Jump to content


Photo

Putting the "Rule of Thumb" to test

  • Please log in to reply
226 replies to this topic

#151 fred1871

fred1871

    Viking 1

  • -----
  • Posts: 871
  • Joined: 22 Mar 2009
  • Loc: Australia

Posted 03 February 2013 - 12:04 AM

Wilfried, thanks for the information about AstroPlanner. The trial version is so limited, now I've looked at it, that I didn't get a sense of how much it could do.

Some observations. You mentioned STF 450 in Taurus being more difficult than you'd expected. I had a look at it a couple of nights ago and found it very easy. Despite modest altitude(~26 degrees) and seeing not great in that direction, I could see the companion at 80x, and it was obvious at 114x (140mm refractor).

I've been working through more Orion doubles and some of these won't be "at the limit" examples. But two I've looked at in recent nights don't appear to be on your Orion list - or I've overlooked them there.

One is BU 15 (0547.8, -0218), mags 7.4 and 11.6 at 2.0". The separation appears to be the same in 1987 (last measure) as it was when Burnham discovered it in 1872 with 6-inch refractor. This one took a bit of magnifying to see the companion, but 285x showed the lesser star as a tiny speck separated but close to the primary. Seeing was very good at that time.

The other one I didn't find on your list was STT 119 in Orion. Again, it's a pair which seems to be at the same separation recently (2009) as it was when discovered (1843). Magnitudes in WDS are 8.08 and 8.93 - separation only 0.6"! 140mm refractor again, this one took a lot of magnifying, being "not single" at 285x, and at 400x a definite pair, a bit uneven, notched but not resolved into separate points. Seeing very good - I looked at this one straight after BU 15. Very pleasing result given the stars' modest magnitudes; and NELM was at least 5.5.

I'll add some more observations as I work through the details of them and cross-match my observing list with your Orion list. My main problem at the moment is late sunsets plus daylight saving (Summer), so it doesn't get properly dark until 9:30pm, and I sometimes get clouds rolling in well before midnight. The weather pattern this Summer has been very unusual and not great for astronomy.

#152 WRAK

WRAK

    Apollo

  • -----
  • Posts: 1151
  • Joined: 18 Feb 2012
  • Loc: Vienna, Austria, Europe

Posted 03 February 2013 - 02:12 AM

Fred, yes - STF450 should be easy in a 140mm refractor. But I had somehow the impression that the companion is fainter than advertised compared with +10mag stars in the same field of view.
Congratulation to BU15 and STT119 - both doubles seem to be very limit observations with a 140mm refractor. Both are not on my list for reasons I don't know - maybe the selection feature of AstroPlanner is not this perfect.
No experience with the trial version - you need access to the full database of catalogues including WDS, so the trial version does not show full performance concerning double stars.
Wilfried

#153 fred1871

fred1871

    Viking 1

  • -----
  • Posts: 871
  • Joined: 22 Mar 2009
  • Loc: Australia

Posted 04 February 2013 - 06:57 AM

A couple more observations. Norme managed a "sort of" split with BU 1190 in Orion (0557.4, +0002)- I've now had a look at it in fairly good conditions, and the close companion was visible at 285x and 333x (7mm and 6mm orthoscopics with 2.5x Powermate). Like Norme, I found it not a clean easy split, but it was a definite one with my 140mm refractor. [mags 6.95, 9.8 at 1.4"]. With averted vision I caught glimpses of the mag 12.1 companion at 6.7" (same magnifications). No moon at the time. A tough combination for 140-150mm.

BU 13 (0534.5, -0429) I also caught up with - mags 7.2 and 9.2 at 1.0", so I thought it'd be not too tough - it was harder than I expected, and took 400x (140mm refractor) to give a clear separation in the steadiest moments. That appeared to be a seeing effect, and my impression was that less power would show it in steadier conditions.

The continuing issue with this project of observing marginal pairs for the telescopes used, is getting "good enough" atmospheric conditions to make it possible. As so often remarked, a pair that's invisible as such one night is clear and obvious on another - SW Burnham went further, and remarked on the difference occuring even with nights that appeared to be equally good. Hence, I suppose, Wilfried, your "50% probability" model.

