed one revolution of the worm, the SS2K has now memorised all the small corrections I made to re-centre the reference star.

I then select the menu command “PEC Mode = Execute” for the PEC recording to be turned on and take effect.

I also find inclined bench presses with a wide grip are best for training my pec’s.

Cheers

Dennis

jase (Jason) · #4 20-05-2007, 07:53 PM

IMHO, I wouldn't waste my time with setting up PEC manually. At the very least used a web cam and guiding software to record PEC. This is far more accurate than the eye/human reaction time.

If you're serious about establishing the lowest PE possible, get yourself a copy of ASCOM and PEMPro. Vixen SkySensor is supported in ASCOM. Download a copy of the new PEMPro 2.0 beta (I think build 28 is the latest). This now supports common web cams, so you don't need a dedicated CCD camera.

rogerg (Roger) · #5 21-05-2007, 03:14 PM

I've always had good results training PEC using an illuminated eyepiece and with my ST7.

Given the choice, I'd use the ST7 over the eyepiece any day, it just has to be that much more accurate. But I found training it manually using the eyepiece produced significantly corrected PEC anyhow.

The Gemini has the ability to do useful stuff like time-offset the PEC corrections (to comphensate for user delay to make the moves more than just reactive), and smooth the data. Does the starbook have useful things like that?

Roger.

ballaratdragons (Ken) · #6 21-05-2007, 03:23 PM

You mean to say that I can PEC train by using my GuideDog Autoguiding Program?

That is cooool. I did mine manually but it would be nice to have it 'spot on'.

I was under the impression that corrections had to be registered inside the Hand Controller, which Autoguiding doesn't do.

JohnH · #7 21-05-2007, 03:51 PM

Ken,

I am still not sure which is better, autoguiding is fast/accurate but can include seeing errors if the rate is fast...and it needs to be otherwise your corrections are offset by the sample rate, you can fix this if you use PEMPro or have a Losmandy Gemini (and I don't) so what to do for the best I wonder. Star of med brightness on the ecliptic is part of it but best sample rate is .1 to 1s according to some sources but would not this run the risk of chasing the seeing? Could a pec train be done on an artificial star?

ballaratdragons (Ken) · #8 21-05-2007, 04:11 PM

Quote:

Originally Posted by JohnH

Ken,

I am still not sure which is better, autoguiding is fast/accurate but can include seeing errors if the rate is fast...and it needs to be otherwise your corrections are offset by the sample rate, you can fix this if you use PEMPro or have a Losmandy Gemini (and I don't) so what to do for the best I wonder. Star of med brightness on the ecliptic is part of it but best sample rate is .1 to 1s according to some sources but would not this run the risk of chasing the seeing? Could a pec train be done on an artificial star?

I run Guidedog a 2 seconds anyway, so 'seeing' errors would be minimalised.

As far as an artificial star goes, how would you get an artificial star to track at exact siderial speed. Sounds like a no go.

rogerg (Roger) · #9 21-05-2007, 04:32 PM

Quote:

Originally Posted by JohnH

... autoguiding is fast/accurate but can include seeing errors if the rate is fast...

That may be true, but you'd always do more than one PEC run, and errors introduced due to viewing conditions will very easily average out after a few PEC iterations.

I typically do about 5 iterations for my LX200, that gets PE down to extremely low levels of movement, barely a few arc seconds judging by how far the star moves visually in the CCD image.

Roger.

JohnH · #10 22-05-2007, 11:41 AM

Quote:

Originally Posted by rogerg

That may be true, but you'd always do more than one PEC run, and errors introduced due to viewing conditions will very easily average out after a few PEC iterations.

I typically do about 5 iterations for my LX200, that gets PE down to extremely low levels of movement, barely a few arc seconds judging by how far the star moves visually in the CCD image.

Roger.

Ah that makes sense! Thnaks Roger. Should have thought of that - the Vixen manual is insrutable on the how to's of PEC.

My idea with the artificial star was - it does not move - your mount does - so if the tracking was perfect the "star" trail recorded would be a perfect arc of a circle. A bit of maths and you could derive the PEC curve I think...perhaps not with enough precision, I have not looked at the numbers?