Optical comparator -- Micro-Vu 400 "front" lamp question

[Torch]

I picked up a used benchtop optical comparator for cheap -- nothing fancy, no electronics or corrective optics, but the price was right. Now I'm starting to wonder if I'm missing something, or if it is. The "front" lamp is actually at the side, 90° to the lens and doesn't really do much over there. In a dark room, it kind of produces an image, but nothing you'd really want to work with. Seems to me that the light would be better positioned beside the lens, so it actually lights up the part of the part that you want to see. (note: the rear light works great and the shadow is clearly visible in normal lighting.) Is anyone familiar with this model?

The other thing I'm wondering about is work holding. It came with two little combination centers and v-blocks. Great for examining screw threads, not so hot for looking at gear teeth. I was thinking a small -- as in tiny -- vice might be in order. I was thinking about making something like a screw clamp with a single mounting bolt (so it can pivot). Any other suggestions? (the table has a single narrow -- as in tiny -- t-slot down the centre. It would probably accomodate about a #10 machine screw with a square nut. )

[GlennW]

Hello Torch,

I have a Micro-Vu as well. (500HP)

The "front lamp" is actually a "surface" lamp and when used will project an image of the surface of the part on the screen. The rear lamp is the "profile" lamp.

If you were to place a block with blind holes in it on the "stage"(table), the image projected on the screen would look like a faint picture of the block so you could measure the hole location from each edge. I found that by using both lamps you get a clear profile of the part as well as a view of the surface of the part.

Another example might be viewing a part with lettering or numbers engraved on it. By using the surface illumination you will see the engraving on the screen in order to measure locaton or character height. It is a fairly faint image, but very accurate. I looked at a QCTP tool holder the first time I tried surface illumination and you could clearly see the grind marks on the dovetails on the screen.

Most of the time I just place the part directly on the stage. You need to place threaded parts on the stage so they are corrected for the lead angle, so using the centers/v-blocks is not always a good idea for looking at threads.

Surface illumination is best used with the 10X lens.

Be careful if you clean the mirror as the reflective coating is on the front of the mirror, not the back as on most mirrors. It will be very easily scratched or damaged. Using methanol and cotton balls is the recommended method. There is also another type of alcohol that is used for cleaning glass that will work and is commonly available. My brain is not awake yet so I forget the type, but will add that when I get to the shop.

[GlennW]

Cotton balls soaked in Denatured Alcohol work well for cleaning the mirror. Don't scrub, just start at the top and swab the mirror starting at the top so the dust and dirt flow down and off of the mirror.

Clean the screen using Ivory soap and your hands, no rubbing with cloths or anything but your bare hands, as thos little markings my just disappear!

Clean the lenses with lens cleaning tissue.

Cleaning makes a big difference in the sharpness and clarity of the image.

You should also do a basic calibration as well by placing a .250" gage block or similar item of known dimension on the stage and adjusting the lens in or out of it's socket until the correct screen measurement is achieved.

[Torch]

Cotton balls soaked in Denatured Alcohol work well for cleaning the mirror. Don't scrub, just start at the top and swab the mirror starting at the top so the dust and dirt flow down and off of the mirror.

Clean the screen using Ivory soap and your hands, no rubbing with cloths or anything but your bare hands, as thos little markings my just disappear!

Clean the lenses with lens cleaning tissue.

Cleaning makes a big difference in the sharpness and clarity of the image.

You should also do a basic calibration as well by placing a .250" gage block or similar item of known dimension on the stage and adjusting the lens in or out of it's socket until the correct screen measurement is achieved.

Thanks for all the tips. The screen actually looks pretty good, but the lens and mirror obviously need cleaning so I'll tackle that tomorrow morning after work.

The image from the side lamp could only be seen in a dark room, and was pretty much completely washed out when the rear lamp was turned on. I noticed there are different bulbs in each, so maybe the PO swapped in a bulb that just isn't bright enough. I'll try swapping them and see what happens.

I'm not sure how the lens moves. I gave it a bit of a tug and a bit of a twist without movement, but ran out of time and didn't really look too hard beyond that. I'm pretty sure it must come out somehow because I understand there were at least 2 other strengths available (this one is a 20x).

[GlennW]

Should be this one...

http://www.bulbworks.com/light-bulbs/ELS

There are cheaper ones out there, but that shows the info. I recently had a lamp quit in my outdoor lighting around my patio. Pretty sure it's the same lamp and available at home improvement stores with the low voltage outdoor lighting.

There should be a thumb screw about 10:00 to secure the lens. Possibly someone replaced it with a set screw.

[Torch]

Should be this one...

http://www.bulbworks.com/light-bulbs/ELS

Cool! The description even specifically references the Micro-Vu 400 Detail. Can't get anymore exact that that, can you?

