HEQ5 Pro Belt mod issue

[Benklerk]

I just recently got a belt mod conversion for my HEQ5 Pro, the guiding is much smoother then it is. But I was expecting a flatter line.
I don't have much mount experience or using PHD2, but can someone point me in the right direction to get this issue fixed? 

The 2 sec graph shows bad RA oscillation while the 1 sec the RA is more smoothed out. 

PHD2 has the correct focal length and pixel size.
EQmod pulse guide is at 0.30 same in PHD2.
The mount is balanced using the 90 deg angle method.
I am using a guide scope. 

Is it a PHD2/software issue, or is it a mount/motor issue?

Ben

[Ronclarke]

There seems to be a lot of issues with PHD 2, I and many others use 1.2 without a problem??

Why not try it and see if it solves your problems??

 

Cheers

Ron

[Hallingskies]

If you look under the “Imaging - Tips, Tricks and Techniques” page, there’s a great practical guide to working with PHD - well worth a read if you haven’t seen it already.  Only thing I do notice is you having the sampling rate set at 1 second.  The guide camera might just be chasing the seeing - try going for two or three second exposures.

If the equipment you have on your HEQ5 is the same as the picture as per your avatar, that looks quite a load and the HEQ5 might struggle.  Having said that, the guiding graph doesn’t seem to bad to me - what focal length are you running at?  Guide scopes, even pretty solid ones, can flex a bit and long focal lengths are sensitive to that.  You might want to try an off- axis guider if all else fails.  I assume you’ve done the obvious stuff like run the guiding assistant and that your polar alignment is good.  You say you’ve balanced the mount.  It helps if you slightly unbalance it and make it a little east side heavy - works for me anyway, gives the gears or belt something to work against.  Apologies if you know all that.  I suppose the bottom line is “are your star shapes OK”?  My scope is running at the moment and my PHD graph is pinging like an ECG, but the stars are just fine so I’m not too worried.  

[Benklerk]

Thanks Hallingskies,

My total weight is around 8kg. Still under the 12kg for imaging. I shoot at 1370mm FL, Polar alignment is good as I can get it. I can get upto 15min guiding with round stars, but im hoping to see if I can get the graph more flatline like a belt drive should be.

[Hallingskies]

15 minute exposures with round stars at 1370mm is very good going!  You could try longer guiding exposures or an OAG I guess, but from my experience I think you have done really well to get where you are now with the HEQ5.

[vlaiv]

Time / examine RA oscillation with longer exposures. If it turns out that it's 13.6s, then you have problem with belt meshing. I had that, and I've seen other people (some) have it with belt mod.

Put more tension on RA belt - that should solve it. I had to really tighten the belt. If you are not sure if it is the problem, there is a simple way to check it. You can do it during the day. Remove cover from mount and expose belt and gears. Use magnifying glass and look at belt / motor gear meshing. You will notice (it is a bit tricky to see since these are small - hence magnifying glass) that belt tooth is coming against gear tooth at the beginning of meshing and then slips into place. This causes belt to move at uneven speed and produces oscillation. It is 13.6s because that is frequency of motor gear mesh.

When you fix this by trying different tensions and find one that meshes properly (you can actually put tension by both moving motor away from RA worm and by "twisting" it a bit - there is bit of slack when you loosen bolts that hold motor in place, and when you rotate it, idler will be closer to other side of belt - more tension, or further away - less tension), you should be able to guide with 3 to 4 second exposures for best results.

[newbie alert]

14 hours ago, Benklerk said:

Thanks Hallingskies,

My total weight is around 8kg. Still under the 12kg for imaging. I shoot at 1370mm FL, Polar alignment is good as I can get it. I can get upto 15min guiding with round stars, but im hoping to see if I can get the graph more flatline like a belt drive should be.

I was going to answer if you have nice round stars then don't get too hung up on guide graphs ..your guiding is great,fantastic at that fl.. if you want to look at flat graphs then change the setting from 400 to 50...be flat as a pancake and guiding will be just as good

[vlaiv]

2 hours ago, newbie alert said:

I was going to answer if you have nice round stars then don't get too hung up on guide graphs ..your guiding is great,fantastic at that fl.. if you want to look at flat graphs then change the setting from 400 to 50...be flat as a pancake and guiding will be just as good

Nice round stars don't mean good guiding. It just means that RA and DEC error is equally spread out for star to form circle.

Good guiding means tight round stars, subject to other conditions, like seeing, focus precision, other optical aberrations.

Guide graph has no relation to imaging focal length. Longer focal length does not mean harder guiding. It is imaging resolution that will reveal to a certain extent guiding errors. Longer focal length does not mean high resolution imaging. It can enable high resolution imaging, but it depends on pixel size, and any binning done. Longer guide focal length just improves guiding and guide error calculation precision (at expense of some SNR loss, and narrow FOV to find guide stars).

Guide graph does indeed provide quite a lot of information on guide performance (regardless of scale used to display it). In above case, it is clear that there is short period oscillation in RA that needs to be sorted out. Next to that, one should look at guide RMS error (ra, dec and combined), and peak guide error. These numbers should be as small as possible, but depend on guide focal length for precision.

[ChrisEll]

See my thread on high frequency periodic error and see if it applies to your situation.

