[Laser] Preamp design ,etc.

[steve kavanagh]

Thanks for your comments Yves and Stuart.

The capacitor between the photodiode (which is a small 20 mA high intensity LED in my current receiver effort) does have an "intended" function.  Whether it is really necessary may be another question.  Like Stuart, I am trying to get both day and night operation from one circuit, and the capacitor is supposed to avoid changing the operating point of the JFET in strong light so much that it no longer works properly.  

In practice, it seems that with normal light levels the gate voltage will follow the photodiode voltage (since the coupling capacitor value is high compared to the capacitance to ground), and so far I have not observed any obvious saturation.  It should saturate for a while, then recover, if this a real issue.  I was a bit concerned that with an LED detector, having a forward voltage limit of about 2 volts, it might be possible to put the JFET into cutoff if the diode was connected as Yves suggests (as in the original K3PGP circuit).  I am thinking now that I may never see voltages from the LED detector anywhere near that high under any reasonable illumination, so it may indeed be unnecessary.  Upon reflection, I have concluded that the value of the capacitance IS too high...if it is indeed necessary, it probably needs to be only about 10 times the combined diode/JFET capacitance, in other words just a few hundred pF, which would improve the transient
 response.

Both of you mention setting the drain voltage to half the supply voltage.  This was roughly my target, though I wasn't very serious about getting very close, since I believed this to be more of a dynamic range issue related to very strong signals - the resistance could probably be a bit higher for best strong signal handling and resistance to bright-background transients.  I would have thought, for weak signals, it should work fine as long as Vgs is at least a few volts.  But Stuart suggests it affects sensitivity....am I missing some understanding of how this works ?  I did measure the drain current before selecting the 1k drain resistance.  It is 2.7 mA under the self-bias condition in my particular example. Idss is about 2.2 mA.

Stuart...I had found your circuit on Tim Toast's site a few days ago while trying to get myself up to date on current front end design.  My 1st receiver dates from several years ago and much has happened since then !  

I have a couple of comments/questions....I see you have two 10 Megohm resistors between the LED and the JFET.  One is for biasing the JFET, as in my older receiver.  My recent design just removes this resistor to try to get less noise (due to Yves' prodding I put some extra effort into analysis which has convinced me that, in theory, this should result in a big improvement at night).  I suppose the other one, for biasing the LED, is unavoidable unless you use KA7OEI's approach, but probably also contributes significantly to the noise.

But what is the 33 ohm resistor at the other end of the LED for ?

I just joined the UKnanowaves group a few days ago (mostly to get a look at M0RJG's circuit, since the files are not public).  I decided that a resistor/capacitor coupling was simpler still, at least for JFETs with low Idss !

73,
Steve VE3SMA