#154 WRAK

WRAK

    Apollo

  • -----
  • Posts: 1151
  • Joined: 18 Feb 2012
  • Loc: Vienna, Austria, Europe

Posted 04 February 2013 - 02:09 PM

Fred, it seems that you have excellent conditions now.
With the STT119 observation you have reached a sensational 0.73 Dawes ratio if the advertised data is correct. Regrettable STT119 is a bit too faint to try to measure the separation with a self made interferometer consisting of several simple aperture masks with two 10mm holes in different distances (minimum contrast of the interference lines at 95mm distance would mean 0.60" separation and a minimum at 85mm distance between the holes would give 0.67" separation with an average error of only 0.05" - but due to the small holes it works only with equal doubles brighter +6mag). I have to admit that I never tried such measures myself but it sounds this interesting that I will do this cerainly on occasion.
Wilfried

#155 fred1871

fred1871

    Viking 1

  • -----
  • Posts: 871
  • Joined: 22 Mar 2009
  • Loc: Australia

Posted 05 February 2013 - 01:28 AM

Wilfried, some nights I have (nearly) excellent conditions for part of the night and in some parts of the sky. STT 119 is not unique among close pairs I've seen as double, whether split (separated) or elongated. A few brighter ones have been even closer - in the 0.5"-0.55" range - only elongated of course.

At 0.5", I'd be getting near-enough to 0.5 of the Rayleigh Criterion. And that would fit with what Couteau gives as the limit for detecting elongation with the Nice 20-inch (50cm) refractor. As 14cm will be less atmosphere-limited than 50cm, I think that's a reasonable target to aim for, under the best conditions.

Taylor (in the Argyle book) detects elongation down to around 0.5 of Dawes, 0.4 of Rayleigh. That's remarkable, but a few other observers are on record for similar feats - the oft-mentioned SW Burnham for one. I think I'm doing quite well as an observer but I've definitely not reached those remarkable levels of discerning close pairs.

One example among the closer pairs I've seen - Zeta Bootis, a binary at present getting closer year by year. I observed it in May 2009 and May 2010 with the 140mm refractor - both times with 400x. In 2009 it was elongated with a hint of notching. In 2010 it was a "rod" without a hint of a notch. I plan to get back to it this year to see if I can get any suggestion of elongation. Separations, based on the grade 2 orbit (~124 years) are just under 0.6" in 2009, 0.55" in 2010, and will be 0.47" by May this year. So, this year if I detect anything "out of round" I'll set a new personal record. Delta-m for Zeta Boo is only 0.1 magnitude.

The interferometer idea sounds interesting, but as you say the stars need to be pretty similar in brightness, and with only 140mm I'd think STT 119 a bit faint, well below mag 6. For measuring, especially for close pairs, I'd think the CCD technique described by Rainer Anton in the 2nd edition of the Argyle book might be more useful.

#156 WRAK

WRAK

    Apollo

  • -----
  • Posts: 1151
  • Joined: 18 Feb 2012
  • Loc: Vienna, Austria, Europe

Posted 07 February 2013 - 01:59 PM

I think I found a solution for my dilemma regarding Dawes criterion and overperforming observations - currently I do not allow the RoT model to result in values below Dawes because I think this is hard enough for a 50% split chance. On the other hand I need at least some overperforming observations as counterbalance to the certainly existing underperforming observations in my data set - but this results in some bias as the statistical analysis has to struggle to minimize errors it cannot avoid.
So I think I will allow the RoT model to result in values below Dawes but in an additional cut off step I will set all sub Dawes values to Dawes criterion (modified for CO if necessary). This will may be result in a greater standard deviation but the quality of the model should be improved this way.
And I will apply a similar approach to the other end of the observations with very faint magnitudes of the secondary - here I will first made a more or less "educated" guess for an additional component of the model but also make a cut off for all results above the required aperture according to the TML (corrected by NEML) required by the magnitude of the secondary.
Wilfried
PS: The CCD would make it necessary to split the double I think. Curiously this is not necessary for the interferometer at all

#157 fred1871

fred1871

    Viking 1

  • -----
  • Posts: 871
  • Joined: 22 Mar 2009
  • Loc: Australia

Posted 07 February 2013 - 06:14 PM

Wilfried, I think allowing for "overperforming" (detecting doubles closer than Dawes) is necessary to allow for - these are mostly fairly bright pairs where a split is not possible, but a notched or elongated - rod or olive - image is visible.

