Ah yes, that makes sense.
Well, guess I'll run the 2SJ74 at 20 Vds. With only 4 mA through them it is well below of that they can handle.
You really sure that it would be a problem? The JFET is rated at 400 mW dissipation.
4 mA @ 20 Vds = 80 mW dissipation.
That is 1/5 of maximum. Should be within reason, even taking derating into account.
Are you building a production amp that needs to last 20-30 years without fault or are you looking for best sound in a self designed amp? I am not saying to violate all paramaters, but there was an suggestion that higher vds may lead to less distortion. If you can move it up to 25 vs 20, achieve better numbers, and still maintain appropriate temperature control measues, why not try. Put appropriate pots in place for adjustment, build it, listen at different levels and decide based on sound and not datasheet
no great problem with 80mW of dissipation. But I would want to check what de-rated power the warm jFET can handle when in an ambient temperature >25degC.
I don't have enough experience to make an informed estimate of where Pq lies with respect to Pmax. That's why at this stage in my learning I have to rely on others and on the arithmetic available to me.
I don't have enough experience to make an informed estimate of where Pq lies with respect to Pmax. That's why at this stage in my learning I have to rely on others and on the arithmetic available to me.
Or I could do something different and go with the 2SK246/2SJ103 combo as input JFETs.
Not as high qualtiy and low noise as the 2SK170/2SJ74 pair, but the maximum rating of 50 V for the P-channel makes it a little mroe flexible with regards to cascode voltage.
But then again, there probably is good reason for people not using 2SK246/2SJ103 for that application.
Not as high qualtiy and low noise as the 2SK170/2SJ74 pair, but the maximum rating of 50 V for the P-channel makes it a little mroe flexible with regards to cascode voltage.
But then again, there probably is good reason for people not using 2SK246/2SJ103 for that application.
It is only for personal use, but I still want to make sure that everything is run within spec.
Besides, if you get THD-20 of 0.007% or 0.008% doesnt make much of a difference.
I would rather take the safe route.
I know, that is why I have used them for cascodes to begin with. However, they also require very precise matching, just as precise as the 170/74 pair.
I'm also a little unsure if the 246/103 pair should be matched to the same Idss as the 170/74 pair they are going to cascode.
Simulation shows that I get low distortion even with special 170/74 pair using models with 10 mA Idss and using them with a cascode of 246/103, using whatever models I could find of those.
Which leads me to believe that the pair of 246 and pair of 103 should just be matched between themselves.
But I'm not sure.
When cascoding a low Vps jFET with a high Vps jFET in a CCS pair, Jung suggests, no insists that, the cascode voltage provided by the high Vps device must be at least twice the Vds of the low Vps device. This correlation determines what the ratio of Idss of the two devices must not fall below.
I suspect we are in a similar situation when the 170 is operating as amplifier inside the cascode voltage provided by the 246. This will likewise determine the Idss required of the amp and cascoding devices.
I would guess that means you do not use same Idss devices for the 170 & 246 pair.
Last edited:
Hi
I think of it in terms of output conductance, goss. The lower goss device is the voltage to current amplifer inside the cascode higher goss device. The higher goss device works as a common gate amplifier. Obviously a higher goss transistor will have larger Idss. Vgs of the top device = Vds of the bottom one. One cascode I use for the P-ch type is J177 and J174. They are essintially the same transistor but the J174 has much more conductance, and thus larger Vp. Vgs will be less for the J177 than J174 at the same operating current.
I think of it in terms of output conductance, goss. The lower goss device is the voltage to current amplifer inside the cascode higher goss device. The higher goss device works as a common gate amplifier. Obviously a higher goss transistor will have larger Idss. Vgs of the top device = Vds of the bottom one. One cascode I use for the P-ch type is J177 and J174. They are essintially the same transistor but the J174 has much more conductance, and thus larger Vp. Vgs will be less for the J177 than J174 at the same operating current.
Last edited:
Well, looked into it some more.
Vds of the special 10 mA Idss 2SK170 JFET model used in my simulation is around 600 mV.
Which means the Vgs of the 2Sk246 cascode is also around 600 mV.
Looking at the datasheet the current through the 2SK246, at the highest Idss shown in the chart(5.6 mA), is just under 4 mA at 600 mV Vgs.
