AUDINE CCD-Camera Project   

After working a long time with the Starlight SX CCD-Camera I decided to get a new camera

with a KAF 400 chip. If you see the following specs for the camera, you can imagine that its

not easy to get such a camera on the market. So I decided to build the AUDINE. 

Specs:

KAF 401e chip
Car battery supply
Fully portable system (Car observatory)
Water cooled 2-stage peltier cooling system.
-35°C chip temperature at 25°C ambient
Water circulator with car heat radiator and 3 PC CPU-fans.


End of 1999, both my friend Gerhard Neumann and I began building our 
own AUDINE cameras

with the KAF 400 sensor.  We've developed a very efficient peltier cooling system. At an 

ambient temperature of 25C we reached -40C at the surface of the CCD-chip. Our estimate is 

that we always can work at -35C with the final camera system. For the maximum cooling the

system needs approximately 2 Amperes at 12 Volts. 

 

Development of the camera

 


 
              

The camera is built up from an Al-frame (88X88mm, with 33mm height) with a thickness of 2mm.    

 

 

a frontplate which carries the M42 Adapter ring and window (still not in). 

 

 

A small Al-frame at the backside which is used for sealing with an O-ring and in which the

4 mounting screws are glued to fix the backplate. 

 

 

The backplate (now all parts are black anodized) with the copper water circulator and the

electronics (not included in this picture!) All parts (without  the backplate) are hold together with

a 2 component epoxy. The backplate with cooling unit, electronics and chip can be removed 

easily from the camera case to clean the entrance window and chip surface from time to time. 

To make the electronic tight I removed the 2 connectors from the original lower board and 

exchanged them with better ones (solid pins). Additionally I made the connectors tight with 

silicon glue. 

 

 

Camera view from the front side after anodizing. The complete camera now has a measure of 

88X88X42mm + 14mm M42 adapter ring(hight). The weight is 450gramme.

 

 

The dual stage peltier cooler is built from a small Cu (electrolytical copper!!) plate with some 

hollow in it as you can see on the picture. This is closed (soldered) with another Cu plate (1mm

 thickness) and 2 water inlets. (see picture backpla1. ) 

 

 

Two Melcor elements CP1.0-127-05L (30X30mm) and CP1.4-35-10L (15X30mm) are used for the 

dual stage peltier cooling system. We also made very good experience with elements from

Peltron GmbH in Germany ( http://www.peltier.de ) Here we used the elements PKE 128 A 1027

and PKE 36 A 0020. They are in series and optimal for the use with a 12V car battery. 

The max. power consumption is approx. 24Watt 

 

 

A copper spacer which transports the heat from the upper element very effective to the lower 

element is glued directly between the both peltier elements. As glue we used a special thermal 

conductive silicon NEE-001 from Dr. Neumann ( http://www.dr.neumann-peltier.de ).

The lower element is glued directly to the copper heat exchanger. 

 

 

The chip (KAF401e) is glued directly on the surface of the upper element with silicon. In my 

system the chip is positioned between the 2 boards. I connect it with thin wires (0.1mm) to the 

upper board. The complete peltier cascade + chip will be isolated with foam. This is the best 

way not to loose too much energy to the interior of the case. This peltier cooling device is very

effective. 

 

 

Here you can see the DACO shutter in one edge of the case (upper left corner). In each edge I

have 10X10X22mm space for some components. To get the shutter blade to the middle of the 

case, I extended it with a needle of a siringe. In the edge where the wires go to is the 

electronics for the shutter control. 

 

 

This is only a PhotoMOS, resistor and diode as you can see in this diagram and was made from

Christian Kuhn, a member of the German AUDINE group. 

 

 

This is the front side of the water cooling device (size 20X20X30cm) which is made from a 

wooden case, a car heating radiator, a small pump and 3 CPU fans. The camera is allways 

connected to the water tubes and transported together with the cooling device. The handling

isn't more complicated than the original camera-with-fan design!

 

 

 

Cooler with backside opened. The chamber is divided into 3 parts by cardboard walls to make 

the fans work better (no turbulence). In the upper left corner you can see the pump. 

Data of the cooling device: 

3 fans 12V/100mA each 
1pump 12V/150mA 
Total energy consumption is only 5.4Watt! 
The heat conductivity value is Rth=0.05K/W 


The low heat conductivity value means that the water temperature is only 1 degree C above the 

ambient temperature if you have an energy input of 20W (peltier cascade) to the system. My 

experience is that with that cooling system I reach a chip temperature which is 6-7degrees C 

below a fan cooling system. That means half noise and double max. exposure time. There are 

two other advantages with a water cooled camera: 


1.no vibrations 
2.system can work inside a telescope tube 

 

 

Power Supply

Peltier supply from 12V car battery



This AUDINE system was optimized for a car battery. We are using the car battery for the peltier

elements, the thermometer and the shutter. We also tried to make the +/-15V from car battery, 

but this is not working till now. 

We did some tests with different peltier setups to get the optimum solution with respect to low 

power consumption. This is a dual stage cooler with the Melcor (or Peltron) elements as 

described in my evolution story. At 11-12V those elements are working at 80% of their dynamic

range which is an optimum. Than you only need 2Ampere which is good enough for the small 

pins of the normal 15pin connector. When you use an CP1.4-71-10L instead of the CP1.0-127-05L

the max. Voltage is only 6.8V at more than 3Ampere. So you give away 5V from the car battery!! 

We did not make tests with a 1stage peltier cooling system, because it is not
possible to reach 

-35C with that solution at 20C ambient, which was our goal. But I can imagine, that it is better to

use in this case a peltier element which is build up from more elements, f.e. CP1.0-63-05L 

instead of CP1.4-35-10L.

 

 

The supply is very easy. We used a Low-Drop 1-30V regulator LT1085CT (3A) or LT1084CP (5A) 

with a small (5X10cm) heat sink. With resistor R2 it is possible to get a voltage from 1.5-11.5V. 

This is a good range to get the final chip temperature at different ambient temperatures. At the

moment we do not use an active regulation combined with the chip thermometer, because we 

do not want to make Photometry. As the ambient temperature is not variating so much during 

night, our solution is good enough and holds the temperature within a range from +/-0.1degree 

(regulated by hand!).

If we find a good solution for active regulation in future, it is planned to build that in our supply.

The problem with those regulations are the oscillations.

 

 


The +/- 15V Supply

Our first try was to use also the car battery in combination with a DC/DC converter. (12V---+/-15V) 

Those converters are working with an 20-100kHz oscillator. This is the reason why they have a 

ripple from 40-100mV which is too high for the AUDINE (not more than 5mV!) Till now we found 

no filter system to reduce this ripple. So we decided to use batteries. From 30 NiMH accus 

(1050mA) we built 2 X 18V accus which are connected to a classical +/-15V supply.

 

Many thanks to Blair Batty from the Canadian

AUDINE group, who helped me to create this page.