Getting the best performance from my AM5: Page 3

Getting the best performance from my AM5

w7ay

ShinyredAM5 end game being 100ms, will drop in increments of 25ms and see what works.

If you are going to slice it that thinly, give it a bit of leeway if your mount is not tracking at King Rate, and your target is near the horizon.

https://canburytech.net/DriftAlign/DriftAlign_3.html

Notice in the second of two charts about 1/3 of the page down that there is a 0.03 arcsec/sec rate added to the sidereal rate near the horizon, which accounts for something like 4 ms of the max rate when autoguiding at 0.5x sidereal rate.

And, as I mentioned earlier, any mechanical flexure too.

Make sure that you don't fine tune where the periodic error curve has the minimum slope. If you do that, the pulses will not be able to keep up at the higher slope regions.

Chen

raawe

@w7ay I did mine, absolute worst case is 0.228 arcsec/sec (violet line). What would be recommended settings for this case?

w7ay

raawe I did mine, absolute worst case is 0.228 arcsec/sec (violet line)

As mentioned multiple times already on this thread, just set the max RA pulse duration to give this much movement for the gude rate that you use. Nobody knows your guide rate, so no one can give you any number.

Chen

Jhaunton

Looking at that PE trace made me think again. All the HD mount guide logs I have seen show short duration, large spikes in RA - even when guiding well. I wonder if these are actually at the point in the PE curve where the error reverses? In the case above this means the error rate goes from approx +0.23" per sec to -0.23" per sec virtually instantaneously - the peaks are sharp. Just a thought.

w7ay

Jhaunton In the case above this means the error rate goes from approx +0.23" per sec to -0.23" per sec virtually instantaneously - the peaks are sharp. Just a thought.

They are probably not instantaneous -- just appears so when the curve is shrunk on a finite graph.

The way to view it is to look at the periodic error curve as a Fourier Series. For good mounts, terms after the third harmonics are very small. For worse mounts, you may see large 5th, and even 7th harmonic terms.

In any case, each of these components of the Fourier Series is a continuous function -- and since they are sinusoidal, the sign change occurs when the magnitude of the first derivative of the curve is small. I.e., you don't go from +0.23" to -0.23", you go through +0.2...+0.10, ... +0.8,... +0.1, 0, -0.1, -0.2, ... etc. And the slopes near 0 are easy to guide away.

Mathematically, the second derivative (how fast the first derivative changes) is actually quite small. I.e., no sudden change in slope.

Chen

prastro

For anyone who is trying to get data points for these calculations from the AM5 graphs, I found this very useful - is very precise.

https://automeris.io/WebPlotDigitizer/

fbitto

prastro It just works great. I took the data into Excel and calculated the first derivative. Then I generated an X-Y plot and obtained this:

The blue line is the PE, and the orange one is the first derivative. So, the maximum slope is close to 0,3 arc-sec/sec.

Thanks a lot for the tip.

prastro

fbitto Sure.

Mine is similar. I get a max first derivative of 0.3 as/sec every 430 sec (surprise!), which lasts for about 20 sec.

Interestingly, when I did it manually, I got 0.18-0.2 as/sec, but this is obviously much more accurate.

Can this also be worked into the sub length, if we know, say, that the guiding will be under stress every 430 sec for 20 sec?

Next step is to do it on the actual graph under load (PHD2 LogViewer) and not the unloaded ZWO chart.

ShinyredAM5

fbitto

prastro

Could I ask a favour and grab your excel formula to calculate the first derivative? Have pulled data from my ZWO report, got it in excel and duplicated the graph with data points.

Whats the next step to get derivative values from existing 2 columns of data?

wookash

Kevin_A @w7ay

I actually used WebPlotDigitizer https://automeris.io/ and I got values that concern me.. even 1.91 arsec/sec (@267.34 degree) and 1.64 arcsec /sec (@264.9 degree) as described in fbitto

I plotted all values, extracted to xls and multiplied degrees (x axis) by 240 to get to seconds (I have AM3 mount). Below are some of the worst numbers.. Am I missing something?

does this mean my mount is terrible? should I use 1.91 arsec/sec as a number to provide corrections against?

Here's my chart with PE again:

[EDIT] I've since then amended the alignment of the points in WebPlotDigitizer and I've got now worst number of 0.60 arcsec/sec and -0.75 arcsec/sec. Does the length of the slope matter? e.g. -0.75 slope lasts for 0.36 second and 0.60 lasts for 0.43 second. If I got the math that is..

and here's the complete chart with derivative

biomedchad

doesnt mean a thing if you cant meridian flip using nina and ascom

prastro

biomedchad Are you saying that AM5 can't do an autoflip with NINA / ASCOM? If so, thats not correct.

fbitto

ShinyredAM5 You have to divide two consecutive values of the error (y axis) by two consecutive values of the time (x axis).

But the time axis is calculated from the displacement in degrees. To do that you'll have to multiply all the values in degrees by 239,3444 (430,82 sec per cycle divided by 1,8 degrees per cycle)

In cell E3 the formula is =+(D3-D2)/(C3-C2), and so on.

Then you can just look at the values in column E or use the formulas =MAX(E:E) and MIN(E:E) to obtain the maximun and minimum slope values

raawe

@StarzLite forgot to ask you what kind of setup are you running? Main/guide scope mainly.

ShinyredAM5

fbitto

Thank you for the formula, just finished off the excel graph and its interesting stuff.

Seems my mount spikes at 0.3 arc secs / sec worst case but mainly sits around the ₀.2 level.

tempus

fbitto In cell E3 the formula is =+(D3-D2)/(C3-C2), and so on.

What about if you are on line D15 or D33? Is there a generic formula that can be used to cover all D values? I am an Excel dummy.