Now, since I am have the input 170/74 pairs biased at around 4 mA though each of them, the cascode 246/103 pairs should also be able to conduct atleast 4 mA at 600 mA Vgs.
This leade me to conclude that if using 246/103 pairs as cascodes the way that I am doing it, the Idss each of the cascode JFETs should be higher than the 5.6 mA shown in the spec sheet as that only guarantees a current of just under 4 mA at 600 mV Vgs.
Somewhere around the 8-10 mA range should be sufficient.
Does my conclusion sound reasonably OK or am I totally off the mark?
Vds of the special 10 mA Idss 2SK170 JFET model used in my simulation is around 600 mV.
Which means the Vgs of the 2Sk246 cascode is also around 600 mV.
Looking at the datasheet the current through the 2SK246, at the highest Idss shown in the chart(5.6 mA), is just under 4 mA at 600 mV Vgs.
Now, since I am have the input 170/74 pairs biased at around 4 mA though each of them, the cascode 246/103 pairs should also be able to conduct atleast 4 mA at 600 mA Vgs.
This leade me to conclude that if using 246/103 pairs as cascodes the way that I am doing it, the Idss each of the cascode JFETs should be higher than the 5.6 mA shown in the spec sheet as that only guarantees a current of just under 4 mA at 600 mV Vgs.
Somewhere around the 8-10 mA range should be sufficient.
Does my conclusion sound reasonably OK or am I totally off the mark?
I was working with k170 & bf244 as cascode.
I came up with Idss of cascode should be ~ double the Idss of the amp device. (actually trying to remember, I think Cascode Idss ~ 2* {k170 Id +1mA}, not idss).
If you are using a 10mA k170 and you bias it inside a 20mA cascode device then you are possibly going to get an Id ~ 60% to 80% of Idss, i.e. somewhere between 6mA and 8mA for that 10mA jFET. (If my alternative formula applies then a 16mA cascode may be about right).
Now to back track from these numbers to Jung's advice.
If you pass say 7mA through a 10mA k170 then that same current must pass through the cascode.
What Vgs of the 20mA cascode will allow ~7mA of Id to pass?
That cascode Vgs must be at least twice the Vp of the 10mA k170.
A 10mA k170 probably has a Vp <1V and more likely towards 700mVp
The only way to be sure that you meet Jung's guidance is to set up the two devices on a plug board and apply voltage and measure the Vgs of the cascode and measure the Vp of the k170.
Unfortunately no one has come in to confirm that Jung applies, when working with other than a Cascoded CCS.
I came up with Idss of cascode should be ~ double the Idss of the amp device. (actually trying to remember, I think Cascode Idss ~ 2* {k170 Id +1mA}, not idss).
If you are using a 10mA k170 and you bias it inside a 20mA cascode device then you are possibly going to get an Id ~ 60% to 80% of Idss, i.e. somewhere between 6mA and 8mA for that 10mA jFET. (If my alternative formula applies then a 16mA cascode may be about right).
Now to back track from these numbers to Jung's advice.
If you pass say 7mA through a 10mA k170 then that same current must pass through the cascode.
What Vgs of the 20mA cascode will allow ~7mA of Id to pass?
That cascode Vgs must be at least twice the Vp of the 10mA k170.
A 10mA k170 probably has a Vp <1V and more likely towards 700mVp
The only way to be sure that you meet Jung's guidance is to set up the two devices on a plug board and apply voltage and measure the Vgs of the cascode and measure the Vp of the k170.
Unfortunately no one has come in to confirm that Jung applies, when working with other than a Cascoded CCS.
Last edited:
I'm using a 10 mA 2SK170 model for simulation, however, the source degeneration resistor is set to bias each input JFET pair to 8 mA, which is 4 mA through each 2SK170 or 2SJ74. That is around 4 mA through each 2SK170.
In my mind, that means that a cascode device with 8-10 mA Idss should be sufficient.
Anyway, that was my reasoning.
Why are you choosing a 10mA k170 to run at your 4mA bias current?
Most designers here, who all know far better than me (I just regurgitate what they tell us) suggest 60% to 80% of Idss is a good bias range and some promote much higher percentages of the order of 90% to 100% of Idss for the operating bias Id.
Most designers here, who all know far better than me (I just regurgitate what they tell us) suggest 60% to 80% of Idss is a good bias range and some promote much higher percentages of the order of 90% to 100% of Idss for the operating bias Id.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.