Datasheet gives you a graph of Yfs vs Id.
You already have all the hints you need from my 2 links to build a high performance source follower with active parts.
No need to use J113 or whatever else.
There are no better N-JFETs than Toshiba, for me at least.
Patrick
You already have all the hints you need from my 2 links to build a high performance source follower with active parts.
No need to use J113 or whatever else.
There are no better N-JFETs than Toshiba, for me at least.
Patrick
Assume that you have shorted the input gate to Gnd, and your rail voltages are symmetrical to say 0.1V.
Then something is likely to be wrong with your matching.
If you have matched to 10µA under the same temperature, I expect your offset to be < 1mV.
That is when you use Fig. 15c with Rs = 0.
Patrick
Then something is likely to be wrong with your matching.
If you have matched to 10µA under the same temperature, I expect your offset to be < 1mV.
That is when you use Fig. 15c with Rs = 0.
Patrick
exact schematic needed
before that , I can only presume that upper source resistor need to be trimmed for offset
before that , I can only presume that upper source resistor need to be trimmed for offset
I meant .3mV. Typing on phone late at night...
I tried the formula for Rs, but getting values under 1 ohm. Not sure I'm doing it correctly. Maybe it's because the idss is under required Id?
I tried the formula for Rs, but getting values under 1 ohm. Not sure I'm doing it correctly. Maybe it's because the idss is under required Id?
Last edited:
After 4 hours, and turning on the DMM, still settles at .9mA. Goes from 1.1mA down to .9 after a few seconds.
I'm using an 18 Ohm source resistor, because that's close to what the article use and thought it would be safe since I'm using 15v+/- source. I'm having trouble calculating the Rs value.
I'm using an 18 Ohm source resistor, because that's close to what the article use and thought it would be safe since I'm using 15v+/- source. I'm having trouble calculating the Rs value.
106.6mV across 18 ohm source resistor on current source jfet and same across buffer jfet source resistor.
I figure that's just under 6mA.
I figure that's just under 6mA.
Last edited:
One more question. I'm not trying to state the obvious here, but not sure of the impact of needing matched quads.
For example, I have a well matched pair of sk170s at 9.76mA for channel 1 and another set at 10.50 and 10.52mA for channel 2.
Will there be, let's call it, a "balance" issue between channels?
I want to assume- Yes they should be quad matched, because they are technically biased at different currents,
but if that's what they require to be at 90% of their Idss, then does it matter if both channels have quad matched JFETs?
Thanks,
Vince
For example, I have a well matched pair of sk170s at 9.76mA for channel 1 and another set at 10.50 and 10.52mA for channel 2.
Will there be, let's call it, a "balance" issue between channels?
I want to assume- Yes they should be quad matched, because they are technically biased at different currents,
but if that's what they require to be at 90% of their Idss, then does it matter if both channels have quad matched JFETs?
Thanks,
Vince
Last edited:
everything ok , at least in my book
that's just buffer , not critical as some other implementations
that's just buffer , not critical as some other implementations
> but if that's what they require to be at 90% of their Idss, then does it matter if both channels have quad matched JFETs?
Yes, because the transconductance (and hence Zout) at 9.7mA is slightly lower than 10.5mA.
So the DC will be low for both, but they are not 100% identical.
Patrick
Yes, because the transconductance (and hence Zout) at 9.7mA is slightly lower than 10.5mA.
So the DC will be low for both, but they are not 100% identical.
Patrick
All very interesting. Now that I understand the parameters a little better, I can play around with the different jfets I have available.
Thanks for everyone's help!
Vince
Thanks for everyone's help!
Vince
- Status
- Not open for further replies.