Chathamnights

Going to chime in and thank Chen and a lot of other people I've gleaned info from the past two weeks. First night out on the AM5 a couple of weeks ago I was averaging .87 RMS total and was completely amazed by that. Coming from a GEM28 I thought I would never regularly see >1"RMS. But after reading/applying some of the recommendations and tweaking the RA/Dec aggression (40/40, I know going lower would lower RMS even more so looking forward to that test), I was able to stay at .59" RMS total last night for over 3 hours. Equip is as follows: AT80EDT main scope, 60mm guidescope, 533MC main, 290mm guide, ASIAIR Plus and AM5. I set the Calibration step to 900ms (per PHD2 recommendation) and RA Dec duration to 200ms with a .5s capture.

When it's a bit warmer I'll do some more tests but I'll take .59 any night of the week. Question on binning. I've seen bin2 recommended before on guiding with my GEM. Would it help or hinder in this case? Apologies in advance if that's been addressed above

raawe

Finally some nice weather so I put those new numbers to the test. Was really conservative with the settings for the first run. Results are amazing. Max RA and Max DEC was set to 250ms and aggression to 35/45%, but I feel this can go even lower. Total RMS was dancing around 0.3" and was maintained for the duration of my imaging session (_3h).

I was using Askar FRA500 at native f/5.6. Off-axis guiding with ASI 290MM mini and main cam is ASI2600 MM Pro.

A big thank you to everyone's suggestions.

ShinyredAM5

tempus

If you start the formula at the same place then you can copy and paste, important thing is that the formula will reference the same cells.

If for any reason you are not using exactly the same cells to start - then you can delete the 'D3-D2' and click each cell you want to use instead, likewise for 'C3-C2'. Important thing is that your formula references the same pattern of cells as given in the example earlier in this thread.

If you don't fancy the excel method, Chen's method is accurate enough for sure.

fbitto

If you don't mind using GNU Octave (a free alternative to MATLAB), you can use a script similar to this:

`clear

format long g

x = csvread('C:/Users/XXXXXXX/Downloads/Default_Dataset.csv');

#Change the following values according to your guide settings and AM5 PE chart
guide_rate = 0.5 # 0.5X, 0.9X, etc.
PHD2_sec_per_frame = 2 ;
Xo = 129.60 #degrees at left side of chart (not needed really because we are interested in differences)

deg = x(:,1);
err_asec = x(:,2);

t = (deg-Xo)*(430.82/1.8); #convert degrees to seconds of time

dt = diff(t);
de = diff(err_asec);

deri = rdivide(de,dt); #divide each element of the de matrix by the corresponding element in the dt matrix

max_slope_pos = max(deri);
max_slope_neg = min(deri);

max_slope = max(abs(max_slope_pos),abs(max_slope_neg));

asec_guide_rate = 15.0 * guide_rate / PHD2_sec_per_frame

max_AR_pulse_ms = max_slope/asec_guide_rate * 1000;`

Where the Default_Dataset.csv file has the data obtained from WebPlotDigitizer.

erossi

I have been following this thread with great interest and would like to express my gratitude to all of you for sharing your instruments and knowledge with us common folks. Thank you! I attempted to apply the workflow to my AM5 and obtained a suggested RA pulse of roughly 78 ms, but I am not entirely confident that I did the homework right... @w7ay does it make sense?

This is the PE of my unit (max/min 24.1/12.0 arcsec):

I also used WebPlotDigitizer to extract the data points from the graph and fed them to the Octave script wrote by fbitto to confirm the 78 ms figure I got on paper.

kaiyung

@w7ay Chen, thank you for analyzing and prescribing the method to improve guiding for these Strained Wave based mounts. Your suggestion to limit the length of the each guide pulse makes sense because of the extreme short and frequent guide corrections.

I can add a few tips to help determine each mount's PE slope by using the app PHD2 log viewer. Based on my mount's performance, I deduced that

1) you cannot trust ZWO's PE curve shipped with each mount. Mine was said to have 24.6/13.1 arc-sec max/min, but the actual measured data for my recent guiding session points to a max ₄₀ arc-sec. Yes, this max error is really bad, but at least the higher harmonics are relatively tamed.

2) the period of the drive error is around 400 seconds. It is quite consistent that I believe we can leverage PHD2's predictive algorithm to proactively reduce these errors.

3) this app is really easy to use and has an intuitive GUI for zoom in and out to measure the slope of the error.

Kai

w7ay

kaiyung you cannot trust ZWO's PE curve shipped with each mount.

If you notice, ZWO's expanded scale ("Partial Zoom") is supposed to be taken where the periodic error is largest. It is not very useful, since they should be showing where the slope of the periodic error is the largest, not where PE itself is largest.

I definitely trust the raw numbers from PHD2 (p-p periodic error of over 40" while showing significant 5th harmonics at places) more than that piece of paper that came with the mount.

By the way, for those who are using ASIAIR instead of the full PHD2, you will notice that you cannot get the "RA Corrections removed" plots that @kaiyung is showing. However, you can get close (how I do drift alignment with the ASIAIR) by setting the Max RA Pulse duration to 1 ms. ASIAIR does not let you set the max pulse duration to 0 (yet another of these brain damaged "easy as 1,2,3" simplificatons in ASIAIR), but you can set it to 1 ms. 1 ms correction at 0.5x sidereal rate comes up to less that .01 arcsec of correction, which is much less than the slope from the mount itself, so you can still get come meaningful results from it.

Chen

tempus

kaiyung 3) this app is really easy to use and has an intuitive GUI

How did you get to the "Drift corrected" and "Frequency Analysis". I looked around and could not get those graphs generated in phd2 log viewer.

Also did you issue correction during that period or you just let the mount track by turning off the guide function in phd2?

« Previous Page Next Page »