Jump to content

Banner.jpg.b89429c566825f6ab32bcafbada449c9.jpg

Best Mount for Imaging (£5k-£6k Budget)


ribuck

Recommended Posts

I guess in the long run, replacing motors in a direct drive system might be easier than replacing worm/screw configuration. Because this is probably going to be a one off purchase that will need to last me a good 10 years, i'm just trying to think of the pro's and con's of each system.

It's a hard one to call.

Link to comment
Share on other sites

  • Replies 367
  • Created
  • Last Reply

I too believe direct drive is the future. The question is if we're there yet. As for gears needing maintenance, well, yes, but more so the belts, which I think will wear down before anything else. On the other hand, replacing a belt must be so much easier than replacing gears and re-shimming the thing.

I am kind of surprised that noone has introduced a mount with one giant toothed belt drive wheel on the axis itself and a smaller one on the motor. After all, belts too are the future and gears are not. I would imagine it being far less sensitive to disturbances that the friction drive... Hmmm... Gather capital?

/per

Link to comment
Share on other sites

Well, in total with encoders the AP1600 is about $21K which is about £13.5K placing it between the GM2000 and GM4000, where it also sits capacity wise. It's tricky to compare across the pond.

/Jessun

EDIT forgot to even mention what mount I was talking about...

Link to comment
Share on other sites

Yeah, but VAT etc is beyond the manufacturer's control, and I think AP generally are in line price wise with the rest - for their home market... I wonder what a GM2000 cost 'over there'. Never looked. (24 volts you know :p ).

This thread will soon encompass everything you ever needed to know about mount design pros and cons, prices, pointing models and even the odd thought on flips. Best thread in a long while :smiley: .

PS the Mesu seems to crawl up in price by the hour...

/Jesper

Link to comment
Share on other sites

I never checked the Mesu prices... What happened?

And yes, it is a good thread. I may come out as a hard-core freak with my mind set, but you should bear in mind that I did a quite extensive research before getting my GM2000HPS - serial number 25 ;)

/per

Link to comment
Share on other sites

25! Epic. I though I was a freak getting no 108 of the Scuderia Ferrari Alonso carbon fibre steering wheel replica!! (Obviously I had to take it a part to fit racing sim electronics!)

I still think you over rate the importance of home position :tongue: ...

We all do research but also demand slightly different specs. If I did market research for this type of upper end amateur range I'd go insane and quit....

/Jesper

Link to comment
Share on other sites

Wish there was some supplier that did equipment show-downs a bit like computer equipment. So you would have showdown, with 2 different mounts both using the same camera, same filters, same length exposure taken at exactly the same time.

Now that would be amazing, as you would be able to see the real world results, and see if all the bells and whistles make a big difference.

Link to comment
Share on other sites

Ah! The dreaded home position! It is THE BASIS of a non-encoder mount. Why do they put homing sensors on mounts without encoders?

Anyway, of course it can be compensated for in parameters. If the mount is orthogonal and perfectly polar aligned it turns into offsets. If the mount is not properly polar aligned, nor orthogonal, it turns into something else, doesn't it? ;)

The nath involves sine and cosine as well as their inverses. Said functions are not as accurate as offsets.

/p

Link to comment
Share on other sites

Why do they put homing sensors on mounts without encoders?

Aha, now that's a new input :smiley: . For mounts with no such sensors the position has to be calculated each time on the basis of your handset inputs. Then it will go on dead reckoning on the basis of this calculated starting point - which btw moves at solar rate rather than sidereal - to find the first star. Is the star off? Probably. Has the universe changed since power down? No. Assumption: could the home position have been out of whack? Probably. So is it of any mathematical use? Err... no.

Anyways, any news on your GM1000 Per? My 1600 in transit is in Ditzingen.... wherever that is :p

/Jesper

Link to comment
Share on other sites

Jesper, you're missing the point! When all is orthogonal, the home position doesn't matter as it turns into a straight Ra offset and a straight Dec offset. When the scope is off polar alignment and/or off orthogonality, the missed home position is anything BUT an offset in Ra and an offset in Dec. It turns into a complex mathematical horror story that the model has to compensate for. Then, it all comes down to the quality of the (most likely) math.h library. Many imnplementations sacrify accuracy for speed. For one arcsecond accuracy, you need a sine function that can deliver accuracy in the 7th digit and some. Add some math and you demand even more.

Now, a normal IEEE floating point number has an accuracy up to 7 digits. It scales, but accuracy during computations are at seven digits (at the most). How much computing power is available in a Synscan handset? I don't know, but it isn't much. If it even is floating point, it is accurate to seven digits. If it isn't, the sine functions are probably implemented in some form of table driven cheat code.

Now, if you can establish a GOOD home position - as good as in OFFSETS ONLY - you are home free. If you cannot, well, the math is going to get you ;)

I've hacked the micro processors since the early 70's, starting with Intel 8008 and moving on. Even today, with the modern micro processors and controllers, I find the math lacking in precision, especially when it comes to trigonometry.