That's different from what is seen with significantly uneven pairs. Pairs with moderate delta-m can show, be detected, as elongated in some cases. But those with a larger delta-m are more difficult to see in this form; usually the primary so overwhelms the dim companion that it's not detectable, unless separated, rather than overlapping or an extension of the primary.

That's part of what makes an RoT hard to establish. There's a gradual change in what can be seen. If delta-m is large, we depend on the fainter star being in a diffraction gap; if it sits on a diffraction ring it immediately becomes much more difficult, how much more depends on the ring brightness and the star brightness.

And it gets messy. A CO changes the relative brightness of rings according to the CO size (% of diameter of main optics). Looking at some of Suiter's illustrations, I notice on p.165 (2nd edition of Star Testing) a set of pictures showing "in-focus diffraction patterns resulting from central obstruction".

Unobstructed, first ring moderately bright, 2nd ring less, 3rd ring less than second.
25% CO - first ring bright, 2nd ring very faint, 3rd ring middling bright.
50% CO - first ring very bright, 2nd ring bright, 3rd ring faded to near invisible, 4th ring increased though very faint.

A very interesting pattern of change, according to CO obstruction ratio!

Low-order spherical aberration is often treated as having much the same effect as CO. However the pictures on p.191, showing the effect on the diffraction pattern of various levels of SA, with obstructed and unobstructed apertures, is different from the pure CO effect.

With an unobstructed aperture, increasing SA firstly reinforces the brightness of the first ring, then the second as well. With an obstructed aperture (33%), the first ring gets brighter, but already with no SA there is some reinforcement of the 3rd ring, visible at 1/8 wave SA as well. By 1/4-wave SA the 2nd ring is broader and somewhat brighter than the third (itself reinforced), and with even more SA the pattern of brightness, 1st brightest, 2nd is 2nd brightest, 3rd is 3rd brightest shows.

What I find interesting here - first, with no CO the diffraction ring brightness pattern of increasing SA is different from adding CO without SA.

Second, 33% CO, as with 25% CO, reinforces the third ring brightness rather than the second, with little or no SA; but with increased SA, the pattern is similar to, though worse than, the unobstructed pattern for increasing SA.

A lot of SCT and Mak-Cass scopes have CO in the 25-35% range. Those that have very little low-order SA may show this effect of faded second diffraction ring and reinforced third diffraction ring. Of course, atmosphere makes such a difference that in the real world this effect is likely lost as a visible effect, merely becoming one of multiple factors mixed together to determine what one sees.

My conclusion is the obvious one - the less SA the better, and smaller CO ratios are better than large ones. :grin: Nothing new there.

How much effect there is on uneven doubles observed in Mak-Casses, where low-order and high-order SA need to be balanced, but some aberration remains, I don't know.

Very faint pairs? - I'm increasingly feeling these are an eyesight test to a much greater degree than uneven pairs of middling brightness. I've got my C9.25 going again and the extra light (a whole magnitude!) over the 14cm refractor is very nice, despite less perfect optics and more sensitivity to atmospheric wobbles. Perhaps I'll try some of your faint Gemini pairs with that. They're certainly beyond my eyes with 140mm - super-vision people like SJ O'Meara might manage them.

Regarding CCD imaging - yes, some kind of split is necessary for measuring. Agreed, an interferometer works differently. You've reminded me to look up information on Finsen's eyepiece interferometer, which allowed him to measure pairs too close for normal micrometers, down to around 0.1" or better with the Innes 26.5-inch (67cm) refractor in South Africa (1940s-1950s). I'll find the reference and give it in this thread - I'm fairly sure it's in one of the journals online through ADSABS.

#158 fred1871

fred1871

    Viking 1

  • -----
  • Posts: 871
  • Joined: 22 Mar 2009
  • Loc: Australia

Posted 08 February 2013 - 12:20 AM

I've now found the reference for WS Finsen's own article about his eyepiece interferometer.
Monthly Notices of the Royal Astronomical Society, 1951, v 111, p 387 ff.

