I will be using the SS DM fet in both F3 and SE, non J2, non balanced Mu follower type schematic. I got the DM fet, simply because it provided the easiest drop in to available schematics with limited tweaking. Being a newb, this seems to work for me. But in reading and conversation i have realized that there may benfits to having similar devices in both positions of PP and was wandering why. For instance, in the graph Nelson provided, is an IRF 240 used as CS. Worked in F3, but is there a disadvatage?
Last edited:
Kimmel says the ideal CS is made from the highest gain device. If this is the objective then neither of these is it. It may be difficult to tell from data sheets due to the lower current we are interested in but, there are higher gain FETs. I've used FQA28N15 as cascode and follower and I beleive it does have more gain but there may be better. that would be for the mu-follower type configurations. I don't know all this is that important when using a "govonor" or "control" transistor as in a plain or Aleph style current source?
****another disclaimer - Mighty ZM is often confused with terms depletion-enhanced ; same way as confused with arrows on Jfet/mosfet symbols ( N pointing in , or out ? )
I know how they're biased , but I just can't associate terms
seems that I must find parallel with triode - that one is carved in my mind .....
**** even if I was lucky this time , with bingo on depletion is same as triode , biasing wise
I know how they're biased , but I just can't associate terms
seems that I must find parallel with triode - that one is carved in my mind .....
**** even if I was lucky this time , with bingo on depletion is same as triode , biasing wise
Kimmel says the ideal CS is made from the highest gain device. If this is the objective then neither of these is it. It may be difficult to tell from data sheets due to the lower current we are interested in but, there are higher gain FETs. I've used FQA28N15 as cascode and follower and I beleive it does have more gain but there may be better. that would be for the mu-follower type configurations. I don't know all this is that important when using a "govonor" or "control" transistor as in a plain or Aleph style current source?
In the article, Kimmel ssays that gain is important for upper fet for reasons of reducing the rp of tube, or dynamic resistance. Isn't this something that is unique to tubes.
****another disclaimer - Mighty ZM is often confused with terms depletion-enhanced ; same way as confused with arrows on Jfet/mosfet symbols ( N pointing in , or out ? )
I know how they're biased , but I just can't associate terms
seems that I must find parallel with triode - that one is carved in my mind .....
**** even if I was lucky this time , with bingo on depletion is same as triode , biasing wise
You are a kindhearted bully!
Well there is rp and Rp? But, the idea is to follow the signal presented to it (the top device) and respond as accurately as possible with a very high dynamic resistance. The higher the load presented to the Triode the more linear the operation.
I suggested there was a way to allow the Aleph CS to operate with the lower threshold SS devices. I couldn't find any of that discussion so I made a picture. I'm not saying this would be better, or not, than the MOSFET solution but it's an idea. I'm sure you recognize most of the circuit below? But, there is an extra transitor.
The problem is that the govonor or regulator (it also sums the output and the Source resistor signals) transistor has to work with the output FETs Vth and RSource drop as its Vce. But at about 1.25V + .6V or so thats not very good. The extra transistor allows the govonor to have about .6V higher Vce. It operates as a follower and should not add much, if anything, bad. I havent actually built or simulated it but it should be close to solving the problem. If not, maybe a ZVN3310 or some small FET in it's place will add 4V to the govonor's Vce.
The problem is that the govonor or regulator (it also sums the output and the Source resistor signals) transistor has to work with the output FETs Vth and RSource drop as its Vce. But at about 1.25V + .6V or so thats not very good. The extra transistor allows the govonor to have about .6V higher Vce. It operates as a follower and should not add much, if anything, bad. I havent actually built or simulated it but it should be close to solving the problem. If not, maybe a ZVN3310 or some small FET in it's place will add 4V to the govonor's Vce.
Attachments
Last edited:
Thank you for the effort. I have considered converting my Aleph J to SS on the bottom, but not interested in that as much now. Looking for different types to compare sounds. I am seeing more and more that the difference between the ACS and the Mu follower is not so great, but still significant enough to warrant different design to optimize parts. Got my R085 SS fets today. I will try them in the F3 before moving on to SE Mu follower. I have FR drivers so SE has an appeal. Maybe no difference. We will see. Interesting to note that Nelson thought F3 was sweeter sounding than J2 or AJ. Guess I will know what he means soon enough. LU F3 then SS F3 then SS MuFollower. Hopefully will learn something before all is said and done.
Sometimes you want a high impedance source, and in that case you will
pick a Fet with a high Drain impedance and/or degenerate it will significant
Source resistance. Sometimes you will play with simply matching the
devices, hoping for a synergistic effect.
The measurements come out similar, and so far I haven't seen a lot of
synergy. If the parts are rare or expensive, IRF240 seems to work fine.
You can get more effect through various adjustments of the load line.
Sometimes you want a high impedance source, and in that case you will
pick a Fet with a high Drain impedance and/or degenerate it will significant
Source resistance. Sometimes you will play with simply matching the
devices, hoping for a synergistic effect.
The measurements come out similar, and so far I haven't seen a lot of
synergy. If the parts are rare or expensive, IRF240 seems to work fine.
You can get more effect through various adjustments of the load line.
I am not sure if you are familiar with BRoskie's varous writings, but do these serve as a good tutorial on the SRPP. He speaks of the top transistor as being an impedance multiplier. Is this what the Aleph CS as well as the mu follower are doing. Making the lower fet see a higher load impedance
I've been reading Broskie's SRPP and Impedance Multiplier articles over and over again. I get the broad strokes, but not the fine detail. While I see elements in NPs work, I can't correlate part for part.
Can anyone try a walk through of the Aleph CS and how it relates to Impedance multipliers?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.