If you ditch the Synscan and go completely EQMOD, you move the mathematical responsibility into a completely different world of opportunities. But you're still giving in to trigonometry, and even the normal math libraries are sometimes inadequate. I believe EQMOD uses the ASCOM astronometric library so that should be safe ;)

My GM1000HPS is, best guess, in München, and is currently being subject to final quality control. It should be here shortly. Ditzingen is en route from Southern Germany to Reims (the route I normally take) so your AP should be along any day now. Ditzingen, I believe, is one of UPS' hubs. All of my stuff from Baader usually pass through there.

/per

Link to comment
Share on other sites

Jesper,

It was the Mesu mount that I was looking at myself some time back, it looks and sounds very good.

I must say I like the idea of doing mount head to heads, you could call it 'mount wars', if I ever win the lottery I will do it, I would need to learn how to use a CCD first though.

Alan

Link to comment
Share on other sites

Good discussion here, and I'm the first to struggle with Pythagora's theorem of us two but these general alignment routines are probably more simple that we give them credit for. They will assume that you stumbled out in a field in the dark, did your best to place the scope in the 'Home' position, and will take it from there. There's not a shadow of a chance that the scope is even near a perfect home position at start up so it lacks any mathematical importance and MUST be disregarded in formulae to find star two and three.

For subsequent runs where you parked at Home, true, you are much closer to home than you were first time around - leaving the setup untouched - but how would the software know if you start from a perfect position or a wonky field setup position? It doesn't have a clue! It will always ask, 'where am I?', 'what's the time of day?', 'Where am I pointing?' (Start from 'Home' - Enter is the answer to that) So it will run the same simple program. Only difference being that you'll be much closer to finding star one.

I know you know your stuff, but I'm talking simple mounts here, with no home sensors, with no clocks that run once you power down etc...

/Jesper

Link to comment
Share on other sites

Nope, Jesper. That's not the way it works. The software assumes you have a perfect home position. It can do nothing else. It is all based on the fact that the home position is good. Then, as it is about to slew to the first alignment star, it calculates the distance and then translates that into a number of stepper motor steps. Fine. If we assume that the home position is perfect then all is fine. The rest of the errors are what they are and will be roughly compensated for. If the home position is not, it doesn't have a clue as to where the error is. It will thus assume that it is all in the alignment and will subsequently struggle to compensate.

I have tried this properly. Start with a close to perfect home and you get good pointing. Start with a less than perfect home and do the same alignment and you get worse pointing.

When you park your scope it goes to the home position. If it was wrong to start with it will park at that (wrong) very position. Nowhere else. Alignment NEVER adjusts your assumed home position, it only adjusts for everything else which, in essence, is fine. When you start up your mount, whatever position it is in will be assumed to be the home position. That, and specifically that, is why you loosen the clutches and adjust your home position to as best it can be on the first aligment star! You assume that you have a good polar alignment and you assume that you have a low othogonality error. The majority of the residual is going to be your home error. Simple as that. Get rid of it and get rid of the dependancy of complex math.

Why don't you try it with an EQ6 the next time you get near one. Misadjust the home position on purpose and use only the hand controller to center the stars in a three star. It will work and the pointing will be so-so to OK to maybe not so OK. Then redo it with the first adjustment being a de-clutch and a handset tweak. The next star will be pretty much on spot and the third as well. You will then find the pointing to be pretty good; as good as can be expected by the equipment in question.

Now, if anyone has a problem getting the EQ-series aligned, I suggest the following:

- Time entry error (or even worse: date error) Doesn't need to be much to throw it off

- Time zone error

- Gross position error (it needs to be signifficant to have any impact)

- Polar alignment error

I have never, ever, missed an alignment with my NEQ6 for any reason but the above ;) Ever.

Sunday night and I'm ranting - Haha!

/per

Link to comment
Share on other sites

I too believe direct drive is the future. The question is if we're there yet.

When the ASA introduced the DDM there were a lot of issues and no support. That period is over. Hardware and software is becoming better. The current software runs without problems, also via ASCOM. Setting up the mount is easy and the software provides lots of tools to do it right (balancing tool, polar alignment, modeling tools, etc.) and fast. I did a lot of research before ordering my DDM60. 10Micron, Mesu, AP, PMX, they were all on my list, but I definitely don't regret choosing the ASA. It tracks with 0.4" rms, pointing is spot on, no sound, no backlash and no periodic error. I think only 10Micron comes close to this performance.

Link to comment
Share on other sites

Per,

The software assumes you have a perfect home position. It can do nothing else.

Yup. It is clue less. So it starts with an assumption - a calculation based on what you tell it about time, position etc.

it calculates the distance and then translates that into a number of stepper motor steps. Fine.

Yup. Simple dead reckoning.

If the home position is not, it doesn't have a clue as to where the error is.

Once you align star one, assuming you only do a one star alignment, it will assume there are no errors. That's why it's so generous about where you started from. Unless you're off by something like 30 degrees or two hours in time, it will happily tell you that alignment was successful. If you're off by more, something is seriously wrong and the safety of your gear is questionable so an alignment failure message will follow.

I have tried this properly. Start with a close to perfect home and you get good pointing.