Bob Argyle, in the 2nd edition of his Double Star book for the Webb Society (1986), remarks :
"with this instrument more than 13,000 examinations of 8,117 stars ...were made. As a result 73 new pairs were found, 11 of which have periods ranging from 21 years down to 2.65 years. In addition, 6,000 measures of pairs too close for the [filar] micrometer were made".
These results appeared in various papers over time.

The interesting thing is the micrometer itself. Whether something similar could be made for amateur use on mid-size scopes? I'll leave that for the instrument designers and engineers to work out.

#159 WRAK

WRAK

    Apollo

  • -----
  • Posts: 1151
  • Joined: 18 Feb 2012
  • Loc: Vienna, Austria, Europe

Posted 08 February 2013 - 12:28 PM

...The interesting thing is the micrometer itself. Whether something similar could be made for amateur use on mid-size scopes? I'll leave that for the instrument designers and engineers to work out.

Look at this: http://www.observatory.org/bfm.htm - may be a bit expensive but seems of high quality.
Concerning eyepiece interferometer: Do you know any commercial products?
Wilfried

#160 fred1871

fred1871

    Viking 1

  • -----
  • Posts: 871
  • Joined: 22 Mar 2009
  • Loc: Australia

Posted 08 February 2013 - 07:57 PM

Interesting filar micrometer, sounds good, price is up where I'd expect as a minimum if it's well made. Should be suitable for long focal length instruments. A 25mm eyepiece suggests very long focal length systems to get decent magnification; I'd hope the thing would still work with shorter eyepieces, or only wide pairs would be measurable (at 4m fl the magnification is only 160x). Personally, I'm more tempted by CCD imaging, maybe with a video style camera for "lucky imaging" in the same way the planetary imagers do it.

Finsen's eyepiece interferometer? - there are photos of the device with the article. Likely made as a 'one-off' by the instrument shop at the observatory. I've never seen an advertisement for this kind of micrometer. Several other types, yes, in the past, but not this type.

An alternative for very close pairs, not too uneven, is speckle interferometry - and the article in Argyle's current book suggests ways for the amateur observer to do it. The potential accuracy is very good. However people like myself, who like simpler techniques, will probably follow Rainer Anton et al into "lucky imaging". That could be next year's project - this year I'm staying visual.

#161 fred1871

fred1871

    Viking 1

  • -----
  • Posts: 871
  • Joined: 22 Mar 2009
  • Loc: Australia

Posted 08 February 2013 - 10:42 PM

Wilfried, I'm still wondering about some doubles that you didn't get on your list with AstroPlanner when you used it to filter the WDS for a list. Have you been able to find why those pairs (I think around 2-3 February the comments on these) did not show on your list?

My reason for asking is that I'm looking for software that can produce lists, by constellation, with data from the WDS - and all the software packages I've tried so far have either edited out too many doubles - perhaps they use default criteria that are wrong - or want to give me lists with unhelpful names for stars (HD, SAO, HIP, etc) instead of staying with the WDS names, which refer to discoverers.

AstroPlanner looked likely to be what I was seeking, until I found it had left out doubles that should have been on your list. So I'm now wondering if there are limitations to it that mean it won't give me what I want - the ability to filter by magnitude, NOT separation, and use the WDS, and have selection by constellation.

Your lists show it can do what I'm looking for EXCEPT that some doubles were somehow missed, when they should have been listed.

#162 WRAK

WRAK

    Apollo

  • -----
  • Posts: 1151
  • Joined: 18 Feb 2012
  • Loc: Vienna, Austria, Europe

Posted 09 February 2013 - 04:13 AM

Fred, I found these "missing" doubles via AstroPlanner without problems with a new search with the accordingly parameters. I do not remember how I exactly did this search for the Orion list so may be it was my combination of parameters leading to this result but I have no real explanation - but I think AstroPlanner is an excellent tool for double star session planning.
Wilfried

#163 fred1871

fred1871

    Viking 1

  • -----
  • Posts: 871
  • Joined: 22 Mar 2009
  • Loc: Australia

Posted 09 February 2013 - 06:51 PM

Thanks, Wilfried - sounds as if AstroPlanner will match up with what I'm looking for. There's a lot of useful software around, but the problem is finding the examples that do what the individual observer wants.

