Mitsubishi ML520G71...Red Holy Grail or Flashlight Fail ??

[flyowynd]

I think you're right, I just hope it won't be too paintfull to adjust the rotation since you don't really have access to the holding part. But I'm sure that wont be too much of an issue if you have two small hole in the back of the holding cylinder to insert and manipulate it.

For the 3D I'm not sure you will like the software I use, it's an inexpensive (free) CAD software provided by Emachineshop.
You do your design with it, you can render it in 3D and once you're happy you place an (expensive) order to get your part done.
Unfortunatly you can really assemble multi part in one file.
But that's enough for me just to do my design.

Flyo

[CDBEAM]

LS...I understand your reasoning behind machining a flat spot on the LD rotational position cylinder. That may be something I will do. I was also thinking of using Copper for this cylinder ??? Copper is difficult to machine however.

Flyo...I will likely drill two (2) small dia. index holes about 4mm deep at 180 degree on the backside of the cylinder. These holes could then be used to rotate the cylinder in suti.

See latest rendering: LDACL V10

BEAM

[LaNeK779]

just a silly quetion, in case i am missing something major here ---> "what is wrong with dave's adjustable mount?"

[CDBEAM]

LaNeK779...Review posts # 612 thru # 632. Daves mount will work...I just wanted a design with improved thermal properties and much improved ease of adjustabillity. Adjusting only two (2) screws to acheive optimal X ~ Y will be superior to the other designs. I assure you....this unit will be more expensive than the other designs....material costs alone !! But...in the never ending quest for DIY perfection....this design is improved...IMNSHO !!!

BEAM

[planters]

CDBEAM,
DO we have any reliable data on the thermal properties of these diodes? No offense,as I am frequently guilty of over engineering "just in case". But, these diodes are being held in separate modules with as I see it at worst 4W of electrical input and probably less. If the heat-sink/base-plate runs close to ambient ( this assumption is unaffected by the efficiency only by the quantity of the upstream thermal load) I don't think there will be much of a delta T. This could be tested. The LOC diodes reliably shift their wavelength 0.25-0.28 nm/C. Put one in one of the available mounts, set it on a large Al heat-sink and load it with 1,2,3W and measure the shift.
Also, copper is a bitch. It breaks tools and the chips are nasty. So, here is a thought. Machine the diode holder out of steel (not stainless) and then mount the C. lens in a disc magnet and use small set screws to adjust the disc position in X-Y. The amount of adjustment should be quite small and so the clearances can also be small. As far as rotation, I have found that fully backing off 3 and slightly on the fourth of the diode retention screws on the back of Dave's mount, using only the leads to manipulate the diode, I can quickly adjust and then fix the rotation of a 445nm diode (especially critical alignment necessary) to within 1-2 degrees. I think this should be enough. What do you think?

[CDBEAM]

Hi Eric...Yikes...I do not own or have access to a Spectrum Analyzer....All I know is that there is a de facto drop off in PO on the ML...that I can measure....so...I assume that with some overdriving...thermal transfer would be somewhat a consideration...and I do over engineer... it is a curse !!

As to the magnetic secure concept....INTERESTING....and simple !!! an off shoot of that basic idea would be to inlay three (3) high energy ceramic magnets into the brass collimation position disc....they sell very small ones...about 4mm in dia...which could be epoxyed into the disc...I must think on this approach...Thanx !!

BEAM

[logsquared]

I finally found the temp probes I had made a few years ago. Later this evening I plan on testing the temperature differential from the diode holder to the mount on my centering mounts. Possibly may also test the temp diff from the mount to an aluminum base as well. This will provide a definitive answer to the question... Do we need the extra engineering? The calculations show it should be around 1deg C/W. Should be around 1/3 that with thermal grease. What that means is... if the diode drops 2.5V at .5A we have 1.25 W going in - ~450mW going out as light = .8 W to be dissipated as heat. Multiply that # X 1deg C = .8 deg. If we figure the same resistance for the the mount to the base plate we get 1.6 deg C above the base plate which should be very close to room temp. There is also the drop from the diode holder to the diode itself but I dont have the means to accurattly measure the small diode's temp. However, it can be calculated with good accuracy because we know the ambient temp and the power going to the diode. In summary, the diode should see only a marginal increase from the ambient. If one plans to use their projector in high ambient conditions then a TEC is necessary. In which case the TEC could be sett a few degrees lower to account for the resistance from the base to the diode.

The magnet idea does work. Laserscope used it on their early KTP lasers in the fiber launch mechanism. I have one around here somewhere and can post a pic If anyone wants?

[planters]

Got an idea ( not ignoring your PO drop measurement CDBEAM, I have a thought about this). Put a sacrificial ( can be dead as long as it resists ) 5.6mm diode in the holder and attach the thermistor directly to the diode and run the power. This is the temperature that matters and I suspect the diode interface is where the money is.
CDBEAM,
You don't need a spectrum analyzer if you can limit this to a null test. Send the beam through a dispersion element and note its far field position. If you have a good amount of dispersion seen as angular displacement and then ramp up the power (heat) you can watch the beam move. See if you can get a correlation between the beam movement and power output. If you can return the far field spot AND the power output to baseline at the same input power by cooling the base plate then it IS temp even if you don't know what that temp is.

[logsquared]

Got an idea ( not ignoring your PO drop measurement CDBEAM, I have a thought about this). Put a sacrificial ( can be dead as long as it resists ) 5.6mm diode in the holder and attach the thermistor directly to the diode and run the power. This is the temperature that matters and I suspect the diode interface is where the money is.

Good call! That much simplifies the test. Find the total differential from diode to base. Will report back later.... Busy milling parts

[flyowynd]

I though the wavelenght shift was not so important on this diodes but the issue cames from a quick output power drop when the temperature raise a bit.
I remember someone doing the test puttin the holder in the freezer and just after doing a test measurement at different power finishing by overdriving the diode.
The result was as soon as the temperature raise over ambiant the output power really decrease a lot.

That may not be a problem with a normal mount and a way to keep the diode at reasonable temperature (even if overdriving it) using a peltier but what will be the outcome when using the laser during summer at 30°C?

My own experience from last week told me for example that running an ouside show last week when the temperature is close to 0°C give you a wonderfull red....but a very little green. (Never seen my 650nm red so bright before)