I've tried it too, and true, you get better pointing for star one the more precise your home was. Only because the dead reckoning steps tallied. The formula for finding star two and three is the same regerdless were you started from. It's a matter of dead reckoning steps from star one and nothing else.

That, and specifically that, is why you loosen the clutches and adjust your home position to as best it can be on the first aligment star!

No, I argue, the software is forgiving; it knows you're stood in a cold, dark field with a dodgy watch and compass. It will accept handset slews to find star one, and never looks back. It will for star one assume everything is perfect, such as orthogonality and PA. Star two and three will make it rethink that. The position of stars two an three though has no bearing whatsoever on where you started from - they are merely coordinates and set motor steps away from star one. If subsequently stars two and three are off it's due to misalignment PA wise or due to cone error which it will try and calculate to compensate for. (I doubt atmospherical diffraction is taken into consideration for cheaper mounts, but I certainly don't know).

Alignment NEVER adjusts your assumed home position

The iOptron does. Since it now knows where things are it can more accurately slew to a home position and if you want weights down, down they go - like a weight on a string. I couldn't tell you for sure what the SW range does, since I can't actually remember - never had alignment issues with the ol' HEQ5 that I miss dearly. :smiley:

Once you have established time, day, position on Earth, and the correct position of star one, it's IMHO nothing short of madness to use the allegedly perfect home position in any way to attempt to find stars two and three.

/Jesper

Nothing like a good Scandi rant :p

Link to comment
Share on other sites

Ah! The iOptron... The whole discussion in the other thread was about the EQ5... I don't know squat about iOptron. Synscan NEVER changes the home position assumption. From that we can deduce that the whole debate is pointless.

Star two and three can never be offsets from star one and two. That would make the math impossible as it would imply that the first star takes out all errors, which it can't. In essence, it would make a home position completely uncalled for, which it isn't ;) I urge you to study the ASCOM code for EQMOD. It contains a very good alignment model which is based on triangular like structures. It is much more complex that you think, even in "simple mounts".

/per

Link to comment
Share on other sites

ASCOM code is best left for clever people... I'm just a fairly uneducated but often contemplating end user. No degrees here, that's for sure.

Star two and three aren't offset as such, that's not what I suggest - they just serve to reflect errors in finding star one.

The perfect setup (pure theory here) would nail one, two and three. Period. All aligned and well.

A less perfect setup will after some help find star one. Both these example mounts of mine use the same database - dots on a sphere turned into coordinates, so both will perform the same number of motor steps to find star two. The perfect setup will find it, the less perfect won't, since let's say the scope is misaligned in the rings or PA is off.

From that the less perfect mount can see that the steps from one to two had to include a compensation - this compensation is because the mount didn't ever 'point' at star one, only the OTA did. One more point in space - a decently large triangle I'd imagine, makes it able to pinpoint where the angular difference is between the pointing of the mount and the OTA on two axis. So for a wonky tube, no clever correction has to be added for subsequent stars, it's pure dead reckoning star hopping from last successful alignment star. It will kick in again though after a flip, as the correction need to switch +/- sign. For PA misalignment it would require a bit of maths for each slew even before flipping sides. The bottom line of my argument (that's all it is - I never programmed Synta mounts or any other piece of kit) is that whilst a spot on homeposition helps to find star one, it's not crucial, and won't ruin your alignment or entire night, and that it doesn't matter if you slew, or move by hand to star one, and that the home position is calculated on the fly and is a nuisance thereafter.

The iOptron has several park positions. A few set ones, and one that you can pick as you like.

I wonder what the latest Synscan handset update brought to the table, I believe it has introduced a few different park positions.

Good discussion anyways! I hope to learn something everyday and this helps :smiley: .

Looking forward to your GM1000 report, and to see what everyone finally ends up buying :smiley:. Sorry for the creap.

(Finally could anyone with a HEQ class mount please fire it up from an slightly offset position in RA, run an alignment and then park to see if the weights point down or turn back to the offset position? Just for fun I'll bet a brew of your choice Per that it will correct the park position so that the CW bar points down :tongue: ). That means you can't test Per, since you're biased. What beer do you like? :grin: ...

/Jesper

Link to comment
Share on other sites

Per when you get your GM1000 setup, could i ask a very big favor, would it be possible for you to do a video of your complete setup process, from first powering on, balancing to completing a 3 star alignment and being ready to image. I know this is a lot to ask but i really want to understand the setup process versus the ASA system, and i would be extremely grateful as it will help my decision immensely.

Jeffery, Would it be possible for you to do a similar video for your ASA, detailing the power on, balance, creating a pointing file to the point that you are ready to take an image.

I know this is a big ask guys, but it would really help me and i'm sure it would also help other in a similar position of choosing either of these mounts. If it;s not possible, then i fully understand as it is a lot to ask for.

Rich.

Link to comment
Share on other sites

I don't mind to do a video, but the problem is that I have a permanent setup. So I don't have to balance, tune, polar align or create a pointing file... I just flip the switch, do a homing routine and go. Online within 1 minute...

What would you like to see in such a video? Maybe I can describe my actions without redoing it.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.