14. Diode Bridge Part Four

Part Three talked already about faster transistors and diodes, in this part I use them.


Removing the Current Mirror and the Switching Transistors.



Middle - Replacing the old transistors with other types.  You can't see them, they are buried under copper. Replacing in such a way is the fastest prototyping, here a small PCB would be of course a better solution, but this is a later step.




Top Trace 5V/DIV Load DAC Oscilloscope triggers on first division
Center Trace 2V/DIV DAC Output 0V to +10V full scale step
Bottom Trace 20mV/DIV Gated Settling Time Waveform  20mV/DIV = 500µV/DIV
Test Equipment - 7A26 amplifier in 20 MHz position
No Oscilloscope Overdrive

Transients are reduced compared to the older devices. Using the new devices was a success.



Layout and shielding is a very sensible task

one example:


Yellow-Arrow shows just a piece of copper with ground potential, it's bended away from a transistor.


Red-Arrow shows now the piece of ground copper bended near to a transistor.


     
Left photo with yellow arrow, (DAC has already settled), after swithing ON the Sampling-Window, there is a small 200ns error on the rising slope. Right photo (red) increased settling error and higher transients.

Such a small piece of copper on the wrong place disturbes the circuit.



Using another Diode Bridge



For experiments replacing the diode array with another diode network - similar results, but a poorer offset stability - due lack of temperature control.



Using other Transistors xxxx in Diode Brigde Connection


for this test I use now another transistor xxxx with a shorted Collector and Base in diode connection.


The four blue arrows indicate the position of the small transistors.
They are mounted on a small PCB. Parasitic board capacitance dominates now.


It's a funny job to solder a piece of copper on such small pads with a patch of 0.15mm. For soldering you should glue the transistor first on the board and then use pre-soldered very thin single copper wires and a microscope magnification of approximately x20.

Up to now circuit don't work, starts to oscillate and shows funny curves. I did a check on the curve tracer of these self-made diode bridge.



Transistors xxxx on the Curve Tracer


   

           Transistor xxxx connected as diode in forward direction.                                               Transistor xxxx as diode in reverse direction 

Unfortunatel< this transitor don't work like a normal small signal diode, in reverse connection the current breaks trough very fast. This is the reason why this diode bridge didn't work in the circuit.



back



Impressum und Haftungsausschluss