Matched 2SK246/2SJ103.
Hi, Lyubin!
Maybe you can better provide Vgs matched sets at 1/2 of Idss.
Those condition are similar to a real working, so your matching would be a great option.
Also, due to transconductance difference between N- and P- devices you can go through EUVL's input stage degeneration sample, see attached article.
BR, Pavel!
Attachments
2sk2013y/2sj313y
Thanks!
This is quite interesting for me.
From the IV curve you have posted it looks that they have about all the same vgsoff ( about 2v or so) since the data sheet indicates a range from 1.4 to 2.8 Vdc at 10 ma for grade y.
If that is so then I do not need matching since the shown vgsoff is perfect in my application. Please confirm the vgsoff value that is within about 1.9 to 2.1 Vdc or so.
Fab
Hi fab,
prices have been updated in the first post. thx
Thanks!
This is quite interesting for me.
From the IV curve you have posted it looks that they have about all the same vgsoff ( about 2v or so) since the data sheet indicates a range from 1.4 to 2.8 Vdc at 10 ma for grade y.
If that is so then I do not need matching since the shown vgsoff is perfect in my application. Please confirm the vgsoff value that is within about 1.9 to 2.1 Vdc or so.
Fab
Hi fab,
I did plotted Vgoff, I remembered it ranges from 1.8 to 2.2. I will show the plots
I did plotted Vgoff, I remembered it ranges from 1.8 to 2.2. I will show the plots
Thanks!
This is quite interesting for me.
From the IV curve you have posted it looks that they have about all the same vgsoff ( about 2v or so) since the data sheet indicates a range from 1.4 to 2.8 Vdc at 10 ma for grade y.
If that is so then I do not need matching since the shown vgsoff is perfect in my application. Please confirm the vgsoff value that is within about 1.9 to 2.1 Vdc or so.
Fab
Hi Pavel,
Thx a lot for the great suggestions! Appreciate.
I can definitely and will be happy to provide 1/2 Idss matching, if anyone requests. It is not difficult to do, as i scanned the whole curve.
For the degeneration of the fets, I actually tried that. It indeed can help match Gm. One thing I did not mention in my previous comment is that the Vgsoff of k246 and j103 are quite different. Therefore, even we manage to match Gm, the two curves are still not matched due to different Vgsoff. The curves are more like parallel curves aparting from each other. I will show some plots later. If anyone is looking for Gm matching but not idss matching, I can certainly provide some.
Thx a lot for the great suggestions! Appreciate.
I can definitely and will be happy to provide 1/2 Idss matching, if anyone requests. It is not difficult to do, as i scanned the whole curve.
For the degeneration of the fets, I actually tried that. It indeed can help match Gm. One thing I did not mention in my previous comment is that the Vgsoff of k246 and j103 are quite different. Therefore, even we manage to match Gm, the two curves are still not matched due to different Vgsoff. The curves are more like parallel curves aparting from each other. I will show some plots later. If anyone is looking for Gm matching but not idss matching, I can certainly provide some.
Hi, Lyubin!
Maybe you can better provide Vgs matched sets at 1/2 of Idss.
Those condition are similar to a real working, so your matching would be a great option.
Also, due to transconductance difference between N- and P- devices you can go through EUVL's input stage degeneration sample, see attached article.
BR, Pavel!
Hi fab,
I did plotted Vgoff, I remembered it ranges from 1.8 to 2.2. I will show the plots
Hi
I Know for the plots you have provided here but is it for all of them of only a sampling done so far. Do you provide the IV plots with the parts ordered or a kind of certificate that they are all within a specific vgsoff range or IV range would be fine for me too.
Thanks
Fab
Last edited:
I actually extracted Vgsoff for all fets (a few hundred) I measured by excel, fairly easy to do.
Hi
I Know for the plots you have provided here but is it for all of them of only a sampling done so far.
Thanks
Fab
Hi Pavel and those who are interested in this topic,
Here are some plots for degeneration of 2SJ103. I tried to use different resistors to decrease Gm of 2Sj103 which has larger Gm. By using a larger resistor, Gm can be matched to 2SK246, but Idss wont be matched due to Vgsoff difference between these two FETs
Therefore, for 2SJ103/K246 matching, I have to sacrify Gm to get Idss matching...
Here are some plots for degeneration of 2SJ103. I tried to use different resistors to decrease Gm of 2Sj103 which has larger Gm. By using a larger resistor, Gm can be matched to 2SK246, but Idss wont be matched due to Vgsoff difference between these two FETs
Therefore, for 2SJ103/K246 matching, I have to sacrify Gm to get Idss matching...
An externally hosted image should be here but it was not working when we last tested it.
Hi, Lyubin!
Maybe you can better provide Vgs matched sets at 1/2 of Idss.
Those condition are similar to a real working, so your matching would be a great option.
Also, due to transconductance difference between N- and P- devices you can go through EUVL's input stage degeneration sample, see attached article.
BR, Pavel!
Here are Vgsoff data from those FETs I measured. Pls take note I will not be able to make sure those I did not measure would have similar Vgsoff
An externally hosted image should be here but it was not working when we last tested it.
Hi
I Know for the plots you have provided here but is it for all of them of only a sampling done so far. Do you provide the IV plots with the parts ordered or a kind of certificate that they are all within a specific vgsoff range or IV range would be fine for me too.
Thanks
Fab
I tried to use different resistors to decrease Gm of 2Sj103 which has larger Gm.
Hi, Liubin!
Check your PM.
Suppose you need to degenerate both N- and P- devices on a different resistance.
By using a larger resistor, Gm can be matched to 2SK246, but Idss wont be matched due to Vgsoff difference between these two FETs
Let me show you simple trick - start from parts with equal measured Vgs(off) and then match Gm's by degeneration, suppose this will resolve the issue.
And remember, you doesn't need full-curve matching! Only in a more or less region around working point - half of Idss.
Strongly suggest, none of us are using overloaded input stage.
One matched pair of 2SJ74V/2SK170V: $33 (first plot)
One matched pair of 2SJ74V/2SK170BL: $30 (second plot)
Get the two pairs at $60 with free shipping.
One matched pair of 2SJ74V/2SK170BL: $30 (second plot)
Get the two pairs at $60 with free shipping.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Here are Vgsoff data from those FETs I measured. Pls take note I will not be able to make sure those I did not measure would have similar Vgsoff
An externally hosted image should be here but it was not working when we last tested it.
Hi
Is it possible to order parts ( N and P channel) based on vgsoff between 1.8 and 2.1 volts?
Thanks
Fab
Transistors are placed on the hot plate with 65C. IV curves are taken using locky_z's curve tracer.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
Interested....
finally have time to sort the measurement data...here are NNNNPPPP (4N4P) matched sets plots. achieved very nice matching as I measured hundreds of devices each
An externally hosted image should be here but it was not working when we last tested it.
Hi Patrick,
You are a true expert on this and know much more than I do this is a test rather than question, lol? Anyway, the operation point depends what ppl build. Current could be 50mA to 200mA for most amp, I think. However, if you want to operate it at higher current of 500mA with heat sink, I don't see there is any issue on this.
You are a true expert on this and know much more than I do this is a test rather than question, lol? Anyway, the operation point depends what ppl build. Current could be 50mA to 200mA for most amp, I think. However, if you want to operate it at higher current of 500mA with heat sink, I don't see there is any issue on this.
What do you think the operating point would be for most applications for 2SK2013/2SJ313 ?
Patrick
- Home
- Vendor's Bazaar
- Selling Zhou Fang’s remaining stocks including some very rare Toshiba transistors