DIY mounts

[Werisit]

Yes it's a combination of PA and speed error that would create a different angle. An interesting thought that something else is moving. Tricky to trace.

Edited April 12, 2014 by Werisit

[Guest peepshow]

The plot thickens. 

 

Last night I did 2 x 15 minute exposures. 

One of them had the 'weird streak' at exactly 45 degrees to the star trailing direction

and the other one was at 90 degrees from it  !!!

 

After checking everything for tightness etc I am coming to the conclusion that the

pre loaded ball races are behaving in some erratic way.....

.........sometimes pivoting OK and at other times not. 

So maybe pre loading needs increasing or quality of them.

[Werisit]

Hmm that's a thought. If the bearings are pre loaded too much they may not move smoothly. Also they need to be perfectly square to the shaft or they may move in a lumpy (technical term) and unpredictable fashion.

Oh what fun!

[Guest peepshow]

Yesterday, I made a clamp to fit the bearings and increased the preloading.

Last night,  again 2 x 15 minute shots with 200mm lens.

 

I hope that I am not kidding myself but on greatly magnifying the results

it did appear that the 'angular streak' was reduced in length last night.

 

I shall move the bearings around to another position and try again.

 

May eventually have to go to angular contact needle bearings to solve it........

.........yet more fun, Alan.

[Guest peepshow]

Here's two more tests for erratic bearing movement.......
.
The first was a 5 minute image using a 500mm mirror lens (not yet in correct focus) at 1600 ISA.

The second as above but 15 minute exposure where the bearing movement is now showing as movement to 2  o/clock position.

Both taken near Celestial Equator where star movement is at its greatest.
Tracking was still OK on both shots though.

So, I have decided to replace the main bearings with two  pairs of angular contact bearings.

This should cure the trouble ??????????  




]

Edited April 21, 2014 by peepshow

[Guest peepshow]

I have now replaced the bearings with angular contacts in pairs but although improved still have that annoying small trial at about 45 degrees over about  5 pixels.

It shows on a big blow up  of  a 30 minute exposure with a 200mm lens at the Celestial Equator so quite a severe test anyway.

This 5 pixel trail at 45 degrees represents about a 1/1000 inch camera movement with this 200mm lens so something very small is happening over a long 30 minute exposure.

So the detective work continues.      "The case of the 5 pixels."

 

[Werisit]

That is a very extreme test Richard. That equates to one pixel in 6 mins. I'd call that a resounding success.

I did a rough assembly job on mine and found that i have a wobble on my screw (oo er). So I'm a bit deflated now as the thing relies on that for its accuracy. I'm working on it!

Edited April 30, 2014 by Werisit

[Guest peepshow]

Sorry to hear about the wobble, Alan.. 

Not sure what you mean though.....is it uneven pitch of screw, as you cut it on your lathe,

or some bearing problem (join the club  ) ?

With your small RA pulley, extremely small errors will give trouble, so everything has to be accurate.

 

Hope you sort it all out soon.  Have lots of fun.

 

It's so difficult to run long exposures these nights for testing with all the cloud.  I only managed a couple of 18 minutes last night.

 

Thanks for kind words but using a 200mm lens gives me approx 9 degree FOV, equating to about 10 second of arc per pixel on my canon.

So these 5 pixel streaks at 45 degrees equate to 50 sec of arc error.

 

My type 3 tracker should track within 1 sec of arc per hour so you can see that my bearing problem giving me a 50 sec of arc error in just 1/2 hour is severely interfering with it's designed performance.  

 

Maybe I am too fussy, but I am now quiite sure it's the bearings causing all my trouble......the main shaft may not be a sufficienlyt tight fit into the i/d of the bearings ? 

With this in mind I have now added 2 semi circular oilite bushes to gently push the shaft in one dirrection......kinda pre loading.

 

We shall see...........I'm sure there's more fun to come.

[Guest peepshow]

Have you sorted that wobble you mentioned out yet, Alan? 

Is it in the lead screw or bearings etc etc? 

You have to be a detective in this lark to establish where the tiny, tiny errors are.

 

My new bearing system is nearly finished, so I am getting rather nervous now to try it out soon.

 

The bearings will now run preloaded on the OUTSIDE of the silver steel shaft in groups of 3 at

120 degrees from each other, each end of the shaft.    

 

Because I don't have a cylindrical grinder silver steel tolerances do not give me a force fit into bearings, hence

the 'outside' experiment.

Edited May 14, 2014 by peepshow

[Werisit]

Hi Richard. Sorry for delay. I've been working on it but earning a living is getting in the way a bit.

I have just been cracking on with the rest of it so I can get it up and running to see how much the wobbly screw affects it. The tensioner pulley is done now but I'm still to mount it. Then it's fitting the cable and aligning it with the screw. Then making the link between the nut and cable. I'm just going to bore a hole through the base block and put a sighting tube through for a rough polar alignment device.

