SS 120R085 Depletion Mode Jfet

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
By looking at the data sheet's "typical output characteristic" graph you can also get an idea of the gain(Transconductance). A 1 Volt change in the gate causes a 10A change in Id with the 085 device were as a 1V change in the gate of the 550 we see about 3A. These numbers would probably be similar to the IRF240 and the IRF610.
:Pawprint:
 
I think the 550 probably would serve very well in the right circumstance. Based on what Pass showed in the graph earlier on the R085, it is hard not to be attracted to it, especially considering the schematics already available to drop it in. As stated earlier, the transistor (irf610) being used as a servo in both the BA's and the PLH amps, seems to offer the easiest drop in possibilty for the 550. I think you know a great deal more than I, but on the issue of capacitance, the part of the article that was interesting to me on a single transistor basis is the following statement. I could be reading this wrong, but it makes me wonder about the actual capacitance you are seeing at operating points of the 550, which would be a Vgs greater than 0.

When you look at the spec sheets on these devices, you will
see typical figures quoted, but at Vgs of 0 volts, which is
not typical of the conditions we will see in a linear amplifier.
 
I think the 550 probably would serve very well in the right circumstance.
And I suggested a couple, one being replacing the LU device :D
Based on what Pass showed in the graph earlier on the R085, it is hard not to be attracted to it, especially considering the schematics already available to drop it in.
Oh, no doubt, I totally agree it has great looking possibilities. What schematic though? The graph was made from a Mu Follower as I recall. A J2 like schematic has been presented or published regarding that graph?
As stated earlier, the transistor (irf610) being used as a servo in both the BA's and the PLH amps, seems to offer the easiest drop in possibilty for the 550.
All good except I'm not sure the word "Servo" should be used? It is quite the traditional VAS satge(produces most of the Voltage gain) and a servo is ussually some type of controller with as much FB as necessary to maintain stability.
I think you know a great deal more than I, but on the issue of capacitance, the part of the article that was interesting to me on a single transistor basis is the following statement. I could be reading this wrong, but it makes me wonder about the actual capacitance you are seeing at operating points of the 550, which would be a Vgs greater than 0.
I'm not sure were that came from but yes, we have data sheets with strange test conditions and we are expected to use that data for our different operating conditions. It's not Apples to Apples but it is a point for comparison of devices. I think generally to much emfasis is on the Ciss figure when very often it is the Coss we should be looking at. I beleive the NP statements regarding C from previous posts some years ago were sited. That exchange I beleive I may have been involved in, mostly discusses the MOSFET as a follower. Something very different happens with a common source circuit. Both configurations however can benifit from Cascoding by subduing the C effects if need be.
:Pawprint:
 
And I suggested a couple, one being replacing the LU device :D

Oh, no doubt, I totally agree it has great looking possibilities. What schematic though? The graph was made from a Mu Follower as I recall. A J2 like schematic has been presented or published regarding that graph?

It could be easily incorporated into the ZV9/F3, De-Lite, ZV4, all needing some changes. I suggested the Kimmel Mu follwer article as a suggested schematic. I believe this was the reason for the posted graph from Nelson.

All good except I'm not sure the word "Servo" should be used? It is quite the traditional VAS satge(produces most of the Voltage gain) and a servo is ussually some type of controller with as much FB as necessary to maintain stability.

As i said, you probably know more than I.

I'm not sure were that came from but yes, we have data sheets with strange test conditions and we are expected to use that data for our different operating conditions. It's not Apples to Apples but it is a point for comparison of devices. I think generally to much emfasis is on the Ciss figure when very often it is the Coss we should be looking at. I beleive the NP statements regarding C from previous posts some years ago were sited. That exchange I beleive I may have been involved in, mostly discusses the MOSFET as a follower. Something very different happens with a common source circuit. Both configurations however can benifit from Cascoding by subduing the C effects if need be.
:Pawprint:

NO offense intended. I have been impressed with yor comments and simply wanted to note what i thought might an example of misleading info on some datasheets. I thin this is why Pass is always building instead of calculating. On the capacitance front, I think i remember reading that Borberly sugggested that Crss was the figure of concern. Could easily be wrong.
 
Last edited:
Thanks, no offense taken or anything like that. Just trying to help everyone be a little clearer if I can. I have some older data sheets of some of these devices that are drastically different. There are areas of the newer ones that look like cut and paste (not actual) data. I could say the same for the IXYS devices too. sometime s the preliminary data sheets are wish lists of what they want but final processing could not actually make the goal.
Was there a refrence to the Kimmel follower that I missed here?
:D
 
Nelson,
I have trafos purchased for the F3 and a pretty good size Variac, will i find the voltage range under which you preformed the test included in the graph. Also, did you have any bandwidth issues. In my reading this is a possible problem with the Mu Follower?
 
FLG,
Went back and read the Diy OpAmp article from Pass when I found it on my desk. Look at the 550 in that context. Pass used 9610's in his original Aleph's. I think he used P-channel on the FE to offer some cancellation to the output stage, but whose to say you couldn't use all n channel, especially considering the quality of the numbers the SS fets are producing.
 
Can't use the 100's in Aleph CS because of necessary gate voltage. I believe this is why Nelson got into quasi complimentary push pull statges like srpp for J2. I by no means am tring to put words in his mouth, this is just me talking. The diferential pair could be done using N-channel fets with N channel ouput. I am thinking in terms of the tube example he used in the OpAmp article. Many tube amps have had multiple gain stage topologies. The ideal scenario is finding one tha takes advantage of the properties of the Jfets. This is why i am trying the Mu Follower. I have been reading on TubeCad a lot, looking at various examples. It is one of the few places that extensively covers tube topolgies, which of course is limited to N-channel devices. Form what little i now, Tubes are good for V amplification and fets are good for I. Jfets seem to come close to merging the two. If you could get performance similar to those found in OpAmp article, you have done something special, I believe. Looking at standalone performance of SS Jfets, you ar already close. I guess it does involve feeback and that is perhaps not ideal, still good. Simple, but not too simple.
 
Last edited:
Can't use the 100's in Aleph CS because of necessary gate voltage...
This idea has been discussed, maybe in the J2 thread, and small modifications may be all that is needed to make the Aleph CS work with a SS part. However, what benifits if any is there to be had by switching out the MOSFET? I don't remember seeing where some one actually tried it yet.
I believe this is why Nelson got into quasi complimentary push pull statges like srpp for J2...
I thought in the J2 article he said the Aleph CS on top was not were the SS JFET did it's best? What exactly that means would be a little open. I'm sure he tried such a configuration but, some ppl actually don't like the sound of the Aleph CS.
 
Official Court Jester
Joined 2003
Paid Member

Attachments

  • Compound-CM-follower.jpg
    Compound-CM-follower.jpg
    62.8 KB · Views: 329
Last edited:
Another question. With an amp like the J2, do you lose some of the SE magic of the output stage with the signal going through the differential pair. Even if used SEnded, doesn't the input pair contribute a sonic signature to the amp, based on the sj109/74 characteristics. How would the J2 driven SE, sound in comparison to just the Mu Follower stage?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.