[SOLVED] Diode connected MOS - reverse current

[HarryH00t]

Hey guys,
if I'm using a Diode connected MOS (Gate and Drain connected), where does the reverse current come from? It is clear that the MOS will conduct when Vgs = Vds = Vd reaches the threshold voltage of the MOS.
If I apply a negative voltage to the "diode", I can switch Drain and Source(since they are actually the same). Then Source is connected to Gate, which results in Vgs being 0V. Vds is positive then(negative for the "diode"). Many formulas give Id = 0 when Vgs < Vth, but I can measure Drain-current anyways. Where does this come from? Is this weak inversion? I always thought weak inversion occurs for positive Vgs which are smaller than Vth.

think you can help me out
Harry

 

[checkmate]

You have not stated how you have biased the body. Could it be possibly a case of body diode conduction?

 

[HarryH00t]

I set the bulk (if you mean that with body) to the lowest applied voltage using a NMOS. I sweeped Vd from 0 to -1V and set the bulk to -1V.

 

[Milad-D]

But you said in your last post that by having Vb=-1, also you get current. How much is the current for vd<0 compared to vd>Vth? also In deep sub-micron tech., for Vgs<vth the current will not be zero ( weak inversion and sub-threshold region)

 

[HarryH00t]

Thanks for your reply. The current with negate input voltages is a lot of smaller than when Vgs > Vth. Also I set Vb to the lowest voltage applied to the circuit (Vb = -1V for Vin > -1V), so the parasitic Bulk-Drain Diode should not conduct. I think this is weak inversion then. I'm only wondering if weak inversion occurs for Vgs = 0 ?

 

[Milad-D]

I'm only wondering if weak inversion occurs for Vgs = 0 ?

For Vgs=0, the transistor works in sub-threshold region. This means that even for vgs=0, the current is NOT zero. This is one of the main issues ( power consumption) for big digital circuits.

 

[HarryH00t]

Allright. I'll try to find some literature about that then. Thanks a lot!
Harry