I'll have some more observing data soon. I'm analysing some recent, and some not so recent, observations, including some with the 9.25" SCT - a good'un for testing CO effects (~36% by diameter!) and for fainter but not very uneven pairs - it makes the faint ones much easier to see merely through extra light, before any other factors are considered.

#164 Asbytec

Asbytec

    Guy in a furry hat

  • *****
  • Posts: 8221
  • Joined: 08 Aug 2007
  • Loc: La Union, PI

Posted 09 February 2013 - 07:20 PM

Hello guys, interesting conversation. I especially appreciate Fred's input on SA and CO. The complexity of which speaks to the difficulty and seems to be a necessary aspect of such a project.

I was really enjoying observing stars associated with this project while testing both my own limits and those of my scope.

#165 WRAK

WRAK

    Apollo

  • -----
  • Posts: 1151
  • Joined: 18 Feb 2012
  • Loc: Vienna, Austria, Europe

Posted 10 February 2013 - 04:21 PM

The effects of spherical aberration for double star resolution seems to me to be a complex topic - anyway as most observers will not know any specific SA value for their scope it cannot be used in a RoT model with benefit, so I tend to see SA as random factor in this regard.
Other factors with some relevance could be the existence of spiders producing spikes at least for brighter primaries and therefore changing the diffraction pattern and also the focal ratio especially of Newtons as "fast" reflectors seem to be of no good use for double star resolution due to coma. But for now I tend to shelve such considerations for possible future extension as I am struggling to come to a decent enough RoT model with the factors in evaluation so far.

Fred - looking forward to your C925 observation reports with great interest.
Wilfried

#166 Asbytec

Asbytec

    Guy in a furry hat

  • *****
  • Posts: 8221
  • Joined: 08 Aug 2007
  • Loc: La Union, PI

Posted 10 February 2013 - 07:40 PM

Wilfried, yes, using SA explicitly would seem too complex and unhelpful. I found assuming Strehl to be interesting for calculations involving MTF. With Strehl, peak intensity is either explicit for unobstructed apertures or implicit for obstructed samples. Strehl and obstruction both determine light distribution, of course. Light distribution seems to be part of the problem. So, while specific measures of SA are both difficult to know and use, assuming a Strehl might offer some utility.

#167 WRAK

WRAK

    Apollo

  • -----
  • Posts: 1151
  • Joined: 18 Feb 2012
  • Loc: Vienna, Austria, Europe

Posted 11 February 2013 - 03:59 PM

Norme, I fear Strehl is in the same leage with SA - an amateur observer knows rarely the Stehl value of his scope I think and it is usually certainly not part of an observation report at least as far as I know. My data set of limit observations would shrink to zero if I would require Strehl as known factor including most of my own observations - I know the Strehl value of two of my scopes but not of my preferred 140mm refractor used with the iris diaphragm.
Wilfried

#168 fred1871

fred1871

    Viking 1

  • -----
  • Posts: 871
  • Joined: 22 Mar 2009
  • Loc: Australia

Posted 12 February 2013 - 01:49 AM

Wilfried, I can only agree that the short Newtonian is generally found to be problematic for doubles - there is the occasional one of exemplary optical quality and with a Paracorr and thermally stable and with a (relatively) small CO, plus very accurate collimation, that might well be good for double stars. But I suspect that it's a minority of short Newtonians that fit the whole description. Pete's experience in going from a 10-inch f/5 to an 8-inch f/9, and the big improvement for doubles, is a useful cautionary tale.

Spherical aberration is a common issue with all kinds of telescopes. In small amounts it need not be a big problem. Indeed, WR Dawes in 1867, in the introduction to his doubles catalog where he also announced what is now known as the Dawes Limit, remarked on SA in some detail: in particular, that a larger refractor with obvious SA outperformed a smaller one without noticeable SA - but the effect of SA was to produce a large amount of "false and scattered light" around brighter stars. Nevertheless, this affected "how a telescope shows a difficult object, than whether it can show it at all" [obviously for double stars, not detail on Jupiter]. His conclusion, that test objects are of "comparatively small importance in the trial of a telescope" because "so much must depend on the eye and habit of the observer, and the circumstances under which the scrutiny is performed", sounds curiously modern. :grin:
But it does suggest that for high contrast objects (doubles) SA is less of a problem than for low contrast objects. Dawes is more optimistic than I am about scattered light and its effect on faint companions.

