[time-nuts] HR2000 Retuning

Peter Gottlieb nerd at verizon.net
Tue Jan 24 00:55:10 UTC 2012


My extreme curiosity eventually led me to acquire a couple ofHR2000 
spectrometers from Roland.  He still has several left, and if you ever want to 
have the chance to grab one of these, I recommend you act fast.

The units he has were from a medical diagnostic device and have a grating which 
gives about 200 nm of range.  The internal sensor is specified to operate over 
the range of 200 to 1100 nm.  My units were adjusted to work over the range of 
about 467 to 671 nm.  Should you need it, Ocean Optics will gladly (for IIRC 
around $500) put a different grating in, or for somewhat less, change the range 
of the existing grating.

I needed to be able to cover the wavelength of a Blu-ray laser at around 405 nm 
so decided to attempt a recalibration.  I was able to download the manual as 
well as their OOIBase32 software which talks to the spectrometer over USB.  I 
got a couple of USB adapters so I could plug the units directly into the 
computer's standard ports.

On the side of the unit you will notice a circular slotted pattern with a center 
screw slot and two screws 180 degrees apart on the circle.  Those outer screws 
have lockwashers and each had a dab of glue.  From the manual it was clear that 
this is the holder of the grating.  I loosened the outer screws, and while the 
spectrometer's input fiber was pointed at the 405 nm laser spot, I slowly turned 
the center slot with a screwdriver.  It didn't take much turning (CCW) before I 
saw a strong response peak.  I moved it to sufficiently within the range and 
tightened the two outer screws.

Now I turned to Appendix A in the HR2000 manual.  I had to locate a Hg lamp (I 
was able to find an antique scientific unit on feebay) as you will need several 
strong spectral lines within your range for the cal procedure.  I actually 
cheated slightly from the procedure, I made an educated guesstimate of the short 
wavelength being seen and loaded that into the unit with the USBprogrammer 
utility.  That made it a little easier for me to identify the Hg lines.

Read the procedure all the way through and understand it before trying it.  Less 
hair pulled out that way.  I followed the instructions, did the linear 
regression as they specified in Excel (yes, an old version will work just fine), 
and loaded the correction factors into the unit's EEPROM (again, using 
USBprogrammer).

I then plugged both my units in at the same time, and if you get SprectaSuite 
from Roland (he usually sends it on a CD) and you have multiple units, you will 
see both of their responses superimposed.  I pointed both fibers at the HG lamp 
(no, I didn't look into an unfiltered lamp!) and where the units ranges 
overlapped, the spectral lines coincided precisely.  I could have gotten a 
larger range, but for now I have two units with a combined range of about 370 to 
670 nm.  I'm just not sure which end I want to expand another 100 nm.

The procedure overall is not that difficult, and I got it right on the first 
try, but you MUST have a source of a multitude of spectral lines!  The good part 
is that you don't have to open the unit up, so no chance of getting any dust 
inside.

One note:  theu nits have a mulitpin connector on one of the edges.  I found 
that ambient light was getting in and raising the overall "dark" floor.  The 
cure was simply to place a piece of black vinyl electrician's tape over the 
connector, flush with the case.

Any questions, feel free to contact me off-list.

Peter

P.S., John, please forward to the microscopes group.  I joined the Tekscopes 
group and ended up with a mountain of scopes, and the time-nuts group and now 
have a pile of time measurement and standards stuff, so no way I'm joining the 
microscopes group.  I have just one, and it's crappy but works, and I'll just 
stop right there while I can still afford rent.



On 1/23/2012 10:32 AM, J. Forster wrote:
> I've been bombarded with emails about retuning the spectrometers. I'll get
> the info and repost it in the near future, unless Peter would prefer to
> put it up himself.
>
> Best,
>
> -John
>
> ==============
>
>
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>
>
> -----
> No virus found in this message.
> Checked by AVG - www.avg.com
> Version: 10.0.1416 / Virus Database: 2109/4760 - Release Date: 01/22/12
>
>



More information about the time-nuts mailing list