How to build an image intensifier?
Bill Reuber
Well Peter if you like a challenge you could give this one a shot,,
the simplest would be a simple diode system(with no luminous gain
to speak of) and you would need to construct a photocathode
from appropriate materials so as to have good efficiency at wavelenght
you desire,, then you will have to deposit the material on the end of the
fibre or whatever that will be inside the intensifier, this must be done
in a very uniform manner , and the equipment and teqnique for doing it
are almost a BLACK ART,, some textbooks will cover it but equipment
for doing it will cost MEGA BUCKS..
Then you will have to design an Electrostatic focusing lens system
to allow the liberated electrons to impinge on a phosphor screen at
the other end of the device (image will be inverted).
with appropriate electrical fields from the power supply(15 kv typical)
you then must evacuate the envelope that will hold all this stuff
together, seal it somehow and test the unit ,, if you did well the device
will produce an image,, you could cascade 2 or 3 of these together
and get some GAIN from the system and you would have a GEN 1 starlight
device,,
From there you could try and produce what the Military spent millions
to develop,, 2nD generation Micro Channel Plate systems,,
they were produced to solve washout of the image in localized bright
spots in the image (high current in the tube)
They are Proximity devices so you will not need to worry about the
electrostatic lens,,If you ever look at a micro channel plate under
a microscope I Don,t think you will want to try and construct this
in your home workshop,, even at full list prices it would be cheaper
to buy the device!!
we will not even talk about 3rd gen devices with specialized purpose
photocathodes and small pitch MCP design,
If you decide to give it a go, let me know,, I had similar thoughts
about 7 years ago.
Lou Boyd
Peter wrote:
>
> I just wonder how difficult this is. AFAIK it is just a tube and a HV
> (very low current) power supply, plus some lenses to give focus at
> infinity at one end, and focus on the eyeball (about 5mm dia) at the
> other.
If you can buy a good used surplus military 18mm gen II or gen III
tube it's pretty simple since the tubes have a built in power supply.
You need a fast lens in front. A nice objective lens I know works well
is the F/0.95 25mm from Universal Kogaku America. (about $125)
without focus or iris, or about $230 with focus and iris.
An F/1.2 or F1.4 50mm 35mm camera lens works well too if you want
a little more magnification. I've found that a 25mm Kellner
eyepiece used for astronomical telescopes makes a pretty good
eyepiece. Some binocular eyepieces work pretty well too. The
tubes need about 3 volts to power them. A single lithium cell
or two AA or AAA alkaline cells work nicely. Most of the
military tubes provide image inversion within the tube so you
don't need a separate erector. The hardest part is making a
suitable housing to keep everything aligned and provide focusing.
The wiring consists only of attaching two wires to the battery
with a switch in series.
Lou Boyd
Fairborn Observatory.
Bill Reuber
If you have priced them from the people that make them you
may get quite a shock,(as I did) they expect to get all the r and d money
back when they sell these things,,
when I started playing with them about 7 years ago i went to
intevac, litton, varo ,philips and they wanted several thousand dollars
for a basic gen 2 tube, over the years I located some sources for surplus
tubes but many times got burned buying lots of tubes THAT DID NOT WORK!
I built, designed and repaired the systems and even considered building
the intensifiers but decided against it due to high startup costs..
so thats a bit of my story ,, If you are interested in experimenting
I would be glad to assist you if I can.
Bill
Morris the Cat
|when I started playing with them about 7 years ago i went to
|intevac, litton, varo ,philips and they wanted several thousand dollars
I've heard that a lot of early Intevac Gen 3 tubes somehow made themselves
onto the surplus market; allegedly, they had seal problems which led to
a short MTBF.
|for a basic gen 2 tube, over the years I located some sources for surplus
|tubes but many times got burned buying lots of tubes THAT DID NOT WORK!
I've heard that a lot of forged papers alleging certain levels of performance
from IITs are accompanying some NV device products...
|I built, designed and repaired the systems and even considered building
|the intensifiers but decided against it due to high startup costs..
Sometimes even IIT manufacturers loose the ability to make tubes properly;
Ni-Tec used to manufacture 25mm Gen II tubes in Skokie, IL years ago. According
to one manufacturing engineer that worked there, they eventually were unable
to produce tubes. I've read that the company moved down to Garland, Texas
but don't know what happened to them. Did Varo (now Litton) buy them up?