More thoughts on SA, including its interaction with CO etc shortly. Meanwhile, thanks Wilfried for the AstroPlanner recommendation. I find it does quite well what I'm looking for, and without needing to do acrobatics to get standard double star designations, such as chasing through multiple tables of equivalence, or manually re-labeling most things, or trying to re-program a program when it doesn't do what it says it does. Some software writers don't appear to know enough about astronomy. And their "preferences" (if one can call it that) then get used by lots of other folk, leading to further confusions and extra work and identification muddles. The Haas project struck this, with a need to translate the original lists into equivalences for some observers for non-standard (in doubles terms) designations such as SAO, a catalog that's effectively obsolete, needed due to the software packages they had and the limitations of those. Yes, some software makes me grumpy.

Strehl? - MTF? - EER?. Comments to come on all that.

#169 WRAK

WRAK

    Apollo

  • -----
  • Posts: 1151
  • Joined: 18 Feb 2012
  • Loc: Vienna, Austria, Europe

Posted 13 February 2013 - 08:47 AM

...
So I think I will allow the RoT model to result in values below Dawes but in an additional cut off step I will set all sub Dawes values to Dawes criterion (modified for CO if necessary)...

This was not such a good idea - the statistical analysis resulted in a parameter value of very near 1 and did not bring any reasonable advantage in terms of standard deviation and correlation.
I think I will simply dump the obvious over- and underperforming observations to get clear of the huge errors they produce.
Wilfried

#170 Asbytec

Asbytec

    Guy in a furry hat

  • *****
  • Posts: 8221
  • Joined: 08 Aug 2007
  • Loc: La Union, PI

Posted 13 February 2013 - 02:53 PM

If your using actual observations with real world Strehl and SA, then tweaking the model to those observations would seem to account for them on average and implicitly.

#171 WRAK

WRAK

    Apollo

  • -----
  • Posts: 1151
  • Joined: 18 Feb 2012
  • Loc: Vienna, Austria, Europe

Posted 14 February 2013 - 10:55 AM

Norme, you may probably be right about Strehl and SA, but I don't have any observations reports including these values - but what I will do is at least eliminating the obvious underperforming observations resulting from using fixed apertures and lack of better competing observations, because these are certainly "wrong". This will give then more room for real limit observations including the overperforming ones.
Wilfried

#172 WRAK

WRAK

    Apollo

  • -----
  • Posts: 1151
  • Joined: 18 Feb 2012
  • Loc: Vienna, Austria, Europe

Posted 15 February 2013 - 10:19 AM

The large number of rather faint pairs in Wilfried's list for Orion got me wondering ...

... me too with the rather moderate proposed apertures for splitting these - as already mentioned there exists certainly a weakness in the current model regarding secondaries fainter than +11mag. As we come here close to the telescope magnitude limit at least for smaller scopes in combination with light pollution we need here an additional component working with the difference between the magnitude of the secondary M2 and the with NEML adapted TML similar to Peterson's approach.
A possible procedure could be to run the current model and then calculate the NEML adapted TML for the proposed aperture and compare this value with M2. If the TML is clearly above M2 then no further action is needed but if it is rather equal or even below then the proposed aperture is to be increased accordingly - what this means exactly is meanwhile not very clear, I will have to work this out.
First step is certainly to have a clear model how NEML works on TML. My first idea was that for a scope it certainly does not mean a full magnitude loss according to NEML, rather may be half of it. But this did not correspond very well to the existing observation data set suggesting a dependance on aperture in terms of less aperture means less magnitude loss.
On next possibility I will make a set of observations for the influence of NEML on TML with different apertures for a better data base for this purpose. Meanwhile I will work with a first approach for delta_TML = LN((6.5/NEML)^2.512)*(TML/13.43)^2.512 with TML = 2.7+5*LOG10(D_mm) and 13.43 as the assumed turning point from which the magnitude loss increases above the assumed half magnitude loss indicated by NEML. The image shows the effects of this algorithm for a NEML of 3.5.
Wilfried

