When I measure Idss, the current various for some time before it get stable. I should that the initial startup current as Idss or to the point when the current becomes stable?
Ji,
if You look at datasheets you´ll find, that several parameters among these the Idss vary with temperature, some varying over a quite wide range.
I´d test preferrably under the voltage and current conditions the part will undergo later when built into the circuit. In any case give the JFET time to warm up. So You could read one set of parameters at turn on, when the JFET is still cold and another set when it has reached and stabilized at working temperature. In any case its a time consuming work to match JFETs.
jauu
Calvin
if You look at datasheets you´ll find, that several parameters among these the Idss vary with temperature, some varying over a quite wide range.
I´d test preferrably under the voltage and current conditions the part will undergo later when built into the circuit. In any case give the JFET time to warm up. So You could read one set of parameters at turn on, when the JFET is still cold and another set when it has reached and stabilized at working temperature. In any case its a time consuming work to match JFETs.
jauu
Calvin
Sunsun,
If you are using your Curve Tracer, you should measure at steady state under working conditions, as Calvin suggested above.
You need to set both the working current range (which is not always Idss) and the Vds in the actual circuit.
Patrick
If you are using your Curve Tracer, you should measure at steady state under working conditions, as Calvin suggested above.
You need to set both the working current range (which is not always Idss) and the Vds in the actual circuit.
Patrick
Thanks for the input. I am travelling in China that I will post details of my question when I returned Hong Kong later this week.
Sunsun
Sunsun
as said above the device Idss varies with junction temperature. That junction temperature is dependent on Ta, ambient temperature, Pd, dissipation at the junction and time since the dissipation last changed.
The standard Idss specified in the datasheets fixes the variables by testing at Tj=25degC and dissipation very short term so that Tj does not change.
We, poorly resourced amateurs, cannot usually replicate that.
If we need precise Id, then in-circuit testing is the best. Next best is modeling so that operating conditions are replicated as close as possible.
Worst is using fingers to place a device in a 3pin socket and applying an unknown voltage at unknown Ta for an unknown time. These measured Idss are almost useless, if precision is required for the circuit to operate properly.
However, many circuits are specifically designed to accept a range of device tolerances and still perform to specification. It is, in my view, bad design to require absolute precision in device parameters to get specification performance.
The standard Idss specified in the datasheets fixes the variables by testing at Tj=25degC and dissipation very short term so that Tj does not change.
We, poorly resourced amateurs, cannot usually replicate that.
If we need precise Id, then in-circuit testing is the best. Next best is modeling so that operating conditions are replicated as close as possible.
Worst is using fingers to place a device in a 3pin socket and applying an unknown voltage at unknown Ta for an unknown time. These measured Idss are almost useless, if precision is required for the circuit to operate properly.
However, many circuits are specifically designed to accept a range of device tolerances and still perform to specification. It is, in my view, bad design to require absolute precision in device parameters to get specification performance.
- Status
- Not open for further replies.