 

[michael8554]

On 05/05/2018 at 23:45, Benklerk said:

EQmod pulse guide is at 0.30 same in PHD2.

Hi Ben

Do you mean guide rate is 0.30? That's too low, you'd be better off with 0.60 or more.

Are your Agr and Min moves as suggested by the Guiding Assistant?

Your star is saturated at 2 seconds so guiding probably wouldn't be as good.

1 second is probably too short for the seeing, so use 2 secs but reduce the camera gain to prevent saturation.

Also your graphs are in pixels, arc-secs would be far more meaningful.

I'm not aware of any issues with PHD2, like any other software if the settings are wrong, so are the results.

Michael

[newbie alert]

16 hours ago, vlaiv said:

Nice round stars don't mean good guiding. It just means that RA and DEC error is equally spread out for star to form circle.

Good guiding means tight round stars, subject to other conditions, like seeing, focus precision, other optical aberrations.

Guide graph has no relation to imaging focal length. Longer focal length does not mean harder guiding. It is imaging resolution that will reveal to a certain extent guiding errors. Longer focal length does not mean high resolution imaging. It can enable high resolution imaging, but it depends on pixel size, and any binning done. Longer guide focal length just improves guiding and guide error calculation precision (at expense of some SNR loss, and narrow FOV to find guide stars).

Guide graph does indeed provide quite a lot of information on guide performance (regardless of scale used to display it). In above case, it is clear that there is short period oscillation in RA that needs to be sorted out. Next to that, one should look at guide RMS error (ra, dec and combined), and peak guide error. These numbers should be as small as possible, but depend on guide focal length for precision.

Thank Vlaiv, your detailed explaination as per norm..i understood it as star bloat if guiding wasn't as good and resolution depended on these figures.. I personally try to get the RMS figures down to as low as I can  but I'm not interested in looking at a flat graph..as in I can get the graph to look flat by altering the setting and doesn't alter the actual guiding...also if I'm getting a flat graph with silly low RMS figures I know not to get too excited  as I'm guiding on a hot pixel..

[vlaiv]

6 minutes ago, newbie alert said:

Thank Vlaiv, your detailed explaination as per norm..i understood it as star bloat if guiding wasn't as good and resolution depended on these figures.. I personally try to get the RMS figures down to as low as I can  but I'm not interested in looking at a flat graph..as in I can get the graph to look flat by altering the setting and doesn't alter the actual guiding...also if I'm getting a flat graph with silly low RMS figures I know not to get too excited  as I'm guiding on a hot pixel..

You are absolutely right that flat graph does not necessarily mean good guiding.

I don't believe that in this particular case guiding parameters are to blame for what is happening. I think that it is indeed mechanical issue that I've described.

I also have belt modded HEQ5 and have experienced that particular issue. Once mechanics are sorted out, pretty standard settings will give good guiding results. Here are mine:

2-3s exposure (sometimes I use 4s, but at that exposure length, PE is starting to show, so it will depend on seeing if I want to push exposure length). I pretty much use stock parameters for aggression and hysteresis. I guide at x0.25 rate, and adjust min mo according to expected guide position measurement error - I set it around 0.1.

I also put in slight delay between exposures (few hundred ms).

With mechanical issue sorted out, OP can expect around 0.5" RMS on a calm (not much wind) night with good seeing. 0.4 will happen very rarely. I believe that anything below 0.4 is almost impossible on HEQ5.

[newbie alert]

.50 on a budget mount is superb.. with a belt modded mount do you still have it east heavy or perfectly balancd..no gears to engage and compensate for backlash..

[vlaiv]

1 hour ago, newbie alert said:

.50 on a budget mount is superb.. with a belt modded mount do you still have it east heavy or perfectly balancd..no gears to engage and compensate for backlash..

I try to get more or less a good balance. In theory one would still need to make it slightly east heavy, and also provide a bit of disbalance in DEC axis (depending on PA error), because there is still a bit of backlash in worm to main gear. This backlash can be adjusted, but I've found in my mount that if I adjust it for one position, about quarter to half rotation away it will have some 'lash again - meaning that main component of worm deviates from perfect circle somewhat.

On the other hand I've not seen any improvement when doing east heavy after beltmod (probably because main backlash was in motor / transfer gear) and that is why I just aim for ok balance - it does not need to be perfect - any focuser position change will throw it a bit off, and also having OAG to the side and not on axis - causes center of gravity to be slightly shifted - I put OAG like that because in that position I can get it the closest to optical axis without cutting into light path.

[Benklerk]

Bit of an update. I think I have sorted out the mechanical issues. Doing a quick test run last night over 3 hrs. I did to objects

The 1st object was around 70 deg alt guiding was good. I got it down to around 0.6 RMS. Doing 5min exp at 2 sec guiding.

The 2nd object was at 87 deg so almost straight overhead this is when problems started to show. Every so and then something moved, I think I have narrowed it down to my guide scope. I heard it creaking and saw the Ra shutter. Don't know if this is the problem or not. I will really tighten up my guide scope now. Doing 2min exp at 2sec.

For those you say I should go to OAG, I haven't had any issues on the worm gears and I'm waiting for my new guide cam and OAG to arrive.