Attached Files



#173 WRAK

WRAK

    Apollo

  • -----
  • Posts: 1151
  • Joined: 18 Feb 2012
  • Loc: Vienna, Austria, Europe

Posted 17 February 2013 - 11:21 AM

I hoped to be able to derive from the formula above the for the compensation of NEML required TML value but failed miserably - may be due to my cold induced loss of clear thinking. Peterson's approach works surprisingly good with M2 fainter +11mag - but not this good for brighter companions and did not provide any help for considering light pollution besides a seeing factor, but this is another topic. But numerical approximation proved again to be a good substitute so I got this fixed. Applied on the mentioned Orion list this added step to the RoT model did not change anything down to a moderate NEML of 4.5 but with lower values of NEML the effects get at least for wide doubles with faint companions quite drastic. With a for suburbs unfortunately usual NEML of 3.5 this gives (with zero obstruction):
SLE832 11.1" +10/11.8mag 150mm instead of 99mm
BAL2147 7.6" +8.7/11.8mag 150mm instead of 113mm
HJ700 11.5" +9.9/11.7mag 140mm instead of 96mm
But there is no change for doubles rather close like for example:
TDS3160 1.5" +9.5/11.7mag 174mm remain.
I will now produce a new spreadsheet for an at the moment final beta version of my RoT model and post it here soon including a list of doubles from the upper part of Eridanus.
Wilfried

#174 WRAK

WRAK

    Apollo

  • -----
  • Posts: 1151
  • Joined: 18 Feb 2012
  • Loc: Vienna, Austria, Europe

Posted 20 February 2013 - 05:17 AM

I have now uploaded the 3rd version of my RoT model and time will show if this is any good. The answer to the question which aperture is required for resolving unequal doubles with faint companions showed to be quite complex. While the single components of the model are based on more or less educated guesses the parameters are calculated with statistical analysis of limit observations (nonlinear regression). Some guesses were evidently not this good because they did not deliver on the numbers and had to be replaced by better ones - I am sure this process could be repeated several times to get even better results but this may be a task for the future. On the other side it was difficult to get enough solid observations to be able to do solid calculations - my data set is still far too small.
The process of developping the model was quite interesting as I gained a lot of for me new insights:
- Light pollution is for doubles with secondaries brighter than +10mag of no real concern
- For doubles with companions fainter than +11mag the by light pollution reduced telescope magnitude limit plays a significant role at least for wider separations
- The effect of central obstruction on the diffraction pattern seems on first look not this significant - but when calculating for required aperture small differences suddenly become a significant advantage for close doubles of similar brightness. With increasing magnitude difference this advantage gets smaller and results finally in an disadvantage for scopes with CO.
The uploaded spreadsheet is filled with doubles from the upper part of Eridanus. Any response with observations would be apprediated.
Wilfried

Attached Files



#175 fred1871

fred1871

    Viking 1

  • -----
  • Posts: 871
  • Joined: 22 Mar 2009
  • Loc: Australia

Posted 20 February 2013 - 07:10 AM

Wilfried, I will provide some more data points soon - I've had too many other things needing doing to be able to go through observations from recent weeks, some of them likely to be limit or near-limit observations.

I can only agree about light pollution being a limiting factor, especially for small and moderate telescopes, in seeing faint companions. Larger telescopes are less affected for 11th magnitude stars UNLESS very close to much brighter primary stars.

Regarding CO - my current impression is that medium CO (25-35%) is not very helpful for seeing close even pairs, compared to unobstructed apertures. The reduction in disc size is very little. Other factors, such as less good optics, collimation being even slightly off, thermal issues, seem to cause more loss than any gains from reduced disc size. And air turbulence unhelpfully interacts with CO as Danjon and Couder noted long ago.

I've worked out some further questions regarding CO and its interaction with optical quality as well. More on that also after tomorrow, because I'm preparing other material for a presentation that will keep me busy tomorrow.

Thanks for the Eridanus list. I'll look through it as well, and see what I can usefully observe, to add to pairs in Orion and Gemini I've already looked at. Gradually the data points will build up to good numbers. :)






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics