Stability of LHires III with new upgrade

[Keith Graham]

Hi,

I am asking this question of anyone who has purchased the new upgrade for the Lhires III.

It is well known that the Lhires has had instability issues that cause shifting of the calibration lines. Recent posts indicate that the instability severity appears to be different for different users. Has anyone noticed any improvement in the stability with the newer calibration unit? I am not sure just where that instability lies, so this upgrade may actually not address that issue. I am asking this because my Alpy600 is quite stable and has a similar calibration unit design with the flat lamp.

Thanks,
Keith Graham

[Robin Leadbeater]

Hi Keith,

I am planning to upgrade my LHIRES at some stage (mainly for the convenience of the flat lamp which will allow flats to be taken remotely and in the same pointing direction as the target which hopefully should be useful for difficult cases of fine scale ripple removal)

I saw an improvement in calibration stability when I automated my LHIRES neon operation but I believe this was more to do with not needing to touch the spectrograph which risked causing small irreversible flexing of the spectrograph rather than any improved repeatability of the lamp operation. The ALPY construction is much more rigid than the LHIRES with no flexure that I could measure.

Cheers
Robin

[Keith Graham]

Hi Robin,

Thanks for the reply. Yeah-I was thinking the shift might also be the caused by weight of the Lhires/camera combo causing flexure as the system tracks or when it is slewed to a different position. As you say, the Alpy is more rigid by nature, but I was interested in knowing if the neon lamp setup in the Lhires might be an issue with the shift.

Cheers,

Keith

[Mike Potter]

Hi Keith:

I did a bunch of testing with the old setup - tests I've yet to repeat with the new. A few comments about my tests of the old setup. Here's a link describing a couple of the tests - the second test being more real-world-like. Here's a link:

http://www.beverlyhillsastronomer.org/s ... ctrograph/

I also did a couple of other tests. In one of them I repeatedly turned the lamp on, rotated the lamp into position, took a neon spectrum, rotated the lamp back out of the light path and turned the lamp off. I waited a few seconds and then repeated. I did this maybe 20 times over the course of about five minutes and found there was no additional movement beyond what I saw just leaving the lamp turned on and taking a series of exposures with the telescope parked (described as test #1 in the link, above.

I also took neon spectra over the course of a few minutes slewing the telescope first to the eastern horizon, then near the zenith, then to the west, back to the zenith, and so on. In that test I found lots of movement.

Now, since doing those tests I found that the doublet lens in my LHiRes was skewed to the light path rather significantly, which might have made my setup more prone to temperature changes. Anyway, what I found was that the LHiRes is definitely sensitive to temperature, but varies rather smoothly; calibration lamp images before and after taking the object spectrum and interpolating the wavelength solutions seemed to significantly lessen the effects of temperature and camera movement for a single "science" exposure, to the point that I've been able to measure radial velocities over several hours with a standard deviation of about 0.3 km/sec. That was with the old setup. There was also something that was causing, at least for a time, a systematic offset of about 3 km/sec. I still have only an un-tested theory as to why that was happening, and it relates to the tilted doublet problem, mentioned above. Note these are all tests I plan to repeat with the new setup, likely this week or the next time we get some clear weather. I'll post the results here.

I can say that, after using the new setup for a few nights, the lines do move by about the same amount as before. I am using the NeAr lamp, and there seems to be some resolvable structure in the lines - not in their width (which does change with y position) but in the strength of the lines - which can, for an individual line, vary by five or more times in intensity. There seems to be a broad "hot spot" at one area along the lines - which perhaps means I'm only seeing a small piece of the actual emission "arc" in the lamp. When first installed I found I had nice evenly-illuminated lines but had to take a 50 second exposure just to get good signal in the strong neon lines. I adjusted the lamp position a bit; it can be moved easily - perhaps too easily as I'm suspecting some movement just from the start/stop of the rotating platform the lamps sit on. But after adjusting all of the lines are MUCH brighter but I do see the aforementioned "hot spot". Maybe I need some sort of translucent diffuser on front of the lamp? I noticed that the old lamp was wrapped in some sort of tape and maybe something similar should be done with the NeAr lamp - though I'm guessing that would absorb too much light and make calibration exposure much longer. Or maybe a reflective screen as is done with the flat lamp?

Mike

[Keith Graham]

Hi Mike,

Thanks so much for the very informative post. I look forward to your update when you get opportunity to do some more testing. It does appear that the lamp update does not improve any shift in the neon lines. I rather expected this would be the case, but it does take some testing as you have done to verify this. If I recall correctly, this line shift has been blamed on the flexure in the Lhires unit, but good testing as you have done when moving the scope from horizon to zenith seems to verify this.

I had never considered any issue with the doublet. Perhaps I need to check my unit to see if this has any adverse effect on results.

Cheers,

Keith

[Robin Leadbeater]

Hi Mike,

I noticed that the old lamp was wrapped in some sort of tape and maybe something similar should be done with the NeAr lamp - though I'm guessing that would absorb too much light and make calibration exposure much longer. Or maybe a reflective screen as is done with the flat lamp?

The manual for the upgrade show a translucent sleeve around the lamp. Does your's not have that?

Robin

[Robin Leadbeater]

I tested a similar configuration to the Shelyak upgrade using the same lamp about 18 months ago. I wrapped a diffuse material round the lamp with a Al foil behind it.
http://www.spectro-aras.com/forum/viewt ... 2380#p2380
It worked well but was just a proof of concept. I planned to try the lamp in place of the Ne lamp at some stage but never got round to it.

Robin

[Mike Potter]

Hi Robin:

Yes, the bulbs look exactly like the picture, but I can actually see the element through the translucent "wrapping". It does a good job of scattering the light along the entire slit, but at least in the position I currently have the bulb there is a section of each emission line that is a lot brighter than the rest of the line - which I suspect is due to the fact the slit is seeing the actual "glow element" through the translucent wrap.

[Keith Graham]

To take this thread one step further, the manual for the new calibration module says it can be operated remotely. I would like to be able to operate this module from my computer, but I have no idea how to accomplish this. Can someone please tell me what I would need to activate/deactivate the neon and flat lamp via computer? I apparently need 2 remote 12v power supplies, but I do not understand how a computer can control them. From what I have seen, some sort of module is also required and can be connected to the computer via USB cable. Is there a source for this module where it can be purchased? Is there software for this purpose? Perhaps someone already has this information on a web site, and this would be extremely helpful.

I do run my scopes, cameras, and focusers remotely using Radmin 3.0. It would be great to be able to run the Lhires remotely as well.

Thanks Much,

Keith Graham

[Paul Luckas]

Hi Keith,

You're right. You need some circuitry between the calibration module and your PC. There are a couple of ways to do this, with Arduino boards becoming the more popular. They are cheap, readily available and provide for the USB interface you need for PC control. Many also allow for adding future 'automation' (e.g., motorised collimator etc.). A single 12V power supply will be sufficient to provide power for the lamps and control for switching between neon and halogen, but with most if not all Arduino boards you'll need a way to convert the onboard 5V control voltage to 12V (required by the new calibration module). This can be achieved relatively easily using a number of means - transistors, relays etc., but you'll need a little bit of electronics knowledge.

I wouldn't be surprised if Shelyak aren't working on something right now - as the new calibration module is only 'half way' towards automation.

If you haven't already seen this, check out Stephane Ubaud's pages on automation. He uses an Arduino (as a few of us have done) to provide the 'brains' and interface.

http://steph.ubaud.free.fr/articles.php?lng=fr&pg=728

I have some less than useful pictures here:

http://jazzistentialism.com/blog/?page_id=1397

The nice thing about the new calibration module, is that you won't need all of that servo malarky. All you need is the Arduino, some 5V - 12V switching components and some time and patience to put it all together.

Once you have the basic Arduino code controlling the calibration module, the sky is the limit. You can create VB code, etc, to really automate things.

Cheers,

Paul