Just started thinking about the twin 555 timer circuit. Fortunately I have a pal down here who can help me work out the component values to get the frequency range I need. My head nearly exploded last night trying to figure it out!

 

Have you solved your streaking problem yet? I'm wondering how you are controlling axial thrust with the bearing arrangement you describe. One interesting thing I did was to make an animation with a batch of quite short exposure single frames. Watching 20 minutes worth of 20 second exposures  squeezed into 20 seconds proved that I had a smooth and predictable error. It might throw some light on what's happening with your rig. 

 

I'll post some pictures soon.

Edited May 25, 2014 by Werisit

[Guest peepshow]

Good news that you are progressing in spite of earning a living.

Your 555 timer cct should have a very very fine frequency control possibly to get the freq spot on.

So it's your screw that is causing the wobble ?  Of course, this screw is a image of your lathe lead screw

and I would be surprised if that had any error.  If it still wobbles maybe you could get another screw from a more accurate source? 

 

I have an old lead screw taken from an Edison dictaphone machine.

It is  about 1/2 inch dia and approx 50 TPI. x 6 inches long actual threaded part. 

Unfortunately, I don't have any nut to fit it. 

You would be welcome to have this if it's any use to you.

Maybe you could source something like this but with a nut.

 

My axial thrust is taken on a brass bush but as I see it axial movement which is up/down the RA axis doesn't cause image movement.  This sounds strange but the camera moves parallel to the earth's axis so the stars do not move on the sensor.  What makes the stars move is an angular camera movement.

(If you place the end of a camera's lens hard onto on a window and watch the image as one slides it about, the image is static.) 

 

I am still puzzling over my problem as the new bearings show no improvement.   So it might be some part of the actual structure moving as the camera rotates about the RA axis.  When static there is no problem, only during tracking movement.   The bearings have no backlash or tiny unwanted play at all now.

 

I do take short exposures and animate them but they only show what I already know is happening....a gradual approx one thou movement over 15 minutes angled off from the star tracking direction.

 

I have bought a Skywatcher AX synscan GOTO head and will mount that on my system.

The idea is to GOTO a DSO etc, then stop the synscan tracking as the barn door motor is engaged.

 

So Alan, the fun continues.

Edited May 26, 2014 by peepshow

[Guest peepshow]

SOLVED

Alan, the streak problem was not caused by a single thing but a combination.

The cheap Tamron zoom lens used I found has a very slight wobble on the zoom barrel

which can move the image.

Also my PA had not been checked out for some months and in messing the tracker around

the PA had moved slightly. 

Also some stability struts I fitted were not really man enough for the job, so have been replaced. 

 

So hopefully now,  it's just some more adjustments and a piece of sticky tape around the zoom barrel.

 

The Skywatcher AZ GOTO mount does not shift when the barn door tracks so that should all work OK

when it is all set up.  In fact, the GOTO once set and parked could have any DSO selected indoors before

taking it outside onto the tracker, because it is always mounted in the same level and radial orientation each time

it will be used.   I shall use a laser to star align it initially.

 

This GOTO is refered to on another thead, together with a photo further down the thread......

http://www.eastmidlandsstargazers.org.uk/topic/8004-another-peepshow-crazy-idea-goto-on-a-barndoor/

 

I hope that the wobble you were having has been cured or at least traced to its source so that you

can do something about it.  

Like my problems,  maybe it is being caused by more than one fault?

It's all too easy to latch onto an idea and never give other possibilities any credence.

Edited June 7, 2014 by peepshow

[Werisit]

Hi Richard. I'm glad to hear you've got to the bottom of your streaking problem. I had a look at your other thread with the goto. A very good idea and looks great. One of the disadvantages of the barn door is the limited run time. Your rig will allow you to get straight onto target and not waste time trying to find and frame subjects. That is a frustration I could do without. I've had to spend, maybe half an hour, getting something framed up how I want it! It's ok with wide field but when you've got a 300 or 600mm lens on and can't see anything through the view finder or live view it gets difficult. No need to tell you that though!

I'm stalled on mine at the moment. Wondering what to do next. Very busy at work and trying to keep up some observing as well. Thank you for the offer of the Edison lead screw. I had a look on ebay. £40 + for a nut!!  I think you should auction it! I may just buy a 8 x 1mm threaded rod and run it micro stepped. I also have fleeting thoughts of using the bits to make an EQ with a type 3 barn door mechanism  .  

[Guest peepshow]

Well, sorry to hear you are stalled at present, Alan.  A 8mm x 1mm rod is very cheap (as are the nuts ! ) so maybe worth a try.

I use a 1/4 BSW rod on mine rotating at exactly 1 RPM, driven from a 60/1 worm and wheel whose worm rotates at1 rev per second on the stepper motor shaft. (crystal controlled.)

 

 Alan, if you ever go the barn door way make sure that your bearings are very good if you wish to take longish exposures with tele lens.  Don't make the mistakes I made.

Also those three type 3 dimensions are very important too for accurate tracking over a period.

They are not given to 3 places of decimals for nothing.

 

Last night I did more tests, for over two hours and it's getting better now..........no weird trailing

 

Here's a 15 minute exposure greatly enlarged of a double star in the constellation, Ophiuchus,

about 4 minutes of arc seperation on a shot 8 degrees wide (200mm lens) 

Slight more adjustment still required though.

 

So slowly getting there but out of focus with a crappy lens still.  

Must buy something better soon.

 

 

[Werisit]

Hi Richard. it's been a long time and for most of it I've been scratching my head!

I am never going to make or buy a thread that's accurate enough for my needs. Why re-invent the wheel? I've come to a decision. The answer is, of course, what you suggested months ago, a Barn Door.

I'm wondering why you went for a type 3 not a 4. The 4 has good accuracy and runs for 2 hours. I've been looking at these plans and performance figures.

p://education.jla...cker/index.html

The plan is to make an equatorial version using the base block, bearings and hub that I've already made, so not all my efforts so far are wasted.

What do you think?

[Tweedledee]

I think this is the properly formed link...

http://education.jlab.org/tracker/index.html

[Werisit]

Thanks Pete, I didn't twig there was a bit missing!

[Tweedledee]

Easily done when copying

 

Trouble is with EMS links, there is a bit missing in the middle, the 3 dots, so you can't just put the "htt" up front to correct it.

[Guest peepshow]

Hi Richard. it's been a long time and for most of it I've been scratching my head!

I am never going to make or buy a thread that's accurate enough for my needs. Why re-invent the wheel? I've come to a decision. The answer is, of course, what you suggested months ago, a Barn Door.

I'm wondering why you went for a type 3 not a 4. The 4 has good accuracy and runs for 2 hours. I've been looking at these plans and performance figures.

p://education.jla...cker/index.html

The plan is to make an equatorial version using the base block, bearings and hub that I've already made, so not all my efforts so far are wasted.

What do you think?

 

Hi Alan again.   That was the site that I based my barn door on.  I was mesmerised by the accuracy one could obtain by type 3 and 4.

 

I went for type 3 as it was smaller and the accuracy over 1 hour seemed acceptable at the time.

But all those dimensions MUST be spot on with no movement at all, anywhere for long focus lenses on long exposures.

 Mine are all adjustable, for trial and error....mostly error as there is only one sweet spot.

 

If you wish to use 200-500mm lenses for exposures of many minutes then there is another fly in the ointment with a barn door.

Your PA MUST be spot on, so precise drift alignment is essential.  

 A star will quickly cover more than  two pixels with the slightest PA error on exposures over minutes with those lenses. 

 

I am learning this the hard way.

 

Another fly is motor drive speed.  Must be correct AND stable if those stars are not to be elongated at all as you have no auto guiding to help you.

 

I do not know how accurate your rig will be, but you did mention once taking 5 minute exposures with a 500mm lens, so that requires good tracking and PA. 

I have attained this but............

 

I am still strugglingl here to get the PA STABLE on a permanent set up.    Having no pier makes this more dificult.

 

I have a large ali tube supported by 3 stays fixed to 3 paving slabs.   There is still movement week by week( I think?), so have recently put in INVAR rods fixed to stakes driven 2 ft into the ground for stability with change of temperature. 

 

So I would be very careful, Alan, before throwing more effort and time into a barn door extention to your rig  if you cannot meet the above and if you wish to use a 500mm lens for 5 minutes or thereabouts.

 

  Also I would never consider a barn door for long exposures with 200mm+ lenses if portability were required.  .......you already have an HEQ5, so get yourself something like a small Skywacher EQ3 PRO guided and just  stick your camera on that for portability. 

Let me know what you do eventually.

 

I am beginning to wonder if all this hassle is worth the candle for me as there are now clouds here nearly every night.  

I am not a late night owl like some so my window is just up to midnight. 

But my short window is always clouded over these days. 

Edited October 27, 2014 by peepshow

[Guest peepshow]

Re PA, Alan...........

  

I do drift alignment using the camera.  I take a 10 minute exposure with the barn door drive motor off

giving me a long 10 minute star trail.

 

My motor, although crystal controlled for imaging speed stability, can be switched into a variable fast forward/reverse mode.

So after 10  minutes I run the motor in fast forward. 

 

This returns the stars to their start trailing position and it shows if the PA is out. This is done both testing east and south as per normal drift alignment.

 

The image shows if I require the barn door axis to be adjusted either N/S or E/W.

I have 4 adjustment screws at the top of the the axis assembly for this.  

 The bottom of the assembly sits on a single  ball bearing. (As the tall Sutton Coalfield TV mast does )  

 

Aw, ain't we still having fun.  

Edited October 27, 2014 by peepshow

[Werisit]

Hmm, plenty to think about there! Thank you Richard.