Standard deviation of FETs parameters (mainly IDSS) will be enough large. Matching (or rather measuring) its for IDSS filters out weaker ones (generally similar IDSS gives similar gfs too) .... and fakes too.
Once I bought -cheap- 100 pcs J310 and measure its. All of them was fake, probably relabelled (below 10mA) K170.
If the gyrator output impedance is important, the lower FET forward transconductance (gfs) is defining parameter.
Approximate data:
J310 (with enough large deviation) 13mS (8-18mA depending of manufacturer);
J113 0.5-30mS;
2SK170 22mS;
BF 862 45mS;
AOT1N60 900mS.
Each of them datasheet shows "from-to" or defined gfs data, but all measured at ten-few ten D-S voltage.
The method of use -in this case- gives -about- 1-2V, so gfs degrading here, but generally higher datasheet value gives lower output impedance.
So AOT1N60 theoretically is better here than J310 .... but a little unpractical.
The first is TO220, the second is TO92 cased, AOT1N60 has -even- 1A capability, the matched/measured J310 is appropriate for 20-30mA application.
Once I bought -cheap- 100 pcs J310 and measure its. All of them was fake, probably relabelled (below 10mA) K170.
If the gyrator output impedance is important, the lower FET forward transconductance (gfs) is defining parameter.
Approximate data:
J310 (with enough large deviation) 13mS (8-18mA depending of manufacturer);
J113 0.5-30mS;
2SK170 22mS;
BF 862 45mS;
AOT1N60 900mS.
Each of them datasheet shows "from-to" or defined gfs data, but all measured at ten-few ten D-S voltage.
The method of use -in this case- gives -about- 1-2V, so gfs degrading here, but generally higher datasheet value gives lower output impedance.
So AOT1N60 theoretically is better here than J310 .... but a little unpractical.
The first is TO220, the second is TO92 cased, AOT1N60 has -even- 1A capability, the matched/measured J310 is appropriate for 20-30mA application.