It must be hard to make image tubes because few firms do it. I recall just
a few years ago the U.S. government successfully suing a tube manufacturing
firm that had submitted some other firm's image tube as an exemplar of the
firm's manufacturing capabilities. When the time came for the firm to
produce these IR imaging tubes, they were unable to ship working devices to
the government.
Gary A Pajer
Morris the Cat wrote:
>
> |when I started playing with them about 7 years ago i went to
[snip snip]
> It must be hard to make image tubes because few firms do it. I recall just
I used to be in the business of making similar devices. Whenever a new
manager came on board, we would describe the products to him (never a
"her". sigh). Invariably, the response would be "That's all?? So
what's the problem?". We'd all just look at each other around the
table...
--
-------------------------------------------------------------------
Gary Pajer gary_...@unison9.com
Unison Fiber Optic Lighting Systems http://www.unison9.com
Route 413 and Route 13 phone: 215 785-8552
Bristol PA 19007 fax: 215 781-0624
G.F. (Jeff) Gwilliam
In article <6lm72a$p...@ssbunews.ih.lucent.com>, Morris the Cat wrote:
>|when I started playing with them about 7 years ago i went to
>
>I've heard that a lot of early Intevac Gen 3 tubes somehow made themselves
>onto the surplus market; allegedly, they had seal problems which led to
>a short MTBF.
>
>|for a basic gen 2 tube, over the years I located some sources for surplus
>|tubes but many times got burned buying lots of tubes THAT DID NOT WORK!
>
>I've heard that a lot of forged papers alleging certain levels of performance
>from IITs are accompanying some NV device products...
>
>|I built, designed and repaired the systems and even considered building
>|the intensifiers but decided against it due to high startup costs..
>
>Sometimes even IIT manufacturers loose the ability to make tubes properly;
>Ni-Tec used to manufacture 25mm Gen II tubes in Skokie, IL years ago. According
>to one manufacturing engineer that worked there, they eventually were unable
>to produce tubes. I've read that the company moved down to Garland, Texas
>but don't know what happened to them. Did Varo (now Litton) buy them up?
>
>firm that had submitted some other firm's image tube as an exemplar of the
>firm's manufacturing capabilities. When the time came for the firm to
>produce these IR imaging tubes, they were unable to ship working devices to
>the government.
>
>|so thats a bit of my story ,, If you are interested in experimenting
>|I would be glad to assist you if I can.
>|Bill
>
>
>
I used to work at Ni-Tec (twice, actually, once '74-'76, then again
from '82-'84 after an intervening stint at Varian (later Intevac). I
can say from personal experience that making image intensifier tubes
borders on art. I was involved in some work which helped make it less
so (automation of the photocathode deposition process for Gen II
tubes), but there are still so many details to be watched and so
little analytical capability for watching the details that it is very
easy to lose the art. Management, engineering, and manufacturing
personnel turnover can result in major yield busts. In some ways, it
is similar to the semiconductor business, but with much less total
investment (since there is not much of a market for the product.)
Ni-Tec did move to Garland, to merge with a sister company who made
the optics for night vision systems (Optic Electronic Corp). I was
working on Gen III tubes at the time, and none of the Gen III team
made the move to Texas. Neither did most of the engineering and
manufacturing staffs -- maybe 5-10% of the people went. I know there
were major problems with transferring the Gen II production line and
with developing the Gen III line after the move. Other problems
emerged later.
I had a lot of fun working on NV tubes, but it's not a business I
would think about going in to -- if you're already in it, you're
probably thinking about getting out. There is an outfit in LaFox,
Illinois called Richardson Electronics -- they are in the business of
buying up manufacturing rights and equipment for old tube product
lines, and continuing to make the tubes. I imagine that many of the
older image tube lines will end up there sooner or later -- I think
some already have.
Good luck.
Jeff Gwilliam
Peter
In sci.electronics.components Peter <z...@ds98.com> wrote:
: I just wonder how difficult this is. AFAIK it is just a tube and a HV
: (very low current) power supply, plus some lenses to give focus at
: infinity at one end, and focus on the eyeball (about 5mm dia) at the
: other.
one small detail: not focus on the eyeball, instead, parallel exit (focus
@infinity).
Why on earth you'd want to do that, I don't know, as yachting mags and other
places advertise reasonable units for under $300.
Anyway, I know that gen 1 & 2 tubes are sensitive to daylight, and must be
mothered considerably with covers, shutters and assorted games to prevent
them from being blinded. Gen 3 are not sensitive but they should not be used
to test powerful lights ;).
and last: a low-cost B/W ccd camera often does better than a cheap image
intensifier in my experience. You may have to add some noise reduction
circuit but that's not so hard. I have a $100 ccd that can shoot very reason-
able pictures with 0.3 lux or so.
(another) Peter
: Peter.
: Return address is invalid to help stop junk mail.
: E-mail replies to zX...@digiYserve.com but
: remove the X and the Y.
Christopher
What we discovered is that when you buy a surplus tube, it's usually the
power supply that is shot. The tube lights up fine when you put a new power
supply. This led to the current problem in the industry: the power supplies
die faster than the tubes, leaving an abundance of unpotted tubes. I spoke
to the original MX9916 power supply manufacturing company some time ago and
found that if I purchased 100 power supplies at a time, they would sell them
to me for $350. Of course, I can still buy complete surplus tubes for
about $200. It might make sense at some point, but more tubes are making
their way out through the surplus channels, so it might not.
Chris
Sam Goldwasser
In article <358330c4...@news.netcomuk.co.uk> z1...@nospam23.com (Peter) writes:
> You can build an unregulated 15kV PSU very cheaply, $50 I reckon. I
> used to be in the HV business. It is an oscillator (50-100kHz), a
> ferrite transformer outputting perhaps 1-3kV, and a diode-capacitor
> voltage multiplier. Even putting feedback round it is easy. The power
> required is extremely low; I think much less than 1W.>
$5 or less if you are a scrounger. See the circuits at:
http://www.repairfaq.org/REPAIR/F_samschem.html
and
http://www.repairfaq.org/REPAIR/F_gadget.html
and the Laser FAQ (HeNe laser power supply design).
--- sam : Sci.Electronics.Repair FAQ: http://www.repairfaq.org/REPAIR/
Latest Sam stuff: http://www.repairfaq.org/sam/
Lasers: http://www.repairfaq.org/sam/lasersam.htm
http://www.misty.com/~don/lasersam.html
Bill Reuber
this type of supply is great for a simple gen-1 tube BUT if you are goung
to power a gen 2 MCP tube, things will get a little more complex
you will nees -1200 for the photocathode, a system to monitor the current
so that as brightness increaces ,gain may be reduced,
they do this by reducing the bias on the MCP,, it operates from
600 to 800 volts DC,
you will have to work out the design of this cirquit AND have a limiting
setting beyond which screen current will not increase,,
THEN you must generate about 6000 volts for the phosphor screen,,
it is actually a little more than that but you will get the idea ...
That is why they want 350.00 for these things,,(gen 3 supplies are
about 600.00 ea,,) oh yes on top of all the above it must be small enough
to WRAP around the intensifier itself..about 1/4 x1/2 inch x2 inch
Bill reuber,,
Bill Reuber
If you can live with a fixed potential on each stage, and as you say
"remote operation of the supply" it would be no big problem to come up
with a working system,,
I was mainly thinking of attempts to duplicate the ORIGINAL supplies
in both form and function.
If you need any tubes ,(18 mm gen2 ) with MCP, i have a few cheapies
with some blemishes,, Ideal for experimenting!
Bill Reuber,,
James Zuzelski
Thanks,
James
Phil Hobbs
ago there were some Russian ones on eBay, but they were complete crap.
Cheers
Phil Hobbs
--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510
http://electrooptical.net
http://hobbs-eo.com
Henry Nebrensky
> James Zuzelski wrote:
...
> > gone through to get these? Know of any still around today?
> >
> > Thanks, James
> >
> There aren't that many surplus image tubes left, it seems. A few years
> ago there were some Russian ones on eBay, but they were complete crap.
Thanks
Henry
Phil Hobbs
sCMOS cameras do not grow on trees.
The EMCCD is a really puzzling case. You can go from starlight to
sunlight with a twist of a knob, with a signal-to-noise penalty of no
more than 3 dB,(*) and the sensor is built on a fairly bog-standard CCD
process--beautiful, right? But they never caught on and the price never
came down to anything vaguely reasonable.
I suspect that part of the issue is that any damage due to inelastic
collisions between hot carriers and Si atoms gets concentrated in the
very small volume of the last dozen or so multiplication stages. I've
never seen anything published about that, so it may be a mirage.
Cheers
Phil Hobbs
(*) The math behind that 3 dB number is actually really pretty.