Unfortunately, mine looks more like this:
http://www.bulbtown.com/1004_MINIATURE_ ... p/1004.htm

You've got a headlight and I've got a taillight. Mine has no reflector, is a standard incandescent rather than a halogen and I'm pretty sure there's no way a bayonet base bulb is going to pull 3 amps. I wonder if that's got something to do with my dim image?

Looks like the PO changed more than just the bulb. Someone swapped in a completely different base too!

[GlennW]

Check with Micro-Vu as there was a mod which incorporated a socket/lamp change. I phoned them just after I bought mine as the surface lamp was dead. They asked which lamp it took as there was a change.

[Torch]

Check with Micro-Vu as there was a mod which incorporated a socket/lamp change. I phoned them just after I bought mine as the surface lamp was dead. They asked which lamp it took as there was a change.

Ok, I sent them an e-mail with the serial number. We'll see what they say.

[Torch]

Well, they replied quite promptly:

"We no longer have user guides for the model 400,
We stopped making the model 400 26 years ago.
By front, I assume you mean the profile lamp?
My records show we used a 1724 lamp for the profile.
See attached picture.

The 1724 lamp was discontinued around 10 years ago,
I think you can still get lamps from some places online,
I found this from a google search, but I have not used this service,
and we do not use the 1724 on any current machines.
http://www.bulbworks.com/light-bulbs/1724"

So, if I understand correctly, he's looked up the wrong bulb, but that one looks like the same base that is in the front/detail/side-mounted one so I'm guessing that I have the older version. Maybe I can retrofit it to something with a reflector.

[Torch]

He got back to me again, but indicated the same ELS bulb that Glenn mentioned. My comparator doesn't have the correct socket for that, the voltage supplied is only about 6v and the arrangement looks factory. So I guess I'll have to retrofit something.

I did play around with it a bit, using my trusty Lunasix F to measure light outputs. The profile lamp produces about 500 foot-candles at the focal point. The surface light that's in there produces 2,000 at the same point, after going through the lens. With the cover off, measuring the spill shows the bare bulb produces 500 foot-candles at the same distance, so the lens makes a substantial difference. The socket mount is on some healthy slots, so by loosening the screws I could move the light and focus the beam. I got it down to a fine spot of 10,000 footcandles, and now it produces a somewhat usable image -- still not brilliant, and the spot is smaller than the field of view, but it does confirm that my problem is insufficient light.

The ELS bulb is a 16v, 50 watt halogen with a built-in reflector, and is out of production. So the voltage doesn't work with my version and the bulbs will only be available for a limited time until stocks run out. I'm thinking I would be just as far ahead to retrofit to something current. Measuring the output of a 50 watt PAR 20 halogen flood at the same distance produced 16,000 fc, and those are rated 550 lumens output. So now all I have to do is find something that puts out at least 550 lumens, has a reflector and can run off 6v.

I think I'm going to look into some of these new high-output LEDs the flashlight hobbyists are playing with. I understand that some of those are putting out up to 1000 lumens and the voltages are similar.

[Torch]

The end result:


After a fair bit of research in a flashlight-builder's forum, I ended up going the LED route. These are not your grandfather's LEDs! At these light levels, even an LED produces considerable heat and they have to be heat-sinked much like a computer chip. They also require a little circuit board called a "driver" to regulate the current. I finally settled on an LED emitter called the SST90 made by Luminus Devices and the H6Flex driver from http://www.taskled.com.



The emitter has been soldered to a copper disk, which in turn has been bolted to a 2-piece copper adjustable bracket with a thermal paste between. The driver, which includes a thermal shut-down sensor, is epoxied to the same bracket. I replaced the existing 6v, 3a AC transformer a 12v, 3a filtered DC power supply from an old external hard drive. The maximum driver output can be programmed in 9 steps, from 1.4 amps to 6.6 amps. In the above photo, it is set to 5 amps, producing about 6,500 foot-candles in a 1/2" square at the focal point. This seems to be about the minimum required to bring up the surface details when combined with the profile lamp. And actually, the different colour temperature of the LED compared to the incandescent profile lamp makes for really good contrast that makes the image pop out very nicely.

Detail Off:



Detail On:



I don't think he is on this board, but I should acknowledge that I had a lot of help and advice from a flashlight builder and fellow home-machinist known as Mick at http://www.candlepowerforums.com.

Total cost was in the $100 range. More expensive than a single projector-type bulb, but should last forever.

[GlennW]

Torch,

It looks like it works well!

Your setup on the HP400 is quite different than mine on the HP500. My surface light is actually inside of the body. The light shines from inside, back out through the lens to illuminate the part. The reflection then passes back through the lens and is projected onto the screen.

I didn't realize they were so different as I though the main difference was only the screen size! (10" vs 12")